A NVMe compatible M.2 M Key connector (like the Rock 5 Model B [0] has) would have been really nice. I'm starting to accumulate old but good, yet small in storage space terms NVMe's which I'm no longer needing, but which would be perfect for a Raspi. Even if the PCIe bus were an older generation and not as performant as the NVMe.
The microSD card slot is really great to have, to get up-and-running, but once it's clear which function the board will serve, being able to move over to a directly connected old NVMe would really be a benefit, also in terms of reliability. These microSD cards scare me, yet I make full use of them.
Finally having a battery backed RTC on it is really great news.
Also looks like PCIe is exposed via an FPC connector. You could probably come up with a scheme to have an M2 device mounted behind the board - if you really wanted.
Only having a single lane of PCIe 2.0 is a little unfortunate. I wonder if the inevitable compute module will get more or not. It seems that their IO chip is also attached via PCIe but it provides a lot of the interfaces that you’d expect to have - so my hunch is that it’ll be included on any CM5.
I think you can have over a hundred different i2c devices connected to the same bus, so it'd get pretty tall if you can power them all. 10-bit i2c (not very common) can do almost a thousand.
Although, you'd need specific rules if you wanted to prevent someone just layering up always-on LED boards.
> you can have over a hundred different i2c devices connected to the same bus
In theory sure, in a world with zero-inductance zero-capacitance busses and infinite-sink-current drivers. In our reality, however, you'll start having issues and start requiring stronger pullups and lower speeds long before you get to a hundred.
One pretty popular home alarm brand uses I2C as “the bus” with somewhat ridiculously large cable lengths for what was designed as a board/enclosure level cheap serial bus. It is in no way a sane design, but it seems to work.
This reminds me of one of the dirtiest things I've seen in this space. Saw i2c buses bridged across multiple chassis by bridging the boards with CAN transceivers. It mostly worked, but I feel like everyone who signed off on that design review should feel bad about the atrocity they created.
A 100kHz I2C with a 10kOhm pullup resistor can only take like... 1nF of capacitance before it all falls apart. (Crude napkin math for RC constant equals 10uS, I'm probably off by a magnitude but I think I'm close).
We're talking about the physical wire and physical electricity. Eventually, the capacitance on the wire grows such that its literally impossible for a 100kHz wave (let alone a 200kHz or 1MHz wave) to be placed upon the wire.
Parasitic Capacitance is real. Each item you add in parallel to a bus will increase capacitance.
--------
Each item you add in serial to a bus will increase resistance and inductance, which also slows down the speed of the bus. Once the speed drops below the specified protocol (ex: 100kHz or whatever you've chosen), it all goes to crap and no one can talk anymore. I don't quite remember the napkin-math formulas for LRC-circuits but... parasitics are hugely important and nH of inductance and nC of parasitic capacitance absolutely can be reached with just a few connectors.
Honestly, I think the earlier estimate of "failing by a hundred connectors" is rather optimistic still. 100kHz I2C and 10kOhm pullup would probably fail within a dozen connectors.
You can have as much addressing as you like and software going on in the background. It don't matter if your literal wires no longer work as you expect.
Well an interesting hack (of a hack) was pirated / imported games couldn't be played because they didn't have a specific chip. They did have splitters so you took the chip signal from a genuine cart, and the other game plugged in the other port.
Fwiw I bought a sleeve for my M.2 SSD and it plugs into usb3 with very little slowdown. It gets 500MiB read and 1.2GiB write. Haven’t measured vs native M.2 to compare, but it’s fast enough that I’d be surprised if the pi were blocked waiting for I/O vs native M.2. And native M.2 comes with increased manufacturing complexity.
Oh, I didn’t test on a pi, just a mac. And yeah the double write speed was super weird. Thanks for answering my internal question about why ~500MiB seemed to be the max.
Same with orange pi5 and 5b. Having an option for a faster and reliable storage is amazing. SD cards are great when you prototype or need just enough storage to netboot a device, but past that it is limiting.
In addition to the ability to adapt M.2 slots into PCIe via an inexpensive adapter, there is a growing ecosystem of M.2 cards for most of the things you'd use a PCIe card for. USB controllers, SATA controllers, Ethernet controllers, FPGAs, even display adapters.
It would not limit you to just SSDs, the m.2 connector on the rk3588 devices generally exposes 4 pcie lanes and an adapter to make this a full pcie slot is already available. So really an m.2 nvme connector is the best of both, you can turn it into any other pcie connector you want via an adapter, but you can just put an nvme ssd directly on the board without a hat.
The CM4 exposes a PCIe 1.0 x1 (Not sure about the revision) lane and can boot from an NVME SSD. It will be nice to have this capability on the Pi 5 with all of the other connectors (except for audio.)
YIL a Pi 4B can run from a directly connected USB/NVME SSD. I've been using USB/SATA SSDs and all I have tried require a powered hub. Do note that I don't just just boot it up and proclaim "It boots - ship it"[1] I thrash it with various disk benchmarks and stress tests to be absolutely certain the SSD does not disconnect due to insufficient power.
[1] I suppose this is the H/W equivalent of the S/W engineer's proclamation "It compiles - ship it!" And I guess this dates me since S/W is pushed rather than shipped these days.
Good luck doing "deployments" on SD drives and still have it run 6 months later, especially in environmentally unfriendly environments. Theses horrible little things die all the time, even when you buy supposedly high quality ones.
Sorry, can you elaborate? Is the death rate of NVMe drives known to be significantly faster than that of other types of drives? I hadn't heard of this.
I can only speak from my own experience, but there are a number of low to mid grade SSDs that use truly awful flash and/or controllers. I've bought many of them over the years for unimportant workloads, and have been unpleasantly surprised by how quickly I can burn them out. It's really worth it to shell out for higher end models just for the increase in durability.
Yeah, I'm not doubting that, and this is consistent with my own experience. But the claim was that it was specific to NVMe SSDs, which I find very odd.
All but the highest models of SanDisk have done this to me, as well as the lowest end of Western Digital, ADATA, Kingston. Same for some lesser brands that I've forgotten (not fakes, just shoddy).
You can get some rather large USB sticks these days. Amazon is awash with 928Gb units for about £10 which are probably not the best. A 256GB or 512GB from a known brand is around £30-50. There are several USB ports on a Pi and you can always boot off a SD card.
Finally, if a network is available then network boot and use NFS or whatever.
The RTC is a cool addition and long overdue. At work I have three Pi 3s with GPS hats and aerials acting as stratum 1 ntp servers. The hats have a RTC included which is handy after a reboot. My use case is "reasonably accurate and stable time" so sub milli second is good enough, I'm not too fussed about nano seconds! I want logs to correlate and desktop clocks to be reliable.
Those are almost certainly 8 to 32GB drives that are modified to report a higher capacity. They are scams that hope your return window expires before you try to load them up with a lot of data. Nobody can profitably manufacture flash drives for less than the wholesale price of the flash chips.
There's no almost about it. They are all scams 100%. The most popular are SD cards with a USB adapter fitted in various housings. The internal adapter or SD cards are modified to do various things like report much larger sizes, allow phantom writes, or even install malware. They can also potentially burn your house down.
FYI I believe all of those ~1TB $10-20 USB drives on amazon are scams- basically set up to trick your computer into showing a terabyte of disk space available, but not actually having that much available if you try to write it all. I was in the market for "largest reasonably priced usb drive" earlier this year and ended up with a 1 TB usb drive for about $80.
I wish some attorney general would buy a bunch of them and then just proceed to sue the shit out of Amazon so hard that Amazon finally gets off it tail and does something.
Flash storage like USB sticks is a crude comparison to an SSD or NVMe with cache and a controller capable of parallel operations.
A USB flash drive is like a dumpster. Big bandwidth when the lid is open but it’s got poor performance for fetching and storing lots of things all the time.
An SSD or NVMe is more like a rolling auto tool chest. Same big metal box, but much more performant for complex and numerous read and write loads.
Or to use a computer analogy: SSD is like a hard disk, USB flash is like a tape drive.
No word on price. When Raspberry Pi first launched, this was the prime feature of the thing. Can we expect same price as the Raspberry Pi 4? (at the respective RAM level)
So the 4 GB model is 60$, which is 5$ more than the 4 GB model of the Raspberry 4 when that was launched: https://www.raspberrypi.com/news/raspberry-pi-4-on-sale-now-... . I guess that is fair, especially with inflation nowadays. So they stay true to the idea of making this available for cheap.
> I guess that is fair, especially with inflation nowadays.
The accumulated inflation in the US since June 2019 is c. 17%.[1] This means US$ 55 in June 2019 is "worth" c. US$ 64.35 today. So it seems that you get a 4GB Raspi 5 for c. 7.25% less today.
The "US Real Average Hourly Earnings" have increase by ca. 22% in the same period, from US$ 27.75 to US$ 33.82.[2] So an average person needs to work c. 12% less to buy a 4GB Raspi 5 today.
However, I think the issue is more complicated: There is inherent deflation in electronics, which is included in the inflation rate. you can observe it when looking at the current price of a 4GB Raspi 4 at Amazone, which is c. US$ 67. So if the introductory price for a Rapsi 5 is really going to be US$ 60, you get something better for more than 10% less now.
Amazon is not a good place to get prices from for a comparison. All the enshittification has ruined the prices. As it forces prices up outside Amazon too even outside prices are not useful. To do a proper comparison that doesn't include changes forced by Amazon IMO you need to look at non-US non-Amazon non-Amazon-sellers prices and work with those. Quite annoying really.
If you were on HN a decade ago, you would have seen "pivot" all over the place. Times change, but we are still human (mostly) and it seems to be part of who we are to mold language to fit our needs and circumstances.
Take "bipping," a word I had never seen before a month ago, but that is now all over the place. (At least in some communities in the Bay Area; I guess all the car break-ins helped it's adoption).
It’s almost as if the English language adapts over time. I’m sure in the 1580s when the word “Gleeful” was seemingly introduced people felt weird about it too, heh.
How so? The vast majority of the English language we use today, that’s “formal,” has been the evolution of what was “slang” back then. I don’t understand your argument here..
It'll always be a subjective argument. On one side you can have rigid formalists who hate any change. On the other there's no rules at all.
I'm not at those extremes. I consider the origin of the word. A few years ago I started seeing "griefer," and it quickly became apparent that these people didn't know there was already a word for that: spoilsport. It's those instances that I bitch about, because it reflects ignorance and an unnecessary obfuscation of the language.
I'm down with adding words, including slang, that didn't have an equivalent. One example of foreign origin, "bokeh," really serves a purpose: to describe the aesthetic quality of blurriness. Cool.
5% is not arbitrary. It is removing massively skewing outliers who also see a massive growth outlier compared to everyone else.
Looking at wage movement as slices of where one falls gives a clearer picture.
Additionally, the reason it doesn’t make sense to remove low wage outliers while removing high wage outliers for average is because the median wage is closer to low wage outliers than it is to high wage outliers.
For example, if you take the median and then +- for your dataset (if median is 48,000, then use 0 thru 96,000), you’ll be removing more than just the top 5%, and yet, this also gives a far far far better picture of the dire economic position and what is happening with real wage movement.
Interesting. Most income graphs I've seen appear to be normally distributed -- could you explain what would cause them not to be, or provide any examples that demonstrate non-Gaussian distributions?
But if wages are non-Gaussian, how would trimming the top and bottom 5% off of your sample be any better than using standard deviation ranges to control for outliers? The assumption that outliers are to be found on the top and bottom of your range is one that seems to apply to Gaussian distributions, and doesn't necessarily hold for others, regardless of whether you are using fixed percentage values or standard deviation thresholds. For example, in a bimodal distribution, outliers might be found in the center.
> 1 a
> : a single value (such as a mean, mode, or median) that summarizes or represents the general significance of a set of unequal values
Average isn't well defined-- if you have to guess, people probably intend "arithmetic mean" when they say it. But if you're using it in the sense of a single number that represents the typical value of a dataset, you may want the median, mode, or midrange, or even a geometric or harmonic mean depending upon circumstance.
So it's best to use that term. And it's best to be charitable to not jump on people saying "average" when the median might be the best measure of central tendency for a task.
It's frequently used in when talking about incomes specifically. [0]
> For example, the average personal income is often given as the median—the number below which are 50% of personal incomes and above which are 50% of personal incomes—because the mean would be higher by including personal incomes from a few billionaires. For this reason, it is recommended to avoid using the word "average" when discussing measures of central tendency and specifically specify which type of measure of average is being used.
I think the median works better in this situation because the huge population means that both the outliers and the skewness have much less effect than they do on the mean. The problem with using the mean “without outliers” is that you have to make arbitrary decisions about what data to exclude as an outlier, unlike with the median.
The answer is always "it depends", which is exactly why I prefer the median in most cases. Once you choose to use the median, there are no more choices/degrees of freedom - it's just the median. On the other hand, "mean without outliers" requires you to make a subsequent value judgement on what exactly is an "outlier".
> "mean without outliers" requires you to make a subsequent value judgement
Do you think that comparison of outliers to interquartile range is not a relatively objective method of determining outliers?
The interquartile range is a number that indicates the spread of the middle half, or the middle 50 percent of the data. It is the difference between the third quartile (Q3) and the first quartile (Q1) . . . The IQR can help to determine potential outliers. A value is suspected to be a potential outlier if it is less than 1.5 × IQR below the first quartile or more than 1.5 × IQR above the third quartile. Potential outliers always require further investigation.
"Real Average Hourly Earnings" is already an adjusted metric to make sure that it's not skewed to the top earners (hence why it's called "Real Average" and not just "Average"). In the linked post you'll note the following explanation of what "Real Hourly Average Earnings" means: "Data relate[d] to production employees in mining and logging and manufacturing, construction employees in construction, and nonsupervisory employees in the service-providing industries," which apparently covers roughly 80% of private sector jobs in the US.
You are right that this metric excludes things like capital gains income, and covers just hourly earnings.
“Real”, however, refers to it being inflation adjusted, not an exclusion of high earners (who are excluded anyway because of the hourly earnings metric).
> The accumulated inflation in the US since June 2019 is c. 17%.[1] This means US$ 55 in June 2019 is "worth" c. US$ 64.35 today. So it seems that you get a 4GB Raspi 5 for c. 7.25% less today.
Included in that 17% figure is the price increase for e.g. corn, rent and steaks.
Why should the change in price of either of these things be relevant to the price of a Raspberry Pi?
That my point: you can’t define how much a dollar is worth in dollars because it’s a circular definition.
You can, of course, calculate the change in price over time of a specific basket of goods (rent, steaks, corn). But there’s no reason this is connected in any way to the change in price of an entirely different basket of goods (Raspberry Pi’s).
In other words, if you try to define how much the value of a dollar has changed over time, you’ll get a different answer depending on the basket of goods you choose. So the change is clearly not inherent to the dollar.
My gripe is that the original Pi cost $35 at launch and while they have made a better Pi...They have not made a Pi at that price ever again, even accounting for Inflation. Furthermore, increased power consumption and features have added big price jumps to the required Accessories. Now you need miniHDMI adapters rather than more common HDMI, you need cooling, you need more expensive power adapters. a fully set up Pi 1 was simple USB, SD, and HDMI All possible in a $50 budget or less if you had some stuff. Now you are $90 in to run it.
I don't see how this is true. $35 in 2012 is roughly $47 today. You can get a pi zero 2 kit today including the adapters, case, and PSU, for $49.95. Just the PSU, which the original didn't come with, is worth more than the $3 difference.
Besides, both the pi 3 and the pi 4 were $35 at launch, so they were actually beating inflation when they were launched.
No need for the wifi and bt dongles anymore, are they are integrated in the Zero 2 W. You can connect keyboard and mouse either via bt or via a single usb otg dongle to connect a wireless keyboard/mouse set or a keyboard with integrated usb hub. In my opinion this is superior to the original B. And far superior to the original A anyway, while still being cheaper. Remember even the original B had just two USB ports, so in practice you needed an extra hub almost always, because of missing functionality that is now integrated.
It depends on the use case. I have several I put around the house with sensors for IoT stuff. ESP32/8266 (and newer) devices would probably be less expensive since they don't need an SD card but are more work to program. The Zero has a full OS with dev tools built in. But for running a desktop, they're not so good.
They're not as cheap any more at $15US for the Zero W. I miss the days when Microcenter had them on sale for $3.14US on Pi day. They did recently have them priced at $10 to celebrate the grand opening of a new store.
The Pi Zero 1 performance is pretty bad and it doesn't support 64bit OSes (so no Ubuntu). The Pi Zero 2 looks pretty decent for the price, performance should be ok-ish and it'll run Ubuntu... if you could actually get one.
Sure, if you want the raw board and nothing else. I spent $100 for a starter kit that had an enclosure and some heat sinks. It came with a controller for a retropi installation too which was a nice touch.
Not to mention it now requires a new PSU. Before you could use your standard $5 (android) phone charger at (5W), then you had to buy a 15W one and now a 27W.
To whoever thinks pi's are cheap, you can get more functionality out of a used laptop for less money, but probably worse specs and probably x86.
The standard charger is 2.5W, not 5W, and RPi3 will happily run from that, for RPi4 2.5W is somewhat marginal, but it will still work. I suspect that RPi5 is going to be somewhere around 3W (non-PD USB3).
The official RPi specs intentionally overblow the power consumption in order to provide an buffer for powering whatever ridiculous stuff people may hang off the USB ports.
Then the other issue is that all RPis need somewhat tighter voltage tolerance than what is in USB specification. So it is perfectly possible that a cable between the power supply and RPi is compliant, but has too large series resistance on power lines for RPi. And well, powering RPi from random aliexpress-grade “Android chargers” is completely another bad idea.
I meant “standard USB charger” from your pile of random chargers, not the one that RPi Foundation sells. Such a thing hopefully provides the 500mA, that the USB2 port is supposed to deliver.
But, you really only need the 27W supply if you need to pump a lot of power through the USB ports. It'll run just fine (even using less power) on the 15W unit with more modest loads.
The value proposition is there for me. I bought a "BMAX B1 Plus" for about 70$ including shipping from AliExpress. It's as small as 2 or 3 CD cases stacked together. It is a fully fledged PC that comes with windows 10 (no support for windows 11 but LinuxMint and other distro work well too), an hdmi cable, and a clever mounting bracket to attach it to the back of a monitor. It is passively cooled, pulls about 4 Watts. Biggest downside is the power supply with a cylindar connector. I use it solely to connect to my main PC using RustDesk and it is great for that.
They link to multiple regional reseller sites [0], where prices are available. I see €73.90 (€60.08 pre-tax) for the 4GB version and €97.50 (€79.27 pre-tax) for 8GB in Poland.
True, it's called Treasure Island because of all the stolen treasure it holds.. but jokes aside (shouldn't be a joke) i'm going to have a friend pick up one for me in UK and send it over to ZA where pricing is so so much worse.
I am curious to know how people here use their RPIs today, and how the RPI5 might help. I have had a model 3 for many years, which I enjoy tinkering with from time to time. I still haven't gotten over the novelty of having such small, cheap computer that runs Linux and does a pretty good job of it. That said, all I do is tinker with it. I turn it on, write a bit of code, marvel that the code runs, and turn it off. I've bought a couple of hats which are fun too, but again it's just tinkering. I'm curious to know who here has found "serious" applications for their RPis.
Until my home automation got complex enough to justify a faster, more capable machine, I ran all my home automation stuff on an RPi.
I use RPIs as "data collection" units. I have one RPi outside with a cheap SDR to pick up all the neighbor's weather stations, which I dump to a MQTT queue and use to populate weather data in my home. I use another to collect GOES satellite images.
I have 3-4 RPis that act as "Digital Ham Radio Hotspots", basically bridging my local ham radio via the internet to other stations. I use an RPi 4 as my "to-go" computer when I do ham radio in the woods. I use an iPad as a screen, and it works just as good as a laptop.
I have an RPi sitting in my garage as a second nameserver. The primary nameserver is in the house "data center".
I have an RPi plugged into my stereo receiver as a streaming device that lets me stream audio from my phone to the stereo.
I have 4 RPis connected togehter in a k3s cluster, for fun. IT doesn't work great. :)
Geographical distribution. If the left half of my house is destroyed by natural disaster, I’ll still have a working name resolver on the other side!
Seriously, it’s just because I may do something else on that Pi that I want it on that side of the house. It’s my only garage Pi. My main internet handoff is in the garage, but my main “datacenter” is in a basement room on the other side of the house.
One is monitoring my trash bins in the backyard and generating visual output of emptying times, current location of the bins, alerts if they are still in the backyard but are scheduled to be emptied tomorrow (in which case I need to move them to the street in the front). This has been running on a Pi with Bluetooth (monitoring uses BLE beacons) for over five years now, with very little maintenance necessary.
Another one runs the home automation hub (Homematic plus some addon stuff). Also very little maintenance, basically just doing backups and an update a year or so. Has been in place for several years as well. I often forget that this thing exists at all, as it just chugs along quietly, never needing reboots or anything. Even the updates are unnecessary unless I want to use some new sensor or actor that the old hub software doesn't yet know about.
And then there are two Pis connected to TVs in the living room and kitchen which run OpenELEC/Kodi for media center tasks. Started doing this when the first RPi came out and frequently used back then, these Pis are rarely used these days, as most streaming now involves commercial streaming services and is done via FireTV sticks. But I still have a private library for the occasional exception of stuff that's not offered on any commercial service, and that library is accessed via the Pis. Fortunately, aside from a reboot every few months and very rare updates, these Pis are also very low maintenance.
Discovering DietPi was kind of a game changer for me. I had the original Pi 1b that basically sat in a drawer after the first month I got it.
I went over to a friends house last year that had a more modern Pi and they had a PiHole on their home network. It was pretty amazing being able to block ads on my phone near totally and not just in Firefox. New Pi's were completely impossible to find but my friend said give DietPi a shot. The benchmarks on my Pi after installing were complete crap, processes took minutes that took a few seconds on the more modern Pis, but AdGuard Home worked flawlessly.
It sent me down the rabbit hole of Tailscale everywhere, self hosting what I can, getting a NAS, and just opened up to me how simple it is to set up these kind of services now that are accessible everywhere.
I recently was able to get a Pi 4 (one week before the Pi 5 announcement of course) and am looking forward to a setup where I can run services that need hard drive access on my NAS and hosting the quality of life apps on the Pi.
I've been using one for https://pikvm.org/ and it's been a rare case of "the Raspberry Pi is neither ridiculously overpowered, ridiculously underpowered, or even beat out by any off the shelf solution at all let alone at the same price or point". It's literally the best IP KVM I've ever used or owned. The use case is almost a perfect match for the exact hardware capabilities of the Pi: hardware encoding, video input, gigabit network (with Wi-Fi alternative, which has saved me a few times), GPIO, USB OTG, the hat system, open source web KVM software which doesn't suck ass and sit untouched for 13 years with endless security vulnerabilities piling up.
Things I've used mine (plural) for at various points in time, in no particular order: Wireguard & ssh entrypoint into my home network. Pi hole. Kodi. Calibre ebook server. Orchestrating turning on/off IP-based lightbulbs or plugs (since I'm too cheap to get a gateway for ZigBee/Threads/$IOT_protocol) - automatically turning on my Christmas lights at sundown daily after querying an API for my local Civil twilight time was good fun. "NAS server" connected to a cheap 5-disk USB JBOD device. Hosting a low-intensity crawler that ran into blanket IP-range black-listing issues when hosted on cloud-providers. Hosting a Gitea/Forgejo server: I no longer star projects on Github - I mirror them locally and keep them synced, storage is really cheap now. Periodical syncing my backups from NAS to cloud. As a digital "tape recorder" for broadcast radio using FM receiver, aux cable, and USB sound card. Twitter Spaces recording for time-shifting conversations I wanted to listen to later; I beat Twitter's "record" feature to the punch! Twitter crawler/archive bot for a niche community.
Wishlist projects: getting alerted when my home loses electric service or internet connectivity, using UPS and LTE modem. "Calendar dashboard" site that displays the household's schedule for the day on a tablet/jailbroken kindle. A lazy-loading reverse-proxy Caddy API server that will keep the connection open while it turns on my workstation in the background if it's off. Archiving tweets using mitmproxy to passively scrape Twitter's API responses while I use the official mobile client.
Basically anything that I wanted to automate and was not demanding on compute would get assigned to a Pi.
I think the Pi 5 would be well positioned to be a free, OSS media box. Android TV and Google TV are very popular today, along with Roku, Apple TV, etc.
I installed PiHole and noticed that every single click of my Roku remote gets sent to Roku’s servers. PiHole blocks this of course, but there was nothing I could do to disable this telemetry on the Roku device itself.
Google TV is slightly better - there’s options to adjust targeted ads, and an “app only” mode, but there’s still usage and other data sent to Google. Also you can’t use it at all if you don’t sign in with a Google account.
I haven’t used other platforms.
But I would like to see an easy to use, easy to configure, OSS streaming box. Now that this can do 4k60 and HDR, it might just work for things like Netflix, Plex, and other services.
Right now the best products on the market for high-bitrate streaming are Apple TV and Nvidia Shield Pro. I wonder if the RPi 5 can compete with that?
There's no shortage of projects for this, particularly enby, jellyfin, and kodi spring to mind. The only real challenges are GPU transcoding support, app selection for integration of services like Spotify, amazon prime streaming, and netflix, and app availability on client devices like TVs.
In terms of specs a Pi4 was already beefy enough for this use case, fwiw. But the software stack is hard just because no one seems to target the "set top box“ space quite the same way. There are client/server media platforms to compete with plex, there are single.box media platforms, there are set top boxes to compete with roku, and every possible mix in between.
You could switch out the Pi 0 W's with ESP32's or Raspberry Pico W's (if you'd like to stay in the Raspberry ecosystem). They're a lot cheaper, require less power, have a smaller footprint, aren't too hard to set up and there's usually a library ready for every sensor.
When the Pi 2 was new, it was one of the few single board computers that properly supported fractional frame rates, which are often needed for smooth video playback. (Many films are encoded at 24000/1001 fps rather than exactly 24 fps, for example.) I assume the newer models still support this, since they're still built around VideoCore chips.
A faster model would allow decoding at higher resolutions and frame rates, even when the codec in use doesn't have direct hardware support.
With the PCIe support in this new model, it could also make a decent home file server.
I have a Pi3 running pihole/dhcp for my home network, and 4 Pis as part of my home k3s cluster (along with 3 cheapo nucs). Yes it's overkill but I'm in infra and I learn a lot from it.
The cluster runs nextcloud, collabora, plex, longhorn storage, deluge, nzbget, some custom projects I built to sync my bank accounts with my finance software and to sync photos from Signal chats to my nixplay frames, backups for my NAS, Grafana/Prometheus/Loki, netdata, etc.
I also run an entertainment company and we use a Pi Zero to control our onstage LED lights via presentation remote or wifi as a fallback.
I keep thinking of other things to do with them, but this is already enough to maintain in my "spare time".
I have a Pi 4 running Pi-hole and connected to a 4TB USB hard drive running a Samba server with a bunch of media. It’s on my Tailscale network so I can access the files away from home, and use it as an exit node if I’m abroad and need to access stuff as if I’m in the UK.
I don’t need to transcode anything so it’s been more than sufficient for me, and if I start to run out of storage I can just hook up more drives.
It also has about a terabyte of PS2 games because a softmodded PS2 can play games off a Samba server rather than the DVD drive. Niche, but very cool imo
I don’t know how serious this is, but I have one in a waterproof box in the backyard and one in the garage to run the irrigation for my garden, etc. They work well enough that I often forget they’re there. There’s another in the office closet running a dynamic dns script and that’s there to use as an ssh tunnel and such.
I use it professionally all the time. I work with arm a lot, and during the testing of customer software that is compiled for arm, I need an arm based Linux os to run and test on. rpi is the obvious choice. I'm very happy with the 4s performance, so I don't require the 5, but I'll still get one.
I also use another one for retro pi, which I imagine will benefit from the increased performance..
lastly, I have a small handheld Bluetooth keyboard with a track pad I use with a pi4 with a 3.5in screen hat and battery pack. the overall size and weight is much less than any laptop I have ever come across, and I use this when I go to cons. (although I'm looking at plantecom to replace this)
Three years ago, my new home came with smart features such as a connected front door lock and thermostat and a wall panel with a sim connected to alarm.com. The intro period ran out, and I didn't want to pay anything ongoing, so I bought a z-wave usb controller and set up HomeAssistant on my old Pi2. With some effort to get everything working in docker with a cloudflare tunnel, I am back to being able to monitor and control my smart home from anywhere.
I'm not feeling compelled to get a more powerful pi. In fact, I'd be more inclined to buy the pi zero or several of them, but they were always impossible to find in stock when I was interested.
I’ve got one of mine showing album art for music I’m playing[1], one with an e-ink display that shows which subways are leaving soon near me, one running Homebridge for a variety of sensors[2], one for WireGuard and Pi-Hole, and one to run Klipper/Mainsail for my 3D printer. I guess these are all tinkering as well but they’ve basically been running 24/7 for a few years now!
A RPi 4 sits near a Teensy 4.1 in a difficult to reach LED installation: its only purpose is to wirelessly re-flash the Teensy firmware in case of updates and for eventual future features. The Teensy is driving ~15k RGBW LEDs.
I’ve owned many Pis through the years. Always for personal/hobby stuff, but they have been workhorses all the same.
I ran a Pihole server on an ancient 1A for several years. It was considered underpowered but worked great. I’ve run Homebridge services which was a great way to integrate various IoT stuff into HomeKit that lacked native support. I’ve since moved these things onto other hardware to consolidate things (not because of anything inherently wrong with the Pi).
For about a year now I’ve been using a Pi as a BLE data collector to slurp data from a weather station I have on my deck. That being said, the onboard Bluetooth on the 3 and 4 is pretty awful in my experience. I needed to use an external dongle to make it reliable.
For over a year I used a Pi 4 to run multiple SDRs to record and rebroadcast trunked police/fire scanner traffic. I eventually had to retire it for that purpose when the trunked system changed to digital and the Pi just couldn’t keep up with the workload anymore.
Those were all applications that I ran for months/years at a time using Pi hardware. I’d consider them “serious” even though they weren’t anything impressive or enterprise-worthy.
Other things I’ve done with them are use them as emulation stations. Even older Pis are more than capable of emulating 8 and 16 bit consoles without a hitch. Their compact size makes them portable and easy to plug into a TV and work with a Bluetooth controller.
I use my RPi4 as a small server having multiple docker containers behind a Traefik proxy. The largest applications are probably my Nextcloud and my Photoprism instance.
The Photoprism instance holds about 114.000 pictures which I took over the past 10+ years and includes features like face recognition and browsing by location. I quite like it and find it super impressive that such a small device can run such a powerful application with that amount of data.
Teams in the FIRST robotics competition community use the Raspberry Pi for computer vision. There's an open source program called Photonvision that provides an easy way for talented high school students to set up CV. Some of the better teams use Orange Pis or mini PCs for this purpose (and some of the best teams have fully custom solutions), but a Raspberry Pi is a lot easier to set up and is cheaper (even at $60). Also, my understanding is that the architecture of the RPi4 CPU/GPU allows for GPU acceleration of classical CV algorithms without copying (and that feature is integrated into Photonvision), so this hardware upgrade will probably make the Raspberry PI perform on par or better than more expensive options, while being easier to setup.
And two MIPI camera ports is also really nice for this application, because the new advancement in FRC is to use April Tags included with the field for full field localization. For this purpose, I've heard you get better results with global shutter MIPI cameras than USB webcams, and having multiple cameras is good for always seeing an AprilTag.
I use one for each of the following. None of which I think are terribly novel anymore:
- Octopi
- PiHole
- Grafana/Prometheus for my local network (I push Speedtest results, ping results, among other things to it)
- home-made SNES/NES minis
I’m excited to try using the new Pi for a more comprehensive option for a retro arcade including many games that didn’t run very well for me last time I tried. (I instead used a home-built small PC)
My pi 4 runs a wide variety of small services for me. Most important is my vaultwarden instance, which is securely accessible everywhere that I go via tailscale. The pi also runs HomeAssistant for my smart home stuff, pihole to block ads, and a resilio sync server. I’ve got a bunch of other small stuff as well that I rarely use. Like “your-spotify” which stores and tracks stats for my entire Spotify history.
I want to also possibly make it a media device? I have it physically located behind a tv, just because I already had Ethernet wired there, but it isn’t hooked up to the display yet, and I haven’t figured what to do with it if I did hook it up. Not sure if that would overstress the cpu. Can’t have my dns crashing if I’m watching a video.
I have a pi running to continuously collect data from my home, temperature, lights, gas usage, electricity usage, weather, etc. It runs Benthos to get all the data (MQTT + http), and stores everything in Postgres.
but, but, but ... the RPi is meant for tinkering, isn't it?
I got an RPi2 connected to a GPSDO acting as time server. Another RPi2 controls test&measurement gear via GPIB (there are GPIB hats, but I happen to have already an USB adapter from Agilent) and TTL PPS input and trigger output for synchronization. A RPi2 is plenty of computational power for those applications.
Three sitting around. They make very convenient small servers. One (a RPi4) has substantial mass storage and is a backup and media server. Another (an RPi3) is an RSS reader service and web proxy for some internal stuff. And the third (RPi3) servers for occasional experiments and what-ifs.
I'm still using a first gen Raspberry Pi Model B to run an adblocking DNS server (github.com/0xERR0R/blocky), and PiVPN (Wireguard). For the most part, it works great except apt-update occassionally maxing out the CPU/RAM which causes my services and sshd to be killed.
i have a raspberry pi zero wifi thats running syncthing and has a few of my syncthing folders/nodes on it. i already have a NAS that acts as the always-on node for all my syncthing folders, but if that is ever down for maintenance or whatever then the pi zero is sort of a backup that will keep syncthing things
i have another running kodi so i can access jellyfin that's on my NAS
pi-hole for network wide ad blocking
currently setting up one to act as a storage location for the restic backups from my NAS instead of having to store them with some storage provider like backblaze. i will be installing tailscale on it as well so i can leave it in a friends house and hopefully be able to reconnect to it even if they move house in the future
I’m using a Raspberry Pi 3 Model B+ to run Steam Link to stream games from my desktop computer upstairs to the TV in the living room. I managed to connect some XBox series controllers to it using Bluetooth by installing xpadneo drivers.
I think it would be really good to give to children so that they can have their own computer for £100. It’s like the modern equivalent of a C64 but also completely capable of being a desktop for everyday use.
TBH the thing I hate most about this category of SBC is the reliance on SD cards. They are both too unreliable to trust and so slow that they often bottleneck the SBC. Buying them is often a crap shoot too, I've purchased cards batches of cards from the supposed reputable manufacturers that were all over the board when benchmarked, and rarely did they hit the claimed speed spec. I would love if there was an alternative that was not as much as a jump as those SSD flash drives or NVME drive. Maybe OS grade eMMC M.2 drives the size of those wifi cards?
"One of the most exciting additions to the Raspberry Pi 5 feature set is the single-lane PCI Express 2.0 interface. Intended to support fast peripherals, it is exposed on a 16-pin, 0.5mm pitch FPC connector on the left-hand side of the board.
From early 2024, we will be offering a pair of mechanical adapter boards which convert between this connector and a subset of the M.2 standard, allowing users to attach NVMe SSDs and other M.2-format accessories."
As you keep spamming this here, did you read the HN Guidelines[0]?
> Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that".
I did read that they were going to support M.2., and I have gotten around my issues with SD cards in the past using SSD-grade USB drives and NVMe adapters. My comment was about how crap SD cards are, and how this class of SBCs (including the PI 5) often use them as their default storage (as in, not needing an adapter or special firmware to boot). My final statement was my wish for a high speed durable storage standard that was better than SD cards without having to spend more money than the SBC itself on storage, although looking today on Amazon it seems that NVMe drives have gotten way cheaper, no idea of those are quality though.
You can netboot a Pi. Every Pi in my house netboots. I have a whole bunch of them. Some play games, some play videos, some play music, and a few other minor things. Because they all netboot, you can change what each one does by renaming a file on the server and then rebooting it. It's great.
Are they using like a RAM disk then or are they mapping a network drive for storage while they're running? I've always been curious about setting up netbooting but beyond a failed experiment a long long time ago I haven't really tried.
I use LibreELEC. I don't watch streaming TV, I have traditional cable and an HDHomeRun Prime box which acts as a cable-ethernet interface, which Kodi will happily connect to.
Raspberry Pi's don't use PXE boot, they have their own proprietary system. The kernel is fetched by TFTP, along with a config text file which species an NFS location to mount as root. Boot then continues from there.
Oh yeah, I've lost quite a few camera projects to these connectors with cables breaking, slipping out of the connector and whatnot. Maybe just a bad choice of cables, not sure.
FPC is used in billions of devices with perfectly fine robustness. As long as it isn't part of some flexible mechanism (e.g., a flex display, or something folding), it should be just fine. If you toss it into a box with the FPC flopping in the breeze then I'm sure it would be terrible, but... don't do that?
FPC is used in billions of devices-- that are professionally assembled, usually assembled only once, well-enclosed, and careful attention has been given to strain relief. It's cheap and compact and allows very small assemblies.
It's not so hobbyist/tinkerer friendly, where you're likely to put a lot of cycles on the connector, bend things back and forth, and end up with an enclosure that does not protect everything as well as one would like. Indeed, you have a sibling comment talking about breaking lots of FPC going to cameras.
Mechanical/connector failure is a small but noticeable share of the SD robustness problems on SBCs. I would expect FPC to be worse.
Again, this sounds overblown. Obviously the cables are somewhat fragile and you can't just be a brute with them... but the alternative is making a PCB that is much larger to accept a M.2 slot. It just isn't possible in the current footprint from what I can tell.
Also, other connectors for this type have surprisingly low durability. Most M.2 slots are rated for extremely low mating cycles. Amphenonol, who I would considered to be an high quality manufacturer, rates their M.2 slots for '25-60' mating cycles total. Less than 100. Most manufacturers do not even specify the number of mating cycles.
> but the alternative is making a PCB that is much larger to accept a M.2 slot. It just isn't possible in the current footprint from what I can tell.
I'm not saying they made a bad choice; they're facing a lot of constraints and have a lot of IO to get out while staying hobbyist friendly.
> Also, other connectors for this type have surprisingly low durability. Most M.2 slots are rated for extremely low mating cycles.
Sure, but I don't -need- as many mating cycles for M.2, as it'd be screwed to the board and done. Whereas if I'm dealing with a Pi stackup and coming in and out of the case, I'm likely to get through the couple dozen cycles I'm allowed with FPC. And if I'm putting it on a vibration-intensive environment like a quadcopter, I need to be pretty dang careful with mechanicals.
> Again, this sounds overblown.
Everything's a tradeoff. Flex is cheap and small and offers versatility. It's also delicate and annoying.
This. I've had an always-on Pi 3 since 2016 and after countless random corruption issues from various micro SD cards, I moved to booting them from old USB2 flash drives, first an 8gb then an 16gb one. Never had an issue with them and they've been solid. I only had to mess with the flash drives when I had to do an OS upgrade.
Also, those micro SD cards were always fine after a format/partition and I can still use them in other devices just fine. I've read before that the Pi has a tendency to corrupt micro SD cards through its reader, and IIRC it's related to power issues.
Agreed. I recently had the rather unpleasant discovery that when samsung called their SD cards “high endurance”, they actually meant 3-6 months, and that half of the video on my dashcam was missing.
I've recently stumbled upon a SBC with a M.2 slot... Then promptly closed the tab, to stop the temptation to get another shiny dust collector. But they exist.
I haven't had much trouble with the SD cards. The thing is, writing to them all the time means your chances of corrupting the filesystem is higher (like any filesystem). I would try to make the filesystem read-only as much as possible. There are settings for this.
I am aware of the write endurance problems of sd cards, but I've had numerous cards that are cooked fresh out of the box, even from 'good' manufactures like Samsung. I don't like having to do binning for the manufacturer.
It would be nice if there was just an improvement to reliability and speed of SD cards. I’ve used SBCs with eMMC built in (non-removable) and using them is a lot more complex compared to just flashing a new SD card image as needed. Based on the forums it takes a lot more effort for support as well, so from this point of view I’m not surprised raspberry pi has kept as cards as the main option.
I've had some really cheap cards fail but have mostly had better luck with name brands.
One of the things I do is to configure hosts to use the overlayfs (read only fileystem) where appropriate and that helps to reduce wear on the SD cards.
I don't use then where I want a responsive system and use USB/SSD or NVME/SSD instead.
If an sd card is too slow and too unreliable for you, considering the full price of a raspberry pi (including case, fan, psu etc) you might be better served by some intel based small computer, like a thin client or some small form factor computer (hp, dell, lenovo have many models, that can also be acquired super cheap on ebay and similar).
I'm fairly sure that none of the earlier SoCs had an IOMMU. It's of course possible that there was one hidden in the GPU or something, but nothing on the CPU side.
Just did a few searches using his name + politics. His blog was the first link. It appears that he is a devout catholic and opposed to progressive policies.
I think it’s notable that the chip powering the RP5 is built on a 16nm process. (The RP4 was on 26nm).
This is a nearly 10 year old manufacturing process and it’s silly to compare the performance per watt to any Intel or Arm chip on the market today. On such an old node, it’s not surprising that the power draw is so high. Of course an M2 would smoke a RP5 at a much lower power. But the RP5 is 60 bucks!
Orange Pi 5, "plus" version also has 2gen 1-lane pci-e (M.2 wifi), and 3gen 4-lane pci-e (M.2 SSD) and 2x2.5Gbit ethernet.
8nm, pretty power efficient. I've measured it to run at 0.7A@5V idle and 1.2A@5V with all 8 cores loaded with md5sum /dev/zero; iirc it had 1 ethernet connected, no other periphery. Running on Armbian.
How is the software support? That’s the main thing keeping me on rPi. I tried Pine64 and it was terrible. Never could get my PineBook Pro to boot reliably.
There are always little things. As far as I know, desktop works pretty well and there are people running PS2 emulation on it, which has always been super heavy.
arm64 has come a long way for that use case.
It's been solid as the VM host so far. I wrote an Armbian SD card and it just worked. Once a VM is booted though, many things become irrelevant outside of arm64 support.
I haven't tried again since February, so there's a decent chance my issues are fixed, but ZFS wouldn't build and VLAN support was disabled for the NIC. Not blockers, but it did make me rethink some ideas I had.
Alder Lake N (N95, N100, etc) is built on a 7nm class process. Many of those systems could be considered SBCs (SoC/RAM/Storage is often all on a single board). Those CPUs are low end for x86, but much higher end than a typical ARM based SBC.
The N100 is quite good all things considered. It generally beats an i5-6500t in pretty much anything while having very low power draw and all the newest media engine encoders.
Side note, lowering the process means smaller wires and more susceptibility to ESD. I've never vacuumed a Pi 4, but also haven't lost any to it in countless sketchy mounting points. That might change on the 5, GPUs built on 10 nm and lower just die if you touch them wrong.
process node only matters that much when you are dealing with the latest system architecture. whether you compare with rockpi's offering or with apple, A76 is a design from 2018.
while i was personally working with 512 mb with my rpi all this time, i wonder if there is a point to having only 1/2 gb with the new soc.
even if you run the sbc headless and never touch the gpu, you risk starving the threads for things where it would be an upgrade over the previous models.
Not yet, but you can pre-order them from some vendors. The 4 and 8gb models should ship in October.
FWIW, the Pi 4 and Pi Zero W are both fairly available these days. The Zero 2 W is still in short supply, but aside from that, the shortage is basically over.
I'm tired of information being vague, under-specified, or only available under NDA (if you're lucky). I'm not stupid enough to hop on this ride again.
Are there any fully open (in terms of schematics, firmware) RISCV rpi-"compatibles" out there? I'd be happy to pay triple the price of this thing for a power-efficient linux-capable sbc that is open.
Genuinely curious: why does their announcement upset you so much?
Most of the tech world announces products, executes a marketing strategy and then releases stock into the market.
You make it sound like someone left you standing at the altar. You didn't know it existed an hour ago. If you were on a long vacation, it might have released before your return. Why get angry?
Not only that, they wont even release all the information needed to it for certain applications. The x86 platform is far more open than raspberry pi ever was.
Why do you assume this specific announcement is what upset me? Why do you assume this has anything to do with their product marketing strategy? Why do you assume I wouldn't have gotten angry even after they released the product?
Any board based on StarFive's JH7110 is currently best in this regard, I think. Datasheet & reference manual for this SoC is available.
Especially their VisionFive 2 board. I've even downloaded a schematic for it (although older revision than actual board I have). And they're pretty good about upstreaming drivers.
That said: what you probably care about is documentation for integrated peripherals (esp. GPU), and existence of open source, mature drivers for those.
RPi is very good in this regard. Afaik the only binary blobs there is some GPU/SoC firmware, and (maybe) some boot code.
RPi's in general are very well supported & documented, and its software ecosystem is very mature compared to anything RISC-V based.
Could you pinpoint what you think is lacking there?
Other ARM based boards may offer more bang/$. Likely at the cost of documentation or driver support (Beagleboard being an exception).
"Raspberry Pi 5 is faster and more powerful than prior-generation Raspberry Pis, and like most general-purpose computers, it will perform best with active cooling. The Raspberry Pi Case for Raspberry Pi 5, with its integrated fan, is one way to provide this."
They pose a question themselves, and don't even answer it. Of course something will perform best with active cooling. Does it need it?
I don't need this wishy-washy marketing language from Raspberry Pi.
I hate it when someone answers a yes/no question with something other than yes or no.
Here’s my proposed edit:
Q: "Does Raspberry Pi 5 need active cooling?
Original A: "Raspberry Pi 5 is faster and more powerful than prior-generation Raspberry Pis…”
Better A:
"For modest workloads, no. For heavier workloads, you will get better performance with active cooling. Raspberry Pi 5 is faster and more powerful than prior-generation Raspberry Pis…”
It literally tells you that you don't need cooling, but if you add it you'll get more performance.
I don't find it hard to understand what this means: the soc limits it's core performance based on thermal conditions and will throttle when hitting limit temperature. That's standard behavior on every computer or smartphone or GPU out there.
Make temperature lower and it will clock and run at higher speeds without throttling.
They can't tell you how the board will behave on any random case you decide to put it on.
The uncased requirement is exactly what I'd expect to see there. Other than that, only if they decide to get really technical (they should) and tell you dissipating power / °C and temperature limits.
The answer is nuanced. It can run workloads somewhat faster and cooler than Pi 4 without active cooling. But it also can't reach close to its peak performance without active cooling.
"The combination of a newer core, a higher clock speed, and a smaller process geometry yields a much faster Raspberry Pi, and one that consumes much less power for a given workload."
pi5 with active cooler is about 1.2-1.5x faster than pi5 without cooler for most workloads that care than without.
Pi5 with active cooler is about 2-2.5x faster than Pi4. So Pi5 without cooler is probably about 1.5x faster than pi4, depending upon workload. (And more than this for quick bursts where thermal mass wins).
And if you dont have a heatsink and fan of sorts just use alternativing fingers on the cpu, they can still absorb about 10-15 DegC off the cpu temp and thats overclocking a 3b in the 1.35Ghz range. Surprisingly robust. Sadly cant get it to idle below .6Ghz yet, that needs more work.
But it makes wonder how much more phone manufacturers could squeeze out of their phones, although Apple are definately overclocking the 15.
It depends on what "need" means. I'm pretty sure you can take a 400W TDP Threadripper and run it without active cooling. It will throttle down to run at whatever speed (well, TDP) that doesn't fry it. The Raspberry Pi does the same thing.
If your goal is to get the highest score on every benchmark, then yeah, you need active cooling. That has been true on every Raspberry Pi, I think. (I don't remember if the 1 needed active cooling. I did not have any. I also remember it taking over a day to recompile Linux! Still faster than setting up a cross compiler at the time ;)
And how hot does the bulb get? Googling yields "The surface temperature of incandescent light bulbs varies from 150 to more than 250 degrees" (https://www.pacificlamp.com/temperature-of-a-100-watt-bulb.a...). Which, googling says, is likely hot enough to damage a CPU.
150F is like 65C which loads of CPUs run fine at. Going past 212F (100C) is usually where you'll start having problems with a lot of common desktop processors.
I think they answer your question in the PSU section “Raspberry Pi 5 consumes significantly less power, and runs significantly cooler, than Raspberry Pi 4 when running an identical workload.”
That's not documentation of any reasonable level though.
An MPU designer expects to see something like "200mA draw from the 1.2V power-domain when running at 400 MHz" or "10mA draw from the 1.2V power-domain when in first level of sleep". (Maybe not this small since Rasp. Pi is a more powerful chip, but... you know... actual specifics).
"Will my Raspberry Pi 4 power supply work with Raspberry Pi 5?
"Raspberry Pi 5 is a higher-performance computer than Raspberry Pi 4, and you may have problems using an under-powered supply. We recommend a high-quality 5W 5A USB-C power supply, such as the new Raspberry Pi 27W USB-C Power Supply."
The question asks about power requirements, but the answer is about performance?
The first time I read that I thought the 5 needs more power than the 4, not less.
But then you get 5V 5A ... 27W that is clearly incorrect too. So my guess is nobody is proofreading the technical specifications, and everybody that cares was kept away from that page.
You don't specify a PSU by its energy consumption. You specify it by output.
Also, that 92% efficiency, is believable, but a bit high for a 5V 5A PSU (this is a difficult combination). I would expect any such unity to be marketed as high-efficiency.
If you frequently work it really hard, it'll have larger temperature swings and may fail earlier, but it'll still probably last quite awhile. The failure is not likely to be catching on fire.
Beaglebone (from Texas Instruments) is more open, but still not as open as you'd probably like. Still, its a better balance than what Rasp. Pi organization has (more documents are available on AM335x, open-hardware for Beaglebone Green and reference designs, full chip specifications and the like). Beaglebone isn't really "more expensive", as much as its just "lower specs at the same price" though.
The "most open" are MPU chips and their associated "System on Module" boards. This isn't quite a SBC, but its easier to use than a BGA. These SoMs are very poor from price/performance perspectives, but instead serve as reference designs and/or prototypes to the $8 or $9 chips. The overall expectation is that you're "supposed" to be building your own PCBs eventually, so the SoM are kind of just a prototyping aid.
Most SoM provide 100+ pins from the chip as well, meaning you absolutely have to build a PCB to use them. However, 2-layer boards solder very easily to a SoM with castilliated edges (even with a hand-soldering iron)... albeit with a bit of flux and technique and practice. Its just the easiest way to deliver the most-pins of customization in the smallest space. So a relative beginer should be able to boot an SoM. The most difficult routing and Power-Delivery-Network details are already solved on an SoM, you just gotta apply power and build out the final interfaces / connectors.
Take the ATSAMA5D27-SOM1, 104-pins in a 40x40mm form factor. $50 from Mouser for 500Mhz and 128MB RAM (though fully open source and fully documented at linux4sam, and processor manual, U-boot process and everything). But the underlying SiP (MPU + DDR2 RAM) is like $15... while the MPU alone is like $8 and 128MB of DDR2 RAM is only like $3.50 in practice. Since in mass-production, you'd probably have a custom PCB anyway, that's the most expected use case. https://www.linux4sam.org/bin/view/Linux4SAM . I'd say that Microchip / Atmel's MPUs seem to be the best documented that I've found, but are unfortunately the lowest specs. Still, they also have some of the lowest power-consumption (like 200 mW or something), so really they're in a low-power class of their own. Still Linux though.
------------
STM32MP1 is the MPU from ST Micro. Like the Microchip SAM-MPU series, the STM32MP1 is available in SOM, SiP, and "raw" MPU form. Except the SOMs are like $100+, the SiP is like $50+, while MPU is $10ish.
-------------
I know NXP has a huge line of MPUs. I haven't researched them yet though.
---------------
I think all the hardware designers at this level just "assume" that their customers, if they care about "open source", are probably making their own PCBs.
If someone "just" wants a SBC (like the Rasp. Pi), there's not much point in publishing a ton of documents. People can just boot the Rasp. Pi and start messing with Linux.
-------
I got no experience with this yet. I'm just curious and am thinking of a simple MPU layout project ever since I discovered that OSHPark has 6-layer boards and KiCAD supports BGAs in practice. Overall, these lower-power lower-end MPUs fill a different niche than a Rasp. Pi ever would. But I feel like there's enough overlap that these might scratch your "open source" and "fully documented" itches.
I always thought the beaglebone had a better hardware design. The thing I first noticed was the female header pins - why would the pi have pins that can be shorted out?
the beaglebone pru is cool too.
But all of that pales in comparison to the huge mindshare the pi has, which makes all the difference.
Hmm. I'm disappointed honestly, I was looking forward to USB-C with display port and "normal" USB-C power.
Is there any reasonable option with software support that comes even close to what RPI offers? I don't want an SBC where I have to use some strange back ported, super old, kernel with little to no chance of getting updates.
Maybe I should just go for an X86 board? Lattepanda Delta, or Khadas Mind whenever that's released. Not even that more expensive.
> Is there any reasonable option with software support that comes even close to what RPI offers? I don't want an SBC where I have to use some strange back ported, super old, kernel with little to no chance of getting updates.
Pine64 makes SBCs with the same goal as their PinePhones to get mainline Linux on them. That effort expands to many boards that use the same SoC. So the rk3399 used in the PinePhone. Their newer board uses rk3588 which isn't completely mainlined yet. But I think these are the best bet for software support.
You can also buy mainboards[1] and expansion cards[2] (usb-c, hdmi etc.) from the same website. Depending on how beefy of a computer you want there are mainboards from 299 USD to 700 USD. So for ~500 USD you can get a very powerful tiny-ish computer. It obviously won't have the IO capabilities of pi-like hobby boards but it'll function great as a thin client for running linux / home automation stuff.
The right choice would've been 9V 2.8A, because it's what USB PD-compliant chargers offer. For a 25W charger powering a 25W device, that's the only combination which is guaranteed to be available.
Using 5V 5A means you have to essentially use the official power brick, because very few third-party chargers support it. It really sucks.
Interestingly the Pi 5 has moved most I/O like Ethernet, USB, MIPI and GPIO into a custom I/O controller chip called the RP1. It talks to the main CPU over 4-lane PCIe. They also have a custom PMIC (Dialog DA9091) with a built-in RTC and support for external backup battery. Everything else seems pretty standard.
> Interestingly the Pi 5 has moved most I/O like Ethernet, USB, MIPI and GPIO into a custom I/O controller chip called the RP1.
For cost-saving reasons, the I/O is located in the RP1 Southbridge (which has a larger process size) instead of the SoC. I had the opportunity to preview the Pi5 and have provided a detailed breakdown of the RP1's components [1]. In summary, what I/O functionalities does the RP1 manage? Essentially, it handles almost all of them.
Good, then the price of second hand RPi4 will go down as it has been ridiculously expensive (even though I'm settled, still makes me happy for other people).
I'll wait for a CM module before I consider to replace my CM4 fleet, but for now I am happy with my CM4 as it is. Although I admit the ability for native NVMe is attractive. Better I/O as well!
Also, the RPi5 needs a better PSU so that is one thing to take into account.
For people for whom a Raspberry Pi Zero 2 is good enough, they should opt for that instead. They're widely available and no longer ridiculous prices.
The audio jack on Raspberry Pi's has always been terrible quality anyway. Ok in a pinch but it's not a huge loss. If you want quality analog output from a Pi you need a DAC hat.
If you're using a pi you're already in a pinch, putting a dac on a pi seems alot like putting lipstick on a pig, the only scenario where it wouldn't be overkill is music streaming and i highly doubt thats even a majority of pi projects with audio
Yeah, that's true, and I did that when I built a 'streamer' from a Pi 2. It was great. At some point though the quality of the built-in jack did improve, and so when I built a sound/lighting effect box for a toy theatre with the Pi 4 it was plenty good enough to use, and made it simpler and cheaper to put together.
There are loads of cool projects for the Pi (synths etc) which chuck out audio and sound good enough through the built-in jack.
You might look into the Teensy boards. There's an audio adapter that adds a DAC and headphone jack, and there's also an audio library. The Dirtywave M8 was built on top of the Teensy stack.
It's a pity about Teensy3 being effectively gone. I've got numerous eurorack modules based on the Teensy3 12-bit onboard analog output.
The trick there is twofold: one, if you're doing traditional digital audio, this output has a very 'old school sampler' sound that takes well to further processing. It's a really primitive output with a lot of character that sits well in a mix or performance: no trouble building an entire mix around sounds coming off this output.
Two, because it's directly built into the Teensy and doesn't have as many intrinsic assumptions about what digital audio is, you can overclock it to sample rates as high as a megahertz or so, while still running the same audio library. That means if you're just putting out a square wave, the ONLY limitation on your quality is aliasing, as bit depth no longer matters. If you're doing multiple square waves you still don't need full use of the 12 bit analog output to do it. And if you're at 1mHz aliasing is next to nothing, so all of a sudden your cheesey 'bad' output is in fact shockingly good. Even waves which do require the 12 bit resolution are a lot better with no significant aliasing. I run four-note chords with two oscs per note, portamento, and additional waveforms, on a single Teensy, at upwards of 300kHz.
From what I can see, no audio hat does this. They're all mediocre CD quality. I'm sure you could build a ladder DAC out of raw digital outputs on a more modern Teensy and do just as well, especially if you ran even lower bit than 12 bit. Teensy3 has that analog pin just right there, though. Lovely :)
I have fallen out of love with Raspberry Pi due to the performance. In desktop mode, I have found it too lacking to be any fun. It even struggles to run the web browser. And things like Mathematica, that you get for free which is awesome, are just much too sluggish to be very much fun.
Yes, although it sounds like the Pi 5 will be a significant step up in this regard. To be honest for what I was doing, Pi 4 was easily good enough. It was just a bit more convenient to be able to get media assets etc on to the device directly, implement a simple UI, etc etc.
that definitely sounds like a whinge. raspi is fundamentally an unfinished, everything-not-included product. most use cases require some sort of dongle, hat or other adapter. an audio jack (especially in a context where you're not using an HDMI device that you can get the audio from) is much more niche than a lot of other things that aren't included on-board, like storage or POE.
if it's a trade between a built-in audio jack or a built-in RTC, i'll take the RTC 100% of the time.
fair enough, i suppose i'm interested in using it for mainly audio/music applications and so am biased. no use for an RTC. demonstrates really how hard it is to keep everyone happy! over the years i think they've traded things off fairly well.
No RTC means your time is incorrect until something sets it. usually NTP, assuming the device connects to the internet right away. if it doesn't connect to the internet right away, then you just get a wrong time. it means that for any device that ever reboots, there's some amount of logging that has untrustworthy timestamps.
A battery-powered RTC means that any logs from the device aren't inherently untrustworthy.
The audio output from all Raspberry Pis has always been crap, to the point anyone looking for decent quality would have been forced to use an external DAC anyway, so not a big loss here, although stopping that PWM audio generation nonsense and adding the DAC into the CPU would have been the right thing to do.
Given that they made custom silicon for the IO module, I'm disappointed that they didn't include a couple PIO cores in there. It would've been great to be able to use the GPIO pins to drive serial LEDs, as extra uart peripherals, or for any of the numerous other things people have developed for the RP2040 PIO.
I've said this before but I feel like the Raspberry Pi Foundation hasn't met its objective of making computers accessible and affordable to everyone (especially children in disadvantaged places). In South Africa, the average child from a disadvantaged background has no chance of buying a Pi.
Arduino on the other hand, and them open sourcing the hardware means very cheap but less powerful SBCs from companies in China (and elsewhere). I hope to see something in a similar vein from them as they have the ecosystem and resources to mass produce very cheap boards, but the Pi itself is prohibitively expensive, and it looks like they cater to companies first and prioritise making a profit.
The Foundation is not the same as Raspberry Pi (Trading) Ltd. While the Foundation is hardware-agnostic, they have a particular fondness for the Raspberry Pi.
> Arduino on the other hand, and them open sourcing the hardware means very cheap but less powerful SBCs from companies in China.
You probably don't know that the Raspberry Pi Pico is now also produced in Africa (I only learned last week). It offers better value for money compared to the ATMEGA328P. Additionally, Raspberry Pi is actively supporting the development of hardware manufacturing skills in Africa.
Raspberry Pi always felt more commercial than Arduino with a stronger focus on a branded range of products than providing a template for others to copy. We've even come to the point where it's no longer the cheapest option for what it provides at the same form factor. There are usually cheaper options with similar or even superior hardware even for things like the Raspberry Zero.
for most countries (besides the uk and the us afaik), getting an rpi almost often meant paying more than the "$35" sticker price, despite accounting for taxes.
in the past, what would have been the prices in South Africa in your experience? for me, usually the final price would have been a 50% mark up minimum.
> We’re also giving every print subscriber to The MagPi and HackSpace magazines a single-use code, giving them priority access to Raspberry Pi 5 hardware.
A subscription is US$43 and up, and gets you a priority code to use at PiShop or The Pi Hut. You pay full price for the RPi 5, but it ships to you first.
The Lichee Module 4A is a CM-style board, though not compatible with CM4, and is generally a bit faster than a Pi 4.
That's the fastest RISC-V you can get right now (but you can get 64 of those > Pi4 cores in one chip!). Chip and boards in RK3588 class coming next year, so that's ~Pi5.
RaspberryPis are a physical good, those have production limitations. They are also quite cheap, sell well, and have kept prices.
It’s quite different from a lootbox that has no production restriction (replication is instant and almost free)
I don’t believe that the low supply of RPIs is an attempt to extract huge profits, logistics are not easy. This magazine move seems more of an honest way to prioritizing people that have been showing interest. They are not demanding new subscriptions, they are prioritizing who has on.
An entusiastic hobbyists, that has been accompanying the project and supporting via these, would be happy that we got his pi before some random guy clicking buy on impulse
> RaspberryPis are a physical good, those have production limitations. They are also quite cheap, sell well, and have kept prices.
I'm not sure they kept prices. $60 for a 4GB board sounds like a price increase compared to the $50 RPi4 4GB model. And didn't the RPi model A sold for around $25?
> Yes RPi foundation, please also spit in my mouth,
Ridiculous take. Seems like a perfectly good balance, prioritising limited stock to supporters of the Raspberry Pi, beyond simply selling to the highest bidder every time.
It's not worth getting a subscription just t get earlier access to RPi 5, but dedicated hobbyists who read MagPi are probably delighted the stores won't immediately be emptied.
> PCIe 2.0 x1 interface for fast peripherals (requires separate M.2 HAT or other adapter)
This could make it appealing for a low-power server, like a home NAS. In previous models, disk I/O had to be over USB, which imposes more CPU load than native SATA, along with various unpleasant quirks typical of USB-to-SATA bridges.
Too bad it's only one lane (assuming that's what they mean by x1), but I think that's almost enough to saturate a SATA bus, so it should nevertheless be useful where NVMe speeds aren't needed. I hope it's implemented well.
I hope they’ll reiterate the computer in keyboard form factor (pi 400). My son is 6 and is learning to read and the pi 400 looks like a perfect first computer.
I really like the form factor, if only there would be some off-the-shelf options for a mechanical keyboard version ..
I know there are plenty DIY projects for this (and the cyberdeck scene is a rabbit hole that I do not dare to go down), but it would be nice something more easily available.
Yeah the 400 made it difficult to fit a replacement keyboard. It will be really nice if the 500 has an RP2040 for the keyboard controller instead of the fixed function Holtek thing in the 400. Then the keyboard FPC connector would be a second secret GPIO bank :-)
Watching the video, is like seeing a commercial for a PC many years ago. Only differences is the Raspberry's tiny size, small price and the few thousand times faster processor.
I was going to post to ask how it was likely to compare with the current crop of RK3588 boards that are marching towards decent mainline support, but this link already answers this to some extent. It looks like it's faster on some benchmarks, slower on others, but basically in a similar ballpark?
The fins on the active cooler are the wrong orientation (the fan is blowing at the flat sides instead of along it). I'm curious why it was designed this way when so much effort was put into using a proper 4 pin blower fan and header.
It will probably work fine but it's something you notice just like how Dell put the heatsink on the IDRAC card in the R420 the wrong way, which would have significant impacts on airflow and cooling.
I suppose you have a point. It was probably extruded along the fins like this (coming straight in/out of screen):
__|||||||___(fan)____
Then they would have to CNC the edges off to create the mounting ears anyway and cut the slots in the fins. I suppose they could just make the gaps larger but this is probably getting into the details...
What is (or would be) the "Tsundoku" equivalent for Raspberry Pi?
I started with Pi-Hole a few years back but ended up with a commercial paid DNS resolver. The timeline usually goes like this -- will stumble on another interesting video of Jeff Geerling, then spend the weekend tinkering with the Pi, keeps running, forget about it, found it to be not needed, plugs off and is lying around.
I think it should be "The art of buying single-board computers and never running them once". I think the "once" is needed in the case of computers. They are as machines not directly comparable to books...
A book, like any media, fulfilled its initial purpose after it has been "consumed" - anything else (looks nice, feels nice, smells nice) are physical, subjective qualities attached to it. You might even buy it solely for these physical qualities, but that's besides the point.
A machine has its purpose in its usage, and that usage requires known resources that should not be carelessly wasted. I personally buy them to enable a utility for myself and have fun discovering it. It fulfilled its initial purpose even if it was powered on just once.
Even those single-board computers that were put to "good use", as building blocks for new devices (eg. [0]), are still not in use the whole time.
> I think it should be "The art of buying single-board computers and never running them once". I think the "once" is needed in the case of computers. They are as machines not directly comparable to books...
I have a few orange pi boards in a drawer that would disagree with you.
I have about 10 unused Raspberry Pis, so I collected unused monitors as well, and now run a mini code club in my local school. The school's laptops are all locked down to the point that running Python on them that this was the easiest way to provide a Python dev environment.
In my case at least, it's because Python has a strong ecosystem. Everybody's heard of it - parents, kids, and teachers. There are lots of good kid-friendly education materials.
Of course, that begs the question "how did that ecosystem develop in the first place?", but I can't answer that.
It doesn't require a person to change how they think about programming from their college C or Java classes, but is a lot simpler than those two languages.
Because every machine you use requires maintenance whereas NextDNS [1] costs 2 EUR/month. The time wasted on maintenance alone is worth more, add to that the electricity bill.
Examples: I also use one with OpenWrt (though I'd prefer OPNsense on it), and I use one as PiKVM. I use one with a portable HackRF (3D printed case), batteries included. I use two with Pimoroni Enviro+, and I have a fun Turing Pi 2 homelab/miniNAS with 3 CM4s (one Jetson Orin Nano). I got two RPi4's one at my mother in law, one at my mother, allowing a VPN connection for tech support and also running Jellyfin with old content for them.
Personally, I still enjoy my Pi's (and no not all of them are on and used 24/7) however I also very much enjoy Proxmox and VMs. But the Proxmox machine is a Xeon... (HP MicroServer 10 Gen 10) the fan is loud af and difficult to replace with say a Noctua due to HP ingeniousness.
I think it’s hard to generalize that into a phenomenon.
I have always bought 2 new raspberry pies with every release. I’ve used some and forgotten, unplugged some, given away some, broke some, and now they’re all used in some way.
I bought them because I knew I’ll use them. I didn’t buy any other random toys or mini computing devices (even though they were alluring) because I knew I wouldn’t use them.
There should be a name for the art of making every thing that randomly happens to some people into a phenomenon.
I do recognize GP, though. In college, I was obsessed with Linux. Nowadays I have a Macbook, and like you, have an Intel NUC (running Windows). I really don't need a Raspberry. But boy, do I need to resist the urge to get one.
I hadn’t enough resources to delve into Linux too much back then, and switched to macOS instead. Fast forward 15 years and I’m back to Linux and don’t enjoy my macOS time any longer. Windows, always despised.
I wonder if anyone here is using RPi as a main PC? Seems to be capable of most office work with a bit of multimedia on the side. My laptop is dying and I would love to get something stationary that is low-power by default yet good enough so that the internet wouldn't lag.
Mounted at the back of an LCD screen could make a nice wireless setup with the other peripherals.
I was using a Pi 400 for my programming PC, basically testing out if I can sell my gaming PC after I got my steam deck. It was perfectly capable for that task, and I’d say I would be happy to use it as my primary desktop PC.
Just sold my Pi 400 after the announcement of the Pi 5, deciding the Pi 5 will be my primary desktop PC. (Also ordered a Pi 4 because I have some other plans for it, but don’t want it built into a keyboard).
Falling back onto my Pi 3 I never really used before — it’s surprisingly capable. Not as a general desktop but gets my programming done fine. Its bottleneck seems to be I/O and limited RAM.
I got one and I don’t touch it. I do also have an orange pi 5 and khandas edge 2 pro. Both are miles ahead of the rpi4 and based on specs they are miles ahead of the rpi5.
Interesting. Can you give more details on your work? I've been on the edge lately over picking a desktop, an intel NUC or something like a Pi. Price to performance and power draw is something I'm considering.
Same here. A great advantage of such devices is they can later be easily repurposed to control home automation, audio system or make a simple DIY project with them. It is much harder with other types of hardware, like a laptop for example.
> I wonder if anyone here is using RPi as a main PC? Seems to be capable of most office work with a bit of multimedia on the side. My laptop is dying and I would love to get something stationary that is low-power by default yet good enough so that the internet wouldn't lag.
Unfortunately, it is too slow to even smoothly power a desktop environment.
It will probably display video at those resolutions and refresh rates just fine, so it could be used for improvised commercial displays like a TV at a conference looping some information, for example.
Surprised by the amount of negativity here. So many comments that are saying the raspberry pi isn't competitive in their market anymore and then post more locked down less open systems at 2x the price.
I mean, keep in mind that the Pi5 is a $60 device with ~$60+ of support hardware (Power supply, cable, miniHDMI to HDMI, Cooling, Case, Storage) You can get considerably more powerful mini PCs for less than a 4GB Pi5 with needed accessories that is upgradable, has further software support, and includes all of what you need to get with a Pi. Also, you are reducing eWaste.
> You can get considerably more powerful mini PCs for less than a 4GB Pi5 with needed accessories that is upgradable, has further software support, and includes all of what you need to get with a Pi.
Name three, please, with emphasis on the "considerably more powerful" part.
The Pi5 is slower than a Core i3 6100, does not support upgradable ram, does not support Pci Gen 3 let alone Gen 3 x16, no actual Sata 6, no upgradable sockets, etc.
There are a TON of uses cases for a Pi 5...but only really if power consumption and size is critical for the application. Almost anyone doing a Pihole, Home assistant, or the like is better served by a used miniPc.
Someone buying used electronics on eBay has already decided that their time is worthless and the hardware doesn't need to be dependable out of the box. Which is fine, those are decisions a person could make, but it's just disingenuous to compare the value of one thing to another without taking the you are using someone's untested trash factor into account.
There are reasons to prefer a Pi over a used mini PC, but it is not reliability. The fact is that with Pis or PCs, _if_ you're still running after a week, you're highly, highly, likely to survive past the point of usefulness.
Reasons to prefer a pi over a mini PC - easy access to GPIO pins; small(er) form factor; power efficiency; lower weight; still cost if you can work with zeros.
Reasons to prefer a mini PC over a Pi - price to performance ratio is often far, far better; size is "good enough" for people just after a small computer (rather than an electronics project, POC, etc); you are, in fact, reducing ewaste.
In short, if you just need something to run Home Assistant or Plex on ProxMox or similar, you would find more reward in a mini PC than a Pi, particularly in performance.
The bias at play is that people see "old" and equate it with "bad performance". That heuristic only works when comparing like for like - yesterday's mini-PC to today's mini-PC.
The biggest reasons to prefer a Pi over a mini PC: the documentation, Raspberry Pi Press, the community, the ecosystem.
Let's put it this way: you pop in your SD card, set up a username/password, a few other settings, and you end up with a system that already has Geany and Thonny installed. Go to Help -> Bookshelf and find all the issues of MagPi and Hackspace for inspiration and learning, and a couple dozen books like "Essentials - Code Music with Sonic Pi" or "Get Started with MicroPython on Raspberry Pi Pico". Search "Raspberry Pi Hats" and get a lifetime worth of reading and learning there, hitting just about any domain you could care about.
The vision of the Raspberry Pi wasn't to create a cheap, tiny PC, it was to create a learning platform, of which the cheap, tiny PC is the basis. More than any other computing device I've seen, it really lives up to that vision.
> The fact is that with Pis or PCs, _if_ you're still running after a week, you're highly, highly, likely to survive past the point of usefulness.
I might be hopelessly biased since you would regard me as someone who keeps using computers when they are "past the point of uselessness," which is not a phrase I'd actually use.
A lot of the stuff I do to keep computers running after a long period of ownership does not even apply to the Pi, which is nice. I won't ever have to replace the fans on one of my Pis (the case design of the prototype Pi 5s we've seen notwithstanding), but that's a standard thing on PCs. There's maybe one capacitor on a Pi's board? I've never needed to think about it. I've guiltily thrown away and replaced soft power supplies on PCs, even though I had a friend once who knew how to fix them and I should really learn to do the same, but there's considerably less guilt when it's a USB wall wart. I've pitched a few SD cards belonging to Pis over the years (never the proper SSDs, though) but that has never been nearly as painful as a drive failure in a PC.
So I guess I'm saying I would factor in the maintenance burden.
> The bias at play is that people see "old" and equate it with "bad performance".
I see "used electronics" and "eBay" and think of "wasted time," "frustration," "incompetence," "shamelessness," and "fraud."
The other thing I would factor in is that my time is worth something and eBay's main goal, when it comes to used electronics, is to waste it.
> In short, if you just need something to run Home Assistant or Plex on ProxMox or similar, you would find more reward in a mini PC than a Pi, particularly in performance.
I'd split the difference here and say I'd be happy to run Home Assistant on a Pi and Plex, quite possibly, on a PC (but I'd want to at least test on a Pi). I'd be happy to buy a new PC for the purpose from a trusted retailer, or build it myself.
That's just silliness. The Raspberry Pi 4 / 5 are made with sub-$60 dollars worth of materials when brand new. You get what you pay for, which is not much and certainly doesn't include dependability. It's a hobbyist and educational plaything whose top design priority is to be low cost.
Little 1-liter office PCs from HP/Dell/etc. have MSRP closer to $1000 when new, meaning a correspondingly larger BOM cost, and are built to last until obsolete by PC OEMs who have years of experience doing that. Even well used units have better prospects for durability and longevity than new Raspberry Pi boards.
I get that you were trying to be edgy, but you landed on goofy. You're addressing someone who owns over a dozen Pis, dating back to the original 256MB Model B. They've been quite dependable for me. It is not something I need to hypothesize about.
The idea that a machine becomes more reliable as the number of parts increases and the cost increases is an interesting one, though.
I've been using RPis and secondhand mini PCs for embedded systems development and testing professionally for the past few years. I've got 3-4 dead Pis already out of a dozen or so that we have, a mixture of Zeros and 4s, which isn't exactly impressive. Meanwhile the couple dozen mini PCs, despite their age and having been bought used, have had only one failure, a RAM module which was trivial to replace. The failed Pis went into the electronics recycle bin because that's all you can do with them when the develop issues.
Sorry, I stand by my opinion that Pis are hobbyist playthings. I'm sure they were reliable for you but you probably weren't trying to do anything serious with them either.
What were you doing with those three boards that failed? I've got the feeling you meant total failure and we're not talking about SD card failures or broken connectors. Heavy use of GPIO?
> I'm sure they were reliable for you but you probably weren't trying to do anything serious with them either.
Of course it was serious. I was frowning the entire time.
> secondhand mini PCs for embedded systems development and testing professionally
I feel like I'm missing some key piece of context when people mention the humble dumpster origins of the PCs that keep coming up when comparisons with Pi are made. Like "we were trapped on an island the entire time," or "our company operates out of South Sudan, the poorest nation in the world," or something like that.
Raspberry Pi 4s fresh out of the factory often have hardware issues themselves (HDMI connector failures, unreliable USB port power supply on reboot etc.), as one can see from the numerous (and unacknowleged) reports on Raspberry Pi forums — so I find the comparison with used hardware fair.
Anecdote: my one attempt to use a Raspberry Pi PoE hat ended in almost immediate failure due to a mechanical failure of a header on the hat. (Also the built in fan is crap.)
Why lie? Pis have fragile connectors and they certainly break more often during use than a brand name laptop's connectors, but they do not have a high defect rate out of the box.
Power and consumption and size are probably significantly more important than the three you list. How often do you "upgrade the ram" in an application where something like a Pi may be used as a component in a larger system?
So double the price for slighly better CPU (3.9GHz turbo speed vs 2.4GHz -or 2.9GHz with OC-, both with 2MB L2 cache), for energy it needs 36W (vs 10W), 2 usb ports (vs 4), no microSD support, and it weights 4 times as much (without including HDD); it does have 256 GB of storage included but for $20 you can get a microSD of the same capacity for the Rpi5 (or 128GB for $10 when that's enough). Yeah, it doesn't look as that much of an improvement.
I just clicked it, and it shows $107 for the 8G/128G version.
But either way, I think your seriously conflating a lot of variables. Things like power draw are not really comparable unless you pick a workload and measure the perf/power, as just clamping the max frequency down, or putting less ram in it can constrain actual power draw, while still possibly being as performant as the rpi.
I don't see pricing for the case + fan + power supply on the rpi, but the 8G model is $80 for the board. So you have to find at a minimum a battery for the RTC, a case, fan, powersupply, and 128G SD for $30. But then SD isn't in the same ballpark as even the cheapest SSD with respect to longevity or at $10 likely perf either. Plus, your shit out of luck if you need 16G of ram on the rpi vs paying an extra $18 on the beelink.
In the end, the rpi is going to be a fun hack device, while the beelink is going to just work with a wide variety of software (windows anyone?) and Jeff G won't have to complain about his m.2->pcie->graphics card not working out of the box either. OTOH, if you actually need to design a HAT/whatever for some special purpose it might be easier than designing a m.2 PCIe device, or it might not be. The GPIO on the rpi might be very handy if it happens to work better than a ESP/whatever attached to the beelinks USB port driving GPIO pins.
Much of this is going to come down to how much one values their time for one off devices, and for larger production runs the fact that the beelink is just one of hundreds (likely thousands) of similar devices that can be drop in replaced when beelink has a supply issue. Particularly because there are devices that are just a $100 more with multiple 2.5G links, sata ports, whatever.
Throw in power adapter, case, and micro sd card for rpi 5. Price difference almost disappears. Do you seriously think that rpi with an sd card going to beat 256gig ssd?
> You can get considerably more powerful mini PCs for less than a 4GB Pi5 with needed accessories that is upgradable
What suggestions do you have for this?
I’m sure some people can find these, but every time I see comments like these (or ones that actually give specifics), it’s always used ones on eBay that aren’t actually better and are in unknown condition.
I’m not doubting that some people come out ahead, but for me, being able to get* the same hardware each time, new, has value. Also, a huge draw for RPIs for me are the GPIO pins.
* Yes, the past few years were hard, but not impossible.
When RPi 4 was going for $200, I got some Orange Pi 5s for about $140 each. My configuration included 8GB RAM, the power supply, and a 250GB NVMe drive.
It seems like it's about on par with the RPi 5 in terms of power. The OS support is lacking, but the M.2 and +4 "Little" cores make it pretty compelling regardless.
The used on ebay is the actual upside. The issue is that most people do not need the size of a Pi and using used hardware is reducing eWaste. That said, my personal grab was an Acer refurbished Acer XC-830 J4125 for $89.
While that looks like a nice small-ish tower PC it doesn't seem like it's in the same class as an rPI on a lot of axes, it seems half way between an rPI and a full desktop tower.
Honest answer: I want systems in my house that have a better power to work ratio than I can get from a used laptop or NUC, but can still burst to high power when needed for workloads that either have bursty compute demands, or on the rare occasion I run a consistent high-demand workload. That said, the RPi5 doesn't fit the bill for me.
Since replacing an old workstation for secondary/media computer in my office, and my laptop with Apple Silicon systems, I've been incredibly pleased with the efficiency one can get from heterogeneous core layouts in traditional compute devices. Not to mention the drop in thermals in my office. The summer I swapped out that secondary/media machine for the Mac Mini, the temperature in my office dropped 5 degrees Fahrenheit, while being just as capable for the things I used that machine for.
I'd like to do the same for my compute nodes in the house, which power my homelab for various workloads. Apple hardware's value retention doesn't make sense to pick up, even used, especially when I would have to trust an Asahi-flavored distro to make that work, so the platform stability isn't there. I don't need GPIO, so most SBC offerings just don't make sense. The only thing that comes close is the LattePanda Sigma, which would get me an Intel platform with efficiency cores. A SBC or even a NUC-like platform with a higher core count, at the $3-500 price point, would honestly be great! It just doesn't seem to be a market yet, and I don't imagine it will be until we see more penetration with ARM as a desktop compute platform.
I'm pretty sure I could do all that with an array of stolen business PCs, after I machined the serial numbers off. It would be cheaper for sure.
Apologies if it wasn't clear, my original comment was a joke. It's becoming a pet peeve that comments about used computers, presumably from users in the third- or even the fourth-world (?) where money is really tight, dumb down just about every thread related to Raspberry Pi.
Because I don't just want a PC, I want a learning/hacking/making platform, with the ecosystem and support to go along with it.
A used nuc or dumpster dived computer doesn't have two magazines, a couple dozen books, official support and documentation and guides, and an entire world-wide community dedicated to teaching and inspiring me.
Same as the people on Reddit claiming that Raspberry Pi has killed itself as a business because it's constantly in such high demand that it's always sold out.
They're claiming that they've killed themselves as a business hobbyists, which isn't exactly wrong. During their supply chain issues they made it clear they were prioritizing their commercial customers over the hobbyist market.
At this point the hobbyist community has moved on to other things (shout out to odroids). I don't think raspberry pi is going anywhere, but they do have a very different market and customer base than they did five years ago.
Also as a company their social media people really are assholes, which doesn't help at all.
> During their supply chain issues they made it clear they were prioritizing their commercial customers over the hobbyist market.
Which is 100% the correct thing to do. They said they were especially prioritising smaller companies that relied on Raspberry Pis, where otherwise those companies would collapse or need to lay people off, and also kept some aside for educational establishments.
Sorry to say this, but your home assistant plant waterer robot dog feeder is less important, especially since hobbyists can just go to another product much easier than a small business can.
> At this point the hobbyist community has moved on to other things (shout out to odroids). I don't think raspberry pi is going anywhere, but they do have a very different market and customer base than they did five years ago.
FWIW the communities I participate in are still RPI focused and I don’t think I’ve seen anyone using an odroid.
> Also as a company their social media people really are assholes, which doesn't help at all.
This is something I’m somehow unaware of (I avoid general social media); what do you mean?
FWIW, the odroid forums have some assholes that will come out of the woodwork if you even mention a competing product. I reported one, and the mods told me that the forum was only for odroid products, so they weren't going to take any action.
Yeahnah, I'm not moving to odroid, its a lot of faff and terrible support.
I have a N100, and some rp2040s for GPIO.
I have a project coming up that needs to drive a small screen, the raspberry pi is totally the platform I'm going for. There is no comparison for hardware and software options.
So what did all those people that couldn't get a RPI do? Stop doing stuff or did they move on? I'm pretty sure the other brands saw a big uptick in sales but if you know otherwise please do share as I have no numbers to back anything up.
Are there any odroid-based PiKVM-like devices? The PiKVM device is really well built and the software is good. Easiest thing for me to LetsEncrypt too.
Of the dozen or so friends of mine with them, only one really uses the rpi for anything "serious" after having played with it, hit some roadblock and tossed it in the drawer. Many of them have soured on it due to something not working right, spending a lot of time debugging it and being stymied by the fact that the docs need to fix some of the problems weren't open, or simply by the fact that it seems every single one is sold as "desktop class" and it turns out they are worse than 10 year old desktops for even basic things like say... playing youtube videos.
So there is a lot of negativity built up around them by people who got frustrated by the previous versions.
I feel like the negativity is coming mostly out of a very small minority that tend to gather here and to a lesser extent reddit. In other places there's much less of that.
I question whether it's wise to launch a new product before the supply chain issues which have plagued Raspberry Pi for years have been fully resolved.
Granted, the situation has improved slightly over the past few months. But you will still find Pi 4s out of stock more often than not.
The CEO said last year not to expect a Pi 5 in 2023, because they wanted to take the time to recover. Why the u-turn?
The point here is likely to pull the rug out from under scalpers' feet.
With the Raspberry Pi 5 out in two weeks, all the held-back inventory of older models will be dumped, prices will plummet, availability will become a non-issue.
Finding older models were also almost impossible in the past two years. It's unlikely that Raspberry Pi 5 will solve the issue. But even so, it's not a wise move because what is the point of bringing a new model when they can't make it available to normal people?
The Raspberry Pi 5 will only be for sale to individuals until the end of this year (no industrial customers competing for inventory like the older models)
Plus, they're only launching the 4/8gb models to start. So there'll be another wave of cheaper ones a little later. Really hoping they still hit the $35 price point on the 1gb model.
I think I've seen Pi Kits at my local target. The issue with those is, they're for niche things I might not care about and now I got tech waste on my hands, but also might not be the exact model I want.
Note I'm not disagreeing, just saying in some cases, the ones in-store are kits.
I have and it is still a pain. Many websites still have limits on how many you can buy. The situation has improved but it is far from what your comment implies.
Many websites still have limits on how many you can buy.
For a hobbyist / individual, is that really a big deal? I mean, how many do you need at one time?
Anyway, the claim all along has been that supplies would be "back to normal" by the end of this year, and so far things seem to be tracking that way. If you look at rpilocator.com now, the entire first two pages are full of green lines, which is a DRASTIC improvement compared to just 6 months ago. And some of the major distributors are getting in shipments of 5,000, 6,000 at a time of some models and having them in stock for weeks on end. So one can clearly see that the situation is improving rapidly.
That said, I will make no claim one way or the other with regards to the question of whether or not shipping a Pi 5 is a "good idea" or not.
I'd be surprised if reselling through authorized sellers isn't much simpler and problem-free than selling them directly.
I also expect that using resellers ensures better odds of protecting the brand/project goodwill. Resellers deal with problems like "I paid a ton of cash for a board and it arrived late and/or broken". Support alone is a nightmare, and I recall that raspberry Pi struggled with PR when they started out. I vaguely recall Liz Upton being behind some ill-advised episodes that didn't improved Raspberry Pi's image and would get anyone other PR person sacked.
I think scalping is more of a supply issue, raising the official price of the product would only require more cash when scalpers are doing their buying.
Raspberry should raise the price until scalping isn't profitable. Keeping the price low is just handing money to scalpers that should be going towards future product development, until they can meet the demand.
I think it's completely acceptable so long that a legacy compatible "Pi SE" options would be available, in Pi 3 and Pi 4 form factors.
Raspberry Pi are used installed as components into third party engineered products, and Raspberry Pi brand holds no value to potential industrial customers if new products did not technologically exchange with existing such products. That is to say the exact mounting details, electrical compatibility, software compatibility, DO provide the value the "Raspberry Pi" brand offers if its competitors offered it.
What I'm saying is, if Espressif brought that new ESP-branded 32bit thing in the Pi 3 mounting dimensions and onboard eMMC and 1/5th performance at $35+9%, that kind of thing could outsell Pi 5 at this rate, and I wouldn't mind watching that happening.
There's nothing stopping a person from putting an ESP on a RPi B board format (or a Zero one), other than certain IO limitation. The upcoming P4 might be a candidate [0]. If you are interested in sbc in Pi form factors (B, Zero, CM3, CM4), then something I follow is CNX [1]. It will typically have notifications of any new Orange Pi, Libre Computer, etc.
It might make sense given pi2040’s PIO capabilities. Additionally, the RPi5’s io chip, RP1, might have some similar tricks inside.
The Raspberry Pi Pico has a fascinating peripheral known as the “Programmable Input/Output” (PIO). This device allows us to write very simple assembly programs to emulate a number of different peripherals and communication protocols.
RP1 is our I/O controller for Raspberry Pi 5, designed by the same team at Raspberry Pi that delivered the RP2040 microcontroller, and implemented, like RP2040, on TSMC’s mature 40LP process.
Through hole components resulted in previous Pi's being slightly slower to manufacture, the Pi5 only uses surface mount components which should increase Sony's build cadence.
Upton talked about this in an interview. It's partly the interplay of manufacturing at this point, too.
When they have parts, they'd like to catch up with the backlog faster. Because the 5 doesn't rely on the robots that place through-hole components, its production can to a degree "overlap" with the Pi 4, basically increasing the total number of boards created per hour.
Having two SoCs that don't require the same machines also helps ease a bottleneck there.
I'm much more surprised that they decided to launch with a GPU that would have been considered awful if it had been released with previous pi four years ago.
The recent crop of unrefined, made-for-experimenters RISC-V chips that have GPUs that are a LOT faster. Add an extra $0.50 to the chip cost and use an out-of-the-box Mali GPU that's several times faster and has better driver support courtesy of ARM (who now support the open source drivers).
My little pi CEO anecdote: Eben Upton visited local hackerspace many years ago, someone asked about Pi2 (or maybe 3) potential release date and should he wait or buy now, Eben answered they arent even planning next version. A week later Pi2 was announced.
Meh, that's pretty standard practice due to the Osborn effect. [1] At the Pi1 phase, Rpi was a pretty small company. It wouldn't have been great for them if rumor got spread far and wide that Pi2 was on it's way in a week or two.
I think the criticism of saying "we're not even thinking of a replacement" is fair, the standard dodge "we don't have anything to announce at this time" which is enough to avoid the Osborne Effect.
This happened a lot sooner than I would have guessed! I thought the 4 was gonna get milked for many more years. It'll still be around no doubt but the speed bump is appreciated.
Interesting to see the emergence of a southbridge chip in SBC land. RPI4 using a Via USB3 controller and a dedicated Broadcom ethernet chip is somewhat against the grain of most embedded chips, which tend to have all this io on package. Here Broadcom now has a single chip that bundles a bunch of the io needs together, and takes back the business from Via, which is a tailored step forward to combine functions. But it's still a secondary chip, which is fairly unusual.
On the upside, it means less pins on the main core. And it physically helps fan out the io, can be placed closer to io connectors. I went looking for a chip identification for RPI4, but didn't find it; I'm curious whether old chips had discrete PHY or whether they were builtin, and what the situation is with the new southbridge.
Around the start of this year they were saying the Raspberry Pi 5 would be released next year, so it’s about a quarter or two earlier than previously expected.
Part of the reason for the separate IO chip is that it can use an older silicon process (40nm) than the main chip (16nm) which makes it easier to have robust GPIO pins.
I think it could avoid some of the issues with gaming on linux because you could focus in on a unified platform and if there was a gaming SDK targeting the pi we might see orginal games being made as opposed to just ports.
Also what exactly would a micro-kernel entail? I'm much more familiar with game development than os development.
Most Raspberry Pi Linux distros take around 10 seconds to boot. It can be made faster by using a tiny distribution or something like https://www.linuxfromscratch.org/ but it would be nice to find an OS geared towards gaming.
You can run PICO-8 on a Raspberry Pi today. Still a "retro" feel, but they are all new games being written for the platform.
https://www.lexaloffle.com/bbs/?tid=3085
It's a really nice idea, I've had similar thoughts myself, but sadly (and strangely) I actually think the Pi is too powerful. PICO-8 and similar works so well because the limitations are severe, 3D is a massive technical achievement on those systems, whereas the RPI can run Minecraft (to my knowledge) without too much issue. It puts it in this odd space where it can do too much to bring game development back to a simpler time, but not enough to make it an actual market.
Aside from that, it'd suffer from all the same problems as PC gaming does; different monitors to account for, no standard input method, varying specs between models, etc...
I could see it working, it's just got more issues to account for than other fantasy consoles
The Pi is probably too powerful for the hobbiest game market just like your saying. I think the Playdate also somewhat fills that niche but maybe the Pi could be something in between the two extremes of modern consoles power and the 8bit fantasy consoles.
For it to work and avoid the issues with PCs it'd need a unified controller, sdk, and a single model that games would be made for.
You're a 100% corect though it'd be a lot harder to make games for compared to fantasy consoles.
I feel like the GPU on the Pi is pretty weak. It was fine in the first model, but they haven't improved it much in subsequent versions compared to the CPU.
It is pretty weak but that might be a plus because it sets the ceiling for what can be made and would somewhat prevent the insanely long dev cycles we have today.
Same could be said for the cpu, but that has seen big jumps in performance.
Faster cpu enables 'heavier' applications, some of which also involves more graphics work. And there's also gpu compute (tbh, dunno if that's even a thing on RPi's?).
A useful general-purpose computing device has a practical ratio for cpu, ram, graphics & storage options. Gpu is definitely the part that's lagging on latest RPi iterations.
The Raspberry Pi CM4 is the main system processor for our open source solar powered farming robot called Acorn. I designed a custom motherboard for it in Kicad. We also use the Raspberry Pi RP2040 processor on our custom dual brushless motor controllers (also designed in Kicad).
* 1 CM4 for a PiKVM on my desktop (my main home server uses a proprietary iLO).
* 1 CM4 as router running OpenWrt, to replace my ER-Lite.
* 3 CM4s in a Turing Pi 2, one Jetson Orin for ML. This is my homelab. The CM4 are running Dietpi.
* 2 RPi4 as remote VPN (Wireguard) endpoints to do tech support for my mother-in-law and mother. Also serves as Jellyfin server containing old movies, and jumphost for NAS backup.
* 2 RPi0 2 W with Enviro+. One indoor, one outdoor (with different, custom 3D printed casings). Runs Raspbian, will switch these to Dietpi.
* 1 RPi3 for with my portable HackRF (with custom 3D printed case). Batteries included!
* 1 RPi0 W for Pwnagotchi. Not really used anymore but too cute to break.
* 1 RPi0 W for Pimoroni grow to track water management in plants (with 3D printed case). Need to get that project moving, it is for my mother.
All my Docker containers run on my main server but I'm considering to switch to my Turing Pi 2. I got solar, so the Xeon is OK for now, but still...
Then I got: 1 RPi3B+, 1 CM4, 1 RPi2B, 1 RPi0 W. I should sell as they're right now 'Raspberry Piles'. Won't use them. To be frank I'd like to upgrade the RPi3 with the portable HackRF at some point, and once I get a 3D printer I might need Octoprint so I suppose I could use one of these machines for that.
I went with Dfrobot [1] because they also sell a case (and it is all in stock). Had the CM4 already. I wouldn't have minded another one if I could've 3D printed the case but I didn't find such. Btw, I did have to go from 3 to 2 ports.
Right now it is hooked up on my switch. I intend to put it between my desktop and my switch and run some iperf3. Because I'll get fiber soon, and I'd like to be able to saturate that.
Yes! I’ve used RPi SBCs for various purposes over time:
- One early RPi I bought when RPi was very new to the market. I played around with it for a while and then I put it in a drawer.
- Months later I pulled it out of the drawer and started hosting my website on it for a while. I have since moved on from using RPi at home for hosting any of my sites. Might do again at some point in time.
- I use an SBC as one of the pieces of electronics that I use for controlling over 9000 LED pixels on a sculpture for which I created the light setup and programming. The LED strips we bought are from China and were supposed to be IP65 but even so we now have a bunch of the pixels showing only red, which is bad. No fault of the SBC. Also the SBC in this installation is not an RPi. But relevant nonetheless.
- I hooked up an image sensor with a lens to an RPi and put it in a surveillance camera housing, and added PoE to it. Currently not in use but would like to deploy somewhere in the future if given the chance to do so.
- I have an RPi that I have connected a 5TB drive to in an enclosure I bought from AliExpress. This RPi I both ssh into from my phone to download things for the future, and I also run a squid proxy on it via which one of my laptops connect.
> If you pair the raspberry pi with cache in cloudfare I think you can host on it
Yup. That’s actually what I did :)
And then the reason I stopped hosting my site on my RPi was because I felt like, well if I depend on Cloudflare anyways, I might as well just use a VPS, or GitHub pages, or Cloudflare Pages :p
But there is something appealing in hosting things from home still. And doubly so when doing it without relying on Cloudflare etc
What I will do however, is that next time I set up some internet reachable hosting of content at home I will maintain one version of it that is directly served from the machine at home itself, and then I will keep an up to date copy of it on one of my externally hosted servers.
I've got one setup at work as a display for some networked security cameras. Displays 4 concurrent 720p streams on a TV for the office 24/7. Cheaper than Ubiquiti's solution [0] as we didn't need more than the 4 streams.
At home I've got one for Home Assistant, one as a TV box because Netflix's sharing blocking check is only on smart TV apps. I've also got multiple as 3D printer controllers (klipper/octoprint).
A Rpi3 as an octoprint server for my 3D oeubter, including canera feed.
A friend uses his as a HomeAssistant server to monitor and manage his home lights, temperature and garden watering needs.
And I have one as a nifty remote KVM (TinyPilot) that we also use for work. Connect it to the Internet and a PC and you can control that machine remotely.
Lots of them are also going into commercial products, eg https://www.raspberrypi.com/news/production-and-supply-chain... from a time of tight supply said "We spend a lot of time on backlog management. We have to balance volume demand from commercial and industrial customers with the demand we see from individuals. Right now we feel the right thing to do is to prioritise commercial and industrial customers – the people who need Raspberry Pis to run their businesses – we’re acutely aware that people’s livelihoods are at stake."
The only one I'm using now is a Pi 3 (B?) with a DVB TV hat. I'm using it to stream broadcast TV to my tablet or phone. I can watch TV anywhere in my house. I very rarely turn on my TV now.
I have another PI in a drawer as backup. I replaced all of them with Odroid since it became nearly impossible to buy Pis. I wonder if they'll start to make them available again to normal people but I don't know if I'm trusting the company anyway.
I have a nice amp (Loxjie A30), but the bluetooth connection is both fidgety and subject to all the usual bluetooth incompetence (year 20 of the alpha test.)
I have a raspberry pi running a spotify connect implementation outputting audio to the amp via usb. It sits there and silently (or, well, not so silently) works; it's great!
There's some mucking with alsa to get it set up, but under 90 minutes.
One for Stratum 0 NTP server (using GSP module with PPS)
One for Home Assistant
One for monitoring the solar power at my cabin, using a 4G USB device to send data home.
The Home Assistant runs on a 4B, the others are using 3B's and a clone (NanoPi Zero).
Got some Pi 4's that I use for various Linux experimentation and such. Sometimes I find it easier to get a Pi 4 up and running than a VM, and other times I need the GPIOs.
As a controller for my lawn sprinklers that adjusts watering time based on historical rainfall data. Better than adjusting it every few months manually.
I used for some years a raspberry pi 2b as a media player with kodi. Then a bit as octoprint server to control a 3D printer.
I will explore a way to use a RP4 or now RP5 for a local NAS that would be extensible and that could run more services and yool that my current QNAP TS-28A but I'm not very confident I will follow
2. Web server (no external hosting hurray); TLS set up via LetsEncrypt.
3. Navidrome for streaming my music collection to my phone/computers. I ripped my thousands of CDs to MP3. I use subtracks on my phone for listening to it, and sonixd on my computers (Mac/PC). https://github.com/navidrome/navidrome
4. mpd for driving some speakers via a USB audio interface so I can use some speakers plugged into the Pi for listening to music in the same room as the Pi (practicing guitar). I control this via the Supersonic app on my phone.
5. Wireguard VPN so I can connect home.
6. PiHole for my network at home. Combined with Wireguard, it means my phone is permanently connected to my home network and gets ad-blocking and stops apps dialing home. I use DroidHole on my phone to see what's going on.
7. xrdp server running, so a usable desktop is always available.
8. miniDLNA running connected to a NAS box so that I can watch all my DVDs easily on my TV downstairs (I spent weeks ripping them).
9. Tuya IoT API for turning some smart plugs in the house on/off via cronjob; I do this instead of using the timer in their app because it means my phone can be off the network/abroad and these plugs/lights still turn on/off.
11. Peer Calls (https://github.com/peer-calls/peer-calls) so I can video conference in decent quality without having to use Google Meet / Teams etc. I also host a STUN and TURN server on the Pi so that Peer Calls works behind NAT.
12. Runs CUPS so that my very cheap Samsung wireless laser printer actually shows up as an AirPrint printer for my wife's iPad/iPhone and shows up in Android printing. (The printer does not natively have AirPrint capability but CUPS means I can provide it to users on the network).
It fetches time over NTP on a cron job. It also blocks various ASNs and IPs by country on a cron job to stop annoying remote pests and cloud providers. It also runs Monitorix so I can see system load, and goaccess on a cron job so that I can see who's hitting my website without having to resort to analytics nonsense.
It boots from USB3 (it has a NVMe in an IcyBox caddy).
It's my Wi-Fi hotspots, gateway, and network servers. I've got more powerful machines but it can route packets, serve http requests, manage dhcp, simple things like that without problem.
I use a USB 3.0 Ethernet cable for the second port. It can keep up just fine.
Mine's reading temperature, humidity and pressure from sensors and uploading it to a VPS. My second one that is, my first one hasn't done more than collect dust.
Absolutely, ESPs are just microcontrollers. A Pi is a fully functional System-On-Chip. That being said the VPN is the only usecase you mention where you couldn't use a flavor of ESP.
Funny thing is that modern 3G/4G/5G modems are full System-On-Chip computers themselves supporting virtualization. The Qualcomm Snapdragon MSM8916 is a popular SoC used for LTE dongles that runs the modem Baseband Real-time OS and Debian/Android OS on the same CPU.
The lifetime of an external USB drive is better than the SD card. My PINE64 after a few years stopped working due to the SD card giving up. It was a pain to recover my database from it but I did manage... that is why I am interested in this PI because of the PCI interface...
Home assistant instance with zigbee usb dongle, Volumio receiver connected to my speakers, screen for calendar, plant monitor, and for a while as a Kodi station.
So a Raspberry Pi 27W USB-C Power Supply and they also showing an active cooler. Seems like this thing is running even hotter than the previous version.
Btw am I the only one who hates to buy their power supply because the USB-C cable is not removable?
.. and their iPads (which is a better size for something like a Pi 4/5)
In fact, all of our USB-C power supplies around the house are Apple - Chromebooks, Thinkpads, etc. They don't break and the cables are detachable; whenever the junk that originally came with the device breaks, the Apple ones we have laying around are there to just work.
The small ones don't even have an Apple logo that you can see when plugged in, so if the idea of using Apple products bother you then go ahead and paint it, plug it in, and forget about it for a decade.
I reckon it must be really hard to keep the raspberry pi relevant. Theres so much competition in this space and the user support/documentation gap must be closing now which has always been the pi’s edge.
It does not appear to be common, or at least, it's not in the spec. Per USB-C spec, chargers only have to supply 3A at various voltages, with 5A used only for the 100W 20V profile.
If you buy a charger advertised to support "PPS" at 65W or so or more, you are more likely to get one that does this, though not all will actually do it.
> When using a standard 5V, 3A (15W) USB-C power adapter with Raspberry Pi 5, by default we must limit downstream USB current to 600mA to ensure that we have sufficient margin to support these workloads.
Therefore 5V/3A can still boot RPi5, but it cannot provide enough power to its USB-A port. Therefore it cannot use something like 12V/2A, since RPi5 doesn't have transformer circuit on its board.
They're not, and it's a very annoying requirement. Even 5V/3A was hard to provide reliably. On the other hand, rock pi supports 12V/2A, which modern chargers will provide without any issue.
Oof, this sucks. I wish they supported the wide variety of Anker/Apple/etc. USB-PD chargers out there that can deliver 100W so I can use my existing power bricks. None of them do 5V5A though, they just created yet another nonstandard piece of "USB-C" equipment and it's going to suck.
25W is easy, but most of them will not do it at 5V. I checked one of my 100W chargers, it'll only do 5A @ 20V. 3A at every other voltage. Looking at many other chargers on Amazon do the same.
Most 65W chargers are 20V/3A. Standard USB cables are built to carry at most 2-3A, and e-marker cable authentication chips are required for 5A, I believe, because people running 100W on a random USB cable is scary.
Chargers that would do 5V/5A probably do PD 20V/5A as well, = 100W. Or it'll be a non-conforming special Pi PSU of sorts.
Unfortunately that's not how USB PD works. In PD Land, 25W means 9V/2.8A. A charger is allowed to offer 5V/5A but at no power level is it mandatory, so even a fully USB PD compliant 100W charger which can do 20V/5A is not guaranteed to power the Pi 5.
I just did a quick review on Amazon of adapters ranging from 30 to 120W. Not a single one claimed to support 5V/5A DC. Only a couple claimed 5V/4.5A and they were expensive. 5V/3A is common in the $15+ category.
If the PCI-e bridge makes PI-powered home-NAS enclosures popular it'll probably squeeze the low-end networked NAS market quite a bit since they're usually powered by less powerful hardware (Arm/Celeron chips with 1-2 gb of mem).
Been looking at some compact/silent replacement for my aging home-server and NAS boxes were at the top of the list but the CPUs were crap or prices high, and everyone was shipping their own weird linux-based distros. Knowing that it'll be likely that software will work on a Raspberry and these hardware specs then it looks like a contender now.
I would love to see a more side by side comparison of specs. 2-3x times the speed pretty good though. The raspi 4 was just on the cusp of being usable as a main computer, I wonder how good this will be.
I've used one in person. I would say that it has crossed the threashold. I've only played w/ it using a microSD card. Whilst the microSD card slot now has support for high-speed SDR104 mode (roughly twice as fast as the Pi4), I can only imagine that speeds will increase when connecting SSDs via PCIe.
They like to use JS benchmarking to do this. I forgot which exact benchmark they usesed (sorry)! The RPI4 was benchmarking ~50, the PI5 was scoring about ~130. They were saying the RPI5 was scoring equivalent to a 2015 MacBook Air.
> raspi 4 was just on the cusp of being usable as a main computer
I use a Pi 400 (running Debian) as my primary dev box and access terminal to the Internet. ^_^
It definitely works for me, but I fear that you will next tell us that you need to run software that isn't compiled for ARM (e.g. Adobe graphics suite) or to connect some specific hardware, such as a centrifuge that works only with Windows drivers.
Phoronix tested the Pi 5 with and without the official heatsink, and without it peaked at around 90C (same temp my Intel Mac Mini from 2018 likes to hover around) and never throttled during benchmarks. You don't need to downclock, undervolt, or turn anything off to run fanless.
They say that the silicon was "designed in-house" for the best possible performance. What does that mean, exactly? Does it have semi-custom ARM cores? That seems unlikely to me.
Its not the main SOC that was designed in house, it is an I/O processor. Which still sounds great. There is much more I/O throughput which has plenty of value for me (robotics).
Does the Pi 4 have hardware video decoding in mainline Linux yet? I gave up on Raspberry stuff after buying a 4 and finding out that the mainline support was so much worse than expected.
It seems the answer is 'yes', especially for formats like h.264. Notably, they discontinued HW support in the RPI5 and chose to process it through software instead. This approach saves money since they don't need to pay for the intellectual property rights in the System on Chip (SoC). Additionally, the BCM2712 handles the video processing w/out breaking a sweat.
This is exciting. I wish this would work without a fan, just with a good cooling system.
I say this because I love the idea of a full linux computer without a fan. I'm loving the Apple silicon machines at the moment for this. Perfect silent computing.
I wish it came as well in a larger board, with proper HDMI ports. There is a market for silent linux machines, I think!
The other day, I found myself in the interesting state of needing an x86 binary, but not having any x86 machines readily available. My (i)phone, laptop - apple silicon macbook, home server - raspberry pi, all run ARM! Previously, that would have been limited to just my phone, but, ah, time marches on
I'd like to see an RPi that includes an Arduino (or maybe an ESP32) on board, with some way to easily connect things to that Arduino's GPIO pins, and with the Arduino programmable and controllable from the RPi.
This would let you run things that need real time GPIO access and real time interrupt handlings on the Arduino, and would also give you DAC and ADC.
I asked Eben about that in an interview earlier this year, and the answer is "probably not too soon" — RISC-V has a lot of designs available, but most have been in the lower-end 'efficiency' class compared to cores like A76 (which the Pi 5 uses) and Neoverse, which are a lot faster.
That sounds really exciting! Do you happen to know if such boards will be competitive with the RPi5 eg quad-core etc. I love the idea of RISC-V resolving some of the drivers dilemma that SBCs seem to face. I recall some talk of extensions that might make it viable as a GPU core one day. Not to rain on the RPi5 parade of course, i'm sure it's a fantastic bit of kit :)
To be fair, the analog output of the RPi delivers a crappy low-pass filtered 11-bit PWM signal. Even the cheapest USB audio adapter will give you a better signal. (Personally, I have been quite happy with https://www.pollin.at/p/logilink-usb-2-0-5-1-audiocontroller...)
Yeah, you're not wrong. I more disappointment with the general trend rather than this specific instance; phones and tablets are where I miss it the most.
With Raspberry Pi's specifically, although I have used the headphones jack to connect speakers, I generally use either HDMI audio, or else I just don't have any audio connected.
Nice to see a new generation but frankly a little underwhelmed. I’d rather spend more for an orange pi plus. Different price class but worth it here. Integrated M2, 16gb ram, newer processor node, and much higher scores on benchmarks
And it seemed to sort of work but not quite enough. There were stutters in the content streamed and strange clicks in the audio. Will this new one be enough? Was that even the likely limiting factor?
Damn, I was hoping the old PoE hats would continue to work with this one.
Looks like the PoE pins have moved around. Another 20+ USD on top of the board cost...
Kinda disappointing, but the upgraded SD speeds will be nice.
Was also hoping they'd boost the core count or memory for some models, I know it's kinda against their targeted group, but would have been nice to have the options
I'd love it if SBC manufacturers would announce their software support along with new hardware specs. I'm using a Pi 4 in a CNC plasma cutter solely because they have the only working realtime Linux PREEMPT_RT build I could find. I would love to use the Odroid N2+ but their ancient realtime branch won't even boot.
I don't see any good reason to buy raspberry pi over something like this mini pc[1] anymore. For $100ish, you get 8g RAM/256g nvme/intel N95. Not mention the wonderful software support.
For me the advantage of Raspberry pi was passive cooling without worrying about any moving objects and running 24/7. Being affordable was another clear advantage. These two does not apply here anymore.
I might be wrong but buying a used mini pc like Nuc or SFF PCs seems to be more reasonable than buying a 80$ SBC + Case and adapter costs.
>Will my Raspberry Pi 5 fit my Raspberry Pi 4 case?
>Raspberry Pi 5 won't fit the Raspberry Pi 4 Case. We recommend the Raspberry Pi Case for Raspberry Pi 5, designed to help you get the most out of our newest computer.
What would someone excited about the improved specs on the 5 be using a Pi for? I have 1-2 of each gen, but to me, they aren't devices I would use if I needed grunt. They're just solid little low power, low heat devices for running odds and ends. Curious what people who are pushing them hard are using them for.
I used a Pi 4 as a CI server for a previous project. While speed wasn’t critical (it was fast enough for me to not care save for one rarer multi-hour job that was usually cached) I certainly would’ve welcomed the improvements the Pi 5 is bringing.
...I'd like to point put that the Pi 4 without any overclocking basically replaced the first non 486 my gamily owned as a family computer.
The Pi 5 will end up replacing my first gaming spec'd E-Machine.
Considering the wide variety of use cases I got out of both of those computers growing up, it takes very little imagination imo to figure out what to do with them.
Even on e I get around to actually getting an x64 server spec'd machine for actually doing grunt work, I figure these little Pi's will be perfect for little oneoffs like mirroring repos I care about. Constrained dev boxes, network jiggery pokery boxes. Hell, might even get around to trying a non traditional form factor builds one of these days
I don't know why people would select a RPi for something that needed raw CPU speed, but the new IO chip is a winner for me. The IO has sucked on all Raspberry Pis to date. I tried to use a Pi 4 connected to a USB3 HDD as a wireless NAS and it was abysmally slow, despite the CPU never breaking a sweat.
Rasbperry Pis are losing their competitive edge for non-embedded applications to sub $150 mini PCs, which support Linux/Windows and will come with enclosure, NVMe slot, power button etc. The mini PCs are also generally much faster.
That's true as long as you don't mind using modified vendor supplied operating systems. Most of these advanced boards will lock you into some patched Ubuntu version.
So far from the ones I've worked only the Pi and a patched Rockpro64 allows you to boot some generic USB installer from UEFI and have the system your way with mainline kernel support.
I agree with you about dev boards but I think their post was focused on things like used NUCs or other small x86 machines that are only marginally larger than a Pi in an enclosure and include literally everything you'd want (enclosure, pwa fan, SATA SSD, NVMe, wifi, etc.) and the software support is there (linux / windows).
Intel's finally getting competitive in this market with the 12th/13th gen chips, because of the heterogeneous efficiency/performance core layout. That said, I pine for an ARM SoC or SBC that has a larger core count, and doesn't have embedded RAM packages, or at least offers at least 32 GB of RAM.
I didn't see any mention of UEFI in my first read through about the Pi 5. Did I just miss it?
I would like to have a quiet and reasonably performant ARM aarch64 box at this price point, but only if it supports UEFI without needing to resort to silly EFI system partition tricks (which the Pi 4 required, last I knew).
The RPi4 has had a functional port of TianoCore for a while now, which is likely what they were referring to. You can use e.g. generic aarch64 UEFI Fedora images out of the box with it.
RasPi 4 requires some special files in the EFI system partition and a special non-stock EEPROM, at least for the older RasPi 4 boards, maybe the newer stock come with the UEFI-enabled EEPROM image? Regardless, in my experience with other EFI-enabled boards, I've never had to put the actual EFI into the EFI system partition, the EFI itself lives in a SPI flash usually so that you can boot "from" EFI with no "disks" attached.
Definitely the RasPi 4 having the ability to run UEFI, even in this way, is good. But what I'm looking for specifically is having UEFI just like every other UEFI-enabled PC and server does it, where I don't need a disk to boot to UEFI.
One of the differences in the Raspberry Pi 5 is the bootloader is in EEPROM. It has 2MiBytes which they doubt would be enough to fit a UEFI implementation. So it sounds to me like there isn’t much chance of UEFI this time.
2MiB could be enough space for an EFI implementation, but some features definitely would need to be disabled and it would definitely depend on how big the code for interfacing to the specific hardware on the board actually is. 4MiB is often big enough to have a functional EFI implementation, so 2MiB doesn't sound crazy to me, but I haven't tried.
hardware consistency and gpio are pretty big factors.
rpi as a compute/media node has always sort of been secondary compared to the software/hat ecosystem -- there have essentially always been better SBC choices for those workloads/media-center-ing.
personally if I wanted a media-center/nas/small-server i'd go with one of the dozens of SBCs that have onboard sata and save myself the hassle of doing it all over an overloaded usb hub.
GPIO is doable on a mini PC. For example, there are Raspberry Pi Pico firmwares out there that present as a normal gpiochip. I recently set up a pico with 3D printer firmware (Klipper, specifically) so I could control power to my actual printer, which actually works surprisingly well.
Where can I find these mini PCs? I am trying to find hardware for an upcoming project and I’m considering the Pi but would love to see some alternatives.
I’ve purchased a few HP Elitedesk 800 G3 Minis (refurbished) for about $125 on Amazon. They even come with Windows 10 if you’re looking for that. They’re a deal considering the Pi’s extra expenses like power, case, storage, etc.
The Dell wyse 5070 idles close to 3-5 watts. A bit more than the pi, but pretty close. It's older now, can be picked up cheaper than a pi off eBay. DDR 4, sata m.2, on board emmc flash. M.2 a-key for WiFi. Max 16GB ram.
When you throw in m.2, lots of ram, all that IO eats power. So the minimum idle power creeps up. Arm boards too.
There is also a general improvement in code support by sticking with x86. I bought a similar 10 year old i5 based workstation dell and so many little things just work _better_. I didn't realize just how much of my battles were based on hardware architecture issues.
I love the principle of the mini elitedesks, but some have a terribly annoying fan. It's not firmly attached to the chassis, so it rattles while spinning (and it spins all the time). The heatsink uses a proprietary 3 point attachment, so you can't use a big aftermarket cooler instead (even though it wouldn't fit in the case, I would have accepted the compromise).
Not sure about the <=G3 and >=G7, but the G4 and G5 have the issue. The G6 seems to have the fan attached more firmly, but I've never tested one in a quiet room.
Hp Elitedesk and prodesk are the two 1L available models they have you want to hunt for.
Lenovo has M700, M75q (AMD) and a bunch others ranging from thin client to workstation performance.
Generally new ones are awesome at about $700 but older ones are absolutely capable for upgrading ram, disk, wifi whatever. there are modules for up to 10g and other things too. Servethehome on YouTube has a bunch of guides.
I did something similar too, just beware of refurbishers putting cheap/trash SSD disks in those machine, they can stop working all of a sudden sooner than later (I experienced that on my skin)
I just stalk eBay's Buy It Now. Got an i7-7700T w/32GB RAM and a 256GB NVMe for $163 shipped+tax. Previously got an i7-6700T with 1 stick of 16GB RAM so it was cheaper to match that for 32GB total.
If you can get friendly with local electronics recyclers or auction buyers you might do way better.
Agreed. Or if one lives near large research universities, they often sell off the equipment they no longer need for a solid deal. Sometimes they do sell computers, servers, monitors and such by the pallet, however, and I’m not sure anyone really needs a cluster that large at home haha
I've been fairly impressed by a couple of Beelinks Mini S12 I purchased a couple of weeks back. Intel N95 CPU, 16 GB RAM, 500GB NVMe. More than enough for checking email.
One was immediately wiped and I installed Debian 12. The other was wiped and I installed Windows 10 Pro. Both seemed to just work.
Main home server/NAS is a Ryzen 7 Beelink with 64GB. Works a treat. Fan is audible when you're hammering the CPU (doing a large backup, unpacking downloaded media, running a Minecraft server with multiple users) but other than that it's not audible from ~5ft away.
This is running a 5257U which is an ancient 5th gen mobile chip and even with the Iris graphics it got (much better than the usual Intel HD) only HDMI 1.4 is supported.
I'm always on the lookout for the beefier chromeboxes, they're pretty small physically and pack a lot of punch for such a little box. Asus chromebox 3/i7 is my favorite so far.
Yes! I picked up a 4GB Acer chromebox with a celeron for $20. This is much better value to me than an RPi because I don't have to worry about ARM. Obviously RPi serves a different use case of tinkering with embedded dev, but tons of people are trying to put docker on these things...
Current Gen Intel are M100… N200 and N305 ... search for that plus mini PC. Older gens will have less expensive models around.
I used several while rpi supplies slowed and scalping had prices where it was the better option. Around $200-250 USD or so. But comes with memory, storage, case and power supply.
I do think these make more sense for most people. But I'll for sure be getting a Pi 5. I sold all my Pi 3 and 4 during the pandemic since I wasn't using them.
I don’t think that’s the only reason to use one. They also have value as a well known and well supported hardware target even in cases where the GPIOs aren’t needed. For example, the hobby 3D printer community uses them extensively for things like Octoprint where the GPIOs aren’t needed - but the Pi continues to be much more popular than other options because it’s very well supported. A user doesn’t need to figure out any CLI magic to get things to work, they can just grab a pre-made SD card image and go to town. And if they do have a problem, it’ll probably be the same problem that a few thousand other people have had, so solutions will be well documented.
> They also have value as a well known and well supported hardware target even in cases where the GPIOs aren’t needed
true, there are cases, but "well known" and "well supported" is the point i tried to make in favor of enterprise thin-clients and the like as they are pc's and run most commodity operating systems and applications.
> hobby 3D printer community uses them extensively
5v/5A seems weird. Can normal power supplies even deliver this? One of the big benefits of rpi has always been to reuse power adapters and isn’t that all PD standard today?
Higher wattage PD chargers and powerbanks that I have here all work with voltages of 9/12/15 at 3A and 20V at up to 5A
No, it is not part of the USB PD specifications, and the Pi 5 is therefore not PD compliant.
The PD specs require a 25W device to operate at 9V 2.8A. A 25W charger is still allowed to offer 5V 5A and a 25W device is still allowed to prefer it when offered, but it is not allowed to require it.
To make it even worse: at no wattage are chargers required to offer 5V 5A. Even a 100W one (which is required to offer 20V 5A) does not need to offer it, and it is 100% PD compliant if it only offers [5V 3A, 9V 3A, 15V 3A, 20V 5A].
as someone who did not upgrade for like a decade, now i feel like i can get on in early and benefit from leaps and bounds of an upgrade!
however, this offering is not really aligned with the foundation started out with imo. their decisions with the changes does not seem to satisfy the answer to "how will this promote computer science among kids?" for the most part.
i feel like the original mission is now relegated to the pi zero w, but it requires soldering for working with gpio so it leaves a lot of room for the hardware side of tinkering.
i am also curious about the m.2 hat and whether it can work with old sata ssd that match the format? otherwise old external hard drives might be a good alternative for relatively reliable long-term operation for now.
When the 4 came out, lots of people wished that power + video would be provided over the same port, just like we expect with almost every modern laptop. Does that exist on the Raspberry Pi 5? Hopefully I'm wrong, but it doesn't seem like it, which would be a shame.
From what I can tell, this is the first version without a composite video output, correct?
It's a bit of a shame. What are the alternatives for someone who needs a computer connected to an old CRT TV? Are there other boards with composite video or does one need to get a HDMI->composite converter?
> We’ve removed the four-pole composite video and analogue audio jack from the board. Composite video, now generated by RP1, is still available from a pair of 0.1”-spaced pads on the bottom edge of the board.
They need to deal with manufacturing and availability too. I wanted to buy a couple of RPIs to help schools across LATAM and it had been a nightmare from pricing to shipping. I'm exploring other rpi alternatives since it feels like they don't care anymore.
I somehow feel RPi5 is getting out of my "small fanless SBC" comfort zone, for embedded projects and products, RPi4 is more than enough for my needs, though I like its RTC and power button.
I'd like to know if the Pi 4 will still be manufactured. It's good enough for my projects and I'm not liking the extra cooling that sounds like will be needed.
Nice. Now, if I could just get an M.2 board for an Intel Edison (I have a few DOS games I want to emulate, but the Pi4 still isn't quite powerful enough for).
Only OpenGL ES is disappointing. ARM's Mali Panfrost drivers get full desktop OpenGL, so do Snapdragon GPUs. Hopefully they implement enough of Vulkan for Zink.
Only quad core, no 16gb or 32gb ram option and pcie only available through a hat (and possibly slower than the usb3 port)? I was expecting a bit more to be honest.
from the product brief: “Raspberry Pi 5 will remain in production until at least January 2035”
I would have guessed that as they move up the performance + power curve, such a long production commitment is less interesting for most users, but maybe I’m wrong.
I wanted to also include their “increased release cadence” - but apparently it’s been 4+ years since the Pi 4 was released, it just doesn’t feel like it to me due to the chip shortage and pandemic…
I'm getting close to line speed with a USB 3.0 Ethernet adapter plugged into a laptop, so the adapter itself shouldn't be the bottleneck. I get 109KB/s ftp transfers to a NVMe drive.
The RP1 chip has some parts of the RP2040, though details are a little scarce. There are apparently some things you can do on RP1 when the main SoC is powered off, which leads me to believe outside of the main IO, you could do some interesting things with GPIO and/or CSI/DSI ports while the Pi is powered down (or in some sort of sleep-like state).
I remember them leaving out some resistors for the USB C port that were needed for some chargers or cables. Wasn't too big a deal if you just used their official power supplies.
The rp2040 was their first foray into custom silicon and has been completely unaffected by supply chain issues. In that case they ordered 10 million chip dies (dice?), I guess it might be similar here.
The VideoCore VII GPU does not have its own VRAM, it's only a compute chip. From the Hackster.io review[1]:
> The chip also features a new Broadcom Videocore-VII graphics processor running at up to 800MHz, and is linked to the buyer's choice of 4GB or 8GB of LPDDR4x memory.
I wonder how does this in CPU performance compare to the new Jetson Orin Nano?
I’d love to find something low power that can be a nice little home server to host various projects. Mac Minis might be the best perf/watt but they’re still coming in pretty expensive in comparison.
At this point, Alder Lakes like N100 processors and multiple sff computers have dominated the market while RPI struggled with the supply chain. now there are multiple clones as well. I’m not sure if The Pi5 will ever reach former glory days.
I suspect the speculator and reseller markets are already abuzz with figuring out how to stockpile and control the distribution like they did during the pandemic and various video game releases. This ought to be fun to watch.
> [...] This is based on the most recent release of Debian (and its derivative Raspbian), codenamed “Bookworm”, and incorporates numerous enhancements, notably the transition from X11 to the Wayfire Wayland compositor on Raspberry Pi 4 and 5.
Price is actually secondary at this point. The primary advantage of the RPi over a lot of other SBCs is how widespread its use is, and all of the addons you can buy for it.
... They really love being in charge of the cabal that sells HDMI dongles, don't they? Mini/microHDMI are terrible for video and just not mechanically robust. :\
It only feels like recently that you can pick up a RPi 4 used for the price they should've been new. Otherwise, scalped listings on eBay or Amazon have took over. I still want my first RPi but I already have associated them with being hard to buy and way more expensive than makes sense.
I'm happy that they're upgrading the pi, but it's not exactly lightweight computing in terms of power (they recommend a 27W PSU) or performance/watt (see: your cell phone)
To me the charm of the raspi ecosystem was always in giving you a very low power, "always on, almost no power draw" linux environment. Oh well, I guess there's always the older versions or the zero for that.
"Raspberry Pi 5 consumes significantly less power, and runs significantly cooler, than Raspberry Pi 4 when running an identical workload. However, the much higher performance ceiling means that for the most intensive workloads, and in particular for pathological “power virus” workloads, peak power consumption increases to around 12W, versus 8W for Raspberry Pi 4."
Alright so 27W with heavy peripheral draw then? But 12 W still means it'll be drawing 2.4 amps instead of 1.6 when a random process decides to 100% it. I guess we can always underclock it.
The real unanswered question is, does it finally have a damn sleep mode so it can save power when idle.
A smartphone has two entirely separate computers in it, the one that runs Android or iOS, and the baseband controller. The latter is built to be very power-efficient while waiting for a radio signal. It will wake up the dormant application OS computer when a push notification comes in.
The same likely can be built on top of an RPi, using, say, wake-on-LAN signaling, or some GPIO as an interrupt source. You'll have to suspend your OS while idle for a prolonged time though.
# Disable Power LED after boot
dtparam=pwr_led_activelow=off
# Disable SD card activity led
dtparam=act_led_trigger=none
dtparam=act_led_activelow=off
# Disable the ethernet LEDs - these are Pi4 specific values.
# Look in the docs for the values for other Pi boards.
dtparam=eth_led0=4
dtparam=eth_led1=4
> The real unanswered question is, does it finally have a damn sleep mode so it can save power when idle.
I dream of DIY'ing a laptop, and RPi looks like a great platform for prototyping that... until you realize there's just no sensible way to put it to sleep. Hibernate + aggressive boot time optimization?
You know it's funny, my current laptop has a weird firmware bug (certified Asus moment) where it refuses to go into sleep mode. If sleep is triggered by any OS it will straight up just shut down completely. I've lost some work the first few times I used it out of habit.
Eventually though it didn't turn out to be much of an issue, SSDs boot real fast these days anyway and I can just do a full power cycle.
I read that the "PC style" power button allows both soft and hard shutdown, so I suppose suspend and hibernate come into the picture as well. It's mostly a matter of OS support, the firmware seems to be there.
Is there a way to limit the processor power so that it never exceeds a certain threshold? Because some people might be okay with slower processor if it means less power usage.
I read a lot of Jeff's blog, but his vlogs put me to sleep. He isn't charismatic and makes the worst jokes. People forget that a visual medium is visual. He channel is an example of great information that is ruined by the visual medium.
The new board supports more power hungry IO. If you want to use all of the IO available and push the new SoC then you're going to need a beefier power supply.
I don't think the newer models are supposed to completely supersede the older models. They are still available, and they still make them. If you have low power needs, you can use those.
I have the above, it's max powerdraw is <5 watts, even at 50% cpu its <3watts. That comes with ram, SSD, case and motherboard.
so to get the pi5 to the same state would need an SD card (boo hiss poor speed.)
From what I've seen the pi5 is 1/3rd faster than the intel j5005. (in pybench at least.)
but comparing to a real intel NUC, of course its going to be faster, the NUC costs an entire order of magnitude more.
I bought an N100 for 160 USD including an SSD off of
AliExpress. Thats 2x the cost of an rpi4. I benchmarked it and cpu perf is roughly 2x the rpi however i/o was easily 10x without any sdcard or USB shenanigans. Bonus because it's an Intel chip I can use the regular x86-64 os builds instead of some goofy fork. That's very compelling if you don't need a gpio or any of the raspberry pi accessories for your use case.
I picked up a Beelink because the Raspberry Pi 4 was unavailable. I use it as a Linux desktop (next to my M1 Air) and a Jellyfin media server. It has replaced the Raspberry Pi for some purposes, but the Pi still has a place when hacking on hardware due to the GPIO.
Yeah, still happy I went with the Odroid M1 for a small home media server. It’s not fast, but it does everything I need out of the box, has a very nice aluminum case, and doesn’t use much power.
eMMC isn't that reliable. It's basically an SD card on a chip.
Difference is they're more optimised for random writes than large files like most SD cards, but you can get such SD cards too ("High Endurance" models)
Exactly the same as every other hardware supplier. For pi there was some kind of scheme for businesses who _actually_ needed the hardware in order to stay in business. No idea how that worked.
Either way, this is all history. Raspberry Pi boards do not cost hundreds of dollars as claimed.
I am so grateful to the folks behind that site. It's the only reason I was able to track down units when I needed to buy them.
you not checking the site != people not being able to buy them
I'm not saying that it was easy, or anything like business as usual. They've done interviews where they talked about the difficult decision to prioritize companies that would go under without new stock over casual hobbyists. It's one of those unenviable situations where there's no good outcome, just a possibly less-bad one.
Only if you are looking for a latest-gen or last gen machine. You can find some old NUCs for cheap, and there are lots of mini PCs or thin clients for around ~100. Yes about twice as expensive but more than twice as powerful (in both processor speed and features). Of course only if you plan to use it as a Linux computer, not for GPIO stuff.
This isn't super-practical for a commercial application which requires 70 identical machines.
Not only are they expensive and relatively large, machines that have had previous owners often have mystery issues which make them great for home tinkering projects, less so for something that can get you in trouble if it breaks down.
There's a reason companies buy new parts instead of employing teams to scour Craigslist for deals.
Faster processor, 16GB RAM, 500GB NVME SSD, with case. $165! That’s damn impressive, considering the RPi5 with 8GB RAM is suggested to sell for $80 (good luck getting it for that little). And Amazon can get it to me in two days.
Yeah it’s definitely bigger, but I wasn’t expecting these systems to be so cheap.
Yes, if I look at those new Intel N100 based mini PCs with a 15W TDP and their power / price / consumption ratio it would seems Intel took notice of the threat and reacted accordingly.
Especially now that the RPI kind of need a fan and you need to buy the power supply, the storage and the case.
Well the RPI has GPIO but for small home server use case nobody cares...
GPIO can be added via tons of USB to GPIO boards out there.
But I agree - lot of people tend to buy the Raspberry Pis for home servers instead of just opting for used mini-PCs from secondary markets. Even a 7-8 year old Intel CPU in those mini PCs will vastly outperform a Raspberry pi. Even the Raspberry 5. Plus, better I/O options and storage with mini PC.
Most of the stuff just flashes LEDs or reads a switch. Doesn't matter for that. And for more intelligent stuff there's i2c or api which have their own interface boards.
There was at least one NUC with a full GPIO header set. The DE3815TYKHE.
I've done motor control, I2C and sensor IO with it and it was rock solid on that little NUC.
There was also the UP Board which was an Atom SOC with PI compatible GPIO. I believe that's still in production?!
If you want the GPIO, that's a good reason to go Pi. Nothing equal to the software support inside the raspberry pi eco system for it's embedded controls.
If you want a small PC for media/homelab server/cheap desktop, they don't make any real sense anymore.
It solves it, as much as adding a trailer to your sports car solves your problem of not having a pickup truck. It works, but it's very inconvenient.
Eventually those boards run into limitations and then you have to just opt to go with serial to a microcontroller which misses the entire point of having a SBC.
I agree that its not the ideal solution. Still would be better to use a SBC like the Raspberry Pi or other boards. But was just pointing out that if GPIO needed to interface with slower hardware like relay boards or sensors , then there is an option.
It's great for my farming robot though. We have plenty of power but want something small and reliable, and the Raspberry Pi Compute Module works wonders for this. More power will be welcome once they release a Compute Module version of the 5.
Is there a reason you're not interested in those NVIDIA Jetson modules? Even excluding the GPU compute, I had thought the Orin boards were significantly faster than a Pi 4.
So for what we’re using it for, which is handling Wi-Fi communications to a central server, handling CAN bus communications to four motor controllers, dealing with RTK GPS and calculating path following parameters for GPS paths, the raspberry pi is perfect. They are cheap, have excellent documentation, and until recently availability was no issue.
Jetson boards are much more expensive. We do plan on making a computer vision add on for our robot which uses the Orin or similar to process images, but I find the raspberry pi perfect for sort of stitching the whole system together.
Would be interesting to see benchmarks of different PI generations at the same power level, assuming the older ones have dynamic power mgmt features too.
I pulled up the actual USB power delivery standard[1] USB_PD_R3_1 V1.8 2023-04.pdf, and 5V 5A is perfectly compliant, albeit optional.
On page 805 you find Table 10-2 SPR Normative Voltages and Minimum Currents, which specifies that a USB PD source with a rating of 15 < x ≤ 27 watts *shall* support 3A at 5V, however it *May* advertise up to RoundUp (PDP/Voltage) to the nearest 10mA. Requires a 5A cable if over 3A is advertised. 27W/5 rounded to the nearest 10mA comes to 5.4A
The problem here is that a power supply offering 5V 5A is compliant, but a device requiring a 5V 5A source is not.
If a device needs 25W it is required to accept 9V 2.8A, if it wants to be PD compliant. This is precisely because 5V 5A is optional for a 25W source, so a device cannot rely on it being present.
It’s worth noting that this is _allowed_ by the USB-PD spec, but you’re right that 5V @ 5A is not common because it is not required. At least it’s better than the Pi 4 was at launch? :D
It doesn't need it, on the blog they say that if you use a standard 15W USB supply, it (by default) limits the board's USB output current so it can't cause a power failure at 100% CPU.
Oh, I see, so 100% CPU with no USB devices can run fine on 5V3A? That's nice to know. I could just use a powered hub for USB devices then.
I wish they had an auxilary DC power connector of some sort though that could just power the Pi on straight up + and - from a DC power supply. Ideally anywhere from 5-12V.
> When using a standard 5V, 3A (15W) USB-C power adapter with Raspberry Pi 5, by default we must limit downstream USB current to 600mA to ensure that we have sufficient margin to support these workloads. This is lower than the 1.2A limit on Raspberry Pi 4, though generally still sufficient to drive mice, keyboards, and other low‑power peripherals.
This is also very uncool, since powering it through the GPIO header with a capable PSU won't trigger the PD signal and makes it impossible to draw any meaningful current through USB? I hope this blockade can be worked around in the boot config.
I mean theoretically dupont connectors are rated for 2.5 A so two of them could get it done, but the current will never be completely equal across both and one may get overloaded. Maybe the safest option would be to find some kind of barrel-jack-to-usb-c dongle.
In practice I doubt it would be a real problem since you'd need to max out both the USB draw and CPU load at the exact same time to get the full draw.
At 27W that is a strong pass vs a NUC. If I go with a low power x86 NUC, I can run all software on it, and probably insert a SSD + additional RAM, for probably ~15W.
That'll be peak power draw though: you have to spec the standard power supply to cope with the CPU running @ 100% & simultaneously feeding the 1G ethernet and all the USB ports at max power.
Jeff Geerling measured 11W peak board power running benchmarks. Idle power was measured at 1.8W.
They claim it will draw less power than the Pi4 for the same workloads.
Won’t you still need an NTP or a GPS time module unless the RTC has a battery backup?
How many applications would you have where timing is critical but you won’t have an external time reference that would be still viable without a guaranteed power backup?
Nobody said anything about not having NTP in the loop!
Here's the deal: RPi<5 runs at the speed of the voltage coming in. The OS creates a fake software clock that assumes it's a perfect 5v, but in reality not only does voltage fluctuate, it drops over the length of a cable run. If you have, I dunno, 70 Pis connected, the 5v on the first one is very different than the 5v on the last one, no matter how frequently you ask NTP to broadcast.
The first thing we did was deliver 12V down the wire with a high quality power supply. Then we used UBEC buck converter modules (intended for RC quadcopters) to deliver power directly into the Pi via its pins.
However, even now you're just chasing physics. There's no way to send a command to all of the Pis to do something right now. Instead, we tell all of them to do something at a very precise moment in about three seconds. In our case, this meant writing a very low-level C program to listen for raw UDP messages and trigger the camera module at the hardware register level.
The only way to verify that it was working as well as we believed it did was to create a reference clock. We had a friend build a 20 LED timer that would count 10ms and 1ms increments.
We achieved < 1ms syncronization across 70 Pis in this manner.
We needed it to be small, low power, low heat, provide hardware level access to a camera module, be ethernet accessible. Oh, and under $100 including a fixed M-mount lens.
At the time, it didn't seem like rolling our own hardware would have been faster.
I would argue - given we succeeded at sub 1ms syncronization - that the Pi was close to the perfect tool.
It just would have been even more perfect if it had an RTC.
You already solved your issue, but still I'll mention it:
For my Pi's I buy a separate RTC (uses GPIO pins). There's various available, the DS3231 is the most expensive but also the best. If you buy it in bulk (70+) I suppose you could shave some money off. If you're OK with a cheaper one (less than half the price) which functions less well (but still does the job) then PCF8523. I got mine from a local Adafruit reseller.
FWIW, the cheapest RTC module we could find that had kernel support was about $31 at the time. Things have gotten a lot better and cheaper since then. Adafruit didn't launch that DS3231 until four years after we did this.
We couldn't justify adding ~$2500 to the bottom line, but you can be confident that we sure did argue about it.
Missed this comment when asking why you didn't add a $2 RTC, but given that you mention Adafruit I assume making your own hardware to stick onto a Pi wasn't an option? The DS3231 you mention is a chip with mainline support since 2009. Cheaper options like a DS1307 cost less than a $0.50 a piece when you buy 100 of them, and they've been supported in Linux since 2006.
I'd say 50/50 that I didn't know what I2C was in 2014.
Still, if we had gone and designed our own RTC daughterboard, it would have replaced the UBEC in our architecture diagram. We were buying those for $3-4 in 1000 unit quantities.
So, had you been on hand a decade ago, you might have won the argument although if I put on my board member hat, fabbing our own PCB sounds a lot scarier than buying boxes of buck converters.
It's hard to express how wild it is that they just pop right on and guarantee close to perfect voltage to each device.
Fair enough! I'm not sure I caught that the project was from 2014.
From your URL, the second part made me wonder if the statement is correct: "UBEC 5V 3A/5A/7A/15A BEC 2-12S Lipo Step-down Module External Power Supply Full Shielding Antijamming For FPV Airplane" would this actually work?
This is an armchair hindsight, but sounds like that needed a way to broadcast trigger GPIO interrupts, which are like JavaScript event handler for hardware.
You can attach a function to a pin, and CPU force-jump into a preprogrammed address for a handler routine, which leads into your code, then return(return(return())) back to where it came from. The trigger can be delivered by, say, a Xenon flash pointed at the ceiling and a light sensor, or a dedicated pair of wires from a master Arduino, or a intentional glitch in power line, a drone receiver per each pi all on a same channel, etc.
> fixed M-mount lens
I assume this is meant to say metric threaded C/CS mount for surveillance cameras, as M-mount is boutique full-frame Leica mount.
Well, if you absolutely needed to use that specific rpi camera module (if I understood your other comment correctly) and you didn't have time to reverse engineer it, then rpi was probably the best option, sure. It just feels weird to assume that, or that every single module needed to be connected by ethernet. Rpi and low power in the same sentence also feels weird. I'd have tried to build little devices out of STM32. But again, I barely know anything about your project.
Ethernet was handy for moving the images off the Pis once they were captured.
The RP2 was indeed low powered compared to the Intel/AMD powered systems we were otherwise familiar with.
Outside of the bubble of people with EE skills, however big the bubble is, stopping to learn expert-level mcu programming from scratch is rarely the way things work in a software startup. Despite the obvious focus of this thread, the hardware we assembled and configured was all in support of our software.
I assume that the rather hostile reaction you've received in this thread stems from the combination of your insistence that you tried everything to "bend the law of physics", but dismissal of any suggestion of simpler methods because you're a software person and couldn't have possibly thought/known of everything and we should just all assume that you found the optimal solution. Said differently, it's a strange combination of "we were a team of geniuses so trust me our solution was the best one" and "we were all software people so stop suggesting hardware solutions, it was too complicated for us to learn".
I didn't perceive a hostile reaction, beyond one dude who declared that something hard would be easy.
On the contrary, I've had a number of solid back-and-forths with folks about GPS, GPIO and homebrew RTC daughterboards that we could likely find success with today - almost a decade after we built what we built.
I'm honestly not sure what else you expect from me. Should I apologize for our relative ignorance at the time? Are you upset that I'm proud we managed to succeed despite doing things differently than others might have?
Given that you can't share the unique details of your use case, what could have been an interesting discussion regarding design tradeoffs is moot.
(And I would politely ask that you don't make assumptions about what I might or might not have actually accomplished in similar spheres, you have zero idea.)
What's with the patching? It sounds like pretty much spot on for what NTP was designed for.
Each clock is monotonically increasing, with its individual drift. NTP will pretty soon find out exactly what the right polling interval is. Even if you run it in broadcast mode (why would you though, with only 70 devices).
It's been eight years, but if memory serves, the primary need was to force it to broadcast as frequently as possible.
Everything you said is totally legit unless you're addressing devices with software clocks that are running at speeds dictated by subtly shifting voltage.
Remember, according to the Pi itself, it's 100% in sync. No problem hereeeeeeee (robot voice as voltage dips to 4.9887 volts).
At some point, and considering how cheap they are these days, the simplest option is to add a GPS module if your use-case allows for satellites in decent view.
Would it have worked to add something like a single GPS module connected to an ESP8266 that sends a pulse every minute to all the Pis' GPIOs so they can sync?
Hm, yeah, you'd need a much more frequent method of synchronization to avoid that. I'm assuming the rate wasn't constant, huh? Because of voltage drops depending on draw?
I want to be a better conversation partner on this, but I am extremely constrained in what I can share and again, it was 8-9 years ago that this was all top of mind.
Can we mutually agree that in a year of not-dumb people trying everything we could manage to achieve the best outcome, we tried a lot of things and we often didn't know what we didn't know?
We cannot, because I'm not trying to find you a solution you didn't think of, I'm trying to learn from your failures so I can skip over them next time I need something similar to this.
First, our failures were not technical. We succeeded at creating a sub-$100 camera node that could be replicated and/or replaced quickly by relatively unskilled talent.
If I was going to do it all over again with the decade of engineering skills I've learned, I would look towards fabbing our own RTC I2C daughterboards and trying some GPIO-based approaches to triggering.
However, given our constraints, the solution I've described above worked incredibly well. I wouldn't be embarrassed to do it again.
I'm not saying you failed at the task, but that you tried a bunch of things along the way (that you've detailed in this thread) and they didn't work for whatever reason. That's valuable to know, it saves me time if I want to do something similar in the future. Plus, it's just nice to hear about the engineering of it all.
I got a u-blox NEO-6m module for a few bucks, add another few bucks for the active antenna. The newer GPS modules are much better (more satellites, faster lock etc) but this is sufficient for me.
We've synchronised multiple Pi's to fractions of a millisecond, with flash systems with accuracy in the microseconds for specialised motion capture. Happy to discuss if you're interested - I'm on <hn username> at 'tessierlabs . com' (without spaces).
Because moore's law is a proxy for RnD for an entire industry. if you want cutting edge features, you need to pay for them.
raspberrypi is about value engineering. Making stuff fast and cheap enough so that you can do useful work, but not worry too much if you break it (on the one hand) Or give access to someone who's never had a programmable computer for a price that's cheaper than most mid range lego sets.
The idea that even 10-20% of users would power their Pis with PoE seems wild to me. Seems like a much smaller niche to me (even though I'm one of those people).
Anyway, I have been wondering whether a headless-oriented SKU wouldn't make sense. Pay for the PoE BOM by jettisoning the video output.
I'm really surprised that PoE isn't more prevalent that is currently the case, but I guess it's because most people prefer WiFi devices. I'd add network jacks to everything and remove the power plugs if the option was presented. Most "normal" people seem to be the other way around.
"Wireless" speakers (they'd be networks speakers then) / home assistant devices / media players and whatever else should always be attached with a network cable, so just power them over PoE. I'm still annoying that the AppleTV isn't PoE enabled. That's not a device I'd use over WiFi anyway.
> I'd add network jacks to everything and remove the power plugs if the option was presented.
You could always use PLC to use your eletrical network as a data network. I use that for IP cameras. Unfortunately, I still haven't found a PCL adapter providing PoE power.
> The idea that even 10-20% of users would power their Pis with PoE seems wild to me. Seems like a much smaller niche to me (even though I'm one of those people).
I'd be surprised if the majority of RPi's userbase didn't used PoE. Having to cart around a power supply isn't a very attractive option when all you have to do to power the device is simply plugging in a yank-proof RJ45 cable. It's also cheaper as you don't have to buy a charger.
I've contemplated trying PoE for powering RPis, but just the HAT alone is more than the official power supply, so it's hardly cheaper and that's not even figuring in the extra cost of PoE network equipment.
I've been powering a Pi 4 for several years now with the official PoE hat (edit: not PoE+), without any problems (other than that the fan is starting to get noisy).
I liked that the Pi lacked features that were taken for granted. Made you think about things in a different way. Now it's just the same as everything else and much less about learning and education.
rPI went from armv6 to armv7 to armv8. CortexA76 does not even support using 32 bit kernels. Rp1 requires an armv6 32 bit kernel. There is NO kernel build on earth that would run on both an pi1 and a pi5 and no amount of efi would help. (I am ignoring the possibility of custom nasty hacks to make a polyglot kernel). We’ll have per-device images until we standardize on a single architecture or at least a subset of one.
And no, you really wouldn’t want a lowest common denominator build anyways. That would leave a lot on the table.
Yeah, I can live with images per architecture; the problem is needing an image specifically for the pi. I want the same thing as we get on x86; sure, there are separate images for i686 and x86_64, but there's not one image for Dell and one for HP and one for Lenovo and...
Well, now that the latest rPi is on the latest ARM arch, it might be possible for future rPis, unless ARM goes and changes the arch around in an incompatible way.
That's beside the point; you can build a UEFI bootloader for at least as far back as the Pi 3, but you have to stick it on an SD card because there's no on-board memory to hold it. I mostly just want enough flash built in to hold it.
rPI is quite cost optimized. That’s one of the stated goals. An extra component to buy and place on the board costs read money. Plus, an implementation of UEFI will not be tiny. It would likely would need a few MBytes of flash. That's a dollar or two already. If everybody’s requests, each of which would add a dollar or two to the board, were fulfilled, you’d be looking at a much more expensive board, and complaining about cost instead. :)
I've been trying for almost 1 year, to get a Pi 4 at MRSP and it's impossible due to production shortages, but they decided it was a good idea to thin even more the production lines with a different model.
Sincerely, it seems a very bad idea from their part.
> I've been trying for almost 1 year, to get a Pi 4 at MRSP and it's impossible due to production shortages
Where are you? Chicago, PiShop, and Digikey have Pi4's at MSRP right now in the US, and have had them for most of the last 6 weeks. Availability in Europe looks even better...
Pi Zero 2W and CM4 are more problematic, but not in the category of "wait a year."
I fundamentally have issues with Raspberry Pi brand. It markets itself as a “open source” company but it is anything but. Sure you can get pcb files, but the real stuff is the binary blobs for qualcomm processors. None of that stuff is open source. You can’t get a datasheet for it.
So you’ve got a company that is taking advantage of OSS ecosystem’s reputation but returning nothing back.
Treat it like any other company and I’m fine with it. Happy to buy their computers. Just like I’m fine with OpenAI admitting they’re anything but “Open”. Just be honest ffs.
All their software and documentation is open source. The only closed source bits are stuff they don’t control from Broadcom. They are easily one of the best SBCs for documentation and continued linux support over time, you can use 5 year old boards with latest kernels
People seem to think it does, and want it, but I don't see that claim (or even the words "open source") on either the trading company or the charity About Us pages. Plenty of companies use OSS, nothing wrong with that. The idea they've returned nothing back is misplaced.
Sadly built on legacy ISA, still ARM instead of the open standard RISC-V. They wasted the opportunity to go RISC-V with Pi 5.
I am surprised they still haven't figured out how to accept USB power with higher voltage than 5V. It'll thus continue to be a pain to power.
Apparently it isn't all that efficient anymore as evident from their active cooler add-on.
It looks like the performance will be similar to the incoming (was meant for release this summer but seems to be late) new SiFive RISC-V development board with P550 CPU.
There are a lot of good reasons to not jump the gun on riscv. You're not going to get the same performance for the same price point on riscv at the moment. A lot of the existing software ecosystem still hasn't been ported over to riscv either.
risc-v isn't going to displace arm overnight. x86 still exists in the same markets it's always done well in despite over a decade of ARM trying to enter its markets. Raspberry Pi is doing the sane thing of delivering a good working product to its customers. It would have an interior product if it tried to do the same with risc-v.The fact there is no existing risc-v that comes close to competing with raspberry pi is evidence of this.
The microSD card slot is really great to have, to get up-and-running, but once it's clear which function the board will serve, being able to move over to a directly connected old NVMe would really be a benefit, also in terms of reliability. These microSD cards scare me, yet I make full use of them.
Finally having a battery backed RTC on it is really great news.
[0] https://wiki.radxa.com/Rock5/hardware/5b