It's always good to see new stuff from Arduino. Especially when I'm not expecting it. This seems to be very specifically for use a controller. Not sure if it's meant for robots or if they are going after the MakeyMakey market since it is based on the Leonardo. The lack of an Xbee or similar RF socket makes it seem more like the latter than for remote control applications. But I'm sure it would be trivial to add your own but that doesn't really seem like the point of this board.
That being said it can be very helpful for prototyping to already have onboard sensors like accelerometer, temp, buttons, and pots.
I tried to buy it, but the link on the page goes to a blank page and I cannot find it in the Arduino store when I look. I wonder what happened with that. I would like to try one out!
EDIT: The "Buy from Arduino Store" button has now been removed from the page and the "Buy from Distributors" button returns a 503 error. Looks like the Esplora may be going the way of the Nexus 4
This is quite a functional piece of hardware. They've provided a package shaped like a game system with joystick and four-way buttons, plus RGB LED, temp sensor, loudness sensor, light sensor, analog slider, and TFT LCD connector. I fully expect Adafruit or another maker company to come out with kits with this, a plastic case, and screen before long.
Does this mean Arduino is stepping away from the all modular, shield based boards towards more complete, all-in-one boards? Also, I'm wondering how the Esplora is going to be better than a standard Arduino board, with added shields. I mean, I can see the point of an Arduino based game controller, but why the extra sensors?
I'm not saying it's a bad thing, just curious about the decision making progress!
I imagine the success of the rasperrypi and other all-in-ones is scaring the arduino people.
Look, I'm the perfect demographic for arduino, but I never bought one. I just don't want to have to buy tons of shields, ribbon cables, lcd displays, etc to get a basic device that I can have some fun with. Its also a little intimidating for beginners.
>but why the extra sensors?
Why not? We're talking a minimal amount of money added to the board. You can always get the vanilla arduino if you want one.
The arduino and the raspi aren't realistically competitors to each other. At all.
The raspi requires ~500-800mA to run, which generally means buying batteries and DC-DC converters meant for hobby planes/cars and butchering a USB cable to power it.
The arduino requires ~10mA (without any accessories) and can be powered by a 9V battery and a LM7805 voltage regulator. All in cost is <$5.
We have 3 raspis in our lab. Mine, and two that have been donated by people who bought them, and then couldn't figure out what to do with them. We have a stack of arduinos that get used constantly.
The raspi is great for "hobbyists", that is: people who are making media centers for their television so that they can stream TV shows to it.
The arduino is made for hackers, interactive artists, etc.
(My current project with the raspi is a telepresence robot. Something that isn't even possible with an arduino. They're really not competitors. The raspi is just a cheap linux box that you run python on. The arduino is a prototyping platform for Atmel AVRs that you write firmware for.)
Not to mention you could delve into bare-metal AVR programming from there, and work with tiny devices that consume literally nothing as long as they have not received a wake-up signal.
Do you have an opinion on which platform is better for projects that need to connect to the internet over wifi? Other than the networking issue, Arduino is more than powerful enough for my purposes. I don't really want to shell out $80 for a wifi shield, so I was thinking it might be better to just switch to RaspPi + a cheap usb dongle. Any thoughts?
It depends on what software you want to use, what software you want to write, how much power you have available, how many production units you'll want, etc.
If you want to do a one-off and have AC power, use the Pi. If you want to make a billion of these and fit them in a tiny box with 4 AA batteries, the AVR is going to be a better choice. But Linux + Python is easier to program than AVR firmware, so it's going to take a lot of units before the AVR pays for its difficult software development with its reduced parts cost. (Really, you should be using ARM, not AVR. It costs the same but it's a modern architecture.)
code for Linux and code for AVR are two totally different beasts. With AVR, I feel that you could just drum up a quick hack and have code running within a couple of hours. With Linux, I always felt that you need at least a couple of days to get things to going (eg, having a root fs structured, choose a kernel, make sure that your memory map is set up correctly etc). Maybe, I have been away from embedded Linux for too long and the goal post have shifted towards a better location.
If you absolutely have to have WiFi, then you might consider the raspi, since it is cheaper.
But...cost on it is going to be ~$50. ($35 + $12 for a wifi chip). That's getting close to the cost of the wifi shield, and if you already know the arduino platform, then I would stick with it.
I don't know, for instance, if PWM is yet support on the raspi (or ever will be, since it's not an RTOS)
I always viewed Arduino as a prototyping platform. Once you finalize your design, you'd buy a $2 AVR and stick it on its own circuit board with its own components, freeing up the Arduino and any shields you have for your next project.
That is quite different to the Raspberry Pi, which just appears to be a small, cheap, general purpose computer that comes without a case.
There is certainly some overlap, but I'm not sure they are directly competing with each other.
I recently just brought both and have been playing with them, as well as reading on the forums. There are some things that raspberry pi is better at, such as video media or computer vision. You will never get to run Quake on Arduino.
However, there is stuff that arduino is better at. Most people have difficult time driving servos with the GPIO pins on the Raspberry Pi. In addition, with the default image of Ubuntu, which is not an actual real-time OS, will cause timing problems in the PWM (pulse width modulation) signals controlling the servos, resulting in jittery servos when an interrupt happens. Currently, the arduinos are better at that, plus they've been around longer, with a larger community, and accessories. Anyway, that's where it's at now.
As for the future trajectory, I can only really guess, but from where I'm standing as a beginner choosing between platforms, while they seem to be targeting education, they're targeting different use cases.
> with the default image of Ubuntu, which is not an actual real-time OS, will cause timing problems in the PWM (pulse width modulation) signals controlling the servos,
CLI / STI (or whatever the CPU specific set is), run as root.
That should solve any timing problems you've got. I've used this to successfully drive a full sized portal mill / plasma cutter from an all-in-one VIA based board.
When you're driving motors you want 100% control. Interrupts you can comfortably do without until you're in a quiet spot, then you check for them and continue.
Probably shouldn't say never. Someone has already managed to get Linux running on the 8-bit micro. Though getting it to run with acceptable performance is a another matter.
Not to quibble with the overall point, but the "default image" is Debian-based, but not derived from Ubuntu, I don't believe. Maybe they thought Raspian sounded better than Rasbuntu...
What I want is a a way to prototype on an Android smartphone. When the design is finished order a custom board based on the particular sensors and parts that are required.
I don't think so. I think they'll still offer the vanilla boards, but they just noticed a pattern in what their users were trying to do, and this gives them a bit of a head start.
This is going to be ALL OVER art games installations (which are definitely picking up steam the past couple of years, and hey, LA Gamespace got a quarter million funding...). I expect we'll see digital arts galleries littered with these within months.
Good idea too. Seems like also-but-not-quite MakeyMakey competition, because the makeymakey was "Make buttons out of anything". This comes with the buttons but also all of the other sensors you'd expect in controllers these days, except you don't have to wait for someone to reverse out everything and then write drivers for your platform.
Anyone working with this is going to have "Dora, Dora, Dora the Esplora" stuck in their heads to the point of insanity. And also any one who reads this. Add a malfunctioning card swiper into the mix and there'll likely be a murderous rampage.
This is cool, but I don't like the fact they've essentially forced you into a particular key layout with this design. I thought the whole point of an Arduino board is how modular it is?
What if I'm trying to build out a controller for a PS2 emulator on my phone? I don't have all the requisite buttons. I also don't have them in the layout I'd expect. That's kind of a bummer.
I think this is for people rapidly prototyping robots and devices that need RC-like functionality (hence the name "esplora") -- judging from that, this is meant to be a time-saver for a very specific purpose: creating remote controls for vehicles, which generally only require a similar button layout. I think it will be a success, mainly because that seems to be a common task for people using Arduino boards for robotics-type projects.
I also think that's their initial idea. What I don't get is this: why would you build an entire Arduino-based robot, only to buy an almost-ready RC-controller. Which isn't actually an RC-controller, because you'd have to have a separate transmitter shield for it...
I don't think it's intended to -just- be a controller, but rather a game console unto itself(hence the socket for color TFT screen) . It certainly could be used as just an input device, though.
I think this might be aimed at classrooms that are teaching Arduino programming. This is a good mix of input and output to get people started without having to break out the soldering irons.
The Tinkerkit connectors would let the students add other physical interfaces, while still not soldering.
But then it is a RadioShack exclusive product‽ Maybe RadioShack just paid them a small pile of money to make something to their specs.
I welcome the addition, but I see the word "fragmentation" has yet to be used in this thread. This board is a true departure from previous designs, but it's my experience that Arduino tutorials are getting harder and harder to "plug and play" thanks to the many various boards out there.
While I agree that Esplora is a true departure from previous designs, I don't think "fragmentation" is really an obstacle here. As always, users just need to choose the board model under Tools -> Board menu. As the underlying library and code, I believe are still the same.
Well, you can't actually drive a TFT with an ATmega. Speeds don't add up.
What they are talking about is connecting a TFT over SPI, which is to say a display that has a very high-level chip on it that does the actual driving and that you can talk to over a protocol to do various things like draw lines or set a pixel.
That being said it can be very helpful for prototyping to already have onboard sensors like accelerometer, temp, buttons, and pots.
I tried to buy it, but the link on the page goes to a blank page and I cannot find it in the Arduino store when I look. I wonder what happened with that. I would like to try one out!
EDIT: The "Buy from Arduino Store" button has now been removed from the page and the "Buy from Distributors" button returns a 503 error. Looks like the Esplora may be going the way of the Nexus 4