Hacker Newsnew | past | comments | ask | show | jobs | submit | sfilmeyer's commentslogin

I feel like there's a bit of a jump from "tech-savvy" to de-soldering things on an expensive piece of home electronics. As it stands now, though, I agree that turning off the smart TV features seems to be the way to go for most people.

Ha, yea it's been awhile since I've done that. Although if I was annoyed enough I might take one apart.

Troy Hunt has been running Have I Been Pwned for years. He even uses the k-anonymity model to allow you to search if a password has been pwned without giving him the password if you don't trust him.

I get your general point, but he's been a leader in this space and walking the walk for a decade. I'm not even into security stuff or anything particularly related to this, and I still recognized his name in the OP domain.


More importantly, since HIBP sells monitoring services to 1Password, if they were maliciously collecting this data they would be immediately sued to oblivion.


I enjoyed reading the article, but I'm pretty thrown by the benchmarks and conclusion. All of the times are reported to a single digit of precision, but then the summary is claiming that one function shows an improvement while the other two are described as negligible. When all the numbers presented are "~5ms" or "~6ms", it doesn't leave me confident that small changes to the benchmarking might have substantially changed that conclusion.


Yeah. When your timing results are a single digit multiple of your timing precision, that is a good indication you either need a longer test, or a more precise clock.

At a 5ms baseline with millisecond precision, the smallest improvement you can measure is 20%. And you cannot distinguish a 20% speedup with a 20% slowdown that happened to get luck with clock ticks.

For what it is worth, I ran the provided test code on my machine with a 100x increase in iterations and got the following:

  == Benchmarking ABS ==
  ABS (branch):     0.260 sec
  ABS (branchless): 0.264 sec

  == Benchmarking CLAMP ==
  CLAMP (branch):     0.332 sec 
  CLAMP (branchless): 0.538 sec

  == Benchmarking PARTITION ==
  PARTITION (branch):     0.043 sec
  PARTITION (branchless): 0.091 sec
Which is not exactly encouraging (gcc 13.3.0, -ffast-math -march=native. I did not use the -fomit-this-entire-function flag, which my compiler does not understand).

I had to drop down to O0 to see branchless be faster in any case:

  == Benchmarking ABS ==
  ABS (branch):     0.743 sec
  ABS (branchless): 0.948 sec

  == Benchmarking CLAMP ==
  CLAMP (branch):     4.275 sec
  CLAMP (branchless): 1.429 sec

  == Benchmarking PARTITION ==
  PARTITION (branch):     0.156 sec
  PARTITION (branchless): 0.164 sec


I also tried myself, on different array sizes, with more iterations. The branchy version is not strictly worse.

https://gist.github.com/Stefan-JLU/3925c6a73836ce841860b55c8...


> I had to drop down to O0 to see branchless be faster in any case

Did you check whether your branchy code actually still was branchy after the compiler processed it at higher optimization levels?


In general, modern compilers will often unroll or inline functions without people even noticing. This often helps with cache level state localization and parallelism.

Most code should focus on readability, then profile for busy areas under use, and finally refactor the busy areas though hand optimization or register hints as required.

If one creates something that looks suspect (inline Assembly macro), a peer or llvm build will come along and ruin it later for sure. Have a great day =3


Doesn’t it also help with branch prediction since the unrolled loop can use different statistics with each copy?


Non-overlapping sub-problems may be safely parallelized, and executed out-of-order.

In some architectures, both of the branch code motions are executed in parallel, and one is simply tossed after dependent operations finish. We can't be sure exactly how branch predictors and pre-fetch is implemented as it falls under manufacturer NDA. =3


Can you link an example of a window unit ERV? I tried searching briefly, and came across some folks hacking together units to make them work with windows or adding their own ducting, but nothing analogous to a simple window air conditioning unit. As a renter of an apartment in a very much not modern home, I don't really see anything that seems like it would work.


http://www.purifresh.com/erv.html

https://swervair.com/

A couple examples I see on Google. I'm not advocating for any of these, because I have no idea if they are any good, but I see no technical reason an ERV couldn't work as a window unit. Maybe it's an underserved market and someone should make a business out of that.

A much more DIY example that's probably closer to what you were talking about with "hacking together" a solution: https://www.mychemicalfreehouse.net/2023/10/window-mounted-p...


Thank you! You're right that's the sort of "hacked together" solution that looks cool but beyond my abilities, and I appreciate the first two links.


That makes sense along the lines of their second proposal, but doesn't address the concerns of the first. Part of democracy means voting for the folks who govern you, but a prisoner might be left unable to vote in an election for the local state or municipal governments.


Fore example, someone with a 10+ year sentence has a compelling interest in local candidates that have different platforms that will affect the parole-rules and phone-call-costs next year.


>2) It's just generally good to diversify into your competitors. Every company does this, especially when the price is cheap.

This definitely isn't a thing that every company does (or even close to every company).


Not every company, but the largest ones do.

Microsoft once owned a decent amount of Apple & Facebook for example.


> What’s Apple’s default time frame for security support?

This isn't thaaaaat far out of support. Their last security update for iOS 15 was just earlier this year, and they only dropped iPhone 6s from new major versions with iOS 16 a few years ago. As someone who has kept my last few iPhones for 5+ years each, I definitely appreciate that they keep a much longer support window than most folks on the Android side of things.


Before I got my first iPhone five years ago, I always noticed that iPhone owners would drag it along for a long time, but really the phones are tanks. I remember switching Android phones every two years, because they quite literally started to decay. I think my last Android Phone I could have probably made last longer than two years, I still turn it on and play random games on it, and its still very responsive.

I assume they know just how long their customers keep their phones and maintain them accordingly.


This... is the opposite of my experience. Friends with iPhones seem to upgrade them unreasonably often, but my (Samsung) Android phones last a loooong time. My first Samsung I retired somewhat involuntarily after 3 years so that I could get a model that would also work overseas, but the phone itself was still fine. My second Samsung (the one I got in 2016 for the overseas trip) I just retired last fall, 2024, and even then only because a job required MS Authenticator and it wouldn't let me download it to the phone. Battery life was still fine, everything I used worked fine.

I fully expect to be using my current Android phone into the 2030s.


Well your experience is maybe more based on your friend behavior than on an absolute rule.

This is the same for absolutely every manufactured goods. The same durable car model will be kept for over a decade by some people while some other opt for a leasing plan that guarantee a new car every two years. But the intrinsic quality of the car remain unaffected.

To ponder this you must consider what become of the phone they replace : did they trash it or did they have a second life with a less edgy owner?


Maybe you use low end phones or crappy vendors?

I'm migrating from my 5 year old flagship (lol) only because vendor decided to stop supporting it. Battery still good for a day, great screen, good enough camera, fantastic sound, ssd card slot...

My next has at least 7 years of mainline support (with all AOSP releases) plus at least couple of years damage control updates.

It's a matter of the choose I think.


The second hand resale market for iPhone is huge, especially in Asian 3rd world countries.

It is in Apple’s interest to keep old iPhones updated, as old iPhones being in active usage is better than them rotting in a drawer.


A relative of mine used their Galaxy Note II until the internal flash died and it stopped booting. It was definitely over 5 years old by that point.


Maybe I'm misreading your comment, but you seem like you're talking about privatizing this research whereas the other commenter seems to be talking about public cuts leading to a reduction of research. Just because something gets cut doesn't mean it gets outsourced elsewhere.


I guess my point is that it's hard to simply cut research that's essential for certifying that the stockpile is safe and works. I'll avoid making any predictions, because who the hell knows what's going to happen, but I think dynamic imaging work may prove a tough target for DOGE.


Yes. It is hard to this honestly and correctly. That would mean that normal people wouldn't make these cuts.

It also has very little predictive power for the loon with the checkbook right now. He might just as likely notice that people care a lot about that issue and hold it for ransom.


I think the overall aim of DOGE is simply to move research into privately controlled entities, especially those that can’t be cut. Its simply a continuation of transferring the national asset base (tax/usd) from democratic control, into private control.

It doesn’t need to be profit making in the normal sense (see SpaceX) it just needs to be the only game in town when the US Gov spends on national security


Why not just rubber stamp the certification and save all that money?


I'm a man, and had wonderful experiences with my many (mostly male) roommates, with only occasional hiccups. Saying adult men cannot live together seems pretty excessive.


they can but it's inherently riskier, especially with a woman involved


On the spectrum of illegality, things can get a lot more extreme than a bit of copyright infringement.


True - I mean, one could try and block based on file-extension/MIME-types, but... nothing stopping a malicious user from renaming a file to an allowed extension, with some sort of malicious/secret payload. (Or... spreading some sort of malware/virus/exploit via media file formats, I have never looked into the possibility of that until just now, apparently it can be a thing - https://cyberpress.org/cybercriminals-exploiting-media-files...)

So yeah - this is probably one of those half-baked ideas that just wouldn't be a good one to actually implement "in-the-wild".


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: