Andrey

[Andrey @justfhs]

@nicbarkeragain Are you saying we need 1x editor? 🤔

[Andrey @justfhs]

@ThePrimeagen Plz include some bouncy balls for visual friendly comparison. Ty bye

[Andrey @justfhs]

@awesomekling @ladybirdbrowser Hello Mr. President 🫡 Seriously though, your progress from fixing serenity bugs to creating a competitive browser engine is a huge inspiration. A great proof that dedication and passion are always fruitful! Go ladybird 🐞

[Andrey @justfhs]

@tunguz I'd go to @__tinygrad__ as the only AI company that is trying to democratize, not silo the AI

[KRAZAM @krazamtv]

NEW VIDEO: Positive Affirmations for Site Reliability Engineers

https://t.co/Jfa9RNVATm

[Andrey @justfhs]

@awesomekling A real yak next year?🤔

[Vjekoslav Krajačić @vkrajacic]

Disk Voyager will be released in alpha on December 12, 2023!
Subscribe: https://t.co/I8PkxUkhWh
Demo: https://t.co/67u4Jy3zaZ

The 5-minute demo video is from the Handmade Cities conference that took place last week, and Voyager received overwhelmingly positive comments!

[Andrey @justfhs]

@Liv_Boeree @ImMollyBloom Just finished listening, it was great! Got motivated to properly try morning meditation! @Liv_Boeree did you find any way to gamify it?

[Casey Muratori @cmuratori]

I'm tired of seeing lazy "software performance doesn't matter" excuses on forums and social media, so I devoted an entire video to the mountains of evidence debunking that ridiculous idea: https://t.co/6mhTXcuGvA

[Casey Muratori @cmuratori]

Using their own example code, I made a simple demonstration of how horrible "Clean Code" ideas are for software performance:
https://t.co/dO3UyaNdf6

[Allan Blomquist @allan_blomquist]

Here's a look at some of our internal programming tools at Tomorrow Corporation!

https://t.co/wsuooryLOg

[Andrey @justfhs]

As time passes and cache fills up, I feel more "usual". Take care of your and your CPUs caches!🤕

[Andrey @justfhs]

I've hit my head while snowboarding. All is fine, but right after the hit, I felt like a CPU that lost all of its caches - I can remember things, but I have no knowledge of what I _might_ know. It felt weird and scary.

[Andrey @justfhs]

A reductionist might say that "choosing a game" is a game itself, thus intelligence let's you win it (choosing only right games) hence wisdom=intelligence, or even that wisdom doesn't exist (since it's just an alias for intelligence)
I don't like reductionists! 🤫

[Liv Boeree @Liv_Boeree]

Intelligence is knowing how to win the game.

Wisdom is knowing which game to play in the first place.

[Andrey @justfhs]

@AlexAltea I think Freenet has most of what you want, with what they call USK for updates. Though no "branches" or direct git mapping afaik

[Andrey @justfhs]

🚿🤔 Apartment size can be measured by how many times you have to unplug vacuum cleaner during entire apartment sweep

[Andrey @justfhs]

@BruceDawson0xB @ChenCravat I'm not really looking for a solution for my specific problem, just wondering why drivers can't set timers with small delay. Maybe some other way to produce an interrupt on specific CPU at a given delay would also work. But that seems like what timers should do 🫤

[Andrey @justfhs]

Is there a good reason for Windows to have hard cap on timer resolution at 0.5ms? That's 2KHz, while even PIT can deliver 1MHz interrupts. Probably sane limit for userland, but why can't driver ask for finer resolution?
@ChenCravat @BruceDawson0xB maybe

[Andrey @justfhs]

@BruceDawson0xB @ChenCravat Sorry, used commonly used function (not instruction) name. I mean VMRUN from AMD Secure Virtual Machine. My intent is to put a limit on how long it can execute, afaik the only way is to get an interrupt. Sometimes that limit is short - hence measuring in TSC units

[Andrey @justfhs]

@BruceDawson0xB @ChenCravat Documented way seemed to be using kernel timers - and it works, except that margin of error ended up very large, due minimum timer resolution being so high (for what I'm trying to do)
FWIW, Linux declared timer resolution is 1ns, and from my brief experiments it is precise enough

[Andrey @justfhs]

@BruceDawson0xB @ChenCravat I understand power/perf implications, and that in general you should use something else. My case is kicking CPU from svmrun, targeting specific tsc value. This requires an interrupt, and should be doable with APIC tsc-deadline timer, but it's owned by HAL :|