the8472

31 ms·

541.
by the8472 3y ago
IBM's quantum roadmap[0] also looks... curious. There's some modest scaling until 2028. Then there sudden orders-of-magnitude jumps in the following years. Perhaps that's backed by technology they have in the labs and plan to
542.
by the8472 3y ago
With modern nuclear weapons design they could probably be a lot lighter. Speculation has it that the US can now build fusion secondaries without the heavy pusher.
543.
by the8472 3y ago
My understanding is that that is a punishable offense in most states, so that's a matter of consistently enforcing existing laws rather than introducing any new ones. Anyway, the mechanisms I have listed can still drive adoption of sel
544.
by the8472 3y ago
It doesn't require government action. Insurance will see computers causing fewer accidents, requiring lower payouts and adjusting rates accordingly. Penalties for drunk drivers might become harsher because they now have an alternative.
545.
by the8472 3y ago
It makes no difference to the pedestrian, cyclist or any other party in a crash. One does not choose whether one crashes with/gets run over by a novice or an experienced driver.
546.
by the8472 3y ago
Sir, our data analysis shows that you and your family would have been crushed under one of our trucks last month if it hadn't been for its autonomous driving software. Based on your internet commenting history you have expressed a pref
547.
by the8472 3y ago
Those are cooler in allcaps.
548.
by the8472 3y ago
There's plenty of CPU-cycle-munching work in the pro segment that does care about performance and can't assume fixed hardware. CI, compression, CGI, ...
549.
by the8472 3y ago
You usually don't design an algorithm for a single CPU. Most software has to run on different tiers and generations from Intel, AMD, a whole herd of different ARM processors. Even the very hottest code paths will mostly do CPU feature
550.
by the8472 3y ago
They prickle differently
551.
by the8472 3y ago
Perhaps we could replace mechanical feedback with tiny electrical shocks. Look ma, no moving parts.
552.
by the8472 3y ago
They don't care when it works. Then they get locked out of their google account and lose a lot of things at once and only have few alternatives to choose from and may return google yet again due to lack of choice. > You can maybe ar
553.
by the8472 3y ago
Not every equilibrium is a good equilibrium.
554.
by the8472 3y ago
https://en.wikipedia.org/wiki/Quasi-star also mentioned in the paper, powered by larger black holes
555.
by the8472 3y ago
> Anyone who's not an Arch/Kali/BSD/suckless enthusiast type Not sure if Arch should be grouped with those. On the Steam statistics it's more popular than ubuntu, and that's not counting SteamOS.
556.
by the8472 3y ago
The client and server need to talk http to the authorization server, that doesn't mean they need to talk http to each other.
557.
by the8472 3y ago
> (Although… we really need a way to do TLS/QUIC to an endpoint without a domain name.) Generate self-signed cert, let the client TOFU. And skip the HTTP part. Just like SSH. There's no reason to do things like a browser.
558.
by the8472 3y ago
> I'm just guessing here; but the reason why it's not implemented in userspace, not written in Rust, not using IOMMU, BP is allowed full DMA, the system is split into AP and BP in the first place, etc etc, is probably because o
559.
by the8472 3y ago
There should be several ways to optimize around this without needing new instructions. Choosing different calling conventions for some functions (e.g. PreserveMost in llvm), amortizing the cost over multiple loops by inlining harder, usin
560.
by the8472 3y ago
BBRv1 doesn't take ECN into account. BBRv2/v3 do, and it's mentioned in the RFC: Scalable variants are under consideration for more recent transport protocols (e.g., QUIC), and the L4S ECN part of BBRv2
561.
by the8472 3y ago
Would a CPU still benefit if it supported both a register-renaming mode for backwards compatibility and a clockhands mode for power saving/a wider frontend? Like ARM thumb or x86 32 vs. 64 bit.
562.
by the8472 3y ago
If this is a per-site issue then maybe just use the devtools to profile what's happening. If it's a whole-browser issue then use the firefox profiler and configure it to profile from startup: https://profiler.firefox.co
563.
by the8472 3y ago
> EDIT: I rely highly on the Restore Previous Session feature. That might be why. As do I, and it doesn't take as long on my machine. Perhaps your hardware is unusually slow or you're using extensions that do something on every
564.
by the8472 3y ago
> It only works on the last window you have closed though. Alt+F, Q exits all windows at once and also restores them the next time you launch firefox.
565.
by the8472 3y ago
260 would be falling short to fall short of the island of stability
566.
by the8472 3y ago
The bug depends on a few factors that aren't really packaging-specific. A) the default socket buffer size B) the polling rate of your mouse C) whether there's anything on your system or in your firefox profile that can cause enoug
567.
by the8472 3y ago
Which is also one of the reasons why WSL1 was so slow. Windows lacks this cache so programs doing unixy filesystem access patterns perform poorly there.
568.
by the8472 3y ago
It's not just Optane. There have been other NVRAM technologies that were hyped as "fast as SRAM, persistent as HDDs, cheap and dense as DRAM" over the years such as FeRAM and MRAM. MRAM exists as a commercial product but it s
569.
by the8472 3y ago
querying the file size can be useful to choose the allocation size for a buffer. but yes, you have to keep reading until you get a zero-length read. https://man7.org/linux/man-pages/man2/read.2.html > On s
570.
by the8472 3y ago
> Of course, the actual movement skeleton is coming from presumably real motion-capture, but still. I recall seeing from-scratch learned realistic motion somewhere. I think was meant for character animation that would more realistically

More ›