crest

18 ms·

271.
by crest 2y ago
They know how to integrate and operate those and all the other pieces required to go from a hardware design to hardware you can touch. If you just had to take delivery of ASML machines Intel wouldn't be struggling to get below a (so ca
272.
by crest 2y ago
This "conclusion" ignores reality. Chat bots like those the article mentioned aren't sentient. They're overhyped next-token-predictor incapable of real reasoning even if the correlation can be astonishing. Withholding in
273.
by crest 2y ago
At that point just run the code inside a chroot with a full /dev and call it good enough. No common GPU driver, firmware or hardware was designed to securely run really untrusted code from multiple tenants.
274.
by crest 2y ago
As I understand that doesn't imply that it's not undefined to pass NULL pointers. While not what most users expect/want it's possible to this is just a wrapper around an memcpy() which will only be correct to call with
275.
by crest 2y ago
Mesh networks are a neat idea, but the reality of is often disappointing e.g. given compatible radios all the IoT devices in a room could form a mesh network of equal peers, but if a few lightbulbs use their position (mains powered, good lo
276.
by crest 2y ago
Google's interests don't align with yours so taking something that infeasible to rewrite yourself like a modern browser and only ripping out the worst parts is a useful alternative?
277.
by crest 2y ago
Only if it runs on a a few microwatts (average power consumption). Any buffering to normal timescale batteries or large capacitors would introduce additional losses. Their self discharge rate might even exceed the power output of such a bat
278.
by crest 2y ago
Maybe if health care wasn't run as a for profit business he would still be alive, but was is one more person to die for corporate profits. ¯\_(ツ)_/¯
279.
by crest 2y ago
IIRC they also implemented some features by reporting them as available and mostly emulating them on the CPU to qualify.
280.
by crest 2y ago
The GUI >99% of users used to interface with the OS required a GPU to composite the different 2d buffers with fancy effects. IIRC if you knew how do disable as much as possible of it the performance without GPU acceleration was not great
281.
by crest 2y ago
Sure but the nice thing about a named pipe is that you can, but don't have to get file descriptor (through inheritance or fd passing). Instead you can e.g. put the path to it into an environment variable. Sometimes that's wanted s
282.
by crest 2y ago
On FreeBSD you also have process descriptors (Linux followed suite a while ago) which provide a race free clean interface to supervise processes even if you're not their reaper (so can't use PIDs reliably). You can add them to kqu
283.
by crest 2y ago
Transfers up to PIPE_BUF are atomic on pipes according to POSIX.
284.
by crest 2y ago
Them *quietly* disabling a feature that few users will notice yet complicates the frontend suggests they pulled this chicken bit because they wanted to avoid or delay disclosing a hardware bug to the general public, but already push the mit
285.
by crest 2y ago
Much more importantly they fixed the MMU support. The original 68000 lost some state required to recover from a page fault the workaround was ugly and expensive: run two CPUs "time shifted" by one cycle and inject a recoverable in
286.
by crest 2y ago
That doesn't explain a 20x regional pricing difference for the same service (outgoing bandwidth). It's either a malicious change to extract exorbitant rent from customers (hostages at that point) or buying enough bandwidth is whol
287.
by crest 2y ago
I would like to know more about their bandwidth costs to learn if: * they have to pay that much more for (outbound) traffic in the US * miscalculated their US expansion and are trying to recoup * they think customers will only bitch
288.
by crest 2y ago
As long as the link doesn't drop or reorder packets too badly a >300MB/s (or probably given we're talking about small virtual machines 300Mb/s) flow doesn't require much tuning. For a single long lived TCP connec
289.
by crest 2y ago
May I recommend the old 27C3 talk "OMG WTF PDF"?
290.
by crest 2y ago
What performance are you looking for alternatively what's you (power) budget?
291.
by crest 2y ago
For a such an interface to be feasible to support in common open source infrastructure it needs a pure software implementation for testing and development purposes. Even better something along the lines of coz to even model performance by t
292.
by crest 2y ago
I assume 1-3 hops of modern switches without congestion. Given 100Gb/s lanes these numbers are possible if you get all the bottlenecks out of the way. The moment you hit a deep queue the latency explodes.
293.
by crest 2y ago
You have to be so far beyond the Balmer peak to understand a Bavarian that I doubt that will happen :-P.
294.
by crest 2y ago
Just wait until Cloudflare decides you can afford to be on a different plan.
295.
by crest 2y ago
I wonder at which point they'll admit they've added back all the complexity of a hierarchical filesystem.
296.
by crest 2y ago
A non-trivial application should not add a dependency for just exponential backoff + proportional jitter, an easy to use wrapper to put around a quick script is a good idea that and the lack of such a "basic" defensive programming
297.
by crest 2y ago
Or give your users a proper streaming API for large or partial reponses?
298.
by crest 2y ago
You can implement a single client NFSv3 server that provides stronger than expected (of NFSv3) guarantees and if you implement the "optional" companion protocols it should come closer to local filesystem semantics than most networ
299.
by crest 2y ago
It depends mostly on the bandwidth-delay-product and packet loss you expect on each connection. A there is a vast difference between a local interactive SSH session and downloading a large VM image from across an ocean.
300.
by crest 2y ago
Afaik the macOS port of OpenBSD's pf firewall is the only firewall used by both Apple's system settings and obdev's LittleSnitch. They're both GUI frontends to the same backend, but Apple supposedly also added internal &

More ›