Sesse__

42 ms·

601.
by Sesse__ 2y ago
“The winner of a game is the one who makes the second-to-last blunder.” (Also this has come up in computer chess; it's more important to improve the quality of your worst moves than your average moves)
602.
by Sesse__ 2y ago
> The tree leaves have no impact on frequency, just amplitude. That's an oxymoron, really. Frequency and amplitude are closely interrelated concepts (e.g. see Heisenberg's uncertainty principle in the context of signal processi
603.
by Sesse__ 2y ago
The decoded frames do not pass through JavaScript. The compressed data may (e.g. if you are using the Media Source API).
604.
by Sesse__ 2y ago
The total internal reflection is to keep it focused, it's in a sense a different question. IIRC, when Corning Co. first started working on optical fibers, the best available glass would be good for sending signals about ten meters. Wha
605.
by Sesse__ 2y ago
It's exceptionally clear compared to e.g. window glass even in the visible spectrum of light. You can shine a red light source into a 10 kilometer standard G.657 fiber (optimized for 1310/1550nm, i.e. deep infrared) and it will st
606.
by Sesse__ 2y ago
Fiber optics also uses _exceptionally_ clear glass. If the ocean were as clear as your average long-distance fiber cable, you would see down to the bottom of the Mariana Trench (also in the range of visible light, AFAIK).
607.
by Sesse__ 2y ago
It _is_ the wrong feel for a concept. The analogy breaks down because the color changes are way too wide in frequency (and thus too robust to noise) compared to what happens in a radio broadcast. If you changed the color from RGB(127, 0, 0)
608.
by Sesse__ 2y ago
It's not that simple, though. The only way you can detect frequency is by measuring the amplitude (and then differentiate; except of course in an analog circuit, you don't do that exactly, you have some mechanism that tries to tra
609.
by Sesse__ 2y ago
blktrace is fairly useful in this regard.
610.
by Sesse__ 2y ago
> I'm not sure how you would optimally use an LLM for a lossless compression task. You have the model output probabilities for each token, and then use arithmetic encoding using that model. (This uses optimally few bits per token, u
611.
by Sesse__ 2y ago
Also, 300µs _per word_, looking up a word in a corpus of 1.3M words. Not 300µs for correcting 1.3M words!
612.
by Sesse__ 2y ago
> Remember, all I/O in NT is async at the kernel level. It's not a "bolt-on". All I/O in Linux is also async at the kernel level! The problem has always been expressing that asynchronicity to userspace in a sane
613.
by Sesse__ 2y ago
Apple's Siri is using MPTCP, so it is presumably viable.
614.
by Sesse__ 2y ago
I am still using my Nikon D4 (2012), but they simply don't make parts for it anymore. These are mechanical devices, so they will simply wear out with use (I got pretty much the last spare parts that existed in my country when it last n
615.
by Sesse__ 2y ago
journald is still very, very slow. As in, getting the entries for a given process can take a minute or more, whereas just grepping a normal text file is nearly instant. It doesn't help that the default rotation settings are insane (lik
616.
by Sesse__ 2y ago
> The article, and Oklab, is not by a color scientist. He is/was a video game developer taking some time between jobs to do something on a lark. As a non-color scientist sometimes dealing with color, it would probably be nice if the
617.
by Sesse__ 2y ago
This is very true. And to make matters even more confusing, certain drivers were at some point capable of constant-folding shaders on-the-fly so that ubershaders are essentially moved back into a bunch of specialized shaders, negating the e
618.
by Sesse__ 2y ago
Basically, not at all. Dolphin's ubershader does one thing; simulate fixed-function blending/texturing using modern flexible hardware. (It was already an old technique when Dolphin adopted it, by the way.) This project is a comple
619.
by Sesse__ 2y ago
Not really. Debian has a Rust policy, which among others prohibits this kind of extreme vendoring. Upstream insists on exact versioning (thus requiring vendoring), including hacked-up versions of some upstream crates, and also insists that
620.
by Sesse__ 2y ago
> You might argue that is wasteful of space pushing these duplicates, but your other options are either to graft this functionality on to a normal priority queue in which case you're using that space anyway, or to use a much more co
621.
by Sesse__ 2y ago
> And investigated strange issues in results instead of drowning readers in data. The basic problem of Phoronix is that it doesn't have the capacity nor competence to do this. Journalism is expensive and takes time, and Phoronix is
622.
by Sesse__ 2y ago
Note that this applies to A5/2; the stronger A5/1 took much longer to break, and this is an attack on A5/3 which was backported from 3G (UMTS) to GSM/GPRS.
623.
by Sesse__ 2y ago
std::priority_queue is sorely missing the operation “change the priority of this element” (you need to do it using a delete and then a new insert, which is rather slow), which comes up all the time in e.g. Dijkstra's algorithm.
624.
by Sesse__ 2y ago
Or stated equivalently: The only operations allowed on elements are binary comparisons and two-element swaps, but both are O(1).
625.
by Sesse__ 2y ago
Extremely sad. There basically is nothing like Anandtech; the depth, the ability to explain, the lack of sensationalism, and the integrity in benchmarking (I still vividly remember when they noticed an issue with HPET in Windows affecting t
626.
by Sesse__ 2y ago
perf or Intel VTune are the two standard choices AFAIK. Both have a certain learning curve, both are extremely capable in the right hands. (Well, on macOS you're pretty much locked to using Instruments; I don't know if Callgrind w
627.
by Sesse__ 2y ago
MySQL doesn't do this anymore; the ONLY_FULL_GROUP_BY mode became default in 5.7 (I think). You can still turn it off and get the old behavior, though.
628.
by Sesse__ 2y ago
AFAIK it's from the old Norse word “grind” meaning fence or gate. It's the entrance to Valhalla.
629.
by Sesse__ 2y ago
If the “a” is supposed to be as in “value” (i.e. closer to ä in some but not all German dialects), that doesn't match typical German pronunciation, does it? (The -grind part is fine.)
630.
by Sesse__ 2y ago
Callgrind is certainly dispensable; just use a profiler. They are much, much faster, and more accurate as well. (Callgrind is using an idealized model of CPUs as they were around 1995, which doesn't match up all that well with how they

More ›