edflsafoiewq

4 ms·

1.
by edflsafoiewq 18d ago
They use their own llama fork anyway, so that shouldn't matter.
2.
by edflsafoiewq 18d ago
I think the general idea is naive quantization falls apart below 4bpw but you can go lower with more sophisticated QAT-adjacent methods. Bonsai's quantization method is proprietary though.
3.
by edflsafoiewq 19d ago
If the s are small the usual geometric buffer growth mitigates that. Of course you can compute the final buffer size in this case, but often you have a bunch of dynamically-generated strings of different sizes.
4.
by edflsafoiewq 19d ago
Does Python actually do that? If the f call throws, you can still observe the (unchanged) binding of acc in reduce.
5.
by edflsafoiewq 19d ago
The binding for acc in the reduce call is still active during the f call, which means there are at least two references to acc.
6.
by edflsafoiewq 19d ago
Doesn't reduce force the accumulator to be shared though? Both the reduce and the lambda are holding onto references to acc, which defeats any "single reference" optimizations.
7.
by edflsafoiewq 21d ago
AFAICT progressive AVIF is actually using the "spatial layers" feature of AV1, which was designed to embed multiple resolutions of a video in one bitstream. The limit on four spatial layers comes directly from AV1. The previously
8.
by edflsafoiewq 22d ago
It has YCgCo too, which should have the efficiency of YCbCr while being lossless.
9.
by edflsafoiewq 22d ago
Being lossless is a function of the whole pipeline from an input image to an output image. The last step being encode to PNG doesn't guarantee anything about losslessness.
10.
by edflsafoiewq 22d ago
AFAIK WebP has never used hardware decoding and no browser uses it for AVIF now.
11.
by edflsafoiewq 22d ago
The DC coefficients in a progressive JPEG are literally an embedded thumbnail (a 1:64 copy of the original) right? Is the question if the second "layer" replaces the thumbnail or layers corrections on top of it? It wasn't cle
12.
by edflsafoiewq 22d ago
Does the difference matter?
13.
by edflsafoiewq 26d ago
I mean Peters himself was German.
14.
by edflsafoiewq 26d ago
I really like oblique projections. Have you seen this neat one from Kerkovits - A Low-Distortion Oblique Map Projection of the World’s Landmasses (2024)? https://files.catbox.moe/k11gqq.avif
15.
by edflsafoiewq 26d ago
The projection you see with your eyes is actually not that great in a sense. You can only see half of the world at a time and it's quite distorted as you move towards the edge. Also on a computer you should be able to "spin"
16.
by edflsafoiewq 26d ago
The article actually maps out the history pretty well. This is basically a recrudescence of the Gall-Peters debate from the 70s, triggered by Boston Public Schools adopting Gall-Peters in 2017. The current wave has taken up the old fight wi
17.
by edflsafoiewq 1mo ago
It was the second most reported problem on the 2025 Rust survey, after compilation times.
18.
by edflsafoiewq 1mo ago
Why is Direct2D so hard to implement in Wine?
19.
by edflsafoiewq 1mo ago
General theme: A programming language's native in-memory object format is typically optimized for random access, uniformity, and mutability (fields at fixed offsets, etc). Serialization formats for network or disk tend to be designed e
20.
by edflsafoiewq 1mo ago
Oh is that why the page is like 50 megs.
21.
by edflsafoiewq 1mo ago
That is the opposite of what GP said. The point is only the total emissions matter, the climate does not care about "fair" allocations over either people or nation-states.
22.
by edflsafoiewq 2mo ago
Sounds like it would be easier to hope for a better UPI experience for tourists.
23.
by edflsafoiewq 2mo ago
1. The output is a permutation of the input. 2. If the comparison implements a strict total order, the output is sorted according to it.
24.
by edflsafoiewq 2mo ago
The DCs giving you a 1/8 downscale is well known, but the interesting part to me was this generalizes so you can get 1/4 (resp. 1/2) downscales by using only the 2x2 (resp. 4x4) block of LF coefficients.
25.
by edflsafoiewq 2mo ago
> Modular found ways to parralize LLVM code generation by introducing novel techniques What techniques?
26.
by edflsafoiewq 2mo ago
There is already Invoke-WebRequest. Powershell 5 infamously shipped with "curl" as an alias for it. So be careful what you wish for.
27.
by edflsafoiewq 2mo ago
I interpreted that to mean "other people should be working because it will enable me personally to buy more stuff (that they produce)", not "each person should be working because working will allow that person to buy themselv
28.
by edflsafoiewq 2mo ago
This was added in 3.14. The zipfile module was updated too, which probably makes Python the easiest way to work with ZIP files with zstd compression.
29.
by edflsafoiewq 2mo ago
You can't use BigDecimal either. There are libraries whose whole purpose is just to exactly reproduce COBOL arithmetic, eg. the IBM Decimal Arithmetic Library.
30.
by edflsafoiewq 2mo ago
"Porcupine" confused with "concubine".

More ›