Hacker News: computerbuster

New comment by computerbuster in "TurboQuant: Redefining AI efficiency with extreme compression"

computerbuster — Wed, 25 Mar 2026 17:02:09 +0000

JPEG XL is mainly based on unique image-specific research, but you're right to say a lot of the techniques are compatible with videos in theory (the XYB color space comes to mind). AVIF is an AV1 OBU in an image-specific container, and required a lot of image-specific engineering to make AV1's tools useful for images; see libaom's tune "iq", and the same in SVT-AV1. The compression gains translated when engineering effort went into creating bespoke implementations, and the same may happen for LLMs if I had to guess.

Same Image, Different Score?

computerbuster — Thu, 05 Feb 2026 01:18:21 +0000

Article URL: https://halide.cx/blog/chroma-handling/

Comments URL: https://news.ycombinator.com/item?id=46894356

Points: 3

# Comments: 0

New comment by computerbuster in "Fast SSIMULACRA2 Implementation in Zig"

computerbuster — Mon, 13 Oct 2025 07:11:29 +0000

Quick follow-up from the original SSIMULACRA2 author:

> The error will be much smaller than the error between ssimu2 and actual subjective quality, so I wouldn't worry about it.

New comment by computerbuster in "Oavif: Faster target quality image compression"

computerbuster — Mon, 13 Oct 2025 05:11:14 +0000

Existing tools really just need to do a better job keeping up.

New comment by computerbuster in "Oavif: Faster target quality image compression"

computerbuster — Mon, 13 Oct 2025 03:02:40 +0000

Very cool use case! 4:4:4 support + 10-bit color make AVIF very compelling here.

New comment by computerbuster in "Oavif: Faster target quality image compression"

computerbuster — Mon, 13 Oct 2025 03:01:52 +0000

I'm a big fan of JPEG XL, but even its most dedicated fans have given up the argument that it is the best for compression efficiency. AVIF's generational leap took place in August 2024 with Tune Still Picture in SVT-AV1-PSY, so much so that Google integrated it into their own encoder and has done very impressive work optimizing it further for the human visual system. JPEG XL's strongest quality is its featureset; lossless JPEG recompression, for example, is really incredible

Oavif: Faster target quality image compression

computerbuster — Sun, 12 Oct 2025 16:21:59 +0000

Article URL: https://giannirosato.com/blog/post/oavif/

Comments URL: https://news.ycombinator.com/item?id=45559372

Points: 44

# Comments: 20

New comment by computerbuster in "Fast SSIMULACRA2 Implementation in Zig"

computerbuster — Sat, 04 Oct 2025 20:51:33 +0000

Hi, author here – the README covers this in the Performance section: https://github.com/gianni-rosato/fssimu2?tab=readme-ov-file#...

If you run the `validate.py` script available in the repo, you should see correlation numbers similar to what I've pre-tested & made available in the README: fssimu2 achieves 99.97% linear correlation with the reference implementation's scores.

fssimu2 is still missing some functionality (like ICC profile reading) but the goal was to produce a production-oriented implementation that is just as useful while being much faster (example: lower memory footprint and speed improvements make fssimu2 a lot more useful in a target quality loop). For research-oriented use cases where the exact SSIMULACRA2 score is desirable, the reference implementation is a better choice. It is worth evaluating whether or not this is your use case; an implementation that is 99.97% accurate is likely just as useful to you if you are doing quality benchmarks, target quality, or something else where SSIMULACRA2's correlation to subjective human ratings is more important than the exactness of the implementation to the reference.

Fast SSIMULACRA2 Implementation in Zig

computerbuster — Fri, 03 Oct 2025 20:36:44 +0000

Article URL: https://github.com/gianni-rosato/fssimu2

Comments URL: https://news.ycombinator.com/item?id=45467503

Points: 30

# Comments: 4

New comment by computerbuster in "The Multimedia Renaissance"

computerbuster — Tue, 19 Aug 2025 06:21:38 +0000

The past two years have seen significant advancements in video and image compression, particularly with the maturation of the SVT-AV1 video encoder and improvements to AVIF image compression. These developments, coupled with faster and more accessible developer tools, have made it easier to produce high-quality compressed media. I have a lot of optimism for the future with AV2 and the potential for further community-driven innovation in open-source compression technology.

The Multimedia Renaissance

computerbuster — Tue, 19 Aug 2025 06:21:38 +0000

Article URL: https://giannirosato.com/blog/post/the-multimedia-renaissance/

Comments URL: https://news.ycombinator.com/item?id=44948770

Points: 4

# Comments: 1

Improving AVIF in Open Source

computerbuster — Mon, 14 Jul 2025 19:21:26 +0000

Article URL: https://halide.cx/blog/improving-avif-in-open-source/index.html

Comments URL: https://news.ycombinator.com/item?id=44564213

Points: 9

# Comments: 0

New comment by computerbuster in "Launch HN: Tinfoil (YC X25): Verifiable Privacy for Cloud AI"

computerbuster — Thu, 15 May 2025 19:00:25 +0000

This is an incredibly robust solution to a really pressing problem for a lot of individuals/orgs who want to use/deploy reasonably powerful LLMs without paying through the nose for hardware. Others have mentioned the hyperscalers have solutions that make some amount of sense (Azure confidential computing, AWS nitro enclaves) but if you read a bit more about Tinfoil, it is clear they want to operate with far less explicit user trust (and thus much better security). This team is setting the standard for provably private LLM inference, and to me, it makes other solutions seem half-baked by comparison. Props to this talented group of people.

End-to-end private LLM inference

computerbuster — Wed, 30 Apr 2025 07:14:23 +0000

Article URL: https://tinfoil.sh/inference

Comments URL: https://news.ycombinator.com/item?id=43842110

Points: 3

# Comments: 1

How to Compare Video Encoders

computerbuster — Sat, 15 Mar 2025 18:22:22 +0000

Article URL: https://giannirosato.com/blog/post/comparing-encoders/

Comments URL: https://news.ycombinator.com/item?id=43374279

Points: 2

# Comments: 0

Indentation Style: Notable Styles

computerbuster — Mon, 24 Feb 2025 08:08:39 +0000

Article URL: https://en.wikipedia.org/wiki/Indentation_style

Comments URL: https://news.ycombinator.com/item?id=43156932

Points: 11

# Comments: 18

New comment by computerbuster in "FFmpeg School of Assembly Language"

computerbuster — Sat, 22 Feb 2025 21:18:21 +0000

Another resource on the same topic: https://blogs.gnome.org/rbultje/2017/07/14/writing-x86-simd-...

As I'm seeing in the comments here, the usefulness of handwritten SIMD ranges from "totally unclear" to "mission critical". I'm seeing a lot on the "totally unclear" side, but not as much on the "mission critical", so I'll talk a bit about that.

FFmpeg is a pretty clear use case because of how often it is used, but I think it is easier to quantify the impact of handwriting SIMD with something like dav1d, the universal production AV1 video decoder.

dav1d is used pretty much everywhere, from major browsers to the Android operating system (superseding libgav1). A massive element of dav1d's success is its incredible speed, which is largely due to how much of the codebase is handwritten SIMD.

While I think it is a good thing that languages like Zig have built-in SIMD support, there are some use cases where it becomes necessary to do things by hand because even a potential performance delta is important to investigate. There are lines of code in dav1d that will be run trillions of times in a single day, and they need to be as fast as possible. The difference between handwritten & compiler-generated SIMD can be up to 50% in some cases, so it is important.

I happen to be somewhat involved in similar use cases, where things I write will run a lot of times. To make sure these skills stay alive, resources like the FFmpeg school of assembly language are pretty important, in my opinion.

New comment by computerbuster in "[dead]"

computerbuster — Sat, 14 Sep 2024 17:42:13 +0000

AvifHash leverages the power of AVIF to create image placeholders that are both compact and efficient.

This Proof of Concept shows promising results: at 27 characters, AvifHash outperforms BlurHash https://blurha.sh/ (using 4x3 components) in quality and detail retention. At a similar quality, BlurHash needs 54 (5x5) to 76 characters (6x6 components).

Given that AVIF decoding is done by the web engine, AvifHash is very small: the entire demo page (including parsing and re-hydration code) is only 2.3 kB gzipped.

New comment by computerbuster in "[dead]"

computerbuster — Fri, 15 Mar 2024 23:42:42 +0000

A deep dive into programming a freestanding QOI encoder using the Zig programming language.

The Kitchen Manifesto: 8 Guidelines to Achieving Chef Nirvana

computerbuster — Tue, 27 Feb 2024 02:21:11 +0000

Article URL: https://www.nateranda.com/blog/post/the-kitchen-manifesto/

Comments URL: https://news.ycombinator.com/item?id=39519346

Points: 1

# Comments: 0