Hacker News: eln1

NVIDIA LLM compression to save money

eln1 — Tue, 14 Apr 2026 19:19:04 +0000

Article URL: https://developer.nvidia.com/blog/cut-checkpoint-costs-with-about-30-lines-of-python-and-nvidia-nvcomp/

Comments URL: https://news.ycombinator.com/item?id=47770127

Points: 2

# Comments: 0

Google Revisits JPEG XL in Chromium After Earlier Removal

eln1 — Sun, 23 Nov 2025 06:05:51 +0000

Article URL: https://windowsreport.com/google-revisits-jpeg-xl-in-chromium-after-earlier-removal/

Comments URL: https://news.ycombinator.com/item?id=46021179

Points: 216

# Comments: 101

Another chance for JPEG XL? PDF will support format as 'preferred solution'

eln1 — Tue, 11 Nov 2025 05:39:43 +0000

Article URL: https://www.theregister.com/2025/11/10/another_chance_for_jpeg_xl/

Comments URL: https://news.ycombinator.com/item?id=45884454

Points: 6

# Comments: 1

New comment by eln1 in "Lossless LLM compression for efficient GPU inference via dynamic-length float"

eln1 — Sun, 27 Apr 2025 06:29:17 +0000

There are lots of materials about ANS, e.g. gathered here: https://encode.su/threads/2078-List-of-Asymmetric-Numeral-Sy...

Unzicker: Forget about Quantum Electrodynamics

eln1 — Thu, 09 Jun 2022 03:18:51 +0000

Article URL: https://www.youtube.com/watch?v=wvz4MRpq6xs

Comments URL: https://news.ycombinator.com/item?id=31677239

Points: 1

# Comments: 0

New comment by eln1 in "Something is wrong in the state of QED - history of physics"

eln1 — Thu, 19 May 2022 10:39:41 +0000

One can get grants in physics nearly only if being mainstream ...

Consa brings concrete arguments regarding g-factor, I still haven't seen any concrete explanation, only saying that it is fringe science because of criticism of mainstream ... but he quotes mainstream papers.

New comment by eln1 in "Something is wrong in the state of QED - history of physics"

eln1 — Thu, 19 May 2022 07:16:26 +0000

Mainstream censors non-mainstream, hence for half a century we are staying in place e.g. quantum gravity, there are appearing new problems ...

https://phys.org/news/2022-05-standard-particle-physics-brok...

The paper claims there are problems already with g-factor, could anybody explain why its objections are not valid?

New comment by eln1 in "Something is wrong in the state of QED - history of physics"

eln1 — Thu, 19 May 2022 06:53:36 +0000

Thanks, I will read, but generally beside e.g. the gravity problem, with increased accuracy there appear inaccuracies all over the standard model, so maybe it is worth revisiting QED? Are you saying the g-factor inaccuracies are not a problem?

https://phys.org/news/2022-05-standard-particle-physics-brok...

New comment by eln1 in "Something is wrong in the state of QED - history of physics"

eln1 — Thu, 19 May 2022 06:02:44 +0000

Indeed, e.g. of Dirac:

    “I must say that I am very dissatisfied with the situation because this so-called ’good theory’ does involve neglecting infinities which appear in its equations, ignoring them in an arbitrary way. This is just not sensible mathematics. Sensible mathematics involves disregarding a quantity when it is small – not neglecting it just because it is infinitely great and you do not want it!. ”

and Feynman:

    "The shell game that we play is technically called ’renormalization’. But no matter how clever the word, it is still what I would call a dippy process! Having to resort to such hocus-pocus has prevented us from proving that the theory of quantum electrodynamics is mathematically self-consistent. It’s surprising that the theory still hasn’t been proved self-consistent one way or the other by now; I suspect that renormalization is not mathematically legitimate.”

Something is wrong in the state of QED - history of physics

eln1 — Thu, 19 May 2022 05:14:26 +0000

Article URL: https://arxiv.org/abs/2110.02078

Comments URL: https://news.ycombinator.com/item?id=31431033

Points: 13

# Comments: 20

New comment by eln1 in "Microsoft granted patent for rANS - used e.g. by JPEG XL"

eln1 — Fri, 18 Feb 2022 03:59:22 +0000

Recently granted patent: https://patents.google.com/patent/US11234023B2/en

Describing rANS used e.g. in JPEG XL: https://en.wikipedia.org/wiki/Asymmetric_numeral_systems

And many others: https://encode.su/threads/2078-List-of-Asymmetric-Numeral-Sy...

Microsoft granted patent for rANS - used e.g. by JPEG XL

eln1 — Fri, 18 Feb 2022 03:58:38 +0000

Article URL: https://www.theregister.com/2022/02/17/microsoft_ans_patent/

Comments URL: https://news.ycombinator.com/item?id=30381990

Points: 10

# Comments: 1

Did Bitcoin collapse Kazakhstan? (18% of world hashrate)

eln1 — Fri, 07 Jan 2022 07:30:02 +0000

Article URL: https://twitter.com/artimusfoul/status/1479068075483873284

Comments URL: https://news.ycombinator.com/item?id=29836126

Points: 3

# Comments: 0

New comment by eln1 in "Lossless Text Compression [pdf]"

eln1 — Sat, 10 Apr 2021 06:39:44 +0000

or replacing synonyms with single representation

New comment by eln1 in "After Google, Now Microsoft Tries to Patent Asymmetric Numeral Systems"

eln1 — Mon, 08 Mar 2021 13:34:34 +0000

Your data is now written with ANS if using e.g. Apple, Facebook, Google, Linux, soon JPEG XL: https://en.wikipedia.org/wiki/Asymmetric_numeral_systems

This patent covers rANS variant which is used for example in https://en.wikipedia.org/wiki/JPEG_XL - if granted, only Microsoft will be able to make its hardware encoders/decoders.

Lots of materials about ANS: https://encode.su/threads/2078-List-of-Asymmetric-Numeral-Sy...

The Google patent story: https://arstechnica.com/features/2018/06/inventor-says-googl...

After Google, Now Microsoft Tries to Patent Asymmetric Numeral Systems

eln1 — Mon, 08 Mar 2021 13:33:29 +0000

Article URL: https://encode.su/threads/2648-Published-rANS-patent-by-Storeleap/page5

Comments URL: https://news.ycombinator.com/item?id=26385484

Points: 5

# Comments: 1

New comment by eln1 in "JPEG Committee releases a call for evidence for image compression based on AI"

eln1 — Wed, 19 Feb 2020 21:21:48 +0000

Also got this ANS coding: https://en.wikipedia.org/wiki/Asymmetric_numeral_systems

New comment by eln1 in "FLIF – Free Lossless Image Format"

eln1 — Fri, 07 Feb 2020 10:41:50 +0000

Jon Sneyers is currently working on JPEG XL, which uses ANS: https://www.spiedigitallibrary.org/conference-proceedings-of...

New comment by eln1 in "Google is trying to patent video compression use of Asymmetric Numeral Systems"

eln1 — Thu, 15 Jun 2017 05:50:40 +0000

Here are 400+ comments on this topic: https://www.reddit.com/r/programming/comments/6h08z5/google_...

Google is trying to patent video compression use of Asymmetric Numeral Systems

eln1 — Tue, 13 Jun 2017 08:37:46 +0000

Asymmetric Numeral Systems ([1]) is entropy coding family currently replacing Huffman and arithmetic coding in data compressors, among others, of Apple, Facebook and Google, thanks to being up to 30x faster [2]. Its author has made it public to prevent pathology of arithmetic coding, which wide use was blocked by patents for many decades ([3], [4]).

However, currently others are trying to patent basic applications of ANS – including Google for AV1 video compressor (initially suggested by ANS author, who has helped them for the last 3 years: [5]) in very general patent application, to prevent others from using it in image and video compression – claims and sources: [6]

[1] https://en.wikipedia.org/wiki/Asymmetric_Numeral_Systems

[2] benchmarks: https://sites.google.com/site/powturbo/entropy-coder

[3] https://en.wikipedia.org/wiki/Arithmetic_coding#US_patents

[4] Charles Bloom comment: http://cbloomrants.blogspot.com/2015/05/05-21-15-software-patents-are-fucking.html

[5] https://groups.google.com/a/webmproject.org/forum/#!topic/codec-devel/idezdUoV1yY

[6] Google claims and sources: https://encode.ru/threads/2648-Published-rANS-patent-by-Storeleap/page3

Comments URL: https://news.ycombinator.com/item?id=14543565

Points: 36

# Comments: 3