frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Half of UK wants to rejoin EU and backing it could drive up Labour vote

https://www.theguardian.com/world/2026/oct/03/half-want-to-rejoin-eu-and-backing-it-could-drive-u...
1•vrganj•46s ago•0 comments

Yves Meyer's wavelet transforms – The Abel Prize 2017 [video]

https://www.youtube.com/watch?v=5Fi6TdH9poU
1•joebig•2m ago•0 comments

Terminal Velocity for Claude Code. Session Orchestrator. Live Action

https://www.youtube.com/watch?v=9tfGn3isfZU
1•quazarzero•4m ago•0 comments

A Design Space Exploration of Async/Await

https://mastodon.social/@tonofcrates/117241380083291749
2•rapnie•8m ago•1 comments

Parsing Latin poetry using constraint satisfaction

https://logical.ai/arma/
1•jruohonen•10m ago•0 comments

Leashterm – a new programming language for agents

https://github.com/jessedorrestijn2-bit/leashterm
1•Jessedor•11m ago•1 comments

Factory Planner

https://factoryplanner.net/
2•mcpcpc•11m ago•1 comments

Defensibility in AI Data: Lessons from Ads

https://twitter.com/gokulr/status/2105838034646077890
1•porridgeraisin•12m ago•0 comments

Show HN: Beautiful form back end for AI agents – Nisuform

https://nisuform.com/
1•haneboxx•18m ago•0 comments

Fastify/busboy DoS via oversized multipart boundary

https://github.com/fastify/busboy/security/advisories/GHSA-xjh9-v7x6-24jw
2•joshcsimmons•20m ago•0 comments

Unikernels are unfit for production (2016)

https://bcantrill.dtrace.org/2016/01/22/unikernels-are-unfit-for-production/
2•mooreds•21m ago•0 comments

"8-pinski" – sierpinski animation and bytebeat in EIGHT bytes x86 code

https://www.pouet.net/prod.php?which=107055
1•HellMood•23m ago•1 comments

An AI agent emailed researchers for help. It told us why

https://www.science.org/content/article/exclusive-ai-agent-emailed-hundreds-researchers-help-it-t...
3•sbulaev•27m ago•0 comments

Why Do Planes Fly?

https://whyplanesfly.com/
1•sbulaev•28m ago•0 comments

Mystery AI Hype Theater 3000 [video]

https://dair-institute.org/maiht3k/
1•mooreds•30m ago•0 comments

A democracy unable to learn cannot endure

https://steady.page/en/cognitive-republic/posts/835a766f-3ecf-4358-90f7-8bb35bd246e2
1•smomara•31m ago•0 comments

Martin Fowler and Kent Beck: Frameworks for reinventing software [video]

https://www.youtube.com/watch?v=CZs8J1ZD0CE
1•mooreds•31m ago•0 comments

A document layer for AI agents

https://www.claix.dev
1•gael_dev•31m ago•0 comments

ChatGPT: Personal Knowledgebase with Custom Content Types

https://community.openai.com/t/rfe-personal-knowledgebase-with-custom-content-types/1403061
1•earlybuilder•31m ago•0 comments

Show HN: UOS – an open-source post-quantum web OS in the browser (Angular 19)

https://github.com/uOS-sh/uOS
1•ultimaos•31m ago•0 comments

GitHub's new dashboard experience now the default

https://github.blog/changelog/2026-10-01-new-dashboard-experience-now-the-default/
2•vanyauhalin•35m ago•0 comments

I track privacy services by what happened to them, not a star rating

https://dontkyc.me/
1•thorarganaut•42m ago•0 comments

Brain-inspired imaging system that can 'see' through fog

https://www.brown.edu/news/2026-10-01/fog-imaging
1•giuliomagnifico•49m ago•0 comments

Kolibri Has Landed: A Sovereign Open-Weight Model

https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/
9•bastitx•58m ago•0 comments

A countermeasure for the counter-cognition web

https://www.armanluthra.me/essays/counter-cognition
1•armanluthra26•1h ago•0 comments

Krillion – Word Game

https://krillion.io/
1•willx86•1h ago•0 comments

Show HN: Algomaya – Paper trading and algo trading simulator for beginners

https://www.algomaya.com
1•yashlohade2502•1h ago•0 comments

Ask HN: Why don't we just use hashcat to evaluate GPU performance

2•nodesocket•1h ago•2 comments

Google Still Can't Make an Accessibility Menu Accessible

https://fireborn.mataroa.blog/blog/google-still-cant-make-an-accessibility-menu-accessible/
3•devinprater•1h ago•1 comments

Hot Flashing Guide Rev. 2.0 (2004)

https://archive.techarp.com/showarticle504a.html?pgno=0
1•userbinator•1h ago•0 comments
Open in hackernews

Ask HN: Why don't we just use hashcat to evaluate GPU performance

2•nodesocket•1h ago
I ran the following on a M4 Pro (20 core GPU) in MacOS vs a 4070 Super in Windows 11 using WSL and was kinda shocked to see the dramatic difference in GPU performance.

    ## generate a random sha256 8 alphanumeric
    python3 - <<'PY' > hashes-sha256.txt
    import secrets
    import string
    import hashlib

    alphabet = string.ascii_letters + string.digits
    password = ''.join(secrets.choice(alphabet) for _ in range(8))
    print(hashlib.sha256(password.encode()).hexdigest())
Hashcat command:

    hashcat -m 1400 -a 3 -O \
      --potfile-path eight-alphanumeric-sha256.pot \
      -1 '?l?u?d' \
      hashes-sha256.txt '?1?1?1?1?1?1?1?1'
Somewhat surprisingly the 4070 Super showed 68,000 MH/s and my MacMini M4 Pro only got 12,500 MH/s. The 4070 Super even in WSL is 467% faster than the M4 Pro (top skew model).

Comments

davydm•1h ago
That's fine if you're looking at running hashcat or some other computation like that, but tells you nothing about gaming performance.
nodesocket•1h ago
Fair, I get the performance of DX12 or OpenGL may vary widely, but I was thinking for AI applications though further research on my part showed a dramatic math difference.

Sha256 is mostly integer additions, bit rotations, XOR, AND operations. While AI workload are matrix multiplication and accumulation. I sort of answered my own question then. :)