Hacker News: hydroreadsstuff

New comment by hydroreadsstuff in "MacBook Pro with M5 Pro and M5 Max"

hydroreadsstuff — Wed, 04 Mar 2026 07:31:36 +0000

The 4x comes from the neural accelerators (tensor core in NVIDIA jargon). It's 4x fp16 over the vector path (And 8x compared to M1 because at some point they 2x'd the fp16 vector path). Therefore LLM prefill(context processing/TTFT), diffusion models (image gen), and e.g. video and photo effects that make use of them can be up to 4x faster. At fp16 that's the same speed at the same clock as NVIDIA. But NVIDIA still has 2xfp8 and 4xnvfp4.

Batch-1 token generation, that is often quoted, does not benefit from this. It's purely RAM bandwidth-limited.

New comment by hydroreadsstuff in "Valve reveals it’s the architect behind a push to bring Windows games to Arm"

hydroreadsstuff — Thu, 04 Dec 2025 12:39:07 +0000

afaia emulators like Fex are within 30 to 70% of native performance. On the fringes worse or better. But overall emulation seems totally fine. Plus emulator technology in general could be used for binary optimization rather than strict mappings, opening up space for more optimization.

New comment by hydroreadsstuff in "FEX-emu – Run x86 applications on ARM64 Linux devices"

hydroreadsstuff — Fri, 21 Nov 2025 12:12:47 +0000

Some companies like to stress the efficiency or performance of Arm SoCs, but really this is a hedge against more expensive x86 hardware. AMD has increased prices of mobile SoCs radically recently. I'm looking forward to having more affordable SoC options for laptops, handhelds and desktops, perhaps from Mediatek or other lower-cost vendors.

The history of the PC is one of commoditization. A fractured multi-polar landscape is detrimental to the ecosystem/productivity and should ultimately fail.

x86 emulation is an important puzzle piece, and I'm happy Valve recognizes this and sponsors it.

New comment by hydroreadsstuff in "Seoul says US must fix its visa system if it wants Korea's investments"

hydroreadsstuff — Thu, 11 Sep 2025 04:34:22 +0000

That's what it seems like. Some people here disagree with you, but I can add anecdata that my employer insisted I do no coding on such a VISA.

New comment by hydroreadsstuff in "AMD claims Arm ISA doesn't offer efficiency advantage over x86"

hydroreadsstuff — Mon, 08 Sep 2025 16:06:14 +0000

Indeed, the memory model has a decent impact. Unfortunately it's difficult to isolate in measurement. Only Apple has support for weak memory order and TSO in the same hardware.

The Next Step for AI – Full Personal Interaction Capture

hydroreadsstuff — Fri, 08 Aug 2025 15:29:22 +0000

Article URL: https://blog.automaton2000.com/2025/08/the-next-step-for-ai-full-personal.html

Comments URL: https://news.ycombinator.com/item?id=44838195

Points: 1

# Comments: 0

New comment by hydroreadsstuff in "The Llama 4 herd"

hydroreadsstuff — Sat, 05 Apr 2025 20:12:26 +0000

This means GPUs are dead for local enthusiast AI. And SoCs with big RAM are in.

Because 17B active parameters should reach enough performance on 256bit LPDDR5x.

New comment by hydroreadsstuff in "Undergraduate shows that searches within hash tables can be much faster"

hydroreadsstuff — Tue, 11 Feb 2025 06:53:30 +0000

Overallcoation has a limit. You only have so much RAM/storage. Beyond that you start swapping. I could really use a hash table (or similar structure) that degrades less with higher occupancy.

New comment by hydroreadsstuff in "Measuring Apple CPU Core-to-Core Latency Without Core-Pinning"

hydroreadsstuff — Wed, 04 Dec 2024 22:28:46 +0000

For other platforms there is already https://github.com/nviennot/core-to-core-latency

This project works around a limitation in MacOS, namely lack of thread pinning, and makes it possible to do it without Asahi Linux, and without using custom MacOS kernel extensions and disabling security features.

Measuring Apple CPU Core-to-Core Latency Without Core-Pinning

hydroreadsstuff — Wed, 04 Dec 2024 21:36:12 +0000

Article URL: https://github.com/hydroo/macos-core-to-core-latency

Comments URL: https://news.ycombinator.com/item?id=42322286

Points: 4

# Comments: 2

Biased AI Code Completion Example: calculateWomanSalary

hydroreadsstuff — Tue, 26 Nov 2024 08:46:24 +0000

Article URL: https://twitter.com/s3nhs3nh/status/1861126547907739821

Comments URL: https://news.ycombinator.com/item?id=42243724

Points: 1

# Comments: 0

New comment by hydroreadsstuff in "xAI announces series B funding round of $6B"

hydroreadsstuff — Mon, 27 May 2024 08:22:01 +0000

Did they raise 6B or are they valued at 6B and don't disclose the raised amount? Probably the latter?

New comment by hydroreadsstuff in "Apple apologizes for iPad 'Crush' ad that 'missed the mark'"

hydroreadsstuff — Fri, 10 May 2024 12:00:25 +0000

If that was the intend an animation that doesn't literally crush things would have worked much better. Let it fall into a black hole and let an iPad emerge or whatever. The dramatic effect of a hydraulic press adds nothing positive.

New comment by hydroreadsstuff in "Apple apologizes for iPad 'Crush' ad that 'missed the mark'"

hydroreadsstuff — Fri, 10 May 2024 11:57:49 +0000

I can tell you that I watched the recording, cringed for a few seconds and skipped it, and moved on with my life. After the outrage, revisiting my 3 seconds of feelings, I tend to agree that destroying nice things isn't a great thing to do in an ad.

New comment by hydroreadsstuff in "Micron Kicks Off Production of HBM3E Memory"

hydroreadsstuff — Wed, 28 Feb 2024 09:12:14 +0000

https://semiconductor.samsung.com/news-events/tech-blog/a-br...

New comment by hydroreadsstuff in "Bungie Devs Atmosphere 'Soul-Crushing' Amid Layoffs, Cuts, Fear of Sony Takeover"

hydroreadsstuff — Thu, 07 Dec 2023 10:10:00 +0000

Egosoft was founded in 1988 and is still doing its thing

New comment by hydroreadsstuff in "HBM’s Future: Necessary but Expensive"

hydroreadsstuff — Wed, 26 Jul 2023 14:44:16 +0000

I wish the author would have put some actual numbers on the cost of the different technologies. Arguing about tradeoffs including that information would be much more meaningful.

New comment by hydroreadsstuff in "Apple Vision Pro: Apple’s first spatial computer"

hydroreadsstuff — Tue, 06 Jun 2023 05:56:34 +0000

Why did they make the front out of glas? This weigh a lot more than plastic (or aluminium if transparency is not needed), and is probably wuite expensive, too.

New comment by hydroreadsstuff in "Intel Explores Transition to 64-Bit-Only X86S Architecture"

hydroreadsstuff — Sat, 20 May 2023 21:51:34 +0000

It’s 4GB per address space / per process. In aggregate you can use more.

New comment by hydroreadsstuff in "GitHub staff are required to use Teams by Sep 1, 2023"

hydroreadsstuff — Mon, 13 Feb 2023 21:52:03 +0000

Webex is worse. Sharepoint & powerpoint integration, large scale broadcast (300+?people uses different techniques) and transcripts are helpful. Haven’t used the chat much, but it seems not good.