hydroreadsstuff

Joined 8 years ago

Karma
235
Submissions
97
Member since
Oct 9
View on Hacker News

Recent submissions

Comment · 7 months ago
I like the Dark Souls model for user input - messages. https://darksouls.fandom.com/wiki/Messages Premeditated words and sentence structure. With that there is no need for moderation or anti-abuse mechanics. Not saying this is 100% applicable here. But for their use case it's a good solution.
Comment · 8 months ago
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.

Comment · 11 months ago
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.
Comment · 11 months ago
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.

Comment · 1 year ago
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.
Comment · 1 year ago
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.
Comment · 1 year ago
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.

Comment · 1 year ago
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.
Comment · 1 year ago
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.

Comment · 2 years ago
Did they raise 6B or are they valued at 6B and don't disclose the raised amount? Probably the latter?
Comment · 2 years ago
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.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton