Hacker News: EnPissant

New comment by EnPissant in "Memory has grown to nearly two-thirds of AI chip component costs"

EnPissant — Sun, 24 May 2026 19:11:19 +0000

Also, electricity isn't free.

New comment by EnPissant in "Memory has grown to nearly two-thirds of AI chip component costs"

EnPissant — Sun, 24 May 2026 18:10:45 +0000

There was only a very brief time it was selling for MSRP (last fall for $2000). Even if you use that as the previous data point, it's only 200% increased.

New comment by EnPissant in "Green card seekers must leave U.S. to apply, Trump administration says"

EnPissant — Sat, 23 May 2026 15:38:46 +0000

> I hear "I'm not anti immigrant, I'm anti illegal immigrant" a lot. To which there is an easy solution: increase the number of legal immigrants we allow.

Being "anti illegal immigrant" doesn't have to imply you let in whoever wants as long as they follow some process. You are taking away the agency of the people to select its immigrants.

New comment by EnPissant in "Green card seekers must leave U.S. to apply, Trump administration says"

EnPissant — Sat, 23 May 2026 15:27:58 +0000

The following things are not in contradiction:

1) Someone can be against illegal immigration and for legal immigration.

2) That same person's idea about who should immigrate to the country may exclude most or all of the people who are currently immigrating illegally.

It's not like you can only be against illegal immigration because they forgot to fill out some form. Legal immigration has a component of deciding who gets in.

New comment by EnPissant in "California to begin ticketing driverless cars that violate traffic laws"

EnPissant — Sat, 02 May 2026 23:18:06 +0000

Now try applying this logic to elevators.

New comment by EnPissant in "SWE-bench Verified no longer measures frontier coding capabilities"

EnPissant — Sun, 26 Apr 2026 20:39:11 +0000

> 1. SWE-bench Verified is now saturated at 93.9% (congrats Anthropic), but anyone who hasn't reached that number yet still has more room for growth.

But if some or all players are bench-maxing it, then it becomes a much less useful metric for comparison.

Also, this doesn't address what OpenAI says about the test suite disallowing valid solutions.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 20:19:21 +0000

But you aren't trying out the model. You quantized beyond what people generally say is acceptable, and reduced the number of experts, which these models are not designed for.

Even worse, the github repo advertises:

> Pure C/Metal inference engine that runs Qwen3.5-397B-A17B (a 397 billion parameter Mixture-of-Experts model) on a MacBook Pro with 48GB RAM at 4.4+ tokens/second with production-quality output including tool calling.

Hiding the fact that active params is _not_ 17B.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 18:18:21 +0000

Yeah, I am mostly only talking about the SSD bottleneck being too slow. No way Apple gets 17GB/s sustained. SSDs thermally throttle really fast, and you have some random access involved when it needs the next expert.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 18:16:08 +0000

I don't mean to be a jerk, but 2-bit quant, reducing experts from 10 to 4, who knows if the test is running long enough for the SSD to thermal throttle, and still only getting 5.5 tokens/s does not sound useful to me.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 06:11:36 +0000

This was perfect.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 06:05:48 +0000

Clearly the top comment should be "Imagine a beowulf cluster of Deepseek v4!"

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 05:53:59 +0000

Streaming weights from RAM to GPU for prefill makes sense due to batching and pcie5 x16 is fast enough to make it worthwhile.

Streaming weights from RAM to GPU for decode makes no sense at all because batching requires multiple parallel streams.

Streaming weights from SSD _never_ makes sense because the delta between SSD and RAM is too large. There is no situation where you would not be able to fit a model in RAM and also have useful speeds from SSD.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 05:50:21 +0000

Those are quants, not distills.

New comment by EnPissant in "DeepSeek v4"

EnPissant — Fri, 24 Apr 2026 05:47:25 +0000

This should not be the top comment on every model release post. It's getting tiring.

9950X3D2 Benchmarks: The Best Desktop Performance for Linux Developers, Creators

EnPissant — Thu, 23 Apr 2026 19:35:09 +0000

Article URL: https://www.phoronix.com/review/amd-ryzen-9950x3d2-linux

Comments URL: https://news.ycombinator.com/item?id=47880583

Points: 1

# Comments: 0

New comment by EnPissant in "Qwen/Qwen3.6-27B · Hugging Face"

EnPissant — Wed, 22 Apr 2026 20:14:18 +0000

Are nvfp4 / mxfp4 even useful without QAT?

New comment by EnPissant in "Landmark ancient-genome study shows surprise acceleration of human evolution"

EnPissant — Sun, 19 Apr 2026 07:48:35 +0000

And yet when you look at corruption perception indexes, they largely track the Hajnal line. As do so many other graphs.

New comment by EnPissant in "Landmark ancient-genome study shows surprise acceleration of human evolution"

EnPissant — Sun, 19 Apr 2026 06:04:42 +0000

There are multiple factors at play:

- Among those who believe in intellectual differences among human groups, very few believe Europeans are the most intelligent group. The prevailing opinion you would find is that both Ashkenazi Jews and East Asians (your second example) are more intelligent on average.

- Northwest Europeans encountered intense selection over a Millennia starting around 300AD through a couple of mechanisms: The Church banning cousin marriage, and Biparte Manorialism. This resulted in the destruction of kinship networks, established the nuclear family, and selected for the high-trust peoples that enabled a kind of society you can still only find among those peoples.

New comment by EnPissant in "Silicon Valley is turning scientists into exploited gig workers?"

EnPissant — Fri, 17 Apr 2026 17:14:45 +0000

> He characterized Stanford and MIT as “mainly political lobbying operations fighting American innovation at this point” and vowed that universities would “pay the price” after “they declared war on 70% of the country.”

Oh? He vowed what? To make them pay the price? Or did he just predict a cause and effect and The Nation (your source) is libeling him?

New comment by EnPissant in "Pijul a FOSS distributed version control system"

EnPissant — Sun, 12 Apr 2026 05:36:21 +0000

git sparse-checkout