Hacker News: chessgecko

New comment by chessgecko in "Samsung is expected to more than double output of its HBM4 and HBM4E DRAM"

chessgecko — Sun, 20 Sep 2026 18:54:00 +0000

I think people might prefer the lower idle power consumption from lpddr over the better bandwidth in hbm in battery powered stuff. That said right now the price is definitely preventing us from finding out.

New comment by chessgecko in "Samsung is expected to more than double output of its HBM4 and HBM4E DRAM"

chessgecko — Sun, 20 Sep 2026 18:51:50 +0000

pretty sure its lpddr not hbm.

New comment by chessgecko in "Trump to impose $100k fee for H-1B worker visas, White House says"

chessgecko — Sat, 20 Sep 2025 18:15:28 +0000

The real issue is that housing is heavily underweighted in the cpi basket. How many people do you know that are only spending 12.9% of their after tax take home on housing, water and fuel? Only people with paid off mortgages.

New comment by chessgecko in "U.S. already has the critical minerals it needs, according to new analysis"

chessgecko — Fri, 19 Sep 2025 01:14:31 +0000

Wouldn’t say never harder, but it’s been pretty flat.

https://fred.stlouisfed.org/series/FYFRGDA188S

New comment by chessgecko in "OpenAI raises $8.3B at $300B valuation"

chessgecko — Sun, 03 Aug 2025 02:56:44 +0000

My girlfriend loves it. She’s an immigrant from Kazakhstan and apparently the Kazakh community on Threads is very active. Reminds her of home.

New comment by chessgecko in "Mistral reports on the environmental impact of LLMs"

chessgecko — Wed, 23 Jul 2025 03:47:04 +0000

if you bought a nvidia h100 at wholesale prices (around $25k) and ran it 24/7 at commercial electric rates (lets say $0.1 per kwh), then it would take you over 40 years to spend the purchase price of the gpu in electricity. Maybe bump it down to 20 for data center cooling.

I don't think the cost of the ai is close to converging to the price of power yet. Right now its mostly the price of hardware and data center space minus subsidies.

New comment by chessgecko in "Why does the U.S. always run a trade deficit?"

chessgecko — Tue, 20 May 2025 17:11:04 +0000

Actually since the qe to fix 2008 foreign investors/governments cut down on bond buying. Treasury has foreign bond holdings up about 2T since 2014 and current account deficits sum to around 7T.

Nowadays we're mostly selling companies and real estate. (though there was a nice jump in bond buying a few years when rates went up)

New comment by chessgecko in "Why does the U.S. always run a trade deficit?"

chessgecko — Tue, 20 May 2025 16:38:17 +0000

Or maybe a lot of it can't deliver returns in line with the current price. The world realizes that and stops sending us stuff for those investments. Then we're stuck with limited manufacturing, high inflation and relatively low ownership of our existing high value exports.

If the pie gets bigger too slowly then I think we still lose.

New comment by chessgecko in "Why does the U.S. always run a trade deficit?"

chessgecko — Tue, 20 May 2025 16:24:37 +0000

It is tracked! there are a few different numbers.

by the BEA: https://fred.stlouisfed.org/graph/?g=gX0f

by the treasury: https://ticdata.treasury.gov/resource-center/data-chart-cent...

It's a little hard to figure out how important it is, but there are "good" numbers.

New comment by chessgecko in "Trillium TPU Is GA"

chessgecko — Wed, 11 Dec 2024 18:05:43 +0000

1800 on the h100s is with 2/4 sparsity, it’s half of that without. Not sure if the tpu number is doing that too, but I don’t think 2/4 is used that heavily so I probably would compare without it.

New comment by chessgecko in "The GPU is not always faster"

chessgecko — Wed, 11 Dec 2024 18:01:13 +0000

I remember reading that it’s too hard to get good memory bandwidth/l2 utilization in the fancy algorithms, you need to read contiguous blocks and be able to use them repeatedly. But I also haven’t looked at the gpu blas implementations directly.

New comment by chessgecko in "Intel announces Arc B-series "Battlemage" discrete graphics with Linux support"

chessgecko — Tue, 03 Dec 2024 20:42:00 +0000

I guess it's hard to know how well this would compete with integrated gpus, especially at a reasonable pricepoint. If you wanted to spend $4000+ on it, it could be very competitive and might look something like nvidias grace-hopper superchip, but if you want the product to be under $1k I think it might be better just to buy separate cards for your graphics and ai stuff.

New comment by chessgecko in "Intel announces Arc B-series "Battlemage" discrete graphics with Linux support"

chessgecko — Tue, 03 Dec 2024 20:06:38 +0000

I think we don't really disagree, I just think that this shape isn't really a gpu its just a cpu because it isn't very good for graphics at that point.

New comment by chessgecko in "Intel announces Arc B-series "Battlemage" discrete graphics with Linux support"

chessgecko — Tue, 03 Dec 2024 19:57:03 +0000

For ai inference you definitely have other options, but for low end graphics? the lpddr that apple (and nvidia in grace) use would be super expensive to get a comparable bandwidth (think $3+/gb and to get 500GB/sec you need at least 128GB).

And that 500GB/sec is pretty low for a gpu, its like a 4070 but the memory alone would add $500+ to the cost of the inputs, not even counting the advanced packaging (getting those bandwidths out of lpddr needs organic substrate).

It's not that you can't, just when you start doing this it stops being like a graphics card and becomes like a cpu.

New comment by chessgecko in "Intel announces Arc B-series "Battlemage" discrete graphics with Linux support"

chessgecko — Tue, 03 Dec 2024 19:07:27 +0000

GDDR isnt like the ram that connects to cpu, it's much more difficult and expensive to add more. You can get up to 48GB with some expensive stacked gddr, but if you wanted to add more stacks you'd need to solve some serious signal timing related headaches that most users wouldn't benefit from.

I think the high memory local inference stuff is going to come from "AI enabled" cpus that share the memory in your computer. Apple is doing this now, but cheaper options are on the way. As a shape its just suboptimal for graphics, so it doesn't make sense for any of the gpu vendors to do it.

New comment by chessgecko in "Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference"

chessgecko — Tue, 19 Nov 2024 03:11:39 +0000

One day is doing some heavy heavy lifting here, we’re currently off by ~3-4 orders of magnitude…

New comment by chessgecko in "The withering dream of a cheap American electric car"

chessgecko — Mon, 18 Nov 2024 16:25:28 +0000

The real reason is that it's basically impossible to produce a cheap new car that is a better deal than a Toyota with 80k miles on it.

New comment by chessgecko in "Tencent Hunyuan-Large"

chessgecko — Tue, 05 Nov 2024 20:32:41 +0000

For decode steps it depends on the number of inputs you run at a time. If your batch size is 1 then it runs in line with active params, then as you get to like batch size 8 it runs in line with all params, then as you increase to 128ish it runs like the active params again.

For the context encode it’s always close to as fast as a model with a similar number of active params.

For running on your own the issue is going to be fitting all the params on your gpu. If you’re loading off disk anyways this will be faster but if this forces you to put stuff on disk it will be much slower.

New comment by chessgecko in "$2 H100s: How the GPU Rental Bubble Burst"

chessgecko — Fri, 11 Oct 2024 06:06:58 +0000

Not just gpus, the k20 was at 3.9 Tflops (fp32) and the new iPhone is at 4.3 (fp16). If you don’t need the precision it got passed by the phones

New comment by chessgecko in "$2 H100s: How the GPU Rental Bubble Burst"

chessgecko — Fri, 11 Oct 2024 05:48:55 +0000

You’d get better perf training on a current gen phone than that gpu, but it probably functions