671–680 of 776 posts
Earlier quoted context omitted.
What is the draw to 300fps?
240Hz or higher monitors. 4K is enough for spatial resolution then it is better to increase temporal resolution. 4K at 240Hz stops looking like looking at a screen and starts looking out a window.
Unfortunately, there are no 4K displays with 200+ DPI on the market. If you want high DPI you either need pick glossy 5k@27" or go to 6k/8k.
Earlier quoted context omitted.
Digital Foundry just covered this. 3x and 4x both add additional latency on top of 2x. https://youtu.be/xpzufsxtZpA?si=hZZlX-g_nueAd7-Q
This is true, but it's worth noting that 3x was 5ms additional latency beyond original FG and 7ms for 4x, so the difference in latency between DLSS 3 FG and DLSS 4 MFG is probably imperceptible for most people.
So Nvidia's example of taking cyberpunk from 28fps to 200+ or whatever doesn't actually work. It'll still feel like 20fps sluggish watery responses even though it'll look smooth
Similar CUDA core counts for most SKUs compared to last gen (except in the 5090 vs. 4090 comparison). Similar clock speeds compared to the 40-series. The 5090 just has way more CUDA cores and uses proportionally more power compared to the 4090, when going by CUDA core comparisons and clock speed alone. All of the "massive gains" were comparing DLSS and other optimization strategies to standard hardware rendering. Som…
I started thinking today, when Nvidia seemingly keeps just magically increasing performance every two years, that they eventually have to "intel" themselves, where they haven't made any real architectural improvements in ~10 years and just suddenly power and thermals don't scale anymore and you have six generations of turds that all perform essentially the same, right?
None of this is magic. None of it is even particularly hard. There's no reason for any of it to get stuck. (Intel's problem was letting the beancounters delay EUV - no reason to expect there to be a similar mis-step from Nvidia.)
Earlier quoted context omitted.
I started thinking today, when Nvidia seemingly keeps just magically increasing performance every two years, that they eventually have to "intel" themselves, where they haven't made any real architectural improvements in ~10 years and just suddenly power and thermals don't scale anymore and you have six generations of turds that all perform essentially the same, right?
it's possible, but idk why you would expect that. just to pick an arbitrary example since steve ran some recent tests, a 1080 ti is more or less equal to a 4060 in raster performance, but needs more than double the power and a much more die area to do it. https://www.youtube.com/watch?v=ghT7G_9xyDU we do see power requirements on the high end parts every generation, but that may be to maintain the desired SKU price p…
Earlier quoted context omitted.
AMD driver quality is crap. I upgraded from GTX 1080 to RX 6950 XT because I found a good deal and I didn't want to support nvidia's scammy bullshit of launching inferior GPUs under the same names. Decided to go with AMD this time, and I had everything: black screens, resolution drops to 1024x768, total freezes, severe lags in some games (BG3) unless I downgrade the driver to a very specific version.
It is an outdated claim. I have both 4090 (workstation) and 7900XT (to play some games) and I would say that 7900XT was rock solid for me for the last year (I purchased it in Dec 2023).
It is an outdated claim.
Good for you though.
Earlier quoted context omitted.
>They are not solely reliant on process node shrinks for performance uplifts like Intel was. People who keep giving intel endless shit are probably very young and don't remember how innovative Intel was in the 90s and 00s. USB, PCI-Express, Thunderbolt, etc., all Intel inventions, plus involvement in Wifi and wireless telecom standards. They are guilty of anti competitive practices and complacency in the last years b…
Those standards are plumbing to connect things to the CPU. The last major innovations that Intel had in the CPU itself were implementing CISC in RISC with programmable microcode in the Pentium and SMT in the Pentium 4. Everything else has been fairly incremental and they were reliant on their process node advantage to stay on top. There was Itanium too, but that effort was a disaster. It likely caused Intel to stop i…
Earlier quoted context omitted.
The price of a 4090 already was ~1800-2400€ where I live (not scalper prices, the normal online Shops) We'll have to see how much they'll charge for these cards this time, but I feel like the price bump has been massively exaggerated by people on HN
MSRP went from 1959,- to 2369,-. That's quite the increase.
https://www.nvidia.com/en-us/geforce/graphics-cards/40-serie...
The cards were then sold around that ballpark you said, but that was because the shops could and they didn't say no to more profit.
We will have to wait to see what arbitrary prices the shops will set this time.
If they're not just randomly adding 400+ on top, then the card would cost roughly the same.
Earlier quoted context omitted.
If you want to run LLMs buy their H100/GB100/etc grade cards. There should be no expectation that consumer grade gaming cards will be optimal for ML use.
Yes there should be. We don’t want to pay literal 10x markup because the card is suddenly “enterprise”.