• 10 Posts
  • 1.96K Comments
Joined 2 years ago
cake
Cake day: July 7th, 2023

help-circle






  • I 100% know what DLSS is, though by the sounds of it you don’t. It is “AI” as much as any other thing is “AI”. It uses models to “learn” what it needs to reconstruct and how to reconstruct it.

    No, you don’t. https://en.m.wikipedia.org/wiki/Deep_Learning_Super_Sampling

    This is blatantly and monumentally wrong lol. You think it’s literally rendering a dozen frames and then just picking the best one to show you out of them? Wow. Just wow lol.

    Literally in the docs: https://raw.githubusercontent.com/NVIDIA/DLSS/main/doc/DLSS_Programming_Guide_Release.pdf

    What it does is allow you to run a game at higher settings than you could usually at a given framerate, with little to no loss of image quality. Where you could previously only run a game at 20fps at 1080p Ultra settings, you can now run it at 30fps at “1080p” Ultra, whereas to hit 30fps otherwise you might have to drop everything to Low settings.

    No it doesn’t. It allows you to run a game at a higher resolution for no reason at all, instead of dropping to a lower resolution that your card can handle natively. That’s it.

    Keep claiming otherwise, and you’re just literally denying reality and the Nvidia link to the docs right in front of you.


  • Like I said…you don’t know what DLSS is, or how it works. It’s not using “AI”, that’s just marketing bullshit. Apparently it works on some people 😂

    You can find tons of info on this (why I told you to search it up), but it uses rendering tables, inference sorting, and pattern recognition to quickly render scenes with other tricks that video formats have used for ages to render images at a higher resolution cheaply from the point of view of the GPU. You render a scene a dozen times once, then it regurgitates those renders from memory again if they are shown before ejected from cache on the card. It doesn’t upsample, it does intelligently render anything new, and there is no additive anything. It seems you think it’s magic, but it’s just fast sorting memory tricks.

    Why you think it makes games better is subjective, but it solely works to run games with the same details at a higher resolution. It doesn’t improve rendered scenes whatsoever. It’s literally the same thing as lowering your resolution and increasing texture compression (same affect on cached rendered scenes), since you bring it up. The effect on the user being a higher FPS at a higher resolution which you could achieve by just lowering your resolution. It absolutely does not make a game playable while otherwise unplayable by adding details and texture definition, as you seem to be claiming.

    Go read up.











  • Actually…not true. Nvidia recently became bigger in the DC because of their terrible inference cards being bought up, but AMD overtook Intel on chips with all major cloud platforms last year, and their Xilinix chips are slowly overtaking the sales of regular CPUs for special purposes processing. By the end of this year, I bet AMD will be the most deployed brand in datacenters globally. FPGA is the only path forward in the architecture world at this point for speed and efficiency in single-purpose processing. Nvidia doesn’t have a competing product.


  • Because they choose not to go full idiot though. They could make their top-line cards to compete if they slam enough into a pipeline and require a dedicated PSU to compete, but that’s not where their product line intends to go. That’s why it’s smart.

    For reference: AMD has the most deployed GPUs on the planet as of right now. There’s a reason why it’s in every gaming console except Switch 1/2, and why OpenAI just partnered with them for chips. The goal shouldn’t just making a product that churns out results at the cost of everything else does, but to be cost-effective and efficient. Nvidia fails at that on every level.