This was actually the sub-headline of the article but I thought was the more important party of the article.
Speaking with developers and artists at studios that have agreed to DLSS 5, including CAPCOM and Ubisoft, Insider Gaming was told that the DLSS 5 tech was revealed to them at the same time as everyone else.
“We found out at the same time as the public,” said one Ubisoft developer.
Developers at CAPCOM tell Insider Gaming that the announcement and the publisher’s involvement were particularly shocking, as CAPCOM has previously been historically very “anti-AI” with projects such as Resident Evil Requiem and other unannounced projects in development. Some at the publisher fear that the DLSS 5 announcement could prompt a change in the publisher’s view on generative AI and its implementation in its games.
The inputs from everything Nvidia has said, are simply the final pixel colour values and motion vector information. It’s meant to sit in the same post-processing stack as the upscale. It’s effectively a screen-space post-processing filter over the final image. Nvidia have said that the artist controls are masking (blocking certain areas from it), intensity (so a slider value), and some kind of colour re-grading (since it destroys the original grading). It’s extremely limited.
And Nvidia are full of shit judging from how it clearly changes geometry in the demos, women’s faces in particular
Edit: It does not use the same inputs as DLSS 4. It only uses motion vectors and 2D color. (See https://sh.itjust.works/comment/24367414) This could change, but most of Nvidia’s marketing is currently bullshit.
If it is the same as DLSS 4 Super Resolution, it seems to use motion vectors, colour buffers, depth buffers, and camera information like exposure. That said, this might change, as, like I said, they’re showing off something they haven’t even got running on the target hardware. Its clearly not even close to being a finished product.
A tech YouTuber ended up following up with Nvidia engineers directly and it’s exactly like I said. Not even the depth buffer is part of the inputs.
Watched through the video, and you’re right. Based on the video, Nvidia’s original statements were, for all intents and purposes, lies. Ironically, since it does seem to be based on the existing DLSS stack (from what I’ve seen), it does have access to things like depth map, it just doesn’t use it. I’ll edit my prior comments.
That said, as I originally said, none of this matters anyway. This technology doesn’t run on desktop hardware. They announced a “”“gaming”“” software that can’t run on gaming hardware. It doesn’t matter what it looks like, because you can’t play games with it. Frankly, at this point, I doubt they’ll even release it.