This is an Nvidia technology running on dedicated hardware. Consoles won't have DLSS.
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Digital Foundry || Control vs DLSS: Can 540p Match 1080p Image Quality? Full Ray Tracing On RTX 2060?
- Thread starter ILikeFeet
- Start date
This is an Nvidia technology running on dedicated hardware. Consoles won't have DLSS.
They won't be getting DLSS, but there's nothing preventing Ms and Sony to come up with their own solutions, in fact Ms is already demoing one implementation that can also increase resolution from 540-1080p, and from 1080p to 4k.
Actually the tensor cores were not being utilized until now. I also thought that they were used before but that is not the case.
And you are mixing hardware acceleration which the consoles (or at least sx and lockarhart) will have. There's no reason for dedicated hardware to inherently perform better than another that has the same hardware instructions save for the number of processors available.
And yes, the tensor cores in rtx have a higher number of processors than what SX will have, but that does not mean a similar tech cannot be used, heck we don't even know how much this implementation is even using.
That's precisely why they added hardware instructions, so they are inherently the same. The number of processors makes a difference, and RTX gpus do have more processors to handle that, but the counter-argument is that on consoles the space dedicated for that is also used for other tasks.
That's exactly what the hardware they added allows. They added support for dealing with int matrix math, and support to perform multiple of these operations at once since a int is a lot smaller than a full fp32 number.
And I'm not basing solely on this quote, I'm talking about the gdc presentation where Ms first said that would be the approach they would take and talked about the performance considerations. With the modifications they made the hardware IS able to use hardware instructions to deal with machine learning workloads.
And yes, the sheer performance won't match that of a 2080ti as they have significant more processors to handle that than what SX has.
Damn, why sony just don't confirm what this hardware has or doesn't XD
And do you know how many TOPs does a 2060 have? Since it's able to run on it, I assume this implementation won't need no where near close to the sheer processing of a 2080ti.
Consoles would have to trade ALU for machine learning tasks, but ALU is usually cheap as pretty much anything is always more bandwidth bound than alu bound, especially if you are already targeting a lower resolution, likely just the bandwidth gains from that will more than compensate the ALU you lose.
But this is machine learning, Nvidia is hardly the only one researching how to upscale images using ML, there are even open source ones already;
And with Ms already researching and demoing something similar, I'd say it's a given they will have an equivalent solution for both LH and SX, even if not as good (but what they demoed already deals with sub pixel details, so it will likely be good enough)
I...I've seen the future and it is 540p!
That's crazy! Here I am sitting here glad that I bought a 1660 Super cause I didn't need ray tracing or DLSS. I still don't care for RT, but DLSS would be amazing! Would allow me to use my GPU for far longer than normal. Now I'll have to wait for the 2060 TI or 2060 Super to drop in price.
That's crazy! Here I am sitting here glad that I bought a 1660 Super cause I didn't need ray tracing or DLSS. I still don't care for RT, but DLSS would be amazing! Would allow me to use my GPU for far longer than normal. Now I'll have to wait for the 2060 TI or 2060 Super to drop in price.
It's Nvidia so maybe you'll also have the 3060 costing more than the highest end AMD card lol
There is no discussion there. I could't have been more specific about what was the intention of the dedicated hardware that was added by Microsoft and how it works in conjunction with the compute units. I provided the quotes and the links that make this clear for everyone to read. This started because someone said "this won't be on consoles." in the context of machine learning upscaling being used on next gen console games. I made clear that this should be happening at least on the XSX because Microsoft made the relevant hardware changes in order to make it possible. This says nothing about the performance relative to the RTX GPUs and how they handle their dedicated hardware solution.
Knowing Nvidia they will probably jack the prices of everything up. :/ Not looking forward to the 3080/3080 ti prices.
I love all these new technologies to influence the final image we see on a screen and the whole topic will be interesting to see evolve in the future for sure. CB and hardware upscalers were only the beginning I feel.
if a game has amd marketing , there are no nvidia effects to choose from afaik.
Loving my nvidia shield... it pretty much does the same thing to all my 480p/720p/1080p movies and media.
Not really, unless DLSS wont be adopted heavily and going by DLSS 2.0 not needed to be trained anymore, and TAA implementation 'replacement', and UE 4 having a beta branch with DLSS implementation, it will be adopted by a lot of games.
AMD's GPU division will be in a really tough spot from now.
My point was there's no dedicated hw working in conjunction with compute units on the SXS. It extends SIMD execution units to support multiple data-widths - ie. it's a building block 'of' the CUs if you prefer.
I think the question mark is more on DLSS itself, NVidia doesn't have a history of sharing this type of tech. MS has clearly demonstrated they are interested in investing in the space though so expectation would be they'll provide their own approach to it.
Assuming linear scaling, yeah this is what we have basically calculated out for ourselves as well.fyi
So if 2080 ti has ~220tops, 2060 rtx has ~100 tops and xsx 49 tops, according to this graph the rtx 2060 will do roughly ~2.83ms @ 4k, hipothetically the xsx with same algorithm should cost about 5.5ms @4k.
5.5 ms btw is still A LOT faster than running the game at 4x the resolution.
Only problem is AMD sponsored games won't use Nvidia tech. It's always a bit of a disappointment when AMD sponsors a game like Resident Evil 2 remake, which desperately needs ray traced reflections & instead AMD offers a logo & nothing more. The one cool thing they ever did was TresFX.
Special hardware means special support for these functions on the CUs of the Series X. NVIDIA has DEDICATED units for them, so they don't need to run them on the CUs.
Thats huge devs can push the XSX at 1080p and use ML to render at 4K. Can't wait for more demos like the path tracing on XSX
Not sure I understand what you're saying here. This is a physical solution added to the CU, in order to allow them to run the machine learning code faster. This physical part of the hardware will be active while it's running the machine learning code. Let's not waste more time talking about how tensor cores are a fully dedicated solution to run machine learning code, while the modified CU on the XSX are not. I'm specifically talking about how they were modified and the reason for that. This was not my point at all and it should be clear after reading what I wrote and the quotes and links I shared.
This is not about the name of the solution, but what the solution actually does and allows. Microsoft has already shown their own machine learning solution with Directml, how you name it is not relevant to what that person was wishing to see on consoles.
The CU will be occupied running the ML code, while on NVIDIA, the CUs will be free to run normal code, while Tensors run the ML code.
Again, what is the problem here? The CUs were modified physically to accelerate machine learning code. Actual hardware was added to the CU to enable them to accelerate machine learning code.
How does this differ from say... checkerboard rendering? Just curious as to what is actually happening here.
No acceleration is done here, they still run on the CUs, which means they consume regular CU TFLOPs available for graphics. While NVIDIA's solution don't use the CUs and don't consume graphics TFLOPs.
Thank you. I will stop talking to you now.
"We knew that many inference algorithms need only 8-bit and 4-bit integer positions for weights and the math operations involving those weights comprise the bulk of the performance overhead for those algorithms," says Andrew Goossen. "So we added special hardware support for this specific scenario. The result is that Series X offers 49 TOPS for 8-bit integer operations and 97 TOPS for 4-bit integer operations. Note that the weights are integers, so those are TOPS and not TFLOPs. The net result is that Series X offers unparalleled intelligence for machine learning."
Inside Xbox Series X: the full specs
This is it. After months of teaser trailers, blog posts and even the occasional leak, we can finally reveal firm, hard …
www.eurogamer.net
Dictator
I have read a couple articles and watched the DF stuff on DLSS, thanks for the hard work and good breakdown. I do have some questions however...
Has Nvidia described how the process of implementing DLSS into your game works? Do you have to apply for a license? Does NVIDIA need your codebase? Does it take a long time? Can this be back-ported to older stuff easily?
This is definitely promising tech, but obviously if it is locked behind some arduous implementation process from Nvidia, it could blunt mass adoption...
I have read a couple articles and watched the DF stuff on DLSS, thanks for the hard work and good breakdown. I do have some questions however...
Has Nvidia described how the process of implementing DLSS into your game works? Do you have to apply for a license? Does NVIDIA need your codebase? Does it take a long time? Can this be back-ported to older stuff easily?
This is definitely promising tech, but obviously if it is locked behind some arduous implementation process from Nvidia, it could blunt mass adoption...
Lol not the only one. I'm almost positive that XSX will do 4k natively 99.7% of the time
You are obviously less informed in this matter, I will refer you to this quote here:
Having quoted me, you should also read my further posts in which my point of view evolved when faced with new data/information.
This is really getting into semantics - but extending SIMD execution units for different data-widths isn't adding so much as modifying things. There's a reason SIMD implementation have done this since late 90s/early 00s - it has a relatively negligible impact on execution unit size/complexity.
But fair enough - it's 'added functionality' so I'll stop the conversation here.
Adding operational modes that process low-precision workloads more efficiently isn't exactly adding hardware (unless you're referring to the shader cores themselves versus their X1/X counterparts, in which case, sure), but that's beside the point at this juncture. Nobody was claiming the XSX GPU doesn't have the means to accelerate DirectML workloads, just that it doesn't have hardware designed for and aimed squarely at this singular purpose a la tensor cores; the latter is what myself and others who've replied to Alucardx23 were referring to with "dedicated hardware". I'm sure you can agree there's a distinction to make between machine learning-oriented shader cores and a pool of discrete machine learning-oriented cores, and that's what we were trying to illustrate to Alucardx23, as he was conflating the two. That's all. The argument was never "The XSX GPU can't do what the tensor cores do."
Late edit: You may also want to read Liabe Brave's posts here and here (the latter especially). As it turns out, the ability to use an FP32 operation to process 4x INT8 or 8x INT4 isn't something Microsoft added; rather, it's inherent to the RDNA architecture. This means, incontrovertibly and unequivocally, that the XSX GPU isn't unique in that regard: said functionality is also supported by the PS5 GPU, any RDNA2-based GPUs AMD has in R&D, and even last year's RDNA1-based Radeon 5700 XT.
Last edited:
I really wish nvidia and ms/sony would patch up their differences,pricing, or whatever. I have way more confidence with nvidia and this type of tech than AMD.
If you see dlss 2.0, native 4k feels just stupid. It's insane how much Ressources you save and can put into stuff like raytracing. It's probably also currently the only way to have raytracing, high IQ and good framerates.
Deliver Us To The Moon, Wolfenstein Youngblood, Control, Mechwarrior, Bright Memory and soon AMID Evil.
Rich is pretty hyped about DLSS being in Switch 2.
This is exactly the kind of power efficient unconventional tech Nintendo would go for.
This is exactly the kind of power efficient unconventional tech Nintendo would go for.
Feels like black magic, and when this gets into all games it'll be an absolute game changer. The improvement in the tech in just under a year makes me very excited to see where it'll be in another years time, another three years time.
I see. Thanks, not interested in those titles except maybe Control.
Hopefully gets more support soon from other games.