Tuesday, March 1st 2022

NVIDIA DLSS Source Code Leaked

The mother of all cyberattacks hit NVIDIA over the weekend, putting out critical driver source-code, the ability to disable LHR for mining, and even insights into future NVIDIA hardware, such as the Blackwell architecture. An anonymous tipster sent us this screenshot showing a list of files they claim are the source-code of DLSS.

The list, which looks credible enough, includes C++ files, headers, and assets that make up DLSS. There is also a super-convenient "Programming Guide" document to help developers make sense of the code and build correctly. Our tipsters who sent this screenshot are examining the code to see the inner workings of DLSS, and whether there's any secret sauce. Do note that this is DLSS version 2.2, so a reasonably recent version including the latest DLSS 2.2 changes. This code leak could hold the key for the open-source Linux driver community to bring DLSS to the platform, or even AMD and Intel learning from its design. Stealing Intellectual Property is a big deal of course and NVIDIA's lawyers will probably be busy picking apart every new innovation from their competitors, but ultimately it'll be hard to prove in a court of law.
Add your own comment

83 Comments on NVIDIA DLSS Source Code Leaked

#51
wahdangun
birdieI've checked the sources and indeed DLSS uses Tensor Cores.

You can probably run this logic on shaders but shaders already do shaders, so any additional load will slow down your game, therefore moving this logic from free to use tensor cores to shaders could result not in a performance gain but a performance loss.

Drivers source code is also there. It's a treasure trove of highly confidential info. NVIDIA might as well make DLSS open source (not necessarily under licenses like GPL but something proprietary e.g. like Epic or Crytek do with their engines).

I've downloaded the entire archive and I'm now repacking it because it was not optimally compressed. I expect it to come off at less than 8GB vs 18GB leaked by the hackers.
but maybe, someone can port it to used avx ? and make that avx512 useful
Posted on Reply
#52
MikeMurphy
RidgeIf not it should be.
IP theft isn't something to celebrate.

The bigger problem that may arise are security vulnerabilities brought on by the release of the source code.
Posted on Reply
#53
wahdangun
mouacykIs there a potential concern for hostile/harmful takeover of remote GPU's, via either drivers or GFE?
MikeMurphyIP theft isn't something to celebrate.

The bigger problem that may arise are security vulnerabilities brought on by the release of the source code.
isn't that a good thing ?
Posted on Reply
#54
lesp4ul
heni87Corporations especially ones like nvidia with monopoloistic as well as manuplative business practices deserve it!
So crime is better than what you're accusing...
wahdangunisn't that a good thing ?
What???
Posted on Reply
#55
mouacyk
wahdangunisn't that a good thing ?
Revelations good. Remote control bad.
Posted on Reply
#56
mechtech
Chrispy_Nvidia have burned away most, if not all of their goodwill from their consumer customers that put them on the map.

In the last decade Nvidia have fragmented APIs to lockout the competition as much as possible, made proprietary shit and charged silly money for it instead of contributing to the sector as a whole (SLI, G-Sync, PhysX, RTX, CUDA, GameWorks to name just a few), contributed as little as absolutely necessary to the opensource community, obfuscated driver features as an anticompetitive measure, plied game developers with black-box, Nvidia-optimised tools instead of contributing to existing tools that were in use for cross-platform development, and all of this is just the headline stuff in the consumer market only.

If you want to know how they've micro-segmented the pro and enterprise market with arbitrary driver limitations for basic stuff like virtual machines, compute, AI, I could rant for days. It's all a massive attempt to cash-grab and monopolise.
Ya, I remember way back when I was younger and dumber, I thought they were great (not sure why), my first AMD A64 S754 had an nvidia chipset, had issues, still got another one for S939, later I got a board with a ULi chipset, then Nvidia bought them out and I thought this should be good, probably get better driver support, etc. etc. Then Nvidia basically just shut them down, and all support. That was when I stopped giving Nvidia my money.........with the exception of a GT1030 not long ago since it was the only card I could get new in the $100 range.
Posted on Reply
#57
nguyen
wolfOn Turing at least RT+ Tensor cores combined account for at most 8-10% of die area, not exactly robbing most people of a bucketload extra performance that could have been CUDA cores. I'm all for improving graphics and adding innovative features instead of strictly more/better/faster of the same.

Native isn't perfection all the time either, far from it, and in fact often DLSS can improve on aspects of Native, especially when paired with meh TAA.

Use your graphics card however you'd like, naturally, but from where I'm sitting it's 8-10% die area put to great use.
Some people just don't realize Turing's useful lifespan are extended greatly thanks to DLSS which is evolving as time pass. My 2070S Max-Q laptop still kicking ass in Dying Light 2 on an external 3440x1440p screen.

Well let hope some AMD engineers can look at the internal working of DLSS from the leaked source code and create a more competitive upscaler to DLSS.
Posted on Reply
#58
Vayra86
MikeMurphyIP theft isn't something to celebrate.

The bigger problem that may arise are security vulnerabilities brought on by the release of the source code.
Security through obscurity is pointless anyway. No harm done, the only message here is Nvidia fix your shit proper.
Posted on Reply
#59
ratirt
I'm like 100% sure, nobody will use NV specific coding for anything. How would they use it if they don't have the hardware backing it up right?
So, most of the accusations are pointless about contaminating the code. Getting an idea of how to do things is more likely although as some developers showed, same effect or goal can be achieved in many different ways. I think in general, people and customers will benefit from that leak. It really doesnt change a thing even if the code somewhat proves they've been dishonest about the hardware etc. Wouldn't be the first time either and people still love them.
Posted on Reply
#60
R-T-B
MikeMurphyThe bigger problem that may arise are security vulnerabilities brought on by the release of the source code.
Good, bugs that would've remained hidden will get patched.
Posted on Reply
#61
Camm
ratirtI'm like 100% sure, nobody will use NV specific coding for anything. How would they use it if they don't have the hardware backing it up right?
So, most of the accusations are pointless about contaminating the code. Getting an idea of how to do things is more likely although as some developers showed, same effect or goal can be achieved in many different ways. I think in general, people and customers will benefit from that leak. It really doesnt change a thing even if the code somewhat proves they've been dishonest about the hardware etc. Wouldn't be the first time either and people still love them.
They apparently have verilog designs, if it gets released I wouldn't be surprised if we suddenly start seeing some 'home grown' Chinese GPU's.
Posted on Reply
#62
ApexG
Meh... Who cares this actually makes life better in the long run for consumers.
Posted on Reply
#63
plonk420
birdieI've downloaded the entire archive and I'm now repacking it because it was not optimally compressed. I expect it to come off at less than 8GB vs 18GB leaked by the hackers.
what about separating images and less-compressables such as PDFs from the actual code?
Posted on Reply
#64
pavle
This is a what happens when a company is screwing with their customers for years, did you all forget FX series (that just mocked 3DFX), det. 44.03, FX5200, GT8400, bumpgate, and many other scams they pulled (not tried to) over us? Nothing will remain hidden!:love:
Posted on Reply
#65
Chomiq
nguyenWell let hope some AMD engineers can look at the internal working of DLSS from the leaked source code and create a more competitive upscaler to DLSS.
Last I checked that's called IP theft.
Posted on Reply
#66
nguyen
ChomiqLast I checked that's called IP theft.
I guess that's the job of the legal team to decide how much similar between 2 competing upscalers before it's considered IP theft :D, Intel XeSS operates very very similar to DLSS too
Posted on Reply
#67
KainXS
Its terrible that the code was leaked but if this gives me open source drivers or unlocked VBios for my 3080 then let it rain, let it rain.
Posted on Reply
#68
Cutechri
bobalazsLinux users prefer
Nobody really cares about them.
human_errorMuch as I dislike how locked down Nvidia keeps a lot of their features this is not something to be celebrated. I work in the software industry and this won't help any competitors - they all have to make sure that engineers don't get contaminated with source code they can't use as even just reading some of it or the overall approach can influence how they would approach a problem in an IP infringing way. Engineers who work on some projects where they can see code the wider business can't use are blacklisted from some projects for years so that no claims of IP theft can be made.

The community won't benefit from this as AMD and Intel engineers won't touch it, and any other groups who do will be sued into the ground if it looks like they've based their work on Nvidia code, even if they haven't, as they'd need to pay for a defense against lawsuits even if they're in the right.

Nvidia have spent money developing these technologies, and frustrating as it is that many are locked to the Nvidia platform it has spurred open competition - G-sync led to Freesync and VRR VISA spec, Nvidia Raytracing led to it being added to DX12 and AMD following suit, DLSS leading to FSR. Some of these may have happened anyway, but Nvidia may well have accelerated their release and pushed competitors to do more which is good for everyone.
Precisely.
Posted on Reply
#69
sith'ari
In some recent article(i've read so many about this hack-attack that i can't remember which) ,nVIDIA had informed their investors that a hacking attempt against them should be expected as a serious possibility.
Does anyone knows any more info about that ? (basically , if this info is accurate? )

Also have we concluded what exactly has happened ?
is it possible that only fractions of the source code had been leaked ,thus ,can this be a gamble-play from the hackers ,trying to profit through deceiving claims ?
(as i said in a past post i find it unthinkable that the hackers managed to overcome nVIDIA 's Morpheus security system developer.nvidia.com/morpheus-cybersecurity without anyone noticing.This system is created to detect anomalies inside a network , so still , my own belief is that these hackers have only claimed fractions of code here and there , and they are releasing deceiving statements to the press to add presure to nVIDIA )

EDIT : By the way , i just noticed this article about the same hackers claiming that they have 190GB data of Samsung's source code and that they have confidential information for Qualcomm : technews.pw/after-nvidia-lapsus-leak-190gb-of-alleged-samsung-data-source-code/
Posted on Reply
#70
R-T-B
CutechriNobody really cares about them.
I mean way more than half the internet runs on on linux or some derivative, so it's kind of foolish to ignore them. The gaming market there is hardly dead as well, making stellar platform strides in recent years thanks to people and companies that decidedly do care about them.
Posted on Reply
#71
Cutechri
R-T-BI mean way more than half the internet runs on on linux or some derivative, so it's kind of foolish to ignore them. The gaming market there is hardly dead as well, making stellar platform strides in recent years thanks to people and companies that decidedly do care about them.
I will ignore them as long as their user base seems to emigrate straight from the tenth circle of hell. OS is okay I guess (not for the average user's desktop - imagine not having drive trimming and automounting on startup by default), user base is the opposite, and I never see it improving. People just get a mad ego boner whenever they switch to Linux and have to look down on anybody who dares to use Windows. It's like people can't switch to that OS without getting an ego surge, it's crazy. Why should I care about them when they don't care about anybody else who doesn't use their OS or uses a ""bad"" version of their OS (Ubuntu, God forbid you mention Canonical next to them)? Is it any surprise I'm so fed up with these people I start to not care about their OS as a whole and leave these smug comments?

After trying that OS for a month, yeah, never again. The only Linux related experiences I'll have will be with my Android phone and my Steam Deck, unless I decide to just put Windows on that thing and call it a day.
Posted on Reply
#72
R-T-B
Cutechrinot for the average user's desktop - imagine not having drive trimming and automounting on startup by default)
Depends a lot on the distro, but that certainly isn't the case with the major kde/gnome based things these days. But being an "alternative" Linux does often find itself behind the gaming tech curve or having to reimplement the gaming-tech wheel (they still don't have HDR support for example).

I think the elitist problem is far more widespread than the linux community though, spilling over into several tech communities... which is why I largely ignore it. Heck I see it here some days. Same response; I've just learned to filter it out I guess. I won't deny it's an issue though.
Posted on Reply
#73
Cutechri
R-T-BI think the elitist problem is far more widespread than the linux community though, spilling over into several tech communities...
Yeah, it's put me off this hobby several times but I always come back. It's indeed especially worse on the Linux side of things, and I wish they would just be more accepting so more people would actually switch. You don't berate the Windows user who chose to ditch their entire workflow and switch to Ubuntu for example just because it's Ubuntu, you encourage them and help them out. But I guess that's too much to ask.
Posted on Reply
#74
R-T-B
CutechriYeah, it's put me off this hobby several times but I always come back. It's indeed especially worse on the Linux side of things, and I wish they would just be more accepting so more people would actually switch. You don't berate the Windows user who chose to ditch their entire workflow and switch to Ubuntu for example just because it's Ubuntu, you encourage them and help them out. But I guess that's too much to ask.
"I hate to say it but sometimes I feel like it's just a bunch of nerds with no social skills in this industry."

Spoken once by me, a nerd with no social skills.

Yes I quoted myself. Told you I have no social skills.
Posted on Reply
#75
Cutechri
R-T-BTold you I have no social skills.
Right there with you.
Posted on Reply
Add your own comment
Nov 21st, 2024 07:21 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts