Denuvo plans to offer independent benchmarks to prove its DRM doesn't cause performance problems

P

polonyc2

Fully [H]
Joined
Oct 25, 2004
Messages
25,860
Steve Huin, chief operating officer claimed that comparisons using cracked versions of games that load faster and run more smoothly are inaccurate, because they're rarely based on the exact same version of a game. "There might be over the lifetime of the game a protected and unprotected version," Huin said, "but these are not comparable because these are different builds over six months, many bug fixes, etc., which could make it better or worse"

Huin is aware that simply saying this won't prove anything to the legion of players who blame Denuvo for every framerate drop they experience. "Our voice is unfortunately not sufficient to convince people because we're not trusted in their mind as a starting point in that debate," he said

Irdeto has a plan, however: a program that will offer media outlets two versions of games to benchmark independently, with and without Denuvo Anti-Tamper, which he believes will prove "the performance is comparable, identical" between both. Apparently they hope to begin it within the next few months...

https://arstechnica.com/gaming/2023/07/denuvo-wants-to-convince-you-its-drm-isnt-evil/
 
Unless the reviewer is respectable like Digital Foundry I am probably not too interested. I have a feeling the benchmarks they provide will only show something like average frame rate. Load time, frame time, etc. would be needed.
 
Flogger23m said:
Unless the reviewer is respectable like Digital Foundry I am probably not too interested. I have a feeling the benchmarks they provide will only show something like average frame rate. Load time, frame time, etc. would be needed.
Click to expand...
I won't care even if the reviewer were respectable, irrespective of the results. I don't want that manure on my system.
 
I believe that DRM is inherently wrong and serves no purpose in the long-term existence of humanity. The next one hundred years will be more kind to non-DRM software than DRM software. People will simply be able to play games and use software, even if something has happened to the DRM-tracking servers or companies that are selling the software.
 
Last edited:
Look at the gamer, he wants to take off his shackles again, how dare he not like our ultra comfortable 10.000 supershackle?

A small benchmark on 10 seconds of gameplay just won't do, I want full analysis of load times, and frametimes for at least an hour of gameplay, including first launch of the game. On multiple different CPU and GPU and OS configurations. Anything less you are not convincing me of shit.

Oh and I want independent verification that the two versions they provide are actually different and one is certifiably DRM free.
 
So if I understand correctly the argument is... cracked versions use a earlier version of a game that has better performance?

That means we gamers and developers for that matter have been doing things backwards all these years... I'm not installing game patches anymore. Sounds like they degrade performance, its that first launch binary we want... the golden performance launch binary.

It is logical though... their DRM bolts on extra calculation cycles that invade the lowest level of your hardware and demand attention every second of execution. Why would that reduce performance? Saying it can do nothing less sounds like a conspiracy theory.
 
M76 said:
Look at the gamer, he wants to take off his shackles again, how dare he not like our ultra comfortable 10.000 supershackle?

A small benchmark on 10 seconds of gameplay just won't do, I want full analysis of load times, and frametimes for at least an hour of gameplay, including first launch of the game. On multiple different CPU and GPU and OS configurations. Anything less you are not convincing me of shit.

Oh and I want independent verification that the two versions they provide are actually different and one is certifiably DRM free.
Click to expand...
This, 100%

Either they show us in detail what's going on that can be independently verified, or I call shenanigans (plus lies, lies, and marketing)
 
From their perspective they are doing the right thing, or rather the only thing they can do.

From our perspective it is very likely that we will find nitpicks with the tests performed, with the nature of the results presentation and overall with the level of detail of documenting what exactly happens during those tests.
 
sfsuphysics said:
so DRM argument aside, why in the absolute F*** did it take this long for any sort of response? Denuvo making games worse/slow/etc has been an argument/observed since it basically came out.
Click to expand...
This is a good question. Why now, and not before?
 
Mr. Bluntman said:
This is a good question. Why now, and not before?
Click to expand...

Well, they now have two products:
- anti-copy
- anti-cheat

Anti-cheat will have more runtime performance impact. They might face customer concerns hindering their sales.
 
Regardless of what they do. I have done my own tests. The DRM screws up performance. It has to hurt sales more than help. I won't buy games with it.
 
M76 said:
Oh and I want independent verification that the two versions they provide are actually different and one is certifiably DRM free.
Click to expand...
On second thought I think this is impossible. It is unverifiable that the exe they provided is drm free. Unless an independent, reliable observer is present when it is compiled and looks through the code that the drm is indeed not included. But the only people familiar with the code are the developers who have a vested interest in the outcome.
 
Denuvo has shown consistently that it hurts the performance of games. This has been going on for as long as games have included Denuvo. It's not just frame rates, but loading times as well. Game executables are usually hundreds of megabytes larger and takes almost twice as long to load a game, compared to not having Denuvo. I'd imagine with the recent release of PC games with terrible performance that many people are blaming Denuvo, and rightfully so. We know Resident Evil 8 has massive performance issues, until the crack community removed Denuvo and the performance of the game was greatly increased. Which then prompted Capcom to fix the issue, but not until after the game was cracked. We have the benchmarks, we don't need Denuvo.


 
M76 said:
Look at the gamer, he wants to take off his shackles again, how dare he not like our ultra comfortable 10.000 supershackle?

A small benchmark on 10 seconds of gameplay just won't do, I want full analysis of load times, and frametimes for at least an hour of gameplay, including first launch of the game. On multiple different CPU and GPU and OS configurations. Anything less you are not convincing me of shit.

Oh and I want independent verification that the two versions they provide are actually different and one is certifiably DRM free.
Click to expand...

I forgot to mention that. You can't just use a top of the line PC. They need to run an "average" build and and a low end build. 5-7 year old CPU and something like a GTX 1060. Even if there is no performance drop on a high end set up, if there is on a low end set up it is essentially raising the minimum specifications which is of course a performance cost.
 
I have severe doubts about this program of theirs. Assuming they actually go through with it, they'd need to do a lot to prove it's actually independent testing and that they're not overseeing or restricting the tests in any way. They would also need to give outlets access to basically any game they want and not just a handful that Denuvo can guarantee are optimized well enough to show no difference. Even then, the results really don't matter all that much. Even if every single game that ever used Denuvo can provably show no system impact, people still are not going to want DRM on their titles. There is nothing Denuvo can do to change that, they will always been seen as scum.
 
ChadD said:
So if I understand correctly the argument is... cracked versions use a earlier version of a game that has better performance?
Click to expand...
I don't think so? I think it's the opposite, comparing release day versions (with DRM) vs the eventually cracked release which usually is a later patch. Or comparing the last release with DRM to the release after the developer dropped DRM, but there's usually other changes too; not a lot of developers are going to go through a release that only drops DRM, because a release is a hassle.
 
toast0 said:
I don't think so? I think it's the opposite, comparing release day versions (with DRM) vs the eventually cracked release which usually is a later patch. Or comparing the last release with DRM to the release after the developer dropped DRM, but there's usually other changes too; not a lot of developers are going to go through a release that only drops DRM, because a release is a hassle.
Click to expand...
I guess that sort of makes sense... seeing as a lot of developers drop DRM after a few weeks.
I was more thinking of the ones that ride with DRM for a lot longer.
One of the draw backs of using a cracked game is not getting update support. I would think its more common for people using cracks to be on out of date versions of a game.

Anyway we all know exactly what DRM does to game performance. :) Canned tests run by the DRM company isn't going to change opinions.
 
They aren't going do this unless they can find a test case that fits their narrative. So even if they have a successful test, why would anyone believe it would apply to anything but their test case?

The only real test case that might even be partially believable is if several game companies without any extra ties or special deals with denuvo put out a game on steam + GoG at the same time with steam having Denuvo and GoG without, benchmarking both, then having a crack group attempt to crack the denuvo version to verify that they actually did implement the DRM in the steam version instead of a version of denuvo that did basically nothing to match performance with the drm-free version. But that test would be impossible to setup because no one would trust any of the sides due to their self interests.
 
Flogger23m said:
Unless the reviewer is respectable like Digital Foundry I am probably not too interested. I have a feeling the benchmarks they provide will only show something like average frame rate. Load time, frame time, etc. would be needed.
Click to expand...
The way the answers were phrased in the OP interview makes it sound exactly like they're just referring to average framerate, which is completely useless. Gamers of new releases are complaining about the stuttering, which is something the average FPS is not going to illustrate. Frametime analysis showing those few frames that spike to 80ms out of the 4,000 others running fine around 10-16ms are what the issue is.
uOpt said:
Well, they now have two products:
- anti-copy
- anti-cheat

Anti-cheat will have more runtime performance impact. They might face customer concerns hindering their sales.
Click to expand...
Their DRM is called Denuvo Anti-Tamper, not anti-copy.
 
LurkerLito said:
They aren't going do this unless they can find a test case that fits their narrative. So even if they have a successful test, why would anyone believe it would apply to anything but their test case?

The only real test case that might even be partially believable is if several game companies without any extra ties or special deals with denuvo put out a game on steam + GoG at the same time with steam having Denuvo and GoG without, benchmarking both, then having a crack group attempt to crack the denuvo version to verify that they actually did implement the DRM in the steam version instead of a version of denuvo that did basically nothing to match performance with the drm-free version. But that test would be impossible to setup because no one would trust any of the sides due to their self interests.
Click to expand...
There isn't even really any need.

The idea that a DRM can be constantly checking authenticity.... and not effect performance is of course silly.

You are asking the system to do work. Work is never free. The best they can prove is that they don't greatly impact performance. For most people even a minor impact is still unacceptable.

Game publishers have already decided that noticeable performance hits are acceptable. Which leads us to the same conclusion why bother trying to prove anything? The publishers are all still going to demand its inclusion, and gamers are always going to know there is an impact.
 
it seems batman Arkham night on steam still has Denuvo, its still listed but not sure that is correct? is also on GOG, if anyone has both do a test :)
 
ChadD said:
There isn't even really any need.

The idea that a DRM can be constantly checking authenticity.... and not effect performance is of course silly.

You are asking the system to do work. Work is never free. The best they can prove is that they don't greatly impact performance. For most people even a minor impact is still unacceptable.

Game publishers have already decided that noticeable performance hits are acceptable. Which leads us to the same conclusion why bother trying to prove anything? The publishers are all still going to demand its inclusion, and gamers are always going to know there is an impact.
Click to expand...
You're right, but their metric IIRC is not "no effect on performance". What denuvo is claiming is the performance delta is insignificant (1%-2% or less IIRC) and unnoticeable.

Newbie_52 said:
it seems batman Arkham night on steam still has Denuvo, its still listed but not sure that is correct? is also on GOG, if anyone has both do a test :)
Click to expand...
I thought it was removed from the steam version at around the same time the GoG version was released?
 
Newbie_52 said:
it seems batman Arkham night on steam still has Denuvo, its still listed but not sure that is correct? is also on GOG, if anyone has both do a test :)
Click to expand...

In 2019, an Ars analysis comparing a Denuvo-free copy of Batman: Arkham Knight on the Epic Games Store to a Denuvo-laden copy from Steam found no difference in performance...

https://arstechnica.com/gaming/2019...vo-drm-doesnt-slow-down-batman-arkham-knight/

another video comparing the Denuvo version to non-Denuvo showing basically no difference...

 
Last edited:
Lakados said:
When I said I thought this was done before this may be the one I was thinking of.
Click to expand...

There are many versions of Denuvo, I assume newer builds may perform differently. That is something like a 25% increase in load time at the minimum. Some developers have also implemented it poorly in the past; it is another thing they must focus on optimizing. Given the state many games release in that they should probably spend time on other things.

If they wanted to do it properly, they would release a modern game with a modern build of Denuvo on Steam (or GOG) that has the option for Denuvo or non-Denuvo. That way everyone can test it and all of the big review websites that are reputable will give us some good insight. For Steam they can do it for a released and patched game, like, Guardians of the Galaxy and remove Denuvo and have the Denuvo build an optional download under the beta option.
 
Games that include Denuvo won't perform at all if the DRM is unhappy for any reason (e.g., it can't phone home). That qualifies as a pretty severe performance problem. Games with Denuvo also won't perform if they don't exist on one's PC in the first place, which could be a problem for any company that wants to sell me a game with that garbage included. There are plenty of alternatives available. I may belong to a small minority, but I'm done dealing with useless bloatware gaming clients and services just to play single-player offline games. They're a nightmare from a security and privacy perspective. Adding additional layers of DRM isn't going to convince me to open my wallet. If a game isn't available from GOG, I probably won't even bother with it. I'm willing to spend more money on a game without DRM, and I'm not alone in that regard. For the record, I don't pirate video games. If a game includes DRM that I'm not willing to accept, I simply won't play it.

Ideally, game developers and publishers that insist on including Denuvo (or any DRM) would remove it after some "reasonable" length of time following the game's initial release, and also be consistent about doing so. While Denuvo has been removed from some games shortly after release, in others it is still present many years later. Dragon Age: Inquisition, which was released roughly eight years ago, is one such example. So there is a lack of consistency. Unforeseen complications can arise in hardware or software environments that weren't anticipated or tested during release (e.g., Alder Lake), and the likelihood of such occurrences increases with the age of the game.

Games that incorporate Denuvo already incorporate some other form of DRM. Usually that means a privileged service running in the background, at minimum, so Denuvo is just an extra layer of DRM with a non-zero performance penalty and the potential to cause problems. Even if I accept that the performance hit is negligible in a given implementation, I don't care. I don't care if it's effectively zero, because it offers no benefit to me. They'd have to demonstrate a performance increase to even claim that their DRM offers anything but a negative value to the purchaser of the video game. If people stop buying games with Denuvo, it will cease to be a problem.

Edit: meant negative, not non-negative value. 😴
 
Last edited:
ElementDave said:
Games that include Denuvo won't perform at all if the DRM is unhappy for any reason (e.g., it can't phone home). That qualifies as a pretty severe performance problem. Games with Denuvo also won't perform if they don't exist on one's PC in the first place, which could be a problem for any company that wants to sell me a game with that garbage included. There are plenty of alternatives available. I may belong to a small minority, but I'm done dealing with useless bloatware gaming clients and services just to play single-player offline games. They're a nightmare from a security and privacy perspective. Adding additional layers of DRM isn't going to convince me to open my wallet. If a game isn't available from GOG, I probably won't even bother with it. I'm willing to spend more money on a game without DRM, and I'm not alone in that regard. For the record, I don't pirate video games. If a game includes DRM that I'm not willing to accept, I simply won't play it.

Ideally, game developers and publishers that insist on including Denuvo (or any DRM) would remove it after some "reasonable" length of time following the game's initial release, and also be consistent about doing so. While Denuvo has been removed from some games shortly after release, in others it is still present many years later. Dragon Age: Inquisition, which was released roughly eight years ago, is one such example. So there is a lack of consistency. Unforeseen complications can arise in hardware or software environments that weren't anticipated or tested during release (e.g., Alder Lake), and the likelihood of such occurrences increases with the age of the game.

Games that incorporate Denuvo already incorporate some other form of DRM. Usually that means a privileged service running in the background, at minimum, so Denuvo is just an extra layer of DRM with a non-zero performance penalty and the potential to cause problems. Even if I accept that the performance hit is negligible in a given implementation, I don't care. I don't care if it's effectively zero, because it offers no benefit to me. They'd have to demonstrate a performance increase to even claim that their DRM offers anything but a non-negative value to the purchaser of the video game. If people stop buying games with Denuvo, it will cease to be a problem.
Click to expand...
I agree that leaving the DRM in is an issue, especially for game preservation. It is depressing that publishers and developers largely don't care if their games are lost to history. I read a report recently that 87% of games ever made are lost to history, and DRM contributes to that number. I think it should be legally required for DRM to be removed from games at least after the game is no longer being sold.

All that being said, I can't recall a single case when Denuvo was causing performance issues. RE8 is always brought up, but in that case it was because Capcom was using their own DRM in addition to Denuvo that literally checked every pixel of every frame as it was rendered. When that was removed, the game performed fine when all that was left was Denuvo.
 
Armenius said:
I agree that leaving the DRM in is an issue, especially for game preservation. It is depressing that publishers and developers largely don't care if their games are lost to history. I read a report recently that 87% of games ever made are lost to history, and DRM contributes to that number. I think it should be legally required for DRM to be removed from games at least after the game is no longer being sold.
Click to expand...
I think DRM should be illegal. Publishers should just sell games and be done with it. If they're afraid of piracy then they should just lower their prices. Shocking, I know, especially when everyone is about "ma inflation" but the reality is the people who can't afford it will find a way to do so, and if that means piracy then piracy it is. Lower prices means higher sales, but does the higher sales make up for lower return? Tears of the Kingdom was piratable immediately at launch and Nintendo can't stop shoveling the money pouring in. Lots of great games are sold for much less than $60 or $70 and do well, like Minecraft and ShovelKnight.

As it stands right now, piracy makes playing games better, including performance. If this is what you've done to stop piracy, especially since Denuvo isn't cheap, then it's time to lower prices and make it better for consumers to buy your games. Instead of selling a game for $60 or $70, make it $40 with ports to Mac and Linux. You should make consumers happy to buy your product, not hate you for giving them $70 while limiting the amount of computers you can install a game you bought.
All that being said, I can't recall a single case when Denuvo was causing performance issues. RE8 is always brought up, but in that case it was because Capcom was using their own DRM in addition to Denuvo that literally checked every pixel of every frame as it was rendered. When that was removed, the game performed fine when all that was left was Denuvo.
Click to expand...
Other than a handful of games, the majority of games are slower with Denuvo. If it isn't frame rates, then it's loading time.

 
DukenukemX said:
Other than a handful of games, the majority of games are slower with Denuvo. If it isn't frame rates, then it's loading time.
Click to expand...
That was like 5-6 years ago games, could have changed quite a bit since.
 
DukenukemX said:
I think DRM should be illegal. Publishers should just sell games and be done with it. If they're afraid of piracy then they should just lower their prices. Shocking, I know, especially when everyone is about "ma inflation" but the reality is the people who can't afford it will find a way to do so, and if that means piracy then piracy it is. Lower prices means higher sales, but does the higher sales make up for lower return? Tears of the Kingdom was piratable immediately at launch and Nintendo can't stop shoveling the money pouring in. Lots of great games are sold for much less than $60 or $70 and do well, like Minecraft and ShovelKnight.

As it stands right now, piracy makes playing games better, including performance. If this is what you've done to stop piracy, especially since Denuvo isn't cheap, then it's time to lower prices and make it better for consumers to buy your games. Instead of selling a game for $60 or $70, make it $40 with ports to Mac and Linux. You should make consumers happy to buy your product, not hate you for giving them $70 while limiting the amount of computers you can install a game you bought.

Other than a handful of games, the majority of games are slower with Denuvo. If it isn't frame rates, then it's loading time.

Click to expand...

Most of those examples in the video are within the range that can be explained by the different variables at play. Doom 2016, in particular, included stability improvements and performance updates in addition to the removal of Denuvo. These comparisons are pointless since they are not apples to apples.

They have data that shows DRM is protecting sales volume, and publishers and developers have a right to protect their investment. The length of time it takes to crack DRM has been consistently going up, with some games not even getting cracked anymore. That isn't to say I agree with the companies utilizing DRM. Games like The Witcher 3 show it's possible to sell a high proportion of games on the PC platform when it is DRM-free across all storefronts, and the myth of the lost sale due to piracy is illogical. However, current DRM schemes are largely not affecting legitimate customers despite the constant haranguing by the loudest voices on the internet. Denuvo isn't actual malware like Starforce was back in the days of disc protection.
 
You must log in or register to reply here.
Back
Top