CPU cache space for code is much smaller than GPU memory for models (and the former is more important for performance since many CPU operations like pipeline parallelism are latency bound, not compute bound).
Why are they bullshit when piracy is a huge problem on the PC? There is a reason why AAA titles that are not multiplayer and subscription based lost developer mindshare.
For a long time now I've found it weird that people who like single player games on PC (and to a lesser extent older consoles which had piracy enabling mods) didn't acknowledge the long game consequences of their actions, or at least were willfully ignorant to them because everyone loves getting something for free. It seems to be a variation on Goodhart's law - you get what you reward - if the reward for a company (big or small) in spending lots of time and money isn't as good as other options, those other options will get more investment in the future and the ones you do like will get less.
The other option I can see for the large companies is that any project involving tens or hundreds of millions of dollars is likely to be insured, and a condition of that insurance is they take all reasonable options available to get the most success out of it that they can. If they don't they need to reduce the risk which probably means less resources allocated which again may not be interesting to the companies capable of making grand experiences versus other options.
> For a long time now I've found it weird that people who like single player games on PC (and to a lesser extent older consoles which had piracy enabling mods) didn't acknowledge the long game consequences of their actions
Isn't historically piracy positive for sales [1]?
That said, I'm pretty sure the real issue is that single / local coop games are just not appealing and so they get weaker sales. Like wtf was with Pikmen 2 not letting player 2 control louie? And then when local games start to sell poorly they get divestment but I'm pretty sure it was just lousey games and not piracy.
It’s hard to see from a US/Euro salary perspective, where not spending $60 is a moral decision, but you can start seeing how someone in a 300/mo salary country doesn’t think “I’ll save a bit and buy it” and instead thinks “I’ll never be able to afford this and this studio made millions anyway” and just pirate it. I’m not that articulate with my words but I hope you get what I’m trying to say.
This looks weird in the context, because the grandparent comment's argument was purely interest-based? You probably mean there's a propensity for tragedy of the commons.
Regardless I'd argue gaming may be the one media category left (after the recent decade's value decline) where piracy remains to seem like more hassle than buying a copy^W license. I would also guess it is more concentrated on a few popular titles compared to music or films. Nowadays I hear more of people collecting games on Steam, to never play them, than of legitimate pirates.
Greenheart Games famously released a "cracked" version of their own game (Game Dev Tycoon) onto torrent sites on launch day. In this version, the player's in-game studio eventually goes bankrupt because "pirates" steal their games.
The Data: Within 24 hours, 93.6% of players were playing the pirated version.
The Consequence: The developer's blog post highlighted the irony of pirates posting on forums complaining that the "in-game piracy" was unfair and "ruining" their fun. The experiment proved that even at a low price point ($8), a massive majority of the PC audience will choose "free" regardless of the developer's size or struggle.
One reason anti piracy companies make a living is because companies that buy it see concrete increases in revenue as a result. It may not be every pirate who converts to a customer but DRM solutions are priced to be below the expected additional revenue. And it's not always cheap.
Unhinged take I checked that was 2013 and the game cost almost as much as you you would pay in a month's rent in India in small towns.
Most pirates aren't people who could pay for this stuff. This is utterly meaningless.
So much in fact I don't even want to link counter examples to it.
No/very few paying user pirates even single player games these days if they can afford it as a luxury please understand that.
I would likemy regular updates bug fixes patches and new feaures ASAP. And on sale at 8$ for a game is less than 0.01% of my income so sure.
But if it costs 800 USD I will get it for free because I am literally too poor for it.
Anyone who thinks otherwise is beyond deluded.
Instead of denuvo you can use simple steam drm, non trivial to pirate for small games cracks will take days or weeks to appear and updates won't be available instantly.
It's safe simple and easy. And doesn't hurt any one.
Denuvo is just invasive bullcrap that deluded people think helps anyone.
Thats playing with statistics and you know it. Why such game?
If they would release only the paid game, there wouldn't be 93% + 7% of the gamers playing, far from it.
Cost is almost irrelevant to pirates, either its free or its not, like it or not. There is mix of folks who do it for the lulz, some do it to have higher performance gaming without denuvo taking resources and computing power, and some are outright poor. Even 8-usd-is-too-much poor.
I've lived like that. Don't judge too easily. Don't do stupid mistakes and count those as otherwise-paying-gamers. Thats PR for denuvo and similar, not a fair discussion.
I've had to take a moral stance and move to just playing games on Gog that I can buy and own the files for. No I can't play the latest and greatest but it's not the end of the world as I've so many classics to still play and enjoy. I can't support lockdown and DRM anymore. If I buy I want to own, otherwise I've not bought. It is true, if buying isn't owning, then piracy isn't stealing.
Right where I've landed as well. I just won't buy titles with Denuvo DRM, ever, no matter how much I want the game.
Was pleasantly surprised to find Doom Eternal is now on GOG a couple of days ago. If you're willing to wait, some AAA titles show up that previously had draconian DRM.
Wonder what will be the consequences of this. I dislike Denuvo for the performance and stability penalties it gives games, but I do wonder if the "security" it gave publishers wasn't a big part of the reason why we've been getting more and more big name games on PC.
This isn't about being right or wrong but about what the publishers will do when they see their games are again getting cracked day one, and if it'll be a catalyst to again return to getting either less PC releases or at least delayed releases compared to consoles.
Denuvo’s market is ‘first 90 days’ revenue protection, not lifelong revenue protection. Lots of games using their crap remove it after a few months to shut down the flood of support issues the DRM causes. If only Microsoft hadn’t fucked up so badly with Windows 11 requiring an account, they’d have a way to stop using it altogether.
>Lots of games using their crap remove it after a few months to shut down the flood of support issues the DRM causes.
No, the overwhelming majority of denuvo games released after ~2020 (when they changed there licensing model to SaaS) have it removed after 2-4 years not because of user complaints but because of licensing costs, contracts and compliance.
If anything with many games it is very clear that the developer/publisher do not care for the user, since even when the DRM gets broken and has lost its purposes, many still refuse to remove it and give paying customers the same better non DRM experience as pirates.
>If only Microsoft hadn’t fucked up so badly with Windows 11 requiring an account
This is not true at all as evidenced by the fact that most games do not get Denuvo removed once they are cracked. And the companies that DO remove denuvo only do so after several years because of licensing costs as denuvo transitioned to a SaaS model.
Untrue, where are all the after-90-days-hacked AAA games? Nowhere, denuvo lives on as long as publisher is willing to pay continuous licence, which is usually years.
And users complaining because denuvo messes up their Windows, sometimes games don't run and so on? Just cost of doing business, as long as enough people buy it who cares.
I honestly doubt it will make much of a difference.
A good percentage of people who would download the cracked games would not have bought those anyway. And with Steam being so convenient it's hard to decide to go for a cracked copy of dubious origin that might install god knows what into your machine.
Run anti-cheat server-side. Give us private servers again. There's no reason we should have to put up with client-side rootkits written by non-kernel-devs to play a game.
Community servers don't want server-side anti-cheat either. Hell they invented client-side anti-cheats. Even current day community servers like Face-IT have additional anti-cheats, not less. Same with modded GTAV FiveM (before the main game added anti-cheats)
Cheating is a social issue, not a technical one. Communities are the solution.
Private servers are a nice way to do this and do still exist in places. My favorite online game uses them along with server side anti-cheat and while cheating occasionally happens, it has never been an ongoing issue. I've maybe seen a cheater once or twice in all my many hours playing the game over 10 years (elite dangerous, in case you were curious).
It's not possible, technically, to run effective anti-cheat server-side. Clients need precise enemy location data for things like sound effects. The server can't tell if the client is using the data for unfair purposes or not.
This. There are a lot of online games I loved playing but the cheating got so bad it made it impossible to play. MW1, MW2, Battlefield, CS, etc... you could see the wallhacks and aimbots taking over every lobby. I eventually stopped playing. I tried using Consoles for online gaming after that but never really got into using joysticks.... still prefer mouse and keyboard. Now I play limited games where the cheating isn't quite that rampant.
Im not a big gamer, but playing GTA Online, and getting taken out as soon as you spawn. Or items just spawning in front of you, like ramps. REALLY ruins the experience
> in late 2025, the MKDev collective and the prolific DenuvOwO came up with a hypervisor-based bypass (HVB) that installs a kernel-level driver to intercept and respond to Denuvo's checks. While that's not an actual crack, it's good enough for piracy work, as the saying goes.
The main difference that Denuvo does nothing to improve the experience of the end user.
I don't like Anti-Cheat solutions with elevated privileges but they have (at least for some time) reduced the number of Cheaters in games like Valorant or BF, for most users this is at least a somewhat understandable tradeoff. Denuvo on the other hand is DRM and a pure tradeoff in favor of the publisher at the cost of the consumed.
There is a user argument for anti cheat as a user = less cheater.
There is no user argument for DRM, if anything there are many against it = higher game price/less money for the actual game and devs, indirect funding of DRM software, worse performance, higher system requirements, worse preservation, worse privacy, longer loading times, online requirements, worse usability, machine activation restriction, bugs...
Kernel level anti-cheat also doesn't introduce a giant performance penalty like Denuvo-style DRM. People just want to play their games without it still stuttering on top of the line hardware.
Interesting to finally see some action from the mouse again. Was kinda sad to see that Denuvo embodies all the worst of DRM but was so thoroughly metastasized that it was nearly inoperable and they had effectively "won".
Once again I'm at odds with TH reporting. Of course you can spoof a server. That happens all the time, especially with videogames. You may not immediately be able to figure out what the call/response is, but without knowing what the check is, it could just be a simple endpoint that returns "true" on every request. Very speculative to say that whatever they do will be impossible to mimic.
> You may not immediately be able to figure out what the call/response is, but without knowing what the check is, it could just be a simple endpoint that returns "true" on every request. Very speculative to say that whatever they do will be impossible to mimic.
It’s trivially easy to use a signed response that is encoding some part of the metadata of your system in the signature to make it impossible to emulate the server. Don’t think the Denuvo devs would be stupid enough to provide a “return true” request for a server call.
Can the underlying function that checks if the server call is correct be bypassed? Sure, but that’s much harder.
Cryptography goes BRRRRR, with a proper implementation of cryptography you'd need to do things like patch out the keys in memory in order to "spoof" messages.
I would hope publishers would take note and remove it, having hundreds of megabytes of junk in the executable is just wasteful to put it mildly
The bigger problem with Denuvo is that it appears to significantly impact game performance as well
Why would they care for a few hundred MBs when the games are in the 10s of GBs?
CPU cache space for code is much smaller than GPU memory for models (and the former is more important for performance since many CPU operations like pipeline parallelism are latency bound, not compute bound).
Don't forget that the guy behing Denuvo is the same person behind SafeDisc, SecuROM and similar bullshit siblings from the past PC gaming world.
Why are they bullshit when piracy is a huge problem on the PC? There is a reason why AAA titles that are not multiplayer and subscription based lost developer mindshare.
Surely, this has nothing to do with the fact that live service and subscription games generate more revenue, whether or not piracy is involved.
For a long time now I've found it weird that people who like single player games on PC (and to a lesser extent older consoles which had piracy enabling mods) didn't acknowledge the long game consequences of their actions, or at least were willfully ignorant to them because everyone loves getting something for free. It seems to be a variation on Goodhart's law - you get what you reward - if the reward for a company (big or small) in spending lots of time and money isn't as good as other options, those other options will get more investment in the future and the ones you do like will get less.
The other option I can see for the large companies is that any project involving tens or hundreds of millions of dollars is likely to be insured, and a condition of that insurance is they take all reasonable options available to get the most success out of it that they can. If they don't they need to reduce the risk which probably means less resources allocated which again may not be interesting to the companies capable of making grand experiences versus other options.
> For a long time now I've found it weird that people who like single player games on PC (and to a lesser extent older consoles which had piracy enabling mods) didn't acknowledge the long game consequences of their actions
Isn't historically piracy positive for sales [1]?
That said, I'm pretty sure the real issue is that single / local coop games are just not appealing and so they get weaker sales. Like wtf was with Pikmen 2 not letting player 2 control louie? And then when local games start to sell poorly they get divestment but I'm pretty sure it was just lousey games and not piracy.
[1]: https://www.engadget.com/2017-09-22-eu-suppressed-study-pira...
It’s hard to see from a US/Euro salary perspective, where not spending $60 is a moral decision, but you can start seeing how someone in a 300/mo salary country doesn’t think “I’ll save a bit and buy it” and instead thinks “I’ll never be able to afford this and this studio made millions anyway” and just pirate it. I’m not that articulate with my words but I hope you get what I’m trying to say.
This single issue convinced me most people have zero moral convictions and will lie to themselves to preserve their self-image.
This looks weird in the context, because the grandparent comment's argument was purely interest-based? You probably mean there's a propensity for tragedy of the commons.
Regardless I'd argue gaming may be the one media category left (after the recent decade's value decline) where piracy remains to seem like more hassle than buying a copy^W license. I would also guess it is more concentrated on a few popular titles compared to music or films. Nowadays I hear more of people collecting games on Steam, to never play them, than of legitimate pirates.
To give you an idea of the scale of the problem:
Greenheart Games famously released a "cracked" version of their own game (Game Dev Tycoon) onto torrent sites on launch day. In this version, the player's in-game studio eventually goes bankrupt because "pirates" steal their games.
The Data: Within 24 hours, 93.6% of players were playing the pirated version.
The Consequence: The developer's blog post highlighted the irony of pirates posting on forums complaining that the "in-game piracy" was unfair and "ruining" their fun. The experiment proved that even at a low price point ($8), a massive majority of the PC audience will choose "free" regardless of the developer's size or struggle.
https://web.archive.org/web/20161118042043/http://arstechnic...
https://web.archive.org/web/20131214165241/http://aussie-gam...
P.S.: It bears repeating that the game cost only 8 dollars.
The number of pirated copies doesn't translate to missed sales.
Someone playing/watching/listening to something for free doesn't mean they would still do it if they had to pay for it.
It’s certainly not a 1:1 loss, but it’s also not zero
One reason anti piracy companies make a living is because companies that buy it see concrete increases in revenue as a result. It may not be every pirate who converts to a customer but DRM solutions are priced to be below the expected additional revenue. And it's not always cheap.
Unhinged take I checked that was 2013 and the game cost almost as much as you you would pay in a month's rent in India in small towns.
Most pirates aren't people who could pay for this stuff. This is utterly meaningless.
So much in fact I don't even want to link counter examples to it.
No/very few paying user pirates even single player games these days if they can afford it as a luxury please understand that.
I would likemy regular updates bug fixes patches and new feaures ASAP. And on sale at 8$ for a game is less than 0.01% of my income so sure.
But if it costs 800 USD I will get it for free because I am literally too poor for it.
Anyone who thinks otherwise is beyond deluded.
Instead of denuvo you can use simple steam drm, non trivial to pirate for small games cracks will take days or weeks to appear and updates won't be available instantly.
It's safe simple and easy. And doesn't hurt any one.
Denuvo is just invasive bullcrap that deluded people think helps anyone.
Brown hands typed these words.
Thats playing with statistics and you know it. Why such game?
If they would release only the paid game, there wouldn't be 93% + 7% of the gamers playing, far from it.
Cost is almost irrelevant to pirates, either its free or its not, like it or not. There is mix of folks who do it for the lulz, some do it to have higher performance gaming without denuvo taking resources and computing power, and some are outright poor. Even 8-usd-is-too-much poor.
I've lived like that. Don't judge too easily. Don't do stupid mistakes and count those as otherwise-paying-gamers. Thats PR for denuvo and similar, not a fair discussion.
I've had to take a moral stance and move to just playing games on Gog that I can buy and own the files for. No I can't play the latest and greatest but it's not the end of the world as I've so many classics to still play and enjoy. I can't support lockdown and DRM anymore. If I buy I want to own, otherwise I've not bought. It is true, if buying isn't owning, then piracy isn't stealing.
Right where I've landed as well. I just won't buy titles with Denuvo DRM, ever, no matter how much I want the game.
Was pleasantly surprised to find Doom Eternal is now on GOG a couple of days ago. If you're willing to wait, some AAA titles show up that previously had draconian DRM.
Do any of the legit scene groups sign their binaries? How do you know a release isn’t tainted?
Wonder what will be the consequences of this. I dislike Denuvo for the performance and stability penalties it gives games, but I do wonder if the "security" it gave publishers wasn't a big part of the reason why we've been getting more and more big name games on PC.
This isn't about being right or wrong but about what the publishers will do when they see their games are again getting cracked day one, and if it'll be a catalyst to again return to getting either less PC releases or at least delayed releases compared to consoles.
I will hope that does not happen.
Denuvo’s market is ‘first 90 days’ revenue protection, not lifelong revenue protection. Lots of games using their crap remove it after a few months to shut down the flood of support issues the DRM causes. If only Microsoft hadn’t fucked up so badly with Windows 11 requiring an account, they’d have a way to stop using it altogether.
>Lots of games using their crap remove it after a few months to shut down the flood of support issues the DRM causes.
No, the overwhelming majority of denuvo games released after ~2020 (when they changed there licensing model to SaaS) have it removed after 2-4 years not because of user complaints but because of licensing costs, contracts and compliance.
If anything with many games it is very clear that the developer/publisher do not care for the user, since even when the DRM gets broken and has lost its purposes, many still refuse to remove it and give paying customers the same better non DRM experience as pirates.
>If only Microsoft hadn’t fucked up so badly with Windows 11 requiring an account
I don't understand how that is related at all.
This is not true at all as evidenced by the fact that most games do not get Denuvo removed once they are cracked. And the companies that DO remove denuvo only do so after several years because of licensing costs as denuvo transitioned to a SaaS model.
Untrue, where are all the after-90-days-hacked AAA games? Nowhere, denuvo lives on as long as publisher is willing to pay continuous licence, which is usually years.
And users complaining because denuvo messes up their Windows, sometimes games don't run and so on? Just cost of doing business, as long as enough people buy it who cares.
I honestly doubt it will make much of a difference.
A good percentage of people who would download the cracked games would not have bought those anyway. And with Steam being so convenient it's hard to decide to go for a cracked copy of dubious origin that might install god knows what into your machine.
We're not in the early 00s anymore.
i think your underwstimating the anticheat value that still exists. many of the online games are trash when theres not strict cheat control.
Run anti-cheat server-side. Give us private servers again. There's no reason we should have to put up with client-side rootkits written by non-kernel-devs to play a game.
Community servers don't want server-side anti-cheat either. Hell they invented client-side anti-cheats. Even current day community servers like Face-IT have additional anti-cheats, not less. Same with modded GTAV FiveM (before the main game added anti-cheats)
Cheating is a social issue, not a technical one. Communities are the solution.
Private servers are a nice way to do this and do still exist in places. My favorite online game uses them along with server side anti-cheat and while cheating occasionally happens, it has never been an ongoing issue. I've maybe seen a cheater once or twice in all my many hours playing the game over 10 years (elite dangerous, in case you were curious).
It's not possible, technically, to run effective anti-cheat server-side. Clients need precise enemy location data for things like sound effects. The server can't tell if the client is using the data for unfair purposes or not.
Once the data is sent to the client, in an untrusted setting, all bets are off. Not your hardware, no control over it.
Too bad. It's not possible for rootkits to be a good idea for a video game.
This. There are a lot of online games I loved playing but the cheating got so bad it made it impossible to play. MW1, MW2, Battlefield, CS, etc... you could see the wallhacks and aimbots taking over every lobby. I eventually stopped playing. I tried using Consoles for online gaming after that but never really got into using joysticks.... still prefer mouse and keyboard. Now I play limited games where the cheating isn't quite that rampant.
Im not a big gamer, but playing GTA Online, and getting taken out as soon as you spawn. Or items just spawning in front of you, like ramps. REALLY ruins the experience
Or everyone in the lobby getting nuked over and over.
Or trying to do heists and having a cheater in every session.
I'd like to play the game again but it's just not fun.
There are still some servers online for games like the first CoD or United Offensive. No hackers as far I can tell anymore. They have all moved on
"Protected" is the wrong word. "Restricted" is much more honest regarding what Denovo does.
Good riddance.
No, it hasn't:
> in late 2025, the MKDev collective and the prolific DenuvOwO came up with a hypervisor-based bypass (HVB) that installs a kernel-level driver to intercept and respond to Denuvo's checks. While that's not an actual crack, it's good enough for piracy work, as the saying goes.
This. It's bypassed, not cracked. All the games released need HVB to work. They use legit Denuvo licenses from other systems.
I find it ironic people mad at Denuvo and yet play games like Battlefield which enforces kernel level spyware nonetheless haha
The main difference that Denuvo does nothing to improve the experience of the end user.
I don't like Anti-Cheat solutions with elevated privileges but they have (at least for some time) reduced the number of Cheaters in games like Valorant or BF, for most users this is at least a somewhat understandable tradeoff. Denuvo on the other hand is DRM and a pure tradeoff in favor of the publisher at the cost of the consumed.
I would say it was wildly successful in Valorant.
There is a user argument for anti cheat as a user = less cheater.
There is no user argument for DRM, if anything there are many against it = higher game price/less money for the actual game and devs, indirect funding of DRM software, worse performance, higher system requirements, worse preservation, worse privacy, longer loading times, online requirements, worse usability, machine activation restriction, bugs...
Kernel level anti-cheat also doesn't introduce a giant performance penalty like Denuvo-style DRM. People just want to play their games without it still stuttering on top of the line hardware.
How are you protecting yourself at the game itself spying on you?
Interesting to finally see some action from the mouse again. Was kinda sad to see that Denuvo embodies all the worst of DRM but was so thoroughly metastasized that it was nearly inoperable and they had effectively "won".
That's all you need to know about DRM - when "pirates" bypass it, paying users are taking the hit.
And I'm not speaking about cost of implementing a technology to actively make the product worse.
A great use of LLM
Are Denuvo using games marked on Steam these days?
I've been getting mostly indies so I feel safe, but maybe I should check...
There's a yellow? box just above payment options that informs you of DRM.
Oh right, it's still there. Nothing on my wish list has it :) I had to go to the store page for NBAsomething to see it.
Yes they are. On the store page.
steamdb.info should have the info too I think?
good riddance. crazy to see game developers hemorrhaging money for malware
Once again I'm at odds with TH reporting. Of course you can spoof a server. That happens all the time, especially with videogames. You may not immediately be able to figure out what the call/response is, but without knowing what the check is, it could just be a simple endpoint that returns "true" on every request. Very speculative to say that whatever they do will be impossible to mimic.
> You may not immediately be able to figure out what the call/response is, but without knowing what the check is, it could just be a simple endpoint that returns "true" on every request. Very speculative to say that whatever they do will be impossible to mimic.
It’s trivially easy to use a signed response that is encoding some part of the metadata of your system in the signature to make it impossible to emulate the server. Don’t think the Denuvo devs would be stupid enough to provide a “return true” request for a server call.
Can the underlying function that checks if the server call is correct be bypassed? Sure, but that’s much harder.
Cryptography goes BRRRRR, with a proper implementation of cryptography you'd need to do things like patch out the keys in memory in order to "spoof" messages.