Developer hacks Denuvo DRM after six months of detective work and 2,000 hooks, allows running Hogwarts Legacy on other PCs

Hogwarts Legacy
(Image credit: Steam)

After six months of work, DRM developer Maurice Heumann has successfully cracked Hogwarts Legacy's Denuvo DRM protection system. Rather than hacking it for illegal reasons, Heumann wanted to hack Hogwart's DRM for the sole purpose of learning its intricacies. As a result, the developer left plenty of the details of his work vague so as not to promote illegal cracking.

Heumann reveals in his blog post that Denuvo utilizes several different methods to ensure that Hogwarts Legacy is being run under appropriate (legal) conditions. First, the DRM creates a "fingerprint" of the game owner's system, and a Steam Ticket is used to prove game ownership. The Steam ticket is sent to the Steam servers to ensure the game was legitimately purchased. Heumann notes that he doesn't technically know what the Steam servers are doing but says this assumption should be accurate enough to understand how Denuvo works.

Once the Steam ticket is verified, a Denuovo Token is generated that only works on a PC with the exact fingerprint. This token is used to decrypt certain values when the game is running, enabling the system to run the game. In addition, the game will use the fingerprint to periodically verify security while the game is running, making Denuvo super difficult to hack.

After six months, Heumann was able to figure out how to hijack Hogwart Legacy's Denuvo fingerprint and use it to run the game on another machine. He used the Qiling reverse engineering framework to identify most of the fingerprint triggers, which took him two months. There was a third trigger that he says he only discovered by accident.

By the end, he was able to hack most of the Denuvo DRM with ~2,000 of his own patches and hooks, and get the game running on his laptop using the token generated from his desktop PC.

Denuvo Performance Impact

The developer also talked about the potential impact on performance as well. Specifically, he wanted to see if he could prove that Denuvo kills performance as much as many people believe.

While he was not able to provide a clear answer by running tests, he instead looked at the frequency of Denuvo calls — noting that if security checks aren't happening very often, it doesn't really matter if they cause a slight slowdown. He discovered that the amount of Denuvo code executed in-game is quite infrequent, with calls occurring once every few seconds, or during level loads.

This suggests that Denuvo is not killing performance, contrary to popular belief. Heumann's discovery seems to back up what Denuvo said officially about its DRM that its DRM protection does not hamper gaming performance.

While Heumann was technically able to successfully hack the Denuvo DRM protection in Hogwarts Legacy, his efforts weren't perfect. The game still crashes at times, as his primary goal was to reach the point where he could launch the game and get to the main menu. He says finding all the Denuvo hooks and patching them is certainly possible, but it would take a significant amount of time to accomplish—which wasn't his purpose. He only wanted to gain a better understanding of how the DRM works.

Aaron Klotz
Contributing Writer

Aaron Klotz is a contributing writer for Tom’s Hardware, covering news related to computer hardware such as CPUs, and graphics cards.

  • garrett040
    Denuvo and others like it are a cancer. You will never eliminate piracy, the answer is to make the product so convienent/priced accessibly that its just easier to purchahse it.
    Reply
  • hotaru251
    This suggests that Denuvo is not killing performance, contrary to popular belief.

    and was proven so it isn't "popular belief".

    Devs released a game (accidentally) that lacked the Denuvo implementation....then they got aware of that and re-added it.

    You literally had an apples to apples comparison as the only difference was the lack of denuvo. and as expected you lost performance w/ denuvo...or fact people will buy a steam game w/ denuvo and just dl a cracked version that stops denuvos crap and get better performance. (this is actually a thing some ppl do)

    I don't support piracy myself, but defending Denuvo's lie is stupid.

    edit: also heres more proof
    https://overclock3d.net/news/software/denuvo-has-been-removed-from-the-callisto-protocol-and-gamers-are-reporting-performance-gains/
    Reply
  • Notton
    From what I've seen, Denuvo only kills performance when it is implemented poorly.
    Which is like 95% of the time.

    It seems like Hogawarts is the rare exception
    Reply
  • DingusDog
    hotaru251 said:
    and was proven so it isn't "popular belief".

    Devs released a game (accidentally) that lacked the Denuvo implementation....then they got aware of that and re-added it.

    You literally had an apples to apples comparison as the only difference was the lack of denuvo. and as expected you lost performance w/ denuvo...or fact people will buy a steam game w/ denuvo and just dl a cracked version that stops denuvos crap and get better performance. (this is actually a thing some ppl do)

    I don't support piracy myself, but defending Denuvo's lie is stupid.

    edit: also heres more proof
    https://overclock3d.net/news/software/denuvo-has-been-removed-from-the-callisto-protocol-and-gamers-are-reporting-performance-gains/
    Just because something has been proven doesn't mean it can't still be popular belief. If you look on gaming forums it's still very much a belief that Denuvo causes slowdowns and instability.
    Reply
  • Hundvd7
    It has been proven to cause performance problems.
    You wouldn't call gravity a popular belief. It's practically a fact.
    Reply
  • nimbulan
    Notton said:
    From what I've seen, Denuvo only kills performance when it is implemented poorly.
    Which is like 95% of the time.

    It seems like Hogawarts is the rare exception
    Rather ironic considering the woeful technical state of the game otherwise. CPU performance is some of the worst I've ever seen, the game stutters heavily when CPU bottlenecked (which is quite common) and these issues have gotten continually worse with every game update. Not to mention every patch breaks the raytracing in different ways...
    Reply
  • atomicWAR
    DingusDog said:
    Just because something has been proven doesn't mean it can't still be popular belief. If you look on gaming forums it's still very much a belief that Denuvo causes slowdowns and instability.
    By proven...I believe he meant that it does hinder performance. So popular belief and fact do actually line up....thus his follow up with Callisto Protocol link which showed performance gains with the removal of Denuvo. Maybe I read it wrong but that was the way I took it.
    And I couldn't agree more with @garrett040 that Denuvo is a cancer for gaming...
    Reply
  • mac_angel
    I can't speak for Denuvo specifically, but this game was cracked over a year ago.
    Reply
  • bujinkanrn
    Admin said:
    A DRM developer successfully hacked Hogwarts Legacy's Denuvo DRM protection system after six months of work and was able to run the game on another machine without proper credentials. He also discovered that Denuvo does not appear to meaningfully affect performance.

    Developer hacks Denuvo DRM after six months of detective work and 2,000 hooks, allows running Hogwarts Legacy on other PCs : Read more
    As long as Denuvo requires online checks for single player offline games and hampers performance, I will never purchase a game that has it until it is removed. Pray tell if developers show data on how much piracy Denuvo actually stops. I doubt they would release such data if they kept it. If they did, they probably couldn't blame piracy for their lack of sales.
    Reply
  • JarredWaltonGPU
    hotaru251 said:
    and was proven so it isn't "popular belief".

    Devs released a game (accidentally) that lacked the Denuvo implementation....then they got aware of that and re-added it.

    You literally had an apples to apples comparison as the only difference was the lack of denuvo. and as expected you lost performance w/ denuvo...or fact people will buy a steam game w/ denuvo and just dl a cracked version that stops denuvos crap and get better performance. (this is actually a thing some ppl do)

    I don't support piracy myself, but defending Denuvo's lie is stupid.

    edit: also heres more proof
    https://overclock3d.net/news/software/denuvo-has-been-removed-from-the-callisto-protocol-and-gamers-are-reporting-performance-gains/
    That's a tinted view of how Denovo may or may not have impacted performance. The reality is that you need to do a lot of systematic testing, with two versions you are absolutely sure don't differ in any way other than the inclusion or lack of Denuvo.

    Based on what has been demonstrated here, if there are thousands of hooks and the calls are only periodic (once every few seconds), that could indeed drop minimum fps at times — a slight stutter for one frame is all it would take. But there are absolutely ways to code around Denuvo (or any other similar function calls) that would not cause such issues.

    Basically, you periodically fire off a low priority thread to check for Denuvo licensing or whatever you want to call it. You keep running everything else and if a check comes back as a failure, then you gracefully exit the rest of the code. You don't do a real-time check and wait for the response, except that's far easier and likely exactly what some games have done.

    The only people who could truly prove Denuvo hurts performance would be the game developers. Take a fully optimized Denuvo implementation, benchmark a variety of hardware, and then remove Denuvo and repeat — with no other code fixes or changes. But to my knowledge, no one has done exactly this. Anecdotes claiming variety of things exist, but "proof" is harder to come by. At best, Denuvo hurting PC performance is a theorem that lacks a concrete proof.

    The real issue is that so many games, without Denuvo, already have massive difficulty getting to the level of being decently optimized. Adding Denuvo on top of the cruft that already exists certainly isn't going to help the situation and certainly can go wrong. But FUD prevents most companies from trying to take a different approach than a tried and despised DRM solution.
    Reply