The “slows down your game” bit has always been hotly contested. There are certainly occasions where a modified exe without Denuvo runs faster, combined with accusations that that specific game integrated Denuvo in a very poor last-minute implementation that calls it dozens of times a second.
I don’t work on video games, but my own experience with software engineering and release management suggests those sorts of murky answers are likely to be the norm.
There’s nothing contested about it. Add a bunch of extra operations to the game loop and you can slow down a game. You only have so much headroom in each frame. Dunova takes up a lot of that time. And let’s not forget you can literally go tests with games that had denovu and then removed it. The testing shows pretty clearly that it does indeed slow down games.
Yes, cracked Denuvo games actually run better because you aren’t running a virus anti piracy software in the background. It runs at the kernel level and Crowdstrike is a pretty good case study on why that’s bad.