You know while some of you don't believe Denuvo causes performance loss.
I wonder if gog.com whenever they get a PC version of a video game that has Denuvo to remove Denuvo from it before selling it on gog.com if they will try some benchmarks on the Steam version and the 100% Digital Rights Management (DRM) free version to see what the performance loss is.
All gog.com employees have a Steam account I'm sure.
I bet Judaliscariot is the gog.com employee who tests this stuff.
He probably did with Inside before it got released for sale on gog.com.
Anyways I fully expect gog.com to use it as a means for them selling 100% Digital Rights Management (DRM) free video games in the future against Denuvo.
Just to be clear, don't include me in the "some of you" above as I haven't stated any opinion about that to date.
What I have read is that the modifications that game developers make to their game where they insert Denuvo are supposed to be done in places in the game code which are not performance critical, specifically so that Denuvo does not have an impact on the game's performance. It is something intended to not be used in inner loops of performance critical code paths, but in inconsequential codepaths that do not impact performance. If this is indeed the case, and it seems reasonable, then if a game developer follows the instructions of how they are told to use Denuvo, then in theory there should not be any performance impact. In other words, whether there is a performance impact or not appears to be a matter of how close the developers of a game follow instructions. This is just what I recall reading previously online somewhere of which I have no reference for and so while it seems reasonable, people should treat it with a grain of salt unless someone has direct information that can be referenced specifically.
Having said that however I can say that GOG was asked sometime in the last month or so in the forums "Does GOG consider Denuvo DRM?" to which JudasIscariot responded something to the effect of <paraphrasing> "Yes, GOG very definitely considers Denuvo to be DRM.". For practical purposes, what this means is regardless of what the Denuvo company claims Denuvo to be or not be DRM, and regardless of whether any other person or company out there thinks Denuvo is "DRM", GOG considers it to be so for the purposes of GOG's own definition of DRM which their store runs by. By extrapolation of that, GOG seeing Denuvo as DRM would mean that Denuvo is considered by GOG to not qualify as being "DRM-free" and thus by definition GOG would not permit games that contain active Denuvo software in them to be sold in the store. I'm not a GOG employee so if anyone doubts this or wants an official word, then they should 100% contact GOG directly because random opinions of people in the forums (including me) are irrelevant to what GOG decides what is what.
I think one thing people may be missing from above, is that the performance comparison is being compared between two different versions of DOOM, one with Denuvo and one without Denuvo in a manner that is not explicitly stated but which appears to assume that there are absolutely no other differences in the DOOM game between the version tested with Denuvo and the one without - other than Denuvo being removed from the game.
Has anyone seeked out a changelog or some other data between the two game versions to determine whether anything else in the game was changed? Bugfixes, performance improvements etc.? Were the performance tests performed one after the other on the same hardware using the exact same video driver version? How was the FPS measured? Was it measured via some scientifically accurate method of recording the data to hard disk at even intervals in an automated and reproduceable manner, with a scripted game sequence to ensure multiple test runs are testing the identical run? Or was it just random off the cuff human playing the game and glancing at the FPS counter from time to time and making a mental note?
If the different versions of the games contained changes other than just Denuvo being removed, then the testing can't be scientific to just testing Denuvo's performance impact. If the testing was not performed in an automated and reproduceable fashion then the results are anecdotal and not scientific either.
From what I"m saying above it may give the false impression that I am defending Denuvo in some way. I am not. I think Denuvo is a black plague of gaming and is anti-consumer software regardless of how one chooses to label it. I do not want to see it in games and want to see it die off. My comments concerning whether or not it might have a performance impact on the game are intended to ensure that anyone who is trying to do a comparison to measure it and draw any conclusions at all either way - are performed in as scientific a manner as is possible so as to not allow personal bias either way to enter into the assessment, and to exclude other factors that may result in different results which might be caused by something other than Denuvo such as different video driver version, other changes made to the game's code in addition to Denuvo being removed etc. In other words, performance testing should be accurate and apples to apples comparison.
As such, I'm not defending Denuvo, nor claiming it has no performance impact on the game, just that if someone is going to do measurements and make claims that "prove" something, then such claims should stand up to some level of standard that is typically used when doing professional performance assessment in order to remove any kind of human bias from the results. Mind you, since I hate Denuvo and want to see it die, if someone threw out bogus data due to bad testing methodology I wouldn't exactly rush to Denuvo's aide to point that out too loudly either, but it would be nice to see an accurate test done that excludes other factors from potentially being a cause for any actual performance differences, and to remove the human brain and eye from the equation. :)