This week I’ve been on a bit of a Denuvo fix. For all that Denuvo does wrong, and I think it’s right to call it out for its overly stringent protection and online check-ins, it’s the claims that Denuvo impacts performance that I keep coming back to.
Anytime anything is written about Denuvo, just about anywhere, it’s immediately followed by a flurry of replies saying Denuvo tanks performance. Curiously, you don’t get these replies when writing about games that use Denuvo and yet have fantastic performance.
If you head here you can read a full list of games which use Denuvo, as well as those that have since had it removed. Of the games that currently have Denuvo, it can be argued that a few of them don’t perform well and will suffer FPS drops, either down to being extremely demanding, unoptimised, or, the third unknown - Denuvo. Games like Assassin’s Creed Origins, Ghost Recon Wildlands, Dishonored 2 and Deux Ex: Mankind Divided have all had the ‘Denuvo tanks performance’ card thrown at them. The vast majority of the list hasn’t though. Not many people are out there complaining about Battlefield 1’s performance, or FIFA 18, or Resident Evil 7, Star Wars Battlefront 2, Mad Max, Metal Gear Solid V, Far Cry Primal, Abzu, Mirror’s Edge Catalyst. I could go on. Suffice to these are games that are all loaded with Denuvo, and yet they all run great. Is Denuvo really the connecting factor that determines whether a game runs well or not?
I’ve done a few tests myself and also done plenty of reading around on the subject, and the actual evidence we have either way is pretty threadbare. Actually performing a true (and legal) test of Denuvo Anti-Tamper’s performance impact is a tricky process. Firstly, we need the hardware to be the same before and after. Secondly, we rely on publishers actually removing Denuvo. Thirdly, we need our benchmarks to be as close as possible to one another to rule out extraneous factors like driver updates and performance optimisation.
Last year I benchmarked DOOM with Denuvo vs DOOM without Denuvo and noticed a few percentage points difference in frame rates. Because I didn’t know Denuvo was going to be removed though, these benchmarks are separated by a year, meaning several patches and driver updates happened in the intervening 12 months, all of which categorically affected performance.
I also downloaded Hitman this week to try and compare it when benchmarks performed back when we reviewed the GeForce GTX 1060 6GB. I ran through roughly the same scenario shown in our gameplay footage here and could discern no difference in performance whatsoever. Frame rates average around 86 fps with occasional dips down to 60fps and peaking in the 90’s.
Out of curiosity I also took a look at any benchmarks I could find that other people had done. We’ve got someone claiming removing Denuvo from DOOM caused a huge uptick in performance, another showing no performance change in Shadow of War, and multiple people claiming Hitman’s performance was unchanged when Denuvo was removed.
I think there’s a growing argument to be made that Denuvo isn’t tanking performance in the way the internet at large would have us believe. It’s just a factoid that’s been repeated enough that it’s become an irrefutable truth.
None of this is to say that Denuvo never has, or never will, affect performance and frame rates. It very may well have, and when it’s poorly implemented, as we saw with Rime, which sent thousands of calls per second, this too almost assuredly affected performance. But, until there’s some evidence to suggest otherwise, I think Denuvo is rarely the culprit for poor gaming performance. It does make for a handy burning effigy when a game doesn’t run as well on your system as you’d hoped though. It’s become the whipping boy for Assassin’s Creed Origins’ PC performance when in reality I think it’s far more likely AC: Origins is just a great looking open-world game that doesn’t run well on a four-year-old graphics card.
As with anything like this though, I think the best way to try and get some info together on this is work together as a community. If we can all pool in with our benchmarks and experiences, hopefully we can get to the bottom of all this.
So, does Denuvo really impact gaming performance? Have you experienced performance improvements when Denuvo’s been removed? We’d love to know your own thoughts on Denuvo, as well as any benchmarks you’ve done that shed some light on either theory!