I don't support these obtrusive DRM systems like Denuvo and I don't own, pirate or play any games that use such systems and will support DRM free where possible. None of the games in my Seam and Origin collections use DRM on top of the very limited DRM that Steam and Origin have and (other than Fallout 3 and Fallout New Vegas which were bought long before GOG had them for sale) nothing in my library is available from any DRM free source.

Yes I could choose to only play DRM free titles but I would rather reward companies like EA for producing awesome titles like the C&C Remasters (which has no protection or encryption at all, just the most basic of checks that Steam is running, no online signup or always-online required and probably the BEST and most comprehensive mod support of any RTS title from a mainstream publisher).

And then we also have devs that've come forward claiming Denuvo has absolutely no impact on performance.

Everything you just mention can also be attributed to other optimizations within the patch, or other environmental factors.

My bigger problem is that people make a big deal out of Denuvo but give consoles a free pass, which are pretty much the ultimate form of DRM, forcing people in a walled garden ecosystem. With a hardware lock that you can't possibly circumvent unless by virtue of the platform holder they offer some back compat.

DRM sucks... people should just stop making it.

You mean companies like Ubisoft whose CEO claims "95% rates of piracy" on PC games, and yet their non-pirateable Denuvo'd games fail to sell even +50% more vs the previous unprotected game in the franchise let alone 1900% more during the period they remain uncrackable?...

Nope. I'm talking about the group responsible for cracking observing that the same build version of a game get sped up 6x when the only difference is removing Denuvo. Nothing to do with comparing different build versions / patches. Likewise, although it's difficult to see what's happening inside the Denuvo VM, the VM container "blob" performance itself can be measured (just like you benchmark decryption performance of an encrypted file without being able to crack it).

The very definition of "obfuscation" literally means to fill up a CPU with junk instructions as padding in an attempt to hide the real ones. It would defy the laws of physics to do that and not have any CPU impact. People who claim "there's zero impact" are often those who basically bought a top-end CPU (Denuvo was introduced roughly the same time the MOAR CORE rate-race started up), and feign ignorance of the fact "only 5% load" on a Ryzen 3900X (12C/24T) chip may well be a 30-40% load on a 4C/4T chip, eg, Ryzen 2200G or i3-9100F (as seen in the disproportionately stuttery mess many Ubisoft games that layer up Denuvo + VMProtect play like on quad-cores vs games of the same franchise made on the same engine by the same devs just 12 months earlier).

Depends on the console. Up until recently, many consoles didn't require an online connection at all (unlike Denuvo) when playing off a local disc / cartridge. And even if the new ones do, two wrongs do not make a right and non-console owning PC gamers can easily criticise Denuvo without needing to mention consoles that don't affect them. Also, many console games of yesteryear (eg, NES, SNES, Wii, Sega, etc) are available to play via emulators. The same isn't said of Denuvo'd games, which speaks for itself which has the greater negative impact on game preservation / longevity.