Better late than never I guess. Back in August, Deep Silver released Agents of Mayhem; a new open-world game from the creators of the Saints Row series. As with pretty much all latest games, Agents of Mayhem was using the Denuvo anti-tamper tech. However, Volition has released a new patch for it that completely removes it.
It’s worth noting that the game appeared to be CPU-bound in some scenarios. As such, we’ll be sure to test this latest Denuvo-free version with the exact same settings on the exact same locations. This may reveal whether Denuvo actually impacts game performance or not.
Deep Silver joins a number of publishers that are removing the Denuvo anti-tamper tech from their games once they are cracked. Bethesda, Microids and Square Enix did the same thing. Electronic Arts has also removed Denuvo from Mass Effect: Andromeda.
Furthermore, publishers like Bethesda are not supporting Denuvo in their latest games. Both Wolfenstein II: The New Colossus and The Evil Within 2 did not use the Denuvo anti-tamper tech.
The only publisher that still supports heavily Denuvo is Ubisoft. In Assassin’s Creed: Origins, Ubisoft has added VMProtect on top of the Denuvo anti-tamper tech. And from the looks of it, the French company was able to protect its game from pirates. At the time of writing, there is no crack yet for this latest Assassin’s Creed game!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
Even though i hate DENUVO and removing it makes me happy, DRM was the least of this game problems.
From Steam users:
“After Denuvo gone, GPU usage/load raised up 20% and fps also went up”
But but but but Denuvo don’t effect performance and hardware
lel
Steam users are not a valid source of information and offer no credibility. Now, if Gamer’s Nexus of DF digs in and states otherwise, I’ll listen. Otherwise, it’s no better than the goof balls on this message board screaming it loudly.
or maybe it is due to the patch which increases performance
if you want real bench about denuvo hitting performance you have to bench a cracked version of the with uncracked version not a patched version with unpatched
Ever heard of a period?
Anyway…A cracked version simple breaks the code, it doesn’t remove it. There is still code running from a cracked version.
Patched vs unpatched is the only way it makes sense.
a cracked version from baldman, the rest still have denuvo in it.
You know nothing,why comment?
*John Snow*
By waiting on actual reliable evidence? Yeah, because that’s what idiots do…use reliable sources.
Go take your “facts” and “reliable evidence” off my site.
We don’t like you “edge-u-medicated” types.
I’m not a frequent visitor of that site so I have no idea but does that site even do articles like comparing a game’s performance with Denuvo and after Denuvo’s been patched off?
They’ve not covered Denuvo (probably because they know it’s ridiculous) but they do similar tests and comparisons using scientifically valid testing. They’re not perfect, but they’re testing is more rational than just about anyone else I’ve come across.
Similar? what do you mean similar? The only test you want is Denuvo vs no Denuvo and they do not do it. So what’s the point of asking for something that won’t happen?
LOL go pirate something.
or maybe it is due to the patch which increases performance stop being delusinal and spreading miss information
because patches increase perfomance so much right? How come ghost warrior 3 patches didnt do jack sht and the game still has denuvo.
and that was a old version of denuvo.
Obviously anti-denuvo gamers will tell you their performance jumped up after Denuvo got removed.
My hair started to grow back on my bald spot after denuvo was removed.
Why would GPU usage go up if denuvo was negatively impacting performance? If anything GPU usage should stay the same or go lower but simultaneously offer better performance.
The evidence just keeps coming lol… Let me quote what you said to me earlier:
“Don’t try to roast people on brain cells when you have none to begin with”
Seems like you don’t have much of a clue on how computer hardware works. So allow someone who (according to you) have no brain cells teach you something:
The higher the utilization of the hardware is, the more performance it pushes out. If a GPU is bottlenecked by a weak CPU, the GPU utilization will go down because the GPU can’t stretch its legs. If a game has performance issues and doesn’t take advantage of the hardware as it should, the utilization goes down. It can also be a case where a software or instruction set or a DMR is occupying CPU performance making the CPU utilization go up higher than what the game itself would require, in that case less performance is dedicated to the game/program from the CPU and thus the GPU can’t stretch its legs.
For maximum performance, the GPU should run at 100% or nearly 100% to push as high framerates as possible regardless of settings. Let’s assume your hardware runs at 100% then if you tweak the settings down then the leftover performance will allow the FPS to rise higher, or if you crank details or resolution up then FPS will be lower. Bottom line, if the GPU is not utilized at or near 100% then there is something wrong, unless of course you or the game has capped the performance in some way, for example with v-sync or a framerate cap.
Now you became a little bit smarter thanks to me 🙂
The moderators still haven’t approved my comment from a day ago. Here’s the comment without the link:
Lol settle down there Mr.Ego, everything you said is basic computer knowledge that is in no way relevant to Denuvo because Denuvo doesn’t bottleneck the GPU by restricting CPU performance. If that were the case, then what you said would be true.
But it isn’t, and I already knew that. That’s the reason I didn’t mention GPU bottlenecking by the CPU. Because that’s not relevant to Denuvo.
Denuvo adds performance drops wrt to disk usage. It adds extra calls for the hard drive to read/write and that can slow-down things such as load times and streaming of assets (or virtual textures if the game has that). It would bottleneck the GPU AND CPU but only to an extent that has negligible impact on frametimes .
If you actually look at the evidence (and I don’t mean random people on Reddit claiming things with no posted stats and benchmarking procedures to back it up):
Since the mod’s are against links, just google “Does DENUVO DRM REALLY HURT Game Performance? by Joker Productions”
Denuvo has an impact on framerate that’s within a margin of error. Denuvo’s real performance issues are stuttering and hitching due to the disk not being able to feed data fast enough. If that bottleneck was removed (by removing Denuvo) it would not affect FPS in any significant way, but it would alleviate some stuttering/hitching issues that may have been occurring.
But to claim that the disk was bottlenecking the GPU by 20% and FPS went up noticeably is utter bullshi*.
Once again, you build up a strawman to attack me for some unknown reason (read: aggressively pathetic and extremely butthurt) only to realize that you are arguing against something I never even alluded to. Keep trying though, one day you might actually grow a brain cell.
LOL
Either you responded to the wrong comment OR you’re dumber than I thought. I wasn’t talking about whether or not Denuvo affects the performance in AC:O. I just responded to your question regarding how CPU and GPU usage works in games.
Calm down Indian boy.
I saw both your comments you idiot. I replied to this days ago and I’ve already replied to your other one. The mods haven’t approved the comment above so I posted it without the link. Also who tf is talking about AC:O?? You replied to me above regarding GPU bottlenecking. What you said was absolutely irrelevant to Denuvo’s impact on GPU/CPU utilization. I replied saying that it Denuvo had no such impact on the GPU/CPU and that the cause of Denuvo’s performance drop was coming from the HDD. This isn’t hard.
Ahaha, try to cool down and perhaps think before you respond to people? I don’t care if you put links so that your comments doesn’t show. You should know that by now and edit those links so that mods doesn’t have to waste time and effort into that just because you’re too stupid to edit those links properly you moron.
Yeah, wrong article let me correct myself:
I wasn’t talking about whether or not Denuvo affects the performance.
Happy now?
“What you said was absolutely irrelevant to Denuvo’s impact on GPU/CPU utilization.”
Exactly, that’s what I pointed out to you, hence my reaction to your long and pathetic comment, why state something again that we already know? By the way, good thing you removed that comment, such a waste of bandwidth!
“I replied saying that it Denuvo had no such impact on the GPU/CPU and that the cause of Denuvo’s performance drop was coming from the HDD.”
Yeah, I already know that from that long and pathetic sh*tpost you threw at me which you later deleted. Again, stop mentioning things we already know.
“This isn’t hard.”
Yupp looks like it is hard for you, to read and understand comments, to know how to respond correctly.
A CPU sends Draw Calls to a GPU. If a CPU is being used to run code unnecessarily during a game then it will not be able to calculate and send as many draw calls to the GPU. FPS will vary depending on the CPU being used. Assuming that a user is using a CPU that is capable of running the game while simultaneously running the DRM then no problem. Some people will experience problems, others won’t.
The CPU submitting more draw calls will not necessarily boost performance. In fact, in most cases, more draw calls reduces performance. The reason being is that a GPU can draw triangles and vertices much faster than the CPU can send drawcalls. If you send too many draw calls with only a few triangles in each call, the GPU sits idle while the CPU keeps sending calls. It’s better to send fewer draw calls with more triangles in each call to prevent bottlenecking of both the CPU and GPU. However, that is not the issue here, that would only be applicable within the game code itself, not 3rd party tools like Denuvo.
As I’ve stated above (with benchmarks to prove it), Denuvo does NOT affect FPS or frame time in any significant way (it is within the margin of error). This mainly has to do with the fact that VMWare/Denuvo doesn’t obfuscate the code on the fly but instead has memory triggers that get checked on the fly.
What it DOES affect is the disk I/O and R/W performance (mainly Read). The obfuscated code sits on the HDD (in the game folder) but get’s parsed on the fly.This causes the HDD/SSD (SSD’s play Denuvo enabled titles very well however, they do get read from a lot) to have to read the code to the CPU at a slower rate than a standard game w/o Denuvo. When too much data needs to be read from at once, you get stuttering and hitching because of the lack of sufficient storage speed.
Basically, removing Denuvo should not free up 20% of the GPU’s resources. Thats utter horseshi*. That kind of bottleneck is insane and completely rebutted by benchmarks.
Denuvo takes CPU cycles as already stated by Denuvo itself, was a nice try to act smart copying articles from wikipedia though.
Can you read? I didn’t say Denuvo doesn’t use the CPU at all, I said it doesn’t bottleneck the GPU by 20% because that’s ridiculous. The benchmarks prove that it has barely any impact on FPS which is impacted almost entirely by CPU and GPU utilization. Nice try at being snarky but leave that to people with an IQ above 40.
Man… RIP. lolololol owned as faq
Yeah lol, Mr Boinpally keeps embarrassing himself over and over on this site. It’s amusing how dumb he is.
Nice copy-pasting a quote from Joakim HĂĄrsman at stackoverflow. The only problem is that you changed his words to try to validate your false claim.
Original comment by Joakim:
“If you submit few triangles with each call, you will be completely bound by the CPU and the GPU will be mostly idle. The CPU won’t be able to feed the GPU fast enough.”
Your edit:
“If you send too many draw calls with only a few triangles in each call, the GPU sits idle while the CPU keeps sending calls.”
That “too many” part is just something you made up lol.
Furthermore he said:
“The main reason to make fewer draw calls is that graphics hardware can transform and render triangles much faster than you can submit them”
That comment doesn’t really make sense, I think it was taken a bit out of context here. Why would you limit the amount of calls when the GPU is asking for more due to it being faster at executing the calls than receiving them. Ideally the CPU should feed the GPU as much as possible thus working at 100% and being dedicated to the game. Ideally the CPU should put all of its efforts and resources to the game, and not to background processes, windows or a DRM. I think you misunderstood that comment and based your point on that.
Quoting YOU:
“It’s better to send fewer draw calls with more triangles in each call to prevent bottlenecking of both the CPU and GPU.”
Wrong, It’s better to send as many draw calls as possible with as much information as possible (to the extent of what your CPU is capable of) in each call in order to keep the GPU busy, the less calls or the less information the GPU gets the more it will “idle” and can’t be utilized to its maximum. That’s why framerate is affected depending on how fast a CPU is and also what clock speed its running. Also, sending less information to the GPU won’t bottleneck the CPU lol, only the GPU. Stop making things up.
Then he said:
“But the main cost of draw calls only apply if each call submits too little data, since this will cause you to be CPU-bound, and stop you from utilizing the hardware fully.”
Which is also true. To sum it all up, you will be CPU-bound if each draw call contains too little data, not necessarily dependent on the amount of calls alone.
Why are you accusing me of copying him? What he stated is common game design knowledge. It’s not “copy pasting” if you said something different.
And yes too many draw calls IS a bad thing. Every major Game Engine has documentation explicitly stating not to have too many draw calls because it hurts performance. Infact, Cryengine has a pretty explicit section saying not to go over 2000 draw calls in any scene (although higher end hardware can handle more).
“Wrong, It’s better to send as many draw calls as possible with as much information as possible (to the extent of what your CPU is capable of) in each call in order to keep the GPU busy, the less calls or the less information the GPU gets the more it will “idle” and can’t be utilized to its maximum. That’s why framerate is affected depending on how fast a CPU is and also what clock speed its running. Also, sending less information to the GPU won’t bottleneck the CPU lol, only the GPU. Stop making things up.”
LOL. If you honestly think that sending as many draw calls as possible is a good thing, you seriously need to understand the basics of game engines. “to the extent of what your CPU is capable of” THAT’S THE POINT. This is why I said “too many”. Developers don’t know what your CPU is capable of. They can’t keep sending draw calls because it bottlenecks the CPU and by extension bottlenecks the GPU, since as we both know, it’s under utilized.
In almost any modern game engine, batching is implemented specifically to REDUCE DRAW CALLS. Every big optimization lecture from GDC talks about using instancing, and batching, and culling ALL so that you can minimize draw calls and geo rendering.
So no, you should keep the draw calls low and keep the info inside each call high. GPU’s are much better at drawing triangles than they are at reading draw calls. What Joakim and I said are one and the same. He even stated other reasons why Draw Calls hurt performance. Namely
“There are some real costs with making draw calls, it requires setting up a bunch of state (which set of vertices to use, what shader to use and so on), and state changes have a cost both on the hardware side (updating a bunch of registers) and on the driver side (validating and translating your calls that set state).”
The whole reason we are having this discussion is because someone thinks that Denuvo is occupying CPU time that could be otherwise used to drive the GPU 20 PERCENT harder. That’s false. I agree that if it actually took up that much CPU time, it would choke the CPU and thereby underutilize the GPU. But that’s not the case AT ALL. (look up “Does DENUVO DRM REALLY HURT Game Performance?”, the benchmarks speak for themselves).
That’s where settings come into play, reduced amounts of draw calls means more performance on weaker systems yes, but after optimization that also means a less complex and less visually impressive looking scene. The 2000 draw calls by Crytek is a mere recommendation and can be higher for higher presets for high-end systems. However you want to to bring down the draw calls by optimization to make room for improvements in visuals and similar areas that requires more draw calls when those settings are bumped up and also to make room for post-processing visuals.
You don’t bring down the draw calls just for the sake of having as low amount of draw calls as possible, but to make room for better visuals on each graphics option preset, that’s where techniques like Culling etc comes into play too, to distribute and direct as much of the draw calls to higher fidelity assets, draw distance, object complexity and detail.
You want the CPU to be utilized at 100% in games to feed the GPU so that it can also perform at 100% too. Then lowering craw calls on lower settings presets to suit lower end hardware. That’s where system requirements come into play too, they build their levels and adjusts the amount of draw calls to suit different CPU’s, you can’t have crazy amount of draw calls on the lower settings because that would hurt performance on lower end hardware, but they should however be kept as high as possible for each preset level and corresponding hardware.
But developers tend to want to keep draw calls low to suit their games for the wider range of gamers with mid-range hardware, that most people use. Crytek didn’t follow that approach when they developed Crysis back in the days, they pushed technology and visuals in a way most other companies didn’t. Performance was bad yes, but the visuals where justified, they were simply on another level. But most people don’t want that, they want a game that runs great and doesn’t care as much for graphics. The same people called Crysis very unoptimized just because it ran bad on their mid-range systems. Today, developers doesn’t push visuals like that, just look at AC:O for example, the visuals are almost the same on PS4 Pro as on the PC version, with the exception of higher resolution shadows and slightly more complex objects in the distance.
Another good example of this is when console games are ported to PC, most of the time developers don’t redo the optimization for a wide range of systems and performance is bad and/or we see examples where CPU/GPU utilization isn’t correctly used. One of the big benefits of developing on consoles is that you don’t have all these different hardware setups to take into account. Also one of the main reasons why there are so many lectures, articles, discussions about optimizing engines to use as little draw calls as possible is because of consoles, to make room for a higher resolution “4K” you know, because that’s the main focus of the current gen consoles, to bump up the resolution draw calls must be kept minimal to give headroom of performance in favor for higher resolutions.
So yeah, to keep draw calls suitable for different hardware and different presets and to use as high amounts information in each call is the key, not to keep the amount of draw calls as low as possible across the board just for the sake of having low amounts of draw calls, at least not regarding the PC version of games or want to push beautiful and detailed graphics. Games like CS, Minecraft, LoL, that’s where low amounts of draw calls are important, to make those games playable on the most amount of systems, sh*tty laptops and even school computers, playability and compatibility is the priority, not the visuals.
The game failed, no coop and boring , doesnt matter at this point.
no customization.
Games flop remove denuvo, what else is there? Oh yes and kim kardashian west is a troll.
to be fair they dropped hitman and thus devs dropped denuvo and something else happened with life is strange, plus not alot of new games from them recently.
Ubisoft though is the definition of idiocy.
So that’s EA, Ubisoft, Square Enix, and Warner Bros. (Injustice 2)
Konami are a chapter on it’s own.
Far too late to garner trust or more sales. You signed your poor sales warrant, the moment you added the DRM.
I tried to watch a playthrough of this on YouTube but the gameplay seemed so dull and repetitive and the action didn’t seem impactful. Gun sounds were meek and there was no punch to anything. And the characters trying to be cool all the time just came off as annoying for the most part. It’s a shame because I was really, really hyped for AoM.
I will not get it because has nothing that made Saints Row fun. No customization, no coop, and not a lively city.
Even removing denuvo wont save this game.
lol I remember when John was one those Pro Denuvo people
What happens when you think patience always means wait?
Well you’re not wrong. And filtering out trolls can be nice. But who else would fight there BS and misinformation and stand up for the truth then?
Let me quote something from the official PcMasterRace homepage:
“To strongly contribute to the spread of correct information as well as to fight against misinformation in technology and in gaming.”
That’s their second main objective and I feel like it’s my obligation as a PC enthusiast to try to follow that, especially on a PC focused site that I love like DSOG. Personally though, I don’t seek endless arguing or flamewars with fanboys and/or console peasants out. I kind of just take it on when it appears in front of me on this site mainly. I don’t venture to sites with a strong infestation of console players and/or fanboys.
I made the mistake to visit one of the comment sections of a Digital Foundry article that’s being hosted on the eurogamer site. The place just crawled with peasants and butthurt fanboys. I saw one guy, a PC user who ranted about how nice it was to play AC:O at 60 FPS and how great it ran on his rig. He was downvoted so badly I don’t even know where to start. After trying to defend him also I got downvoted to oblivion, and as soon as I started to call the fanboys out, some moderator removed my posts. Pathetic!