• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Devil May Cry 5 Benchmark Performance Analysis

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,665 (3.74/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
Devil May Cry 5 is one of the most anticipated titles this season. The third-person hack-and-slash action adventure comes with outstanding graphics thanks to DirectX 12 support and incredibly detailed character models and textures. Optimization is excellent, too, especially for AMD Radeon cards.

Show full review
 
Last edited by a moderator:
DX12 shows that when done properly, it helps muchly to allow gamers to play such visual impressive games close to 60FPS without needing a close to $1K GPU, even at 4K resolution.
 
Just because there's no reference GTX 1660 Ti doesnt mean a reference clocked GTX 1660 Ti shouldn't be in the charts....

:(
Yeah I noticed after leaving for the weekend when I wrote on the train, will add on Monday. And you can't set reference clocks to a 1660 ti
 
Just because there's no reference GTX 1660 Ti doesnt mean a reference clocked GTX 1660 Ti shouldn't be in the charts....

:(
Or you can use reference 1070 data, as they perform the same.
 
DX12 shows that when done properly, it helps muchly to allow gamers to play such visual impressive games close to 60FPS without needing a close to $1K GPU, even at 4K resolution.

Well, DMC always ran on a toaster, it even ran at 60 FPS on the consoles ever since day / part one.

The engine uses a very efficient way to show environments, lots of pre-cooked elements, fixed camera perspectives etc. Its built to run at perfect fixed FPS because that is what the gameplay also demands (timed button presses) much like beat-em-ups. Bayonetta has this too. Another interesting feature is that lighting and shadowmaps are also pre rendered most of the time. As a result you will see a very limited dynamic lighting system, if any.
 
Last edited:
Or you can use reference 1070 data, as they perform the same.

Overall that is true, but there is game to game variance with that. RE2 runs a little better on Turings so 1660ti might have little edge on this game too. But yeah, overall gaming experience will likely be the same between the two.
 
Thanks for the write up. It's refreshing to play a game that perform so well and play is so smooth. On my system I'm playing full ultra settings 170% resolution scale 1080p averaging high 80-90fps
 
Looks like a really good engine for this style of game. Performance is good from Nvidia and AMD with rather low vram usage. I actually think the bandwidth is holding back the GTX 1060 more than anything. The rest of the nVidia lineup does well for their weight class.

Curious to see how well the Fuji X does as well as how poorly a 1050ti does due to its 128 bit GDDR5.
 
Really? Rx 570 beating gtx 1060 6gb? Another day,another AMD sponsored and gimped game (like resident evil 2 remake). I bet some people will use it to prove their mythical 'AMD FineWine' agenda. Lol. If nvidia did this,there would be rioting in the streets already!
 
Last edited:
Really? Rx 570 beating gtx 1060 6gb? Another day,another AMD gimped game (like resident evil 2 remake). I bet some people will use it to prove their mythical 'AMD FineWine' agenda. Lol. If nvidia did this,there would be rioting in the streets already!

It seems you have a hard time telling the difference between a gimped brand A and a well optimized brand B. Even the so called "gimped" GTX 1060 still manages over 70 fps at 1080p with impressive visuals. It is clear that the game likes bandwidth as the GTX 1070 does very well itself.

The same can not be said by other visual turds like Just Cause 4 where a Vega 56 can't even hit 60 fps at 1440p.
 
It seems you have a hard time telling the difference between a gimped brand A and a well optimized brand B. Even the so called "gimped" GTX 1060 still manages over 70 fps at 1080p with impressive visuals. It is clear that the game likes bandwidth as the GTX 1070 does very well itself.

The same can not be said by other visual turds like Just Cause 4 where a Vega 56 can't even hit 60 fps at 1440p.

It seems like you don't know what 'gimped' means. Just bcz it gets high fps,doesn't mean it's not gimped for other brand. If rx 570 4gb gets 60fps,then gtx 1060 should be getting 68-70fps,not 55fps. Also Vega 56 is doing just fine,beating gtx 1070Ti on higher resolution even though this game doesn't require any more than 4gb vram. Not to mention this game is AMD sponsored. And as Vayra86 said,DMC games can run even on toasters so the high fps on mid range gpus doesn't have anything to do with 'not gimping'.
 
It seems like you don't know what 'gimped' means. Just bcz it gets high fps,doesn't mean it's not gimped for other brand. If rx 570 4gb gets 60fps,then gtx 1060 should be getting 68-70fps,not 55fps. Also Vega 56 is doing just fine,beating gtx 1070Ti on higher resolution even though this game doesn't require any more than 4gb vram. Not to mention this game is AMD sponsored. And as Vayra86 said,DMC games can run even on toasters so the high fps on mid range gpus doesn't have anything to do with 'not gimping'.
Ever noticed that most AMD sponsored titles run well on both NVIDIA and AMD (not AC Odissey...) while most NVIDIA sponsored titles run like crap on both (with nvidia taking the lead)?
 
It seems great performance and nice visuals on both brands is not enough for some. It must be artificially gimped when their precious doesn't beat the competition.

I guess it is like people having a hard time telling the difference between a stellar offense and a crappy defense in a high scoring game.

Rendering cows with hairworks 2 miles away from view, such as in FFXV, would be an example of a "bad defense."
 
Ever noticed that most AMD sponsored titles run well on both NVIDIA and AMD (not AC Odissey...) while most NVIDIA sponsored titles run like crap on both (with nvidia taking the lead)?

Ever noticed most of the AMD sponsored titles are linear close quarter single lane game with static geometry and sub-par visual with an overlayed filtre to make it look cool. While most Nvidia titles are huge open world titles with complex and dynamic geometry which ends up being too heavy for gpus? AMD tried that with AC Odyssey,what happened? Boom,even shittier result than nvidia titles.

It seems great performance and nice visuals on both brands is not enough for some. It must be artificially gimped when their precious doesn't beat the competition.

I guess it is like people having a hard time telling the difference between a stellar offense and a crappy defense in a high scoring game.

Rendering cows with hairworks 2 miles away from view, such as in FFXV, would be an example of a "bad defense."

It seems like my comment went right over your head.
 
Ever noticed most of the AMD sponsored titles are linear close quarter single lane game with static geometry and sub-par visual with an overlayed filtre to make it look cool. While most Nvidia titles are huge open world titles with complex and dynamic geometry which ends up being too heavy for gpus? AMD tried that with AC Odyssey,what happened? Boom,even shittier result than nvidia titles.



It seems like my comment went right over your head.
I think you wrote something about AC Odyssey and AMD GPUs having lower performance than nVIdia...

assassins-creed-odyssey_1920-1080.png
assassins-creed-odyssey_2560-1440.png
 

He's right.
Nvidia cards are generally faster in AC Odyssey, different scenes in the game can cause different results, but nvidia is better overall at this one.
Even if we ignore this fact, the vega 64 which is AMD's second fastest gaming GPU is losing to a 2060 which is not even nvidia's third fastest card.
 
Last edited:
Hmm... Whenever I see a debate about which is better I cringe a bit. Numbers don't lie. If analysis is done right it can be treated as established truth unless proven otherwise.

I think a proper way to differentiate video cards would be to do an analysis of the instructions supported by each and take all engines and study them architecturally and observe what the strong/weak points of each are when ran on different GPU's. Saying one is gimped because it's optimized for the other proves nothing.

The numbers that we have from W1zzard and other reviewers are at the level of better than nothing. They are not the full picture. That's why we see differences like this happening. AMD taking the lead when in most titles NV is the clear winner. There are clearly differences at the hardware architecture level that outreach into software performance based on usage patterns. That cannot be mitigated by software updates, it takes a whole new hardware arch.

So unless people try to understand the exact methods by which GPU's and instructions on those GPU's function and how different ways of using them affects performance we will never really know the truth of which is better than which (or maybe one is better than the other in certain use-cases, which is what I would tend to agree with). It's just an estimation and we need to treat it as such.

P.S An estimation is better than nothing.
Cheers.
 
Ever noticed most of the AMD sponsored titles ...

Ever notice that most forum warriors like to give only anecdotal evidence to try and prove their point? One game a trend does not make. There are plenty of open world games that are AMD sponsored as well.

This game is just too much for 192 bit GDDR5.
 
Ever notice that most forum warriors like to give only anecdotal evidence to try and prove their point? One game a trend does not make. There are plenty of open world games that are AMD sponsored as well.

This game is just too much for 192 bit GDDR5.

Ever noticed that most forum warriors don't properly read the comment/understand the comment and jumps to BS argument? I guess you overlooked the part 'most of the'. Yes there are some open world AMD titles like kingdom come or recent fc games and division games, but they also run like crap,just like AC odyssey. The only exception would be fh4. So yeah,try to read the comment carefully first and also try to understand what's been trying to indicate.

Also noticed that people will come up with a made up theory and try to shove it down to other's throat as a fact. "This game is just too much for 192 bit GDDR5" - where's the proof of that?? You just made up a conclusion based on what you 'think' is the case. It's like saying - "ohh that dimly lit light bulb is giving same amount of light as the moon,hence it's proven that the moon and the light bulb are situated at same distance from me"

It's easily noticable that AMD gpus are getting severely unfair advantage compared to nvidia gpus(turing aside) regardless of memory bandwidth across the board.

Ohh and also, this game is too much for 192bit bandwidth? Then how come rtx 2060 with 192bit bandwidth managing to beat Vega 56?? You might say ohh i said gddr5,not gddr6; but that does really make up to get advantage against the 2048 bit and 409GB/s bandwidth of Vega 56 if the bandwidth was actually a factor as you said?? Definitely not.

Also you are mixing up replies from one thread with another thread.


Actually i said AMD sponsored huge open world games also runs like shit/shittier than nvidia titles. I think you thought i said AMD runs shittier than nvidia in ac Odyssey. I didn't mean that.

But since you brought that up, AMD is actually getting worse fps than Nvidia in their own sponsored game which also runs like crap on both brands. So much for 'amd games always run better on gpus from both brands' LOL. Reminds me of deus ex mankind divided and kingdom come. Most AMD games aren't graphically intensive(i said most,not all, don't get triggered like that other guy plz),and whenever they are graphically intensive,they run like crap on gpu from both brands,with exceptions like fh4 ofcourse.
Nvidia also have exceptions where their games run much better on both brands, like Witcher 3,bf5,sottr,hellblade and many more. But AMD loyalists always tend to overlook that segment and try to generalize the negative part of nvidia and positive part of AMD. A game being heavy/not heavy doesn't always depend on which brand sponsored it. Most of nvidia titles are graphically demanding open world/semi open world,so most of nvidia games end up being heavy. Most of AMD games are close quarter normal games,so most AMD games end up being much less heavy. Yes, nvidia intentionally gimped only a few games with their gimpworks,but other demanding games are heavy for pretty fair reasons. Most people don't know that.
 

Attachments

  • fymjqmAqesFmnXPp395DZ5-650-80.png
    fymjqmAqesFmnXPp395DZ5-650-80.png
    144.8 KB · Views: 561
  • FLAK4DevwGZwyoKqocQ9d5-650-80.png
    FLAK4DevwGZwyoKqocQ9d5-650-80.png
    134.2 KB · Views: 626
  • wTa2Xb3E6iCjn7prSBDbL5-650-80.png
    wTa2Xb3E6iCjn7prSBDbL5-650-80.png
    134.5 KB · Views: 644
Last edited:
I hate to digress, but you made this point twice. I have to ask: AC Odyssey runs like crap? :confused:

I failed to notice this in a 221 hours so far.

Looking at the fps numbers of current/latest 1080p/1440p cards,yes it runs like crap. May be you have OP desktop setup/running on lower settings, I don't know. But unable to hit even 50fps on 1080p high with rx 580! Nothing to say.
Even gtx 1080Ti can't even hit 60fps on 1440p. If this isn't the definition of 'running like crap',then I don't know what is!
 

Attachments

  • vcSaEjDkbBkoBwC9QK8Jb5-650-80.png
    vcSaEjDkbBkoBwC9QK8Jb5-650-80.png
    143.4 KB · Views: 625
Even gtx 1080Ti can't even hit 60fps on 1440p. If this isn't the definition of 'running like crap',then I don't know what is!
Except I do, regularly. It varies between 53 and 60. And no low or medium settings at all. Lowest setting is shadows on High. So, reviews are not all they are chalked up to be. Personal experience is.
 
Ohh and also, this game is too much for 192bit bandwidth? Then how come rtx 2060 with 192bit bandwidth managing to beat Vega 56?? You might say ohh i said gddr5,not gddr6; but that does really make up to get advantage against the 2048 bit and 409GB/s bandwidth of Vega 56 if the bandwidth was actually a factor as you said?? Definitely not.

Well done. You answered your own question. I didn't say bandwidth was an absolute limit as you only need so much at a certain frame rate. Just look how the 1070ti closes the gap on the 1080 as the frame rate is reduced.
 
Well done. You answered your own question. I didn't say bandwidth was an absolute limit as you only need so much at a certain frame rate. Just look how the 1070ti closes the gap on the 1080 as the frame rate is reduced.

Every gpu closes gap when the framerate is lowered,lol,that doesn't prove rx 570 beating 1060 6gb due to bandwidth or anything. And my comment didn't fall in line with any of your pseudoscience.
Certain framerate you say? Then how come rtx 2060's 192bit bandwidth beating Vega 56's 2048bit bandwidth in both higher and lower framerate? How come radeon 7's 1tb/s bandwidth being beaten by gtx 1080Ti in higher fps but in lower fps, radeon 7's higher bandwidth beats gtx 1080Ti? That's completely the opposite of what your pseudo theory suggests about 1070Ti closing in with 1080 and also your base pseudoscience damage control for rx 570 4gb beating 1060 6gb. Instead of cherry picking one gpu to prove your pseudoscience narrative to defend AMD's gimping,try to look at broader picture here.

Except I do, regularly. It varies between 53 and 60. And no low or medium settings at all. Lowest setting is shadows on High. So, reviews are not all they are chalked up to be. Personal experience is.

53-60fps? So, almost 57fps on average. Same as the given review said. So,1080Ti struggling just to get 60fps on average, perfect example of running like shit. Same for other gpus.
Just to clearify,i said ac Odyssey runs like crap in the context of that other guy saying nvidia titles run like crap. If nvidia titles run like crap,then AMD titles like ac Odyssey also runs like even more crap which he overlooked. I'm not singling out ac Odyssey out of context. I guess you missed the initial comment that i was replying to.
 
Last edited:
Back
Top