Corsair H115i RGB Pro XT liquid cooler review
XFX Radeon RX 5600 XT THICC 3 review
Fractal Design Define 7 XL review
Guru3D Rig of the Month - January 2020
Palit GeForce RTX 2070 SUPER GamingPro review
Team Group T-Force DarkZα 3600 MHz DDR4 review
Corsair Scimitar Elite RGB gaming mouse review
Cougar 700K Evo keyboard review
Aorus Liquid Cooler 360 review
AMD Ryzen Threadripper 3990X review
Deus Ex: Mankind Divided PC GPU (DX11 and DX12) performance benchmark review
Posted by: on: 09/08/2016 12:28 PM [
164 comment(s) ]
We take a look at Deus Ex: Mankind Divided for the PC in our usual tech savvy way of graphics performance articles. We'll test the game on the PC platform relative towards graphics card performance with the latest AMD/NVIDIA graphics card drivers. Multiple graphics cards are being tested and benchmarked. We have a look at performance with the newest graphics cards and technologies.
Read article
Advertisement
pages 1 2 3 4 > »
Undying
Senior Member
Posts: 12510
Senior Member
Posts: 12510
Posted on: 08/24/2016 06:50 PM
Im looking forward to dx12 benchmark and comparing the two.
Fury X looks impressive.
Im looking forward to dx12 benchmark and comparing the two.
Fury X looks impressive.
narukun
Senior Member
Posts: 189
Senior Member
Posts: 189
Posted on: 08/24/2016 07:01 PM
Just because you dont believe, doesnt mean that other people can't. Everyone can have their own point of view.
Yeah, ignoring not because the DX12 renderer isn't out -- it's because it's all some conspiracy against AMD. Can you stop posting nonsense or at least edit your other useless post and add this to it?
Just because you dont believe, doesnt mean that other people can't. Everyone can have their own point of view.
Dygaza
Senior Member
Posts: 535
Senior Member
Posts: 535
Posted on: 08/24/2016 07:09 PM
I don't know if you are totally ignorant, or just trolling. But obiviously DX12 path ain't available for the game yet. Also DX11 tests will still have their place in the near future.
I fully expect more and more dx11 tests done ignoring the new API`s to try and limit the damage of the upcoming AMD hardware that handles dx12 and vulkan so much better due to there inherent nonlinear architecture design.
I don't know if you are totally ignorant, or just trolling. But obiviously DX12 path ain't available for the game yet. Also DX11 tests will still have their place in the near future.
Dazz
Senior Member
Posts: 866
Senior Member
Posts: 866
Posted on: 08/24/2016 07:15 PM
Only on that demo. In real game when you get further in game, AMD cards start getting cpu bottlenecked and nvidia cards start performing better in comparision. Benchmark only shows kinda worst case scenario graphically, but not by cpu load.
Still overall, I'm enjoying the game, and there's plenty of option to get fps higher if needed without really sacrificing anything.
True, although benchmarks are really there for worse case scenarios rather than typical game play.
Only on that demo. In real game when you get further in game, AMD cards start getting cpu bottlenecked and nvidia cards start performing better in comparision. Benchmark only shows kinda worst case scenario graphically, but not by cpu load.
Still overall, I'm enjoying the game, and there's plenty of option to get fps higher if needed without really sacrificing anything.
True, although benchmarks are really there for worse case scenarios rather than typical game play.
Reddoguk
Senior Member
Posts: 1807
Senior Member
Posts: 1807
Posted on: 08/24/2016 07:17 PM
I never really take these built in game benchmarks too serious because they are all Reference models only and those of us with G1's or SSC or any other top branded card will totally smash HH's scores.
The tests are for reference only as i know for a fact that my card would probably get equal or higher scores than the reference 980Ti in this test.
Then again it's hard for me to get a medium between 1440 and 1080 because i still love my old Dell ultrasharp 24" monitor that still to me looks amazing @ 1200p.
Still think this is too early even for DX11 because i bet you can add 10-20 extra fps once the game gets proper optimization along with better driver optimization.
Not 100% sure if DX11 will even get any optimization but i'll be interested in seeing DX12 results and then see if it gets better over time or not.
I never really take these built in game benchmarks too serious because they are all Reference models only and those of us with G1's or SSC or any other top branded card will totally smash HH's scores.
The tests are for reference only as i know for a fact that my card would probably get equal or higher scores than the reference 980Ti in this test.
Then again it's hard for me to get a medium between 1440 and 1080 because i still love my old Dell ultrasharp 24" monitor that still to me looks amazing @ 1200p.
Still think this is too early even for DX11 because i bet you can add 10-20 extra fps once the game gets proper optimization along with better driver optimization.
Not 100% sure if DX11 will even get any optimization but i'll be interested in seeing DX12 results and then see if it gets better over time or not.
millibyte
Senior Member
Posts: 1152
Senior Member
Posts: 1152
Posted on: 08/24/2016 07:19 PM
Good showing for AMD cards (comparatively) but crappy overall.
Good showing for AMD cards (comparatively) but crappy overall.
Primey0
Senior Member
Posts: 121
Senior Member
Posts: 121
Posted on: 08/24/2016 07:36 PM
Gee I wonder why. Hm maybe because the game and engine favours amd over Nvidia?
RX480 beats GTX1060 once again, but this time in a DirectX11 game. Glad to see that.
Gee I wonder why. Hm maybe because the game and engine favours amd over Nvidia?
Undying
Senior Member
Posts: 12510
Senior Member
Posts: 12510
Posted on: 08/24/2016 07:39 PM
Still, difference is there, it will probably become even bigger with dx12.
Gee I wonder why. Hm maybe because the game and engine favours amd over Nvidia?
Still, difference is there, it will probably become even bigger with dx12.
primetime^
Senior Member
Posts: 897
Senior Member
Posts: 897
Posted on: 08/24/2016 07:40 PM
Cant wait for the dx12 follow up!
Cant wait for the dx12 follow up!
millibyte
Senior Member
Posts: 1152
Senior Member
Posts: 1152
Posted on: 08/24/2016 07:49 PM
I'm ready for the 480 to beat the 1070 in dx12. I'll laugh...and then I'll cry...
Still, difference is there, it will probably become even bigger with dx12.
I'm ready for the 480 to beat the 1070 in dx12. I'll laugh...and then I'll cry...
eclap
Senior Member
Posts: 31497
Senior Member
Posts: 31497
Posted on: 08/24/2016 07:55 PM
Who here honestly cares about the dx11 results except people who own nvidia cards.
DX12 is what matters this test is MUTE and pointless but no surprise it was tested here.
Mute?
And yeah, who cares about dx11? Oh, apart from those who bought the game.
Who here honestly cares about the dx11 results except people who own nvidia cards.
DX12 is what matters this test is MUTE and pointless but no surprise it was tested here.
Mute?
And yeah, who cares about dx11? Oh, apart from those who bought the game.
RotatingFan
Senior Member
Posts: 31
Senior Member
Posts: 31
RotatingFan
Senior Member
Posts: 31
Senior Member
Posts: 31
Posted on: 08/24/2016 07:58 PM
wont happen
Na the 1070 is to strong for the 480. the 1070 is 2 classes above the 480 but the 1060 card is a different matter.
wont happen
I'm ready for the 480 to beat the 1070 in dx12. I'll laugh...and then I'll cry...
Na the 1070 is to strong for the 480. the 1070 is 2 classes above the 480 but the 1060 card is a different matter.
Carfax
Senior Member
Posts: 2678
Senior Member
Posts: 2678
Posted on: 08/24/2016 08:08 PM
So this guy tests the GTX 1060 vs RX 480 and found the former to beat the latter, and everyone cries foul.
Then he does a retest, but with a different CPU that lacks hyperthreading, and now the RX 480 wins, but with inferior minimums
How ridiculous is that? If he's going to retest, he should have stuck with his original platform..
So this guy tests the GTX 1060 vs RX 480 and found the former to beat the latter, and everyone cries foul.
Then he does a retest, but with a different CPU that lacks hyperthreading, and now the RX 480 wins, but with inferior minimums
How ridiculous is that? If he's going to retest, he should have stuck with his original platform..
Undying
Senior Member
Posts: 12510
Senior Member
Posts: 12510
Posted on: 08/24/2016 08:13 PM
So this guy tests the GTX 1060 vs RX 480 and found the former to beat the latter, and everyone cries foul.
Then he does a retest, but with a different CPU that lacks hyperthreading, and now the RX 480 wins, but with inferior minimums
How ridiculous is that? If he's going to retest, he should have stuck with his original platform..
Changing the CPU to i5 could only give the 480 disadvantage. Higher clocked i7 would probably even improve amd performance if he did the test on his original system.
So this guy tests the GTX 1060 vs RX 480 and found the former to beat the latter, and everyone cries foul.
Then he does a retest, but with a different CPU that lacks hyperthreading, and now the RX 480 wins, but with inferior minimums
How ridiculous is that? If he's going to retest, he should have stuck with his original platform..
Changing the CPU to i5 could only give the 480 disadvantage. Higher clocked i7 would probably even improve amd performance if he did the test on his original system.
ObscureangelPT
Senior Member
Posts: 541
Senior Member
Posts: 541
Posted on: 08/24/2016 08:22 PM
FYI, the in game benchmark is a lot different, and also tends to inflate AMD performance relative to the actual game itself. AMD seems to be doing this a lot recently with their G.E titles like Total Warhammer, where the scripted benchmark will return higher values for AMD hardware compared to the game itself.
PCGH.de tested the game using the Prague area, and there the performance was not only much better overall for both NVidia and AMD, but NVidia took the lead over AMD:
PCGH.de test results
The worse results are due to DX11 overhead bottleneck, not lack of AMD GPU Power.
The game makes use of a lot of drawcalls, that is clealry evident on the benchmark of the russians GameGPU.RU, where AMD GPUs makes CPUs lag behind with much ease than Nvidia.
It's a very well known problem that AMD can't deal with DX11 well and who suffers with that is the CPUs in CPU Bound areas.
DX12 should solve the difference on the hub area for AMD.
Anyway the benchmark sequence is very GPU Bound due to the volumetric lightining, not sure why, but Quantum Broken had tons of Volumetric Lights, and it was exactly there where nvidia used to had less frames than the usual.
FYI, the in game benchmark is a lot different, and also tends to inflate AMD performance relative to the actual game itself. AMD seems to be doing this a lot recently with their G.E titles like Total Warhammer, where the scripted benchmark will return higher values for AMD hardware compared to the game itself.
PCGH.de tested the game using the Prague area, and there the performance was not only much better overall for both NVidia and AMD, but NVidia took the lead over AMD:
PCGH.de test results
The worse results are due to DX11 overhead bottleneck, not lack of AMD GPU Power.
The game makes use of a lot of drawcalls, that is clealry evident on the benchmark of the russians GameGPU.RU, where AMD GPUs makes CPUs lag behind with much ease than Nvidia.
It's a very well known problem that AMD can't deal with DX11 well and who suffers with that is the CPUs in CPU Bound areas.
DX12 should solve the difference on the hub area for AMD.
Anyway the benchmark sequence is very GPU Bound due to the volumetric lightining, not sure why, but Quantum Broken had tons of Volumetric Lights, and it was exactly there where nvidia used to had less frames than the usual.
0blivious
Senior Member
Posts: 2639
Senior Member
Posts: 2639
Posted on: 08/24/2016 08:35 PM
This game runs like complete a$$! It doesn't look even remotely good enough to bring a GTX1070 to it's knees, let alone the countless other good cards this game thrashes.
Hopefully those suffering through it get some serious patching and soon.
This game runs like complete a$$! It doesn't look even remotely good enough to bring a GTX1070 to it's knees, let alone the countless other good cards this game thrashes.
Hopefully those suffering through it get some serious patching and soon.
Paulo Narciso
Senior Member
Posts: 1219
Senior Member
Posts: 1219
Posted on: 08/24/2016 08:55 PM
The benchmark is mostly cpu limited. I get the same average running at 1080p and 1440p. Ingame framerates are much better.
I think DX12 will help the benchmark in those cpu limited scenarios.
The benchmark is mostly cpu limited. I get the same average running at 1080p and 1440p. Ingame framerates are much better.
I think DX12 will help the benchmark in those cpu limited scenarios.
Stormyandcold
Senior Member
Posts: 5287
Senior Member
Posts: 5287
Posted on: 08/24/2016 09:08 PM
Easy choice for me, wait and see and if nothing changes, I don't buy it. Job done.
Easy choice for me, wait and see and if nothing changes, I don't buy it. Job done.
Dygaza
Senior Member
Posts: 535
Senior Member
Posts: 535
Posted on: 08/24/2016 09:15 PM
This game runs like complete a$$! It doesn't look even remotely good enough to bring a GTX1070 to it's knees, let alone the countless other good cards this game thrashes.
Hopefully those suffering through it get some serious patching and soon.
How do you define that it doesn't look even remotely good enough to bring GTX1070 on it's knees? How about shadows or lightning? For humans it's easy to see only texture levels on objects and ground. But for example complex shadows and lighting is very hard to see, yet they also are ones that give extreme performance penalties.
There are settings to tone down to get performance for a reason.
This game runs like complete a$$! It doesn't look even remotely good enough to bring a GTX1070 to it's knees, let alone the countless other good cards this game thrashes.
Hopefully those suffering through it get some serious patching and soon.
How do you define that it doesn't look even remotely good enough to bring GTX1070 on it's knees? How about shadows or lightning? For humans it's easy to see only texture levels on objects and ground. But for example complex shadows and lighting is very hard to see, yet they also are ones that give extreme performance penalties.
There are settings to tone down to get performance for a reason.
millibyte
Senior Member
Posts: 1152
Senior Member
Posts: 1152
Posted on: 08/24/2016 09:21 PM
How do you define that it doesn't look even remotely good enough to bring GTX1070 on it's knees? How about shadows or lightning? For humans it's easy to see only texture levels on objects and ground. But for example complex shadows and lighting is very hard to see, yet they also are ones that give extreme performance penalties.
There are settings to tone down to get performance for a reason.
Right, keeping with the theme, you need optical augs to truly appreciate Ultra settings, lol.
How do you define that it doesn't look even remotely good enough to bring GTX1070 on it's knees? How about shadows or lightning? For humans it's easy to see only texture levels on objects and ground. But for example complex shadows and lighting is very hard to see, yet they also are ones that give extreme performance penalties.
There are settings to tone down to get performance for a reason.
Right, keeping with the theme, you need optical augs to truly appreciate Ultra settings, lol.
Glasofruix
Member
Posts: 96
Member
Posts: 96
Posted on: 08/24/2016 09:22 PM
Quick tip, don't use the ingame sharpen as it is too strong, use reshade/sweetfx to add some instead (at 1.60 strength it looks fine to me). The ingame temporal AA is very good at smoothing jaggies (way better than the injectable smaa solution) but it blurs the heck out of everything and the exagerated sharpen doesn't exactly help.
Quick tip, don't use the ingame sharpen as it is too strong, use reshade/sweetfx to add some instead (at 1.60 strength it looks fine to me). The ingame temporal AA is very good at smoothing jaggies (way better than the injectable smaa solution) but it blurs the heck out of everything and the exagerated sharpen doesn't exactly help.
Bleib
Senior Member
Posts: 371
Senior Member
Posts: 371
Posted on: 08/24/2016 09:22 PM
Personally I'm fine with the performance, it did a good job on setting proper settings from the get-go, but I removed motion blur, upped the texture resolution to high and apart from a few hick-ups it works just fine at 1440p.
Happy to see that there are so many settings in place.
Personally I'm fine with the performance, it did a good job on setting proper settings from the get-go, but I removed motion blur, upped the texture resolution to high and apart from a few hick-ups it works just fine at 1440p.
Happy to see that there are so many settings in place.
airbud7
Senior Member
Posts: 7725
Senior Member
Posts: 7725
Posted on: 08/24/2016 09:36 PM
I feel ya bro!...my setup looks weak too, something gotta be fixed or this game won't sell very well on pc platform.
eh if 1060gtx barely does 35fps 53 fps i guess my 660gtx will not cut it for the game
I feel ya bro!...my setup looks weak too, something gotta be fixed or this game won't sell very well on pc platform.
pages 1 2 3 4 > »
Click here to post a comment for this article on the message forum.
Guru3D.com » Articles » Deus Ex: Mankind Divided PC GPU (DX11 and DX12) performance benchmark review
Senior Member
Posts: 12576
Yeah, ignoring not because the DX12 renderer isn't out -- it's because it's all some conspiracy against AMD. Can you stop posting nonsense or at least edit your other useless post and add this to it?