Kingdom Come: Deliverance PC graphics performance benchmark review

#5519792

GREGIX

2018-02-14 08:42

Y...years behind...thats why dx12versus11 is almost always slower... I want vulcan games...

#5519795

pegasus1

2018-02-14 08:45

Solfaur:

1440p. I have a strong feeling that the performance will improve a lot with patches though, because the frame drops are inconsistent. For example I had drops in 50s in visually unimpressive places like dark cellars and 100+ fps outside with loads of vegetation.

Cheers, just seen the price, il hold off for a while until its patched and the price drops, ive got REs Evil and Tomb Raider to play first.

#5519803

Keesberenburg

2018-02-14 09:05

3820oc 4C 8t, does a great job withthe gtx 1080 Ti.

#5519808

JonasBeckman

2018-02-14 09:40

Kaarme:

Alright, now I know I'll definitely wait until it hits a sizable discount. As a consumer I need to do something to send a signal to the studios to get with the times.

It's been updated a bit here and there though, still not too sure if there's all that many games actually using features from the later version but we're up to DirectX 11.4 as of I think it was Windows 10 RS2 though most games use 11.1 and it's only been recently that games have outright required support for it.

#5519812

Hilbert Hagedoorn

Administrator

2018-02-14 10:03

I just started with the CPU scaling results. Check the article for the first chart live, the results are MASSIVELY different for more-core processors scaling from two to 8 cores ATM. I'll perform platform results (Ryzen etc) later today. http://www.guru3d.com/articles_pages/kingdom_come_deliverance_pc_graphics_performance_benchmark_review,8.html

#5519818

xIcarus

2018-02-14 10:38

I'm surprised to see it scale so well up to 6c/12t at 1080p considering it's a DX11 title. If anything, it makes me question DX12. So far I haven't seen it show any impressive performance. Vulkan on the other hand seems to have proven itself, at least somewhat.

#5519824

Vananovion

2018-02-14 11:32

Tried it out and I noticed my GPU dropping clock by 20-40MHz occasionally, which drops the FPS somewhat. Which is weird since I get like 45-55 FPS on 1440p medium on my 290. I would expect 100% load and 100% clock all the time - as The Witcher 3 does for me.

I'm surprised to see it scale so well up to 6c/12t at 1080p considering it's a DX11 title.

It was always possible to offload some tasks to CPU cores, even before DX12 (like Skyrim did with shadows). DX12 just makes it easier to implement properly.

#5519854

sverek

2018-02-14 13:19

mikev190:

No he was hoping his ageing card was going to get a couple FPS more.

CPU gets more FPS, not GPU. Not sure if dx12 performs better overall.

#5519858

Hilbert Hagedoorn

Administrator

2018-02-14 13:30

Update Feb 14th: article updated with CPU scaling results including Core i5 6600K, Core i7 8700K, Core i7 5960X, Ryzen 7 1700X and Threadripper 1950X.

#5519862

Ridiric

2018-02-14 13:51

Hey Hilbert, any chance we could get 8600k CPU scaling results? or maybe 6c6t on your 5960x? kinda interested to see how real cores VS threads works with this game.

#5519863

Hilbert Hagedoorn

Administrator

2018-02-14 13:55

Ridiric:

Hey Hilbert, any chance we could get 8600k CPU scaling results? or maybe 6c6t on your 5960x? kinda interested to see how real cores VS threads works with this game.

I do not have a 8600k here anymore but can quite easily do 6c6t on the 5960x. Check back in an hour or so and I'll have them added.

#5519870

Ridiric

2018-02-14 14:05

Hilbert Hagedoorn:

I do not have a 8600k here anymore but can quite easily do 6c6t on the 5960x. Check back in an hour or so and I'll have them added.

Nice! great work as usual Hilbert, look forward to the 6c6t results!

#5519871

sverek

2018-02-14 14:07

Hilbert Hagedoorn:

Update Feb 14th: article updated with CPU scaling results including Core i5 6600K, Core i7 8700K, Core i7 5960X, Ryzen 7 1700X and Threadripper 1950X.

Thanks Boss! Look at that 6600k difference. Pretty much sums up why it's not good idea to buy 4c4t processor anymore.

#5519873

Hilbert Hagedoorn

Administrator

2018-02-14 14:10

Ridiric:

Nice! great work as usual Hilbert, look forward to the 6c6t results!

added

#5519875

Ridiric

2018-02-14 14:14

Hilbert Hagedoorn:

added

Nice, so looks like 6c6t is definitely better than 4c8t but 6c12t is definitely getting some use out of those extra threads (well at least it does when things become more cpu bound at 1920x1080 and lower)

#5519917

Raider0001

2018-02-14 16:12

sverek:

Game development started at 2011. Do you even know the difference between DX11 and DX12, beside CPU utilization and that DX12 only works with Win10? Or you just barking at DX11 cause its 1 digit behind DX12?

Well I am just stating a fact to all of those ppl thinking it is a good idea to use this game for benchmarking just because its a new game... it is not, it is unfair to use it while it doesn't offer new features from DX12 as well as Vulkan - both of which use more of Radeon powahh... Enlarging virtually this problem... you would not want a DX8 title as an example for testing TITAN V

#5519919

pegasus1

2018-02-14 16:18

Raider0001:

Well I am just stating a fact to all of those ppl thinking it is a good idea to use this game for benchmarking just because its a new game... it is not, it is unfair to use it while it doesn't offer new features from DX12 as well as Vulkan - both of which use more of Radeon powahh

Its a benchmark of a game, why complicate things, a new game comes out, people want to know about it and how well it may run on their gfx hardware, simples, who cares what DX variant it uses, it is what it is, play or do not.

#5519923

Raider0001

2018-02-14 16:29

pegasus1:

Its a benchmark of a game, why complicate things, a new game comes out, people want to know about it and how well it may run on their gfx hardware, simples, who cares what DX variant it uses, it is what it is, play or do not.

a benchmark of a game yes... a game used for benchmarking NO

#5519924

pegasus1

2018-02-14 16:31

Raider0001:

a benchmark of a game yes... a game used for benchmarking NO

Well done, you are trying to win a game only you are playing.

#5519927

Raider0001

2018-02-14 16:39

pegasus1:

Well done, you are trying to win a game only you are playing.

Well I did saw in couple of places some ppl saying it is a good looking game and should be used for benchmarking, I have no time in my life to respond to all of them so I posted here a fact that game is probably running an ancient engine