CIV: BE benchmarked

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Ha? nVidia cards are 50% faster than AMD with DX.
And right now SFR is not really working so you can either use DX with Crossfire support and be 33% slower or using Mantle with nearly no improvement.

Whoa. What a load of BS. Several sites already tested AMD vs. NV in DX. 280X and 770 and 290X and 780Ti are close. Go look at GameGPU data where 7990 beats 690 and 290X CF is only beaten by 980 SLI in DX. Even on AT, 280X and 290X are very competitive given their market positioning. If we look at DX in this game, 970 and 980 are crazy overpriced for the performance they deliver against the 290/290X.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
What are you talking?
This here is about CIV and not some other games. And nVidia cards have a 50% advantage with DX.

And with Mantle there is literally no improvement over a single card and with DX you are limited by the dx driver overhead.

And nVidia cards are overpriced? Right, after AMD needed to cut their prices to stay relevant.
 

KaRLiToS

Golden Member
Jul 30, 2010
1,918
11
81
What are you talking?
This here is about CIV and not some other games. And nVidia cards have a 50% advantage with DX.

And with Mantle there is literally no improvement over a single card and with DX you are limited by the dx driver overhead.

And nVidia cards are overpriced? Right, after AMD needed to cut their prices to stay relevant.

Come again?

be-4k-fps.png


be-25-fps.png



This is GTX 980 (Brand spanking new 550$ ) versus R9 290x (One year old 350$)
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
From PCPer, computerbase.de and a few others, it definitely seems like Mantle boost R290X performance massively.

But on AT, the boost was weak, is it due to them using reference R290X that runs hot, loud and throttling? Other sites use 1ghz custom models or 1050mhz models that maintain its clock rate & run cooler.

http://www.pcper.com/reviews/Graphi...mance-Maxwell-vs-Hawaii-DX11-vs-Mantle/2560x1

be-25-fps.png


That is completely contrary to AT's results, with R290X in Mantle a clear winner in average FPS and massively in min FPS, as well as smoother in frame latency.

Clearly Mantle is working great and now I understand the hype from the Firaxis engineers... but reading AT's results, I was pretty underwhelmed.

The 290X is faster that the 970 in the game per GPU performance. Results are to be expected.

The results are pretty similar except for a few large spikes. This is why absolute terms of min fps are not useful.

be-25.png


At 4K the two are identical.

Look at the min frames - that is crazy good

50.4 vs 38.3

that is crazy seeing the R9 290x is a year old !

Somebody needs to test time between turns on a giant map at turn 150 or something on a load of cpus

Two spikes. Mantle is better but its not a terribly huge difference. I would like to see 1080p results.

civbe_cpu_860k.png

http://pclab.pl/art59998-18.html

Hopefully new AMD cards dont need a application patch to work with Mantle...

Interesting but the difference isn't huge. With such a small difference between the 970 and the 290X mantle isn't really needed.


What I mean is that a 860k can get 90 fps. Almost any CPU on the market will achieve 60 fps; the game is not terribly demanding on the CPU.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
And here with Crossfire and Mantle:
be-25-mgpuini-fps.png

48787

http://www.pcper.com/reviews/Graphi...mance-Maxwell-vs-Hawaii-DX11-vs-Mantle/2560x1

11% improvement for twice the money.

there is a day zero patch aimed at better multi GPU support and other tweaks. Both vendors are going to be having better performance on this game. so its still too early to say that GTX 980 is a clear winner in this title.

http://www.anandtech.com/show/8640/benchmarked-civilization-beyond-earth/4

"We're still early in the release cycle for Civilization: Beyond Earth – in fact, our testing was done with pre-launch code, and there's already talk of a day zero patch to fix a few glitches (e.g. with multi-GPU configurations that have more than two GPUs) – so by no means is this the final word in performance. It's not too surprising to see NVIDIA's GTX 980 taking the single GPU performance crown for Beyond Earth, but additional driver tuning may change things a bit."

btw you forgot to mention the R9 290X Mantle beating GTX 980 in pcper benchmarks :whiste:

I could show you a few more which show the R9 290X on par and trading blows with GTX 980

http://www.computerbase.de/2014-10/...diagramm-grafikkarten-benchmarks-in-2560-1600

http://www.computerbase.de/2014-10/...diagramm-grafikkarten-benchmarks-in-3840-2160

http://www.legitreviews.com/civiliz...rmance-benchmarked-amd-versus-nvidia_152260/3

http://pclab.pl/art59998-12.html
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
No, i haven't. That was not the point of my posting.
The reason why nVidia is much faster under DX has nothing to do with the graphic cards.
 

n0x1ous

Platinum Member
Sep 9, 2010
2,572
248
106
Karlitos did you get it running in Mantle? I read somewhere that the game wont even start in Mantle with more than 2 gpus installed in the system - sounds right up your alley......
 

KaRLiToS

Golden Member
Jul 30, 2010
1,918
11
81
Karlitos did you get it running in Mantle? I read somewhere that the game wont even start in Mantle with more than 2 gpus installed in the system - sounds right up your alley......

Thanks my friend for the suggestion.

Unfortunately, I just tried with only a single card after you told me and it still doesn't work. So you helped me conclude that we need Driver 14.9.2 beta . :thumbsup:
 

JarredWalton

Member
Aug 23, 2004
53
0
66
Unlike Anandtech reviewer, it is the first thing I did in the .ini file.

Reviewers got 14.9.2 to test with because I guess it brings Mantle support to CIv: Beyond earth. Maybe they sent the drivers to reviewer to test them beforehand

To be fair, this was not documented until after my look at performance had already gone live. We got pre-release access to the game on Tuesday (though I was unable to start benchmarking until Tuesday night due to other things I was working on). We even received a reviewer's guide with documentation on all of this stuff. No mention of the need to manually set "Enable mGPU = 1" anywhere. Time to retest....
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
No, i haven't. That was not the point of my posting. The reason why nVidia is much faster under DX has nothing to do with the graphic cards.

Why is the GTX 780 Ti slower than R9 290X if Nvidia is much faster in DX11 due to its so called superior drivers ?

http://www.hardwareluxx.de/index.ph...rth-im-grafikkarten-vergleich-mit-mantle.html

http://www.legitreviews.com/civiliz...rmance-benchmarked-amd-versus-nvidia_152260/3

http://pclab.pl/art59998-12.html

Funnily enough even without mantle R9 290X minimum fps is slightly ahead of GTX 980 at 1440p and 4k

http://www.anandtech.com/show/8640/benchmarked-civilization-beyond-earth/3
 

KaRLiToS

Golden Member
Jul 30, 2010
1,918
11
81
To be fair, this was not documented until after my look at performance had already gone live. We got pre-release access to the game on Tuesday (though I was unable to start benchmarking until Tuesday night due to other things I was working on). We even received a reviewer's guide with documentation on all of this stuff. No mention of the need to manually set "Enable mGPU = 1" anywhere. Time to retest....

Hey Jarred, I didn't meant to criticize. Sorry, just wanted to point out that it is the first thing I did but to no avail.

Your review is awsome and I wish I could do those kind of things to.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
Why is the GTX 780 Ti slower than R9 290X if Nvidia is much faster in DX11 due to its so called superior drivers ?

http://www.hardwareluxx.de/index.ph...rth-im-grafikkarten-vergleich-mit-mantle.html

http://www.legitreviews.com/civiliz...rmance-benchmarked-amd-versus-nvidia_152260/3

http://pclab.pl/art59998-12.html

Funnily enough even without mantle R9 290X minimum fps is slightly ahead of GTX 980 at 1440p and 4k

http://www.anandtech.com/show/8640/benchmarked-civilization-beyond-earth/3

Because these numbers are not real from the internal benchmark tool:
http://www.pcper.com/reviews/Graphi...mance-Maxwell-vs-Hawaii-DX11-vs-Mantle/3-Diff

And here are minimum frames within the game and DX:
civbe_1080p_ultra.pngl
civbe_1080p_ultra.png

http://pclab.pl/art59998-10.html
 
Last edited:

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
Yeah, because using 4K for a discussion about the DX driver overhead makes absolut sense.
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Apr 6, 2004
6,283
5
81
If this turns into an AMD vs. NVidia war with price bashing over a single game benchmark, I will be closing this thread and infracting/warning everyone above that took the bait.

From my post forward, get this discussion back on track, now.

-Moderator Rvenger
 
Feb 19, 2009
10,457
10
76
civbe_1440p_ultrab.png


Average & Min included. The other chart does not have Mantle.

"AMD Catalyst, oczywi&#347;cie w trybie DirectX"

They even tell you specifically on the 1080p tests, they did not run Mantle for their campaign benchmark, but with DX. http://pclab.pl/art59998-10.html

civbe_1080p_ultrab.png


Seems to vary from a few review sites, but definitely majority now show clear gains with Mantle.

Sadly, CF Mantle needs a manual edit of the cfg file to enable so all the sites did find either crashing or it wasn't scaling.
 
Last edited:

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
The other chart doesnt have Mantle because there isn't one single programm which can show FPS with Mantle.

So all these sites using the internal benchmark. And especially nVidia's number doesnt seem to be right.
 

Grooveriding

Diamond Member
Dec 25, 2008
9,107
1,260
126

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
And here are minimum frames within the game and DX:
civbe_1080p_ultra.pngl
civbe_1080p_ultra.png

http://pclab.pl/art59998-10.html

Their results are fabricated.

Why would the same hardware in the same test give 30% different result? Just look at thhe 970 + i7 4700@4.5GHz

civbe_1080p_ultra.png

civbe_cpu_gn.png


108 FPS vs 140 FPS in the same test for the same hardware config? They do something terribly wrong, or didn't check if the numbers match after adjusting them accordingly to subjective Gaming Experience.

Whatever they did, they are bad at it and shouldn't be a valid source of anything.
 
Last edited:

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
Mantle seems to work just fine with 14.9.1.
 
Last edited:

Grooveriding

Diamond Member
Dec 25, 2008
9,107
1,260
126
pclab is an amateurish site. They tried to show DX11 differences by- vendor using benches they ran on the Battlefield 4 beta client, that was broken, and kept rehashing those results in later reviews. I don't think they have a clue what the hell they are doing.