• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

***THE BF4 BENCHMARKS THREAD***

humbug said:
Blanket statements Roff ^^^^
Click to expand...

Whats that supposed to mean? anyhow the facts are blatantly obvious and reproduceable for anyone with a recent i7 and a 780 let alone 780ti - something IS wrong with their numbers.
 
Sorry humbug but its of no use to me using that.. My Metro video was done with OBS and Quick Sync and the hit is minimal. Probably less than 5% as its using the IGP on my 2700K :)

Heres another video on Paracel Storm with Vsync disabled this time.

Same settings as before..

Apologies for the poor brightness quality. Will have to sort that out if i do another video.

Oh.. Max VRAM was 2.6GB for this one.

 
You guys need to understand the inconsistencies of multi player benchmarking. Those results are likely correct. If they're lower than typical for a 780TI, they'll also be lower than typical for the 290X. Server load plays such an important part in MP that can define how many fps you get and how heavy a toll it takes on the cpu/ your pc. We also don't know what they benched, how they benched it and the load the server was under at the time. What we do know is the 780TI/290X results are close, within a couple of fps of each other so you can be assured they're correct for the testing method they used. Stop trying to write them off.

Anyway heres some tests i quickly pulled from 64 man Siege servers. Classical CPU/API bottlenecking going on in my first screenshot. Check out that gpu low usage. :eek:


[email protected]
290 Pro crossfire @1000/1250 - 2.5% overclock from stock


1080P%20settings.png


DX
DX1080P.png


Mantle

Mantle%201080P.png


To further highlight the inconsistencies of MP benchmarking check this out. 1440P DX getting higher fps than 1080P DX on Siege. Reason? I was on a different server and there was less action focused on this point at the time. The mantle effect is stark in that previous screenshot. The server was under heavy load during both but Mantle does not bat an eyelid.


1440P%20settings.png



DX
1440%20DX%201.png



Mantle
Mantle%201440P%201.png


6 core cpu's help a lot in BF and trump 4 core 8 threaded i7's comfortably.
 
Last edited:
@Matt.. just for comparison i got 71fps on Ultra 4xMSAA @ 1080p on my system standing in the same spot u guys do on Siege. Just didnt know how to screenshot it or more likely just cudnt be arsed :D
 
MyBrains said:
@Matt.. just for comparison i got 71fps on Ultra 4xMSAA @ 1080p on my system standing in the same spot u guys do on Siege. Just didnt know how to screenshot it or more likely just cudnt be arsed :D
Click to expand...

Inconsistencies of multi player benchmarking. That's why its wrong to call results false in this instance. Or say but i get 5 times that much fps with my gpu in 2d mode etc. As it was all benched on the same system/server you can be sure its correct.

As for taking a screenshot, bring up the console and type screenshot.render and hit enter. The screenshot will be in your documents folder. I will try a different server and see what i get again at 1080P DX on Siege.
 
MyBrains said:
@Matt.. just for comparison i got 71fps on Ultra 4xMSAA @ 1080p on my system standing in the same spot u guys do on Siege. Just didnt know how to screenshot it or more likely just cudnt be arsed :D
Click to expand...

LtMatt said:
Inconsistencies of multi player benchmarking. That's why its wrong to call results false in this instance. Or say but i get 5 times that much fps with my gpu in 2d mode etc. As it was all benched on the same system/server you can be sure its correct.

As for taking a screenshot, bring up the console and type screenshot.render and hit enter. The screenshot will be in your documents folder. I will try a different server and see what i get again at 1080P DX on Siege.
Click to expand...

Nope just too cpu/api bottlenecked at this point. Different 64 man server, 1 slot free. Gpu usage around 40-80% fluctuating.


1.png
 
MyBrains said:
Sorry humbug but its of no use to me using that.. My Metro video was done with OBS and Quick Sync and the hit is minimal. Probably less than 5% as its using the IGP on my 2700K :)

Heres another video on Paracel Storm with Vsync disabled this time.

Same settings as before..

Apologies for the poor brightness quality. Will have to sort that out if i do another video.

Oh.. Max VRAM was 2.6GB for this one.

Click to expand...

No problem. :) just though i would suggest it as your recording is from a CamCorder.
 
I get a noticeable performance hit with Mantle when i go into borderless mode. Its a shame as its stopping me from recording footage. However Dice have crossfire working in Mantle at least in Windowed/Borderless mode somehow.
 
LtMatt said:
Do you know that or just a hunch? Cos that sure would be an awesome feature. Maybe i should give Sam a nudge. :p
Click to expand...

Just recorded Mantle in Borderless mode for you to see matt. For some reason since switching my MB or Dice last Patch its now very playable with Crossfire enabled.

Post link once finished uploading
 
humbug said:
Go ahead. reproduce it, if only you knew what they did and where in the game. right?
Click to expand...


LtMatt said:
You guys need to understand the inconsistencies of multi player benchmarking.
Click to expand...

Not saying that the 780(ti) v 290(x) performance would be any different but the point is as I said - using the settings they claim to be using you simply can't find a scenario under normal circumstances that you can get result that low below the normal for a 780 let alone 780ti on that map - no matter how hard to try - the busiest 64 player server with most of the action crammed into where your benchmarking from the worst possible bits of the map etc. you will still see results comfortably above what they have.

Its literally impossible when everything is work as it should be to see those kind of numbers from a 780(ti) with those settings on that map under any normal benchmarking circumstances (and likewise the 290 results are probably low for what they should be but I can't answer for that as I don't have one to hand to test on the map).

RE Humbugs point it doesn't matter where they did it or how they did it in the map - the point is no matter how hard you try to find a scenario that pulls the numbers down to the levels they have posted - you won't - even with the varied numbers between different servers, player loads and activity levels you will still see numbers however varied that are higher than what they have by a fair margin.
 
Last edited:
Rroff said:
Not saying that the 780(ti) v 290(x) performance would be any different but the point is as I said - using the settings they claim to be using you simply can't find a scenario under normal circumstances that you can get result that low below the normal for a 780 let alone 780ti on that map - no matter how hard to try - the busiest 64 player server with most of the action crammed into where your benchmarking from the worst possible bits of the map etc. you will still see results comfortably above what they have.

Its literally impossible when everything is work as it should be to see those kind of numbers from a 780(ti) with those settings on that map under any normal benchmarking circumstances (and likewise the 290 results are probably low for what they should be but I can't answer for that as I don't have one to hand to test on the map).
Click to expand...

Did you not see the screenshots scotty posted with a stock water cooled titan? They were right around the fps of that bench. The results look correct to me.
 
You must log in or register to reply here.
Back
Top Bottom