Take the 2-minute tour ×
Arqade is a question and answer site for passionate videogamers on all platforms. It's 100% free, no registration required.

I want a way to get a consistent average FPS reading in BF3 for benchmarking purposes. How do guys like tomshardware.com get a reliable FPS reading? In the Quake series you could always replay a demo and get the average FPS - anything similar in BF3?

share|improve this question
    
So you already have a program to perform the benchmarking, such as 3DMark or somthing? If so, what one are you using? –  Taz Apr 16 '12 at 8:55
    
No I don't. I'm asking how guys like Toms Hardware can say "Graphic card X yields Y FPS in BF3". –  Nilzor Apr 16 '12 at 10:32

3 Answers 3

up vote 3 down vote accepted

I will assume that you have FRAPS(if not you can get the free version from FRAPS.com;the main difference between the full and the demo AFAIK is that in the demo you can record videos of upto 30 seconds at a time)/

  1. Run FRAPS before starting BF3
  2. Load a section of a mission that is graphics intensive(plenty of smoke,destruction,enemies and explosions).
  3. Press F11 or whatever the hotkey you have set for benchmark.
  4. Play your normal game while FRAPS benchmarks for <insert time here>. Press F11 again to stop benchmark. The min, max and average FPS will be stored in a spreadsheet file.
  5. Repeat this to get an average min,max and average.

Note: Make sure you have the MinMaxAvg dialog ticked under benchmark settings.

enter image description here

share|improve this answer
    
Ok so you're saying i actually have to play the game manually, right? When I repeat the process with different settings, there are more factors than just the settings that differ because I'm actually playing a live game. If this is the only/best way, I accept that. –  Nilzor Apr 16 '12 at 10:31
    
Absolutely.This is the best way IMO to get a consistent result of a card X giving Y FPS in Z game .Nothing beats real life testing . –  GTX OC Apr 16 '12 at 10:46
1  
What he means is different things will happen.. For example maybe the second time you see more explosions etc creating more factors that effect the fps. But as you suggest, running a section of a mission is best as things will tend to stay the same. Just run from A to B without stopping. And things should be mimicked as fair as possible. –  Doomsknight Apr 16 '12 at 10:52
    
Exactly, @Doomsknight . I'm am bit surprised if this is the way "professional benchmarkers" do their instrumenting, but again - if it's the only way then thats what we've got. –  Nilzor Apr 16 '12 at 11:27
1  
This is not benchmarking.. –  DrFish Apr 16 '12 at 12:34

Battlefield has a built in FPS reading output.

You can turn it on by first pressing the tilda key ' ~ ' (top left of most English keyboards - to the left of the '1') and then type "render.drawfps true"

You can then turn it off by changing the true to false

share|improve this answer
    
Again, this is not benchmarking. –  DrFish Apr 16 '12 at 12:34

Personally, I don't think is it entirely possible to perform a legitimate benchmark within Battlefield 3. There are way too many variables to account for as each run or play through of single player and/or multiplayer, as a benchmark should include the same graphical rendering every time.

Your best option to gather a solution close to a benchmark within Battlefield 3 would be to manually play a single player mission (less variables in comparison to multiplayer) with FRAPS running, and, if possible, at the same time run GPUz or similar software to gather information on the vitals of your VGA(s). There is an option within GPUz to log the information on the selected VGA.

It is always good practice to watch or be aware of your system temperatures (as seen in the GPUz example below).

enter image description here

Be sure to these steps multiple times to gather an average FPS.

Keep in mind, this is not a valid benchmark, but rather a sense of an average FPS you obtain in Battlefield 3.

For legit Benchmarking, this article and links provided within it may interest you.

share|improve this answer
    
Thanks for the links and additional tips on getting best possible quasi-benchmark, as I suppose we've now established it is. –  Nilzor Apr 16 '12 at 14:22
    
Anytime. Hope it helped. –  Zero Stack Apr 16 '12 at 14:27

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.