• 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.

In order to prove the limitation of 1GB vram under 1920 x 1200

harmony

harmony

harmony

Permabanned
Joined
12 May 2011
Posts
1,119
I might be able to get one of the following two configurations for a comparison based on frametime analysis, and hopefully to be able to show that vram shortage can increase the probability of lag spikes / stuttering. In many cases, the traditional benchmark results of "average fps" and "adhoc min fps" fail to show such phenomena. Techreport used such methodology to investigate micro-stuttering. (Don't confuse the old skool stuttering I'm saying here, with micro-stuttering caused by AFR. That's two different things.)

1) GTX 560 Ti 1GB x 2 SLI vs GTX 560 Ti 2GB x 2 SLI
2) HD 6950 1GB x 2 CF vs HD 6950 2GB x 2 CF

Which comparison would you guys prefer, if I can manage to get?

Also, which games would you guys prefer? Would BF3 (final release, not beta) be a good, persuasive option?

The tests would be done under 1920 x 1200 (or 1920 x 1080) resolution. Would those defenders of 1GB demand me to test on 4AA instead of 8AA under native resolution? (i.e. does 8AA improve IQ over 4AA?)

What framerate would you guys regard as playable? 30 fps or 60 fps?
 
Last edited:
Just looked at steam and these are the most played games according to those stats, so maybe try as many of these as you can:

Team Fortress 2
Counter-Strike
Counter-Strike: Source
Call of Duty: Modern Warfare 2 - Multiplayer
Sid Meier's Civilization V
Call of Duty Black Ops - Multiplayer
Football Manager 2011
Left 4 Dead 2
Dead Island
Total War SHOGUN 2

You could try Metro 2033, but at it's peak today it was less played than the original "Knights of the Old Republic", so might be best to worry about the games people are actually playing (according to statistics).

Just my thoughts.
 
^Nope, max everything including maxing out the Control Panel too.

@GoogalyMoogaly out of all the games you pointed out, only Shogun 2 is remotely demanding.

It would be a total waste of time testing the likes of any of the source games at all as they are designed to run on low specs.

Doesn't matter whether it's team green or red for me(although I do have the reds in my rig), they should both prove it either way.

Metro, Crysis 2, BF3 and if you do max the CP, then Witcher 2 as well.
 
OP in my case at full hd res for BF3 beta, even without using AA and keeping shadow low, the vram on my 1Gb 5850 was around 990MB. When I applied AA, frame rate hit low 20
 
tommybhoy said:
@GoogalyMoogaly out of all the games you pointed out, only Shogun 2 is remotely demanding.

It would be a total waste of time testing the likes of any of the source games at all as they are designed to run on low specs.
Click to expand...

Yes I suppose that makes sense assuming you're trying to see if with the games that do use more than 1GB VRAM how much impact it makes when run on 1GB cards.

In which case why not use a bunch of the games on the list of games that use more than 1GB of VRAM?
I'm sure there was a big thread floating around with a decent size list on wasn't there?

EDIT:
This one and this one. (Although a similar selection of games are used I think)
 
Last edited:
Shogun 2, crysis 2, and BF3 are the obvious ones. Don't care which brand, it's all the same to me. I'd be interested in seeing the results on a 2560x1440 monitor (more chance of maxing vram). With fast enough cards the impact of low vram should become apparent pretty quickly.
 
So the fanatic defenders of 1GB still haven't shown up yet :) They're lurking somewhere and preparing to attack flaws that they're not going to mention here now :D
 
harmony said:
I've been taking photos in the sun light for 2 days, tired, so back to forum :D
Click to expand...

Let's be fair, they usually don't defend 1GB. They usually are taking the Mickey because the only thing you bang on about on this forum is VRAM. I have no problem with this, mind. I like how you're raising awareness of the issue. When I used my 5850, I noticed how GTA's VRAM usage screwed with my FPS horrifically.

If you can, test both ATI and Nvidia.

Also, can somebody tell me what the software is called that displays your GPU temps, usage and VRAM usage on the top-left of the screen in that pink text, please? :p
 
^ MSI Afterburner.

Given the OP's retoric on VRAM I see no point in these tests. I believe he has already made his conclusions and will manipulate the tests in such away to support he's believes making this a rather pointless exercise.

If you want to see 1Gb vs 2Gb then there are plenty of reviews on legitimate websites that can test the effectiveness of extra VRAM.
 
Ill be interested to see the results of this, I never had an issue with vram until bf3 beta was released. I still think its something to do with drivers ot the fact that bf3 is still in beta.
 
Freddie1980 said:
^ MSI Afterburner.

Given the OP's retoric on VRAM I see no point in these tests. I believe he has already made his conclusions and will manipulate the tests in such away to support he's believes making this a rather pointless exercise.

If you want to see 1Gb vs 2Gb then there are plenty of reviews on legitimate websites that can test the effectiveness of extra VRAM.
Click to expand...

That is low level troll's behavior. If 1GB is perfectly fine for SLI/CF then I'll say it is. You could continue adding me to your ignore list and problem solved ;)
 
I really don't get what the issue is with VRAM - if you're buying a new card then get one with more than 1gb as the price difference is negligible. If you have a decent card with 1GB of VRAM (as I do) then enjoy it on full for most games and turn down AA on those those it doesn't and you can still tweak settings till it does. Viva la PC - this is why we have graphics options.
 
Devrij said:
Shogun 2, crysis 2, and BF3 are the obvious ones. Don't care which brand, it's all the same to me. I'd be interested in seeing the results on a 2560x1440 monitor (more chance of maxing vram). With fast enough cards the impact of low vram should become apparent pretty quickly.
Click to expand...

Running BF3 at that res with 3x 470's @ 750/1500/1700 and ram is totally maxed out all high apart from textures which are on ultra NO aa getting about 50-70 outside and 70-100 indoors so i must just be on the vram limit as if i enable AA it goes down to 5-10 fps :(

But still getting the screen flashing like a bad oc but still does it with no oc and also still feels stuttery despite the high fps i think it down to multi gpu's not being implamnet well even though iam getting 95% usage on all 3 card and and do notice a jump up in fps each time i added a card

Just hoping the final release feels smoother :o
 
Of course, testing a whole range of settings will prove valuable, just please make sure to include the commonly seen standard settings in reviews, x 4 AA and x 16. :)
 
harmony said:
So the fanatic defenders of 1GB still haven't shown up yet :) They're lurking somewhere and preparing to attack flaws that they're not going to mention here now :D
Click to expand...

No fair! I was suspended! :p

I would prefer Nvidia ;)

If you conduct these tests fairly and they show something more than lag spikes then I will admit I am wrong/come up with another theory.

Also, BF3 is slightly harsh.

Can you test some older games aswell? Heaven benchmark etc etc.
 
You must log in or register to reply here.
Back
Top Bottom