[H] Battlefield 3 Open Beta Performance and Image Quality

[Grooveriding]

I'm going to test my OC'd 580 at 2560x1600 using hardocp's recommended settings.

That will help us see how of a difference over clocking makes in this game.

Should I just record using fraps? Any benchmarking tips to come close to hard's methods?

Just make sure once you adjust your game settings to match hardocp's that you quit the game and restart it. The changes don't take effect on the fly, you need to exit out and rejoin another game.

I tried it myself with a single overclocked 480 and was getting less than he did, but I was trying to run it with AA and everything was on Ultra. My results were more inline with the techspot benches at the same settings; low 20s outdoors, it wasn't playable to say the least. :\

 

[SirPauly]

Let's do a SirPauly post to see how much value is added to the thread...

2.1 average fps faster than the 2GB Radeon 6950! ZOMFG THE 560TI STOMPS A F'ING HOLE IN THE RADEONS CRANIUM ALL THE WAY DOWN TO IT'S ANAL CAVITY AND THEN RIPS OUT IT'S ENTRAILS AND STRINGS UP ITS MOTHER SO THE JACKALS CAN SERVICE AND THEN CONSUME HER !11111!!

24.48979591836734693877510204082 % faster than the 6870 OMFGGAGGG!@!!

Thanks for that! As your data is a perfect illustration to my point. If you look at the GTX 560 TI, this site is getting 42.7 frames with ultra and with the 6870, this site is getting 34.3, with ultra, but with HardOCP, the findings offer a dramatic shift, where the 6870 is offering a noticeable difference win over the GTX 560 TI, with settings disabled for the TI. The key is why?

 

[WMD]

Thanks for that! As your data is a perfect illustration to my point. If you look at the GTX 560 TI, this site is getting 42.7 frames with ultra and with the 6870, this site is getting 34.3, with ultra, but with HardOCP, the findings offer a dramatic shift, where the 6870 is offering a noticeable difference win over the GTX 560 TI, with settings disabled for the TI. The key is why?

Without MSAA and with post AA, AMD has the edge. Nv is better with 4XMSAA which actually works much like SSAA in deferred render. This is consistant with Hard Reset Benchmark as well. Nvidia hardware is just faster with SSAA.

 

[SirPauly]

If that was the case, then why is the 6950 winning easily over the GTX 570 in HardOCP's findings as well with AA enabled? And these don't match other sites as well.

 

[GaiaHunter]

People need to understand that [H] generally uses slower CPUs than everyone else in their reviews - that generally favors AMD GPUs that require much less CPU power to achieve peak performance.

I've also seen that techspot review.

Even at 1680x1050 the 6870 has the same performance has the GTX560 Ti

 

[SirPauly]

I don't think HardOCP is biased and there has to be a reason why the performance differs.

 

[notty22]

I don't think HardOCP is biased and there has to be a reason why the performance differs.

They told you why. They said

Performance in the BF3 open beta varies to a great extent. The two available maps are large, and performance can vary significantly between one part of either map and any other part. Thus, testing the BF3 open beta was a tremendously frustrating experience. Due to the nature of this being a multiplayer only map without the full-game's features, we were unable to reproduce a testing run with anything remotely resembling consistency. In the end, we were forced to just play the game for a while on each video card while adjusting settings to make sure it was playable everywhere we tested to find the highest playable settings in the maps offered.

So the question is why, they went to press with the numbers they did.
They made a choice.
Imo, one that would be controversial and if AMD cards were on top 50% of the time, they went with who butters the bread the most (advertising).
All without losing any integrity, because of the said inconsistency.

 

[SirPauly]

But what stayed constant was AMD winning and pretty easily, too.

 

[blackened23]

They told you why. They said
So the question is why, they went to press with the numbers they did.
They made a choice.
Imo, one that would be controversial and if AMD cards were on top 50% of the time, they went with who butters the bread the most (advertising).
All without losing any integrity, because of the said inconsistency.

This is getting ridiculous... Anyone claiming HOCP to be biased needs to get their head examined, they have been around for years and have been nothing but a quality site.

 

[notty22]

Pretty sure you're just POed because your favorite brand didn't win. This is getting ridiculous Anyone claiming HOCP to be biased needs to get their head examined, they have been around for years and have been nothing but a quality site.

I'm not PO about anything, not that I care what you think about me. I gave my opinion. I have a opinion about you , but I won't share it publicly.

 

[happy medium]

Even at 1680x1050 the 6870 has the same performance has the GTX560 Ti

Funny my gtx460 overclocked beats the crap out of both these cards @ 1600x1200//high settings even. ?

I'm starting to think all the performance numbers are not very accurate.
I guess it would depend on exactly where and how long you run your bench for.

 

[happy medium]

Anyone claiming HOCP to be biased needs to get their head examined,

Well look over this thread and you will find quite a few/ unbiased/smart/ long time members that need there heads examined,starting with Russian.

If they are not being biased ,they are the very least decieving the readers with contraversal information for some extra page hits.

One thing is for sure, for some odd reason they are CONSISTANTLY THE ONLY SITE THAT DIFFERS FROM 90% OF THE OTHERS.

Example: If 20 people review the smell of IDC's feet and say they stink and Kyle says they don't stink, do you believe Kyle? It would sound to me Kyle just mabe don't like IDC too much. :

 

[blackened23]

Well look over this thread and you will find quite a few/ unbiased/smart/ long time members that need there heads examined,starting with Russian.

If they are not being biased ,they are the very least decieving the readers with contraversal information for some extra page hits.

One thing is for sure, for some odd reason they are CONSISTANTLY THE ONLY SITE THAT DIFFERS FROM 90% OF THE OTHERS.

Really? Prove it. You're claiming HardOCP is biased based on the fact that your favorite brand isn't winning in 1-2 benchmarks. If you want to provide tangible proof, take all of their reviews, compile the information and present a compelling argument.

They have certainly done many ATI/AMD reviews that weren't flattering, and thats a fact whether you want to believe it or not.

Crying foul over 1-2 benchmarks certainly makes it appear that you're not happy with your favorite brand not winning. If thats not true, compile some evidence spanning 2+ years of their ATI and nvidia GPU reviews and get back to us.

 

[badb0y]

So I guess my game is just being retarded using 1.4 GB of VRAM? There can't be no ramifications of maxing out your VRAM.

 

[SirPauly]

I don't think it is bias and was the site's findings. I don't think there is any malice or nefarious intent with HardOcp, or wild conspiracy against nVidia in any way. It may be the platform that it is evaluated on.

 

[happy medium]

You're claiming HardOCP is biased

You quote me but say nothing about what I posted? I think IDC calls that a stawman or something like that.

 

[The Ultimate]

Ah ok. I'm using eVGA precision. Ok at 1680x1050, Ultra Settings with 4xAA 16AF on a 560Ti 1GB card. Memory usage was it's highest at 1011MB. Never went higher than that. So I think Happy Medium is totally right on this one and Groover is not.
Also some FRAPS min avg and max coming up for Metro Level. BRB.

I think that both are right, you can't use more memory than you don't have. If you have a game that's constantly pushing the RAM limits, means that the game is VRAM hungry. If you play a game on a GTX 560 Ti and uses 1011MB, and the same game uses nearly 1,4xx something on the GTX 480, means that the game is VRAM hungry. With Crysis 1 and several mods, AT 1280x1024 it used 1,2GB of my 2GB of VRAM, at 1920x1080, it increased to 1,5GB of my 2GB VRAM.

 

[happy medium]

So I guess my game is just being retarded using 1.4 GB of VRAM? There can't be no ramifications of maxing out your VRAM.

I thimk if you read the entire thread you would see there is no ram limitation with this game engine.

I really dont want to over the last 20 posts to sum it up for you ,no offence.

The game engine uses the memory avalable,but somehow a 1gb card gets no performance hit.

I'm sure this will be explained by a site or someone with more patients then me.

 

[Grooveriding]

So I guess my game is just being retarded using 1.4 GB of VRAM? There can't be no ramifications of maxing out your VRAM.

From what we've seen so far it looks like appropriate cards for BF3 from what is currently on the market are :

1680x1050 - GTX 560 / 6950 1GB / 6870
1920x1080 & 1920x1200 - 6950 2GB / 6970 / GTX 570
2560x1600 - 580 SLI / 6970 CF / GTX 590 / 6990

 

[badb0y]

I thimk if you read the entire thread you would see there is no ram limitation with this game engine.

I really dont want to over the last 20 posts to sum it up for you ,no offence.

The game engine uses the memory avalable,but somehow a 1gb card gets no performance hit.

I'm sure this will be explained by a site or someone with more patients then me.

Dynamic image degradation to support available VRAM LOL.

(I joke).

 

[happy medium]

I think that both are right, you can't use more memory than you don't have. If you have a game that's constantly pushing the RAM limits, means that the game is VRAM hungry. If you play a game on a GTX 560 Ti and uses 1011MB, and the same game uses nearly 1,4xx something on the GTX 480, means that the game is VRAM hungry. With Crysis 1 and several mods, AT 1280x1024 it used 1,2GB of my 2GB of VRAM, at 1920x1080, it increased to 1,5GB of my 2GB VRAM.

Nobody said the game was not ram hungry, we were showing that a 1gb card @ 1080p ultra setting 4xaa is NOT memory limited and does not suffer any performance loss compared to a 1.25gb/2gb card/s.
The conclusion seems to indicate that a 1gb card is not ram limited but might give lower image quality vs a card with more memory but that has not been looked into.

 

[Grooveriding]

I think that both are right, you can't use more memory than you don't have. If you have a game that's constantly pushing the RAM limits, means that the game is VRAM hungry. If you play a game on a GTX 560 Ti and uses 1011MB, and the same game uses nearly 1,4xx something on the GTX 480, means that the game is VRAM hungry. With Crysis 1 and several mods, AT 1280x1024 it used 1,2GB of my 2GB of VRAM, at 1920x1080, it increased to 1,5GB of my 2GB VRAM.

Exactly. It's maxing out cards with less memory but not ones with more, so with cards with over 1GB there is enough VRAM, ones below, not enough under the most intensive settings. None of these sites have benched a 1GB card with 4xAA on ultra at 1080p, they probably tried, got a slide show and said forget it. Hard was showing 38fps for the 560 and 43 for the 6870, techspot 36fps for both the 560 and 6870. Those are already on the border of what could be called playable, adding AA would only tank the cards.

I also think it may explain the low minimum of 25fps for the 570 with AA, while the 6950 2GB has a 35fps minimum at the same settings. Those poor minimums on the 570 could also be what drags down its overall average against the 6950.

 

[crisium]

I wish someone would just test a 1GB 6950 against a 2GB 6950 for we can end this bickering.

 

[The Ultimate]

Remember the issues you had with Dragon Age 2 on the GTX480? Its performance was disastrous.

NV worked hard and now a GTX580 easily beats the HD6970 in that game.

I think we should wait for the full game and another 2-3 driver releases from both camps.

It better be, the GTX 580 is more expensive than the HD 6970.

My gtx460 runs dam close to 5870 speeds at 950 core. 1gb cards do not run out of memory at 1080p ultra settings in BF3.

I dont see much difference (if at all) between ultra and high anyway.
This game seems far from finished, I can even remap my control keys.

Certainly I doubt that, a stock GTX 560 Ti is consistently faster than an overclocked GTX 460, and the GTX 560 Ti barely matches the HD 5870 and HD 6950.

Damn glad i got my cash back on the gtx560ti that failed on me last month...i would be kicking myself in the ass about now.

Most obvious choice if your looking to buy is amd,no question about it....but there isn't no bench's as of yet showing how well a 3gb gtx580 can do...only how a 2gb 6970 wages against a 1.2gb gtx570 and 1.5gb gtx580 gpu...

What happened to your GTX 560 Ti?

 

[Genx87]

Intel + AMD + Nvidia please poop out your next generation products. Looks like my i5 2500K + 470 may hit a wall with this title.