Borderlands 2 GPU/CPU benchmarks [TechSpot/HardOCP/Others]

[f1sherman]

Shall I start linking Nvidia demos?

Not true? That wall destruction you linked before is a huge script if I ever saw one.


NV might use some scripting and more agressive aproximation themselves, but Havok is a huge aproximation and that's why it performs "better".
If not, show us some Havok game(s) with BL2 comparable amount of physics.

 

[WMD]

Shall I start linking Nvidia demos?

Not true? That wall destruction you linked before is a huge script if I ever saw one.


NV might use some of the scripting and more agressive aproximation themselves, but Havok is a huge aproximation and that's why it performs "better".

If not, show us some Havok game(s) with BL2 comparable amount of physics.

You are saying Havok is just scripted animation and no cloth/ fluid capabilities. I posted a video to show otherwise. Feel free to post Nvidia demos if that proves your earlier points.

 

[f1sherman]

OK cloth is not exclusive to PhysX, but you posted Havok cloth DEMO.

Now feel free to show us Havok fluid GAME IMPLEMENTATION.

You also linked that archaic wall destruction as a proof that Havok performs "better".

PS take your time, im out to get some exercise

 

[SirPauly]

The particle effects are not bouncing around -- even goes back to one of the first titles with GPU Physics, where the particles effected by the environment and don't disappear -- Sacred 2. For the most part, physics on the CPU has been scripted and disappear before they hit anything.

http://www.youtube.com/watch?v=JK7BkiNboRU&fmt=22

http://www.youtube.com/watch?v=gkVwo-fohJo&hd=1

 

[wand3r3r]

Just played BL2 multi. Everybody is familiar with the army of robots battle? We were eight fighters against them. There was so much (expletive) flying around I almost didn't know which way was up. The lowest my fps went was 47 fps. I kept a close eye on it. Again, this is with all settings highest/farthest/PhysX High 1920x1200. Is there a specific area in this game I should be investigating? Because no matter how I try, I can't find any areas that make my fps tank. Not a one. I could be always missing it, so tell me where to go.

Wand3r3r, Grooveriding? Where to?

8 vs. the robots? The PC has 4 player coop, what were you playing.

I've posted a few screenshots, it's just random fights with enough (4) characters and all the special effects. You keep talking but no-one has posted proof of what a gtx 680 does, while I have shown the 690 can get drops. I'll try get a game going and record some longer runs (I wish a pro site would do a real min fps comparison and how record actual heavy gameplay). The drops aren't non-stop >40 fps, but happen quite frequently with enough action/effects. In single or 2 player I didn't notice quite as big of hits.

 

[blackened23]

Thinking things over, I have to do a comparison to Batman: AC. For any physx performance anomalies, I would say wait for a new driver or patch. Some people are having no issues while others are, so I don't know what the common denominator is. Yet, I remember Batman: AC received an absolutely huge physx performance jump from an updated patch earlier this year - when that game was released the performance hit for physx high was absolutely huge. A couple of months later a patch was released that caused physx high performance to skyrocket upwards - so the game is actually very playable and smooth now with physx high. I'm sure nvidia can do some magic to optimize performance a bit.

Physx will always incur a performance hit. I haven't found any game to be an exception to this. But for any performance issues (if you're experiencing them - some aren't) I would say a patch or driver will address it - and improve it.

 

[Keysplayr]

8 vs. the robots? The PC has 4 player coop, what were you playing.

I've posted a few screenshots, it's just random fights with enough (4) characters and all the special effects. You keep talking but no-one has posted proof of what a gtx 680 does, while I have shown the 690 can get drops. I'll try get a game going and record some longer runs (I wish a pro site would do a real min fps comparison and how record actual heavy gameplay). The drops aren't non-stop >40 fps, but happen quite frequently with enough action/effects. In single or 2 player I didn't notice quite as big of hits.

Ah. I see how this is going to turn out already. It's going to be one section of one map played a certain way where performance tanks. Gee, I'd sure like to know how to replicate this situation i keep hearing about. I am not saying it doesn't exist. I am saying i can't replicate it. You saying that because i haven't posted any screenshots is kind of pointless a railven already has. And he states he backs up my findings.

 

[Gunslinger08]

Can anyone recommend a quiet 7970 ghz? I'm still trying to decide between 7970 ghz and 680. Seems like just about everything I read says the 7970 (even with custom cooler) is loud. I'm not interested in a custom fan profile that is only quiet at idle.

 

[Lonbjerg]

Are you serious? You are either blind or a troll.

^
You are comparing a lesser effect to a superoir (PhysX has more INTERACTIVE debris, INTERACTIVE sparks ect...plus the debris in the havok clip you provided dissapear (CPU limitation)).

It's like comparing Gouraud shading to textures :whiste:

So let me aks you once again:
Are your serious?!

 

[blastingcap]

Can anyone recommend a quiet 7970 ghz? I'm still trying to decide between 7970 ghz and 680. Seems like just about everything I read says the 7970 (even with custom cooler) is loud. I'm not interested in a custom fan profile that is only quiet at idle.

You may want to start a new thread or search for an old one. This thread is full of off topic posts already...

 

[KingFatty]

I watched the Borderlands 2 videos mentioned above and it's pretty damn cool to see those PhysX effects!

Just curious, after reading how the Batman AC patch improved PhysX performance, do you think that patch would only be applicable to Physx implemented on the GPU? I'm not sure if Batman lets you do physx on the CPU like Borderlands 2, so maybe if Borderlands 2 gets a patch, it will help Physx across the board, and not just improve the performance on the GPU?

 

[blackened23]

The problem with Batman: AC pre-patch: GPU usage was terrible in physx heavy scenes prior to the patch. In particular, during the in game benchmark there is one scene where glass shards are flying everywhere and in that scene pre-patch my framerate would tank to about 25 fps (680 sli, 2560x1600). I never tried CPU physx, but keep in mind that Batman: AC had a far different physx implementation that would never work on AMD GPUs (as you can in BL2 - i'm not sure about "hybrid" phsyx) - you can force it to run on the CPU via nvidia control panel but it would be super slow.

Batman: AC later had a patch that increased physx performance quite a bit, especially for SLI configurations. The ingame benchmarking sequence immediately became faster post patch, including the scene with glass shards. I know that my framerate jump in physx heavy scenes was substantial - the game has been smooth for me ever since with physx high. Note that there is still a performance drop for physx - but it is very much better now in Batman.

So anyway my point is, if there is anyone with a performance issue related to physx in BL2 i'm sure nvidia is tweaking things. There will always be a performance hit for physx -- but a future driver and/or patch should help quite a bit.

 

[railven]

The problem with Batman: AC pre-patch: GPU usage was terrible in physx heavy scenes prior to the patch. In particular, during the in game benchmark there is one scene where glass shards are flying everywhere and in that scene pre-patch my framerate would tank to about 25 fps (680 sli, 2560x1600). I never tried CPU physx, but keep in mind that Batman: AC had a far different physx implementation that would never work on AMD GPUs (as you can in BL2 - i'm not sure about "hybrid" phsyx) - you can force it to run on the CPU via nvidia control panel but it would be super slow.

Batman: AC later had a patch that increased physx performance quite a bit, especially for SLI configurations. The ingame benchmarking sequence immediately became faster post patch, including the scene with glass shards. I know that my framerate jump in physx heavy scenes was substantial - the game has been smooth for me ever since with physx high. Note that there is still a performance drop for physx - but it is very much better now in Batman.

So anyway my point is, if there is anyone with a performance issue related to physx in BL2 i'm sure nvidia is tweaking things. There will always be a performance hit for physx -- but a future driver and/or patch should help quite a bit.

I'll answer that

You can force PhysX to run on the CPU in Batman: AC and I can tell you right now - you don't want to do that. For the canned demo, the sparks alone would tank you <30 FPS. That scene black describes with the flying shards, you're looking at <15 FPS.

Actual game play, before I really fixed the offloading in my PhysX Hybrid setup (forgot one *.dll) one of the first scenes involves you saving the Catwoman from Two-Face. That room has a lot of fog, and that area alone tanked me to ~14 FPS.

Once I fixed the offloading issue, was average ~40 FPS in DX11 Max settings, FXAA High, and PhysX High on a Radeon 5870+9800GTX+ and this was before the official Batman:AC Patch (which helped with the DX11 issues) and the nVidia PhysX update (which helped with the PhysX issues.) I was running the game BETTER than most nVidia users since AMD released a hotfix before nVidia that improved DX11 performance haha. I had posted some pics on another forum documenting my progress.

While trying BL2, I re-ran Batman:AC with my updated hardware (7970+GTX 460) and I averaged 62 FPS with 8xMSAA, everything else max.

 

[wand3r3r]

Ah. I see how this is going to turn out already. It's going to be one section of one map played a certain way where performance tanks. Gee, I'd sure like to know how to replicate this situation i keep hearing about. I am not saying it doesn't exist. I am saying i can't replicate it. You saying that because i haven't posted any screenshots is kind of pointless a railven already has. And he states he backs up my findings.

I just played the section with the little girl pyro area and I tried the 60fps limit with 4 players but it was under that regularly. There were helicopters flying and the machine gun post shooting rockets and huge mosquitos etc. It's not 1 area. I tried to benchmark with fraps but when I got out of the game nothing had saved. Will try again later.

 

[Aikouka]

Can anyone recommend a quiet 7970 ghz? I'm still trying to decide between 7970 ghz and 680. Seems like just about everything I read says the 7970 (even with custom cooler) is loud. I'm not interested in a custom fan profile that is only quiet at idle.

I have to admit that the noise level is one thing that I really like about my GTX 680. Over the years, I've grown more and more tired of loud computers, and the deep whirr from a high-end Radeon GPU when it's really pushin' the pixels doesn't help (I own a 5870 and a 6950). The reference-style cooler on my 680 just sounds like a loud gust of wind (but not too loud). It's a nice change of pace.

EDIT:

In the spirit of fairness, I don't know what a high-end 7000-series sounds like under load. Maybe it lacks the whirring noise that made my 5870 and 6950 somewhat annoying.

 

[railven]

In the spirit of fairness, I don't know what a high-end 7000-series sounds like under load. Maybe it lacks the whirring noise that made my 5870 and 6950 somewhat annoying.

My Radeon HD 7970 is quieter than my GF's GTX 680

But of course I'm cheating because I'm using a kick ass Accelero Extreme cooler while she is using stock. :whiste:

 

[KingFatty]

I just played the section with the little girl pyro area and I tried the 60fps limit with 4 players but it was under that regularly. There were helicopters flying and the machine gun post shooting rockets and huge mosquitos etc. It's not 1 area. I tried to benchmark with fraps but when I got out of the game nothing had saved. Will try again later.

Your efforts are appreciated. I put a greater weight on actual hard numbers that you are going after, as opposed to seat-of-the pants eyeballing that is prone to user bias.

 

[wand3r3r]

Here's a pretty mild run through caustic canyon or whatever it was called. The x is seconds, and the y FPS. I had the frames limited to 60 or there would have been jumps up to 200 most likely. This was only with 1 and then 2 other people. (1 joined somewhere in the timeframe)

I still need some more data from 4 person because single/2player "feels" smoother then with the 4 person chaotic battles.

 

[chimaxi83]

Lol this thread is still arguing about PhysX

 

[wand3r3r]

Lol this thread is still arguing about PhysX

Most of the arguing is a few people claiming physx works like a champ, and basically only my data to show that it isn't so, at least in my case. Perhaps it's my rig, perhaps SLI bugs, but I'd like to see some comparable stats from fellow 680/690 owners with everything maxed out just to know.

 

[Lonbjerg]

Most of the arguing is a few people claiming physx works like a champ, and basically only my data to show that it isn't so, at least in my case. Perhaps it's my rig, perhaps SLI bugs, but I'd like to see some comparable stats from fellow 680/690 owners with everything maxed out just to know.

That is borderline lying.

MOST people don't have YOUR problem...aka they experinece fine gameplay.

Cart before horse.

Have you checked your system for DCP?
Or do you just want to whine up and do no searching for the cause?!

 

[wand3r3r]

That is borderline lying.

MOST people don't have YOUR problem...aka they experinece fine gameplay.

Cart before horse.

Have you checked your system for DCP?
Or do you just want to whine up and do no searching for the cause?!

Do you have an agenda here? I am posting my experiences, nothing more, nothing less. I still haven't seen a single person post proof of anything, to say they experience "fine gameplay" is borderline shady without benchmarks if anything is. I'm not saying they do/don't, however I have proven and only spoken facts with evidence, including MY/ME/MYSELF/MY RIG in everything (not yet sure if it's my rig or nvidia/sli).

I am asking others to please post their experiences to see if there is something funny in my setup/situation and isolate the cause (sli, drivers etc.). Prove me wrong. (Although I can't be wrong to show my experience, it could be proven if it were only my setup)

About the DCP, you provided some link to a foreign site with an .exe. I don't install just anything shown by random people much less from random sites. I did google DCP but didn't see anything I was comfortable installing.

 

[Termie]

Do you have an agenda here? I am posting my experiences, nothing more, nothing less. I still haven't seen a single person post proof of anything, to say they experience "fine gameplay" is borderline shady without benchmarks if anything is. I'm not saying they do/don't, however I have proven and only spoken facts with evidence, including MY/ME/MYSELF/MY RIG in everything (not yet sure if it's my rig or nvidia/sli).

I am asking others to please post their experiences to see if there is something funny in my setup/situation and isolate the cause (sli, drivers etc.). Prove me wrong.

This was an honorable response to an unwarranted attack. If someone is experiencing a problem with the game and is willing to post results, I can't understand why anyone would insult him for it.

Provide suggestions without the attacks or keep it to yourself.

 

[Lonbjerg]

Posting post after post whitout doing any search for the anomaly is useless...it's not even informative, it's just a waste of bits.

You have an issue.
You post about your issue...alll the time.
But you won't take the time of day to look into what is at fault at your system...:whiste:

I'm not the one trolling here.

Make a new thread about YOUR personal issue...as this thread is not about YOUR issue

 

[wand3r3r]

This was an honorable response to an unwarranted attack. If someone is experiencing a problem with the game and is willing to post results, I can't understand why anyone would insult him for it.

Provide suggestions without the attacks or keep it to yourself.

Thanks. So much for a nice atmosphere around here.

If this is not the right place to post about Borderlands 2 GPU/CPU benchmarks then I'm not sure where is. I can create a thread but this sure seems like a fitting place to discuss and inquire about BL2 benchmarks.

Is this thread here for you to be unfriendly and not even discuss BL2 performance? (How does that work?)