• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Tomb Raider benchmarks

Rauelius said:
Then why not use the graphics engine to render the cutscenes? I hate looking at pre-rendered video when the machines I'm using to play the game is several quadrillion orders of magnitude more powerful than the machines involved in making the pre-rendered video.
Click to expand...

I dont know, Ive always hated seeing sucky bitrate cutscenes that look worse than the normal game along with changing the guns/outfits of the character. The worst Ive seen is Arkham City, they use console graphics, low bitrate, and if you are using another Batman skin it doesnt change his outfit. Maybe they just cant animate whatever they are doing with the game engine (like explosions or debris).
 
I guess i'm lucky, running latest beta drivers with my GTX 570. Only thing not on max is TressFX and DoF (I turn DoF off all the time anyways, don't like the look) @ 1080p. I get a Vsync locked 60 with the ingame benchmark and 60 most of the time while in game. It does drop to 40ish if a lot is going on but without a frame counter I can't tell. Also the game is stable for me no crashes.
 
When a new game is released and AMD cards fall behind:

"Haha, AMD's driver team sux! Nvidia rules, what rubbish AMD drivers, wow AMD can't get it together! stupid AMD!"

When a new game is released and Nvidia cards fall behind:

"This is unacceptable, the developers are horrible! AMD is paying them off! This is poorly optimised! the developers of this game are rubbish!"
 
KazeoHin said:
When a new game is released and AMD cards fall behind:

"Haha, AMD's driver team sux! Nvidia rules, what rubbish AMD drivers, wow AMD can't get it together! stupid AMD!"

When a new game is released and Nvidia cards fall behind:

"This is unacceptable, the developers are horrible! AMD is paying them off! This is poorly optimised! the developers of this game are rubbish!"
Click to expand...

Because it's all true, lol.
 
w00t, with new patch (3/9) performance on Ultimate settings went from 4 min 25 max FPS to 25 min 31 max FPS. Gonna tweak the settings a bit (AA, FOV and TressFX) to see what my I can get the game to balance out at now. System in sig.

EDIT To switch back and forth, right click TR in Steam library, go to BETA tab. BETA is old game, pre patch.
 
RangerXML said:
w00t, with new patch (3/9) performance on Ultimate settings went from 4 min 25 max FPS to 25 min 31 max FPS. Gonna tweak the settings a bit (AA, FOV and TressFX) to see what my I can get the game to balance out at now. System in sig.

EDIT To switch back and forth, right click TR in Steam library, go to BETA tab. BETA is old game, pre patch.
Click to expand...


How can I tell if I got that patch on Steam?

I just played and had some really weird issues. FPS would randomly drop to like 2- 3 fps for like 20 seconds and then go back to normal for no reason. Even in the menus it would lag extremely bad.

Didn't happen before just this last session. I noticed one of my Titan's drop to like 600mhz when it was doing this then the next time it happened the other card did the same thing.

I also noticed it happen when both cards were running @ 1130Mhz so not sure what is going on here...
 
Check if you have the beta option in the Tomb Raider properties from the Steam game library (do not use it, because its the old version), but if its there then you updated.
 
This game looks fun, wish my GTX 680 came with this game instead of that free2play thing.
 
If you want the no frills version, I think you can still get it on GMG for $40 (if you bought a $500 video card, that is not much) with coupon code: GMG20-P4DLK-FKYRS

Otherwise, I wouldn't doubt we're are going to see it discounted on Steam in the next month or so. Most games I've gotten have had some kind of a heart breaking discount within a month of release. At least the PC version isn't $60 like the consoles are being forced to pay, which I still think this game is worth.
 
RangerXML said:
If you want the no frills version, I think you can still get it on GMG for $40 (if you bought a $500 video card, that is not much) with coupon code: GMG20-P4DLK-FKYRS

Otherwise, I wouldn't doubt we're are going to see it discounted on Steam in the next month or so. Most games I've gotten have had some kind of a heart breaking discount within a month of release. At least the PC version isn't $60 like the consoles are being forced to pay, which I still think this game is worth.
Click to expand...

Or if you wanna get Bioshock Infinite too, you could get a Radeon 7850 1GB (cheapest I saw was an HIS for $165), and then sell the Card for say $120-$130. You'd get two awesome games, and help out someone looking for a great video card.
 
Lol, that seems like too much work with flipping the video card. I might check out that coupon code Ranger. Thanks!
 
Zarich said:
It really appears that AMD has turned a positive corner and maybe the leader, but tressfx just seems really dumb.. all the things we need graphically realistic hair did not ramp up to the top of my list.
Click to expand...
More realistic boobies would be my first..... ;)
 
Or you can go to Ebay and buy the game with infinite for $45.
That seems to be the going rate.
 
KazeoHin said:
When a new game is released and AMD cards fall behind:

"Haha, AMD's driver team sux! Nvidia rules, what rubbish AMD drivers, wow AMD can't get it together! stupid AMD!"

When a new game is released and Nvidia cards fall behind:

"This is unacceptable, the developers are horrible! AMD is paying them off! This is poorly optimised! the developers of this game are rubbish!"
Click to expand...

AMD is definitely paying them off to get the benchmark to run well.

However in game testing shows the cards are on par with them, as I showed in an earlier link. Considering Nvidia has performance on par with AMD in a gaming evolved title on release basically(again not the benchmark), credit has to be given to Nvidias driver team for doing so well with so little time with the game.
 
tajoh111 said:
AMD is definitely paying them off to get the benchmark to run well.

However in game testing shows the cards are on par with them, as I showed in an earlier link. Considering Nvidia has performance on par with AMD in a gaming evolved title on release basically(again not the benchmark), credit has to be given to Nvidias driver team for doing so well with so little time with the game.
Click to expand...

You basically just represented EVERYTHING I just said.
 


Uploaded with ImageShack.us

As I have said, performance for the built in benchmark Nvidia doesn't do that well because AMD ensured that they would perform better in the benchmark. If the coding was generalized into the game itself the same as the canned benchmark, then AMD would have better performance in game compared to Nvidia. However this isn't the case.



Uploaded with ImageShack.us

Notice this is from the exact same review and they actually now test in game.



Uploaded with ImageShack.us

This is from another review website that tested in game which mirror the same results.

The point I am trying to make is although this is a gaming evolved title where AMD should have every advantage in the world since they are now putting themselves in Nvidia shoes like the way it was meant to be played program. They do not have an advantage beside a canned benchmark. AMD basically performs as well as nvidia in real world testing which should be shocking considering how well it does in the canned benchmark.

And thus, this doesn't represent anything you post because people that own Nvidia hardware have no reason to bitch about this title besides losing in the built in benchmark(which is important for marketing reasons only). And considering Nvidia had so little time with the code compared to AMD since they didn't get early access, it pretty impressive of Nvidia. When was the last time you seen a Nvidia sponsored title, where they didn't have a big advantage initially on release? Pretty much never. Both company are guilty of paying developers money to get games prioritized for their hardware. But when was the last time you seen this happen at launch and the team that did the sponsoring didn't get any performance advantage? AMD sure got it with sleeping dogs and the latest drift, but it sure ain't happening here. That why I made the comments I made.

Its a certainty AMD paid Eidos to get code early so they could perform well in the benchmark, otherwise why would it be even be a gaming evolved title. However because this difference doesn't show up in the real game and is equal to nvidia, we have one of two options. AMD driver teams sucks if they cannot beat Nvidia driver team when they have early developmental access with the game or Nvidia's driver team is really good. Of course Nvidia wants to get better performance in the benchmark(hence the apology) because for lazy reviewers, its how they do a review. However, if we take real world performance seriously like HardOCP, its meaningless to people who are playing the game rather than looping a time demo.

In summary there should be no reason why this game should perform equally on both platforms equally on release. If AMD could only translate performance into a time canned demo and not the game itself from having early access to the game, it is simply a poor reflection on their driver development.
 
Last edited:
I think this game looks excellent and runs very well for the most part [im about half way thru the single player campaign]

I have it maxed [with TressFX] and for the most part it sticks to a constant 60fps v-synced. Although some larger regions cause it to dip a bit which i wasnt expecting, especially after my benchmark results without v-sync. [about 110-115fps]

Interestingly I noticed that steam did a small update to it the other day, I ran the benchmark again and the result actually had a lower average score but didnt exhibit a weird clipping issue with the camera going thru Lara as I had before and also my low fps increased dramatically, cause earlier the game reported like a single digit low fps...which i call BS on as it looks silky smooth from beginning to end.

If anyone is interested in seeing the original benchmark with these unusual anomalies here is the link to the youtube upload I copied offscreen:

http://www.youtube.com/watch?v=qQaEq-kPcpw

And this is my first post, just joined, so would like to just say hi to everybody ;)
 
sammyflinders said:
I think this game looks excellent and runs very well for the most part [im about half way thru the single player campaign]

I have it maxed [with TressFX] and for the most part it sticks to a constant 60fps v-synced. Although some larger regions cause it to dip a bit which i wasnt expecting, especially after my benchmark results without v-sync. [about 110-115fps]

And this is my first post, just joined, so would like to just say hi to everybody ;)
Click to expand...

Hello,and welcome to the no.1 enthusiast forum:D

Nice scores btw only getting about 45 here (no crossfire/sli) but im happy with it:)
 
Thanks for that :)

yeah 45 is perfectly acceptable, I drop down to it occasionally and its only notable for the fact that it surprised me but it is still highly playable. Dumps all over one of the console versions any day of the week hey ;-)
 
Here is what I get at 5760x1080 with sig rig (images downsized to 40% to not blow up the thread)

2013-03-11_00002.jpg


And more importantly the settings:
2013-03-11_00003.jpg


It is suprisingly playable - I have done about 7 hours with only 1 crash. If I turn off post processing I get about double that framerate. If I enable TressFX I get between 1 and 5 FPS.
 
That seems low for a 690... my single titan does 35-40 FPS at 2560x1600 2xSSAA with only tessfx off which is effectively double the resolution, about 2 million more pixels than your res. Maybe it's a VRAM thing, but could be that SLI isn't working properly.
 
x3sphere said:
That seems low for a 690... my single titan does 35-40 FPS at 2560x1600 2xSSAA with only tessfx off which is effectively double the resolution, about 2 million more pixels than your res. Maybe it's a VRAM thing, but could be that SLI isn't working properly.
Click to expand...

You are running 2500x1600, which is 4 million pixels...He is running 5760x1080, which is 6,220,800 pixels...I think you meant to say that HE is running 2 million MORE pixels then you, not LESS...

EDIT:

Are you saying that you are running more pixels do the fact you are running 2X SSAA? I am just trying to figure out if the first statement was a typo, or if the SSAA is the reason behind it..
 
ccityinstaller said:
You are running 2500x1600, which is 4 million pixels...He is running 5760x1080, which is 6,220,800 pixels...I think you meant to say that HE is running 2 million MORE pixels then you, not LESS...

EDIT:

Are you saying that you are running more pixels do the fact you are running 2X SSAA? I am just trying to figure out if the first statement was a typo, or if the SSAA is the reason behind it..
Click to expand...

Right, I'm saying with SSAA, that is effectively 8 million pixels - so close to 2 million more pixels than his res...

Edit: And without SSAA I get 62 average in that benchmark (may be held back by CPU since I've got only an i7 @ 3.5ghz) so the performance he's getting from a 690 isn't right, should be double what he's getting currently. I also have High Precision on and he doesn't.
 
Last edited:
If I should be getting more I'd love too. Not sure what the deal is though. I am curious if the raid 0 ssd's I run it on cause a problem at all. Using the onboard intel raid for a pair of intel 600gb ssds.
 
Looks like I made a goober move - I didn't have Exclusive Fullscreen set to on...DOH!

New Settings:
2013-03-12_00002.jpg

2013-03-12_00003.jpg


With new results:
2013-03-12_00001.jpg


If I turn Anti-Aliasing from 2x SSAA to FXAA:
2013-03-12_00004.jpg
 
6048x1080 completely maxed with 2x SSAA. 7970 @ 1125/1575

TombRaider_Bench.jpg
 
GTX 670 DC2 Top @ 1254mhz (stock boost cl.), 1080p, Ultra

4xSSAA - 40.9

2XSSAA - 50.5

FXAA - 62.1

Ultimate (TressFX) + FXAA - 44.7

Ultimate + 2XSSAA - 29.2

Ulitimate + 4xSSAA - 11.9

TressFX is takes almost 30% off FPS at the best caes. It is nice but not worth. Boobs wolud be different story. I play with ultra + FXAA and game is 90% off time at 60fps.
 
Last edited:
For people who have the latest high end Intel cpu's, since this game appears to be cpu limited in certain areas, like Shanty Town, are you able to get 100% gpu utilization in these much more demanding sections of the game?
 
Little update..
Nice jump in performace with new beta drivers..Round 20% with treesFX.

Ultimate + 2XSSAA - 36.6

Game is still not fixed. Patch from developers is needed to fix crashing and some flickering. Every time treesFX hair is zoomed FPS drops by half, below 20. Didint notice GPU usage problems but i am still at the beginning of the gameplay.
 
Last edited:
roki977 said:
Didint notice GPU usage problems but i am still at the beginning of the gameplay.
Click to expand...

Yeah, gpu usage doesn't really become an issue until you get further into the game. When I get to Shanty Town, gpu usage nosedives, and my fps plummets. If you're able to verify whether or not that happens on your cpu, that would be greatly appreciated.
 
I started to play bit more, now that driver is out so i probably will. Hate that left, right meshing and stuff like that. Wulf just rape my in the cave.
 
I just ran a bench on the system in my sig and I actually get better results with TressFX on then off with 314.21 and new patch. Everything maxed except FXAA and DoF normal.

TressFX off:
17 min
70 max

TressFX on:
26 min
40 max

I should mention I prefer less flux between min, max and closer to a min of 30fps+. TressFX still looks horrible.
 
2500K @ 4.6, (2x) Powercolor Myst 7870XT

2560x1600 Ultimate TressFX enabled 52 ave, 37 min
2560x1600 High + TressFX + Tessellation + Ultra textures 63 ave, 45 min

The image quality difference isn't very dramatic. Cards show 95% usage each.

1920x1080 Ultimate TressFX enabled 84 ave, 66 min
1920x1080 High + TressFX + Tessellation + Ultra textures 91 ave, 70 min

Driver 13.3 beta 3.
 
http://www.hardocp.com/article/2013...deo_card_performance_iq_review/5#.UUqCgldqks8

Basically the hardocp results mirror my earlier posts. AMD doesn't have an advantage in this game and after the new drivers for Nvidia, they see to have a performance advantage now.

Considering this is a gaming evolved title, it makes Nvidia driver team seem pretty boss.

Nvidia before the new driver basically had performance that was on par with the 7970 ghz.

http://www.hardocp.com/images/articles/13632141234v2TkTbPdM_5_4.jpg

http://www.hardocp.com/images/articles/13632141234v2TkTbPdM_9_1_l.jpg

After the driver update, Nvidia leading by more than 20 percent.

At launch I find it ironic Nvidia is apologizing for having the same performance as AMD in a gaming evolved title.
 
Fooshnik said:
Depends on the resolution.

13632141234v2TkTbPdM_5_3.jpg
Click to expand...

At that point it isn't even about drivers anymore though, just video memory. Take off the SSAA at 2560*1600 or use it with less resolution and your ahead again, which shows the Nvidia series doesn't have enough video memory at that resolution.
 
So my buddy has this game too and I was watching him try to play it with a 360 controller ( he can't aim for shit) and I noticed something about the game while watching him. I don't think the Nvidia cards are doing Tessellation, or it didn't look the same as when I play on my 7970. He has EVGA 670 FTW, and he was on Ultimate with FXAA + Tress FX and Tessellation, etc. I think Nvidia lowered/dropped/messed with the Tessellation in the game to get a good FPS boost.

It looks too different to not be noticeable.

EDIT: Wasn't Tessellation one of the problems with Nvidia cards and this game, causing crashes? It would make sense.
 
You must log in or register to reply here.
Back
Top