GTX 770 - Tomb Raider - Are these FPS normal? + GPU noise

bourbonster

Reputable
Jun 27, 2014
8
0
4,510
I'm getting the following FPS in TR's benchmark (1080p, everything maxed out, TressFX on, FXAA):

*V-sync on: min 40 max 62 avg 55
*V-sync off: min 46 (?) max 87 avg 61

I have a i5 4670k and asus gtx 770 direct cu ii at stock settings, 2x8gb corsair vengeance, msi z87-gd45 gaming, corsair cx750m PSU and samsung ssd (evo 840 250gb). All drivers up to date.

Are these FPS acceptable for my rig? I'm asking because this guy http://www.youtube.com/watch?v=WPTRNi87lYI and these guys http://www.maximumpc.com/10_most_graphically_demanding_pc_games?page=0,1, among others, seem to have better results with similar or slightly inferior hardware. In fact, if I turn on SSAAx2 + TressFX the games becomes unplayable in most areas.

Also, when playing this game my GPU occasionaly makes a strange electrical/buzzing noise that last for about 10/15 seconds and then fades away, I don't know if it is coil whine, as it comes only in short bursts (it's not constant) and the pitch seems lower... should i be worried about this? I think I'm a little paranoid that my new rig is underperforming or that the GPU is defective in some way :)

Thanks.
 
Solution
Yes, v-sync should be OFF for a realistic result. I just ran the bench with the standard Ultimate settings @ 60HZ. But I left exclusive full screen enabled. W/o that, it basically runs in a window... a pseudo "full screen" window. That may skew the results, I'm not sure. Anyway, my results with std Ultimate settings @ 1080p are:

Min: 58.0 fps, Max: 62.3 fps, Avg: 60.0 fps

As per HWiFo64:
CPU usage: 47.6%. GPU usage: 100%
CPU Core (core #1) temp max: 49C GPU max temp: 70C

So, your results with a better CPU and a slightly lesser gfx card look to be pretty good.

clutchc

Titan
Ambassador
I've been running that benchmark a lot lately, benchmarking some rigs I have been working on. What exactly are your settings; Ultimate w/o making any other changes? 60HZ? The only rig I have that would come anywhere close to yours would be the one in my sig below. If it would help any, I could run the TR bench at the same settings.
 
That is about right for the benchmarks for your equipment in that game.

as for the noise, you will need to be searching for the noise when it happens to pinpoint it. With out being there we can only guess and give you ideas to what it can be. Like a bad fan, replaceable by contacting the manufacture. They will send out a new fan. Coil wine on the other hand is usually not RMA'able. Tee only thing besides coil wine that can give a buzzing sound from the card is a bad fan. this should give you ideas on where to look for problems.

Don't forget that this game runs best on AMD GPU's also.
 

bourbonster

Reputable
Jun 27, 2014
8
0
4,510
Firstly, thank you for your answers.

clutchc, my exact settings are:

BASIC:
1920x1080p
V-sync - double buffer when on
Refresh rate 60 hz (fullscreen ON, exclusive fullscreen OFF)
Monitor aspect ratio - auto

ADVANCED:
Quality - Ultimate
Texture Quality - Ultra
Texture filter - Anisotropic 16X
AA - FXAA
Shadows - Normal
Shadow Res - High
Level of Detail - Ultra
Hair - TressFX
Reflection - High
DOF - Ultra
SSAO - ULTRA
Post processing, High precision and Tessellation ON

Also, what is the point of running a benchmark with V-sync on, like the guy on the youtube video did? Shouldn't it be always off when benchmarking, so even if you don't "see" the FPS beyond 60 on a 60hz monitor you at least know that the game is performing better than 60 FPS? Or am I missing something?

bgunner: it's pretty difficult to pinpoint it exactly. I've run the game with my case open to see the fans spin but everything seems normal, also I've been monitoring the fans speed and temp with GPU-Z and everything seems ok... and I would not like to send it back except if really needed, as I ordered it from another country here in Europe and so it would be a lot of trouble (and a a lot of days of no gaming) to replace it ;)

 

clutchc

Titan
Ambassador
Yes, v-sync should be OFF for a realistic result. I just ran the bench with the standard Ultimate settings @ 60HZ. But I left exclusive full screen enabled. W/o that, it basically runs in a window... a pseudo "full screen" window. That may skew the results, I'm not sure. Anyway, my results with std Ultimate settings @ 1080p are:

Min: 58.0 fps, Max: 62.3 fps, Avg: 60.0 fps

As per HWiFo64:
CPU usage: 47.6%. GPU usage: 100%
CPU Core (core #1) temp max: 49C GPU max temp: 70C

So, your results with a better CPU and a slightly lesser gfx card look to be pretty good.
 
Solution

bourbonster

Reputable
Jun 27, 2014
8
0
4,510
thanks, clutchc! You have v-sync OFF, right? My average fps is about the same as yours, but sometimes when i run the benchmark, just like i did now after your reply, my min fps are very low (this time 19) and my max very high (74). The average ran at 61, similar to yours.
 

clutchc

Titan
Ambassador


V-sync is "application controlled" in Catalyst Control Center. I can only assume it is off in the game since I didn't see an option for it. But you bring up a good point. But My max result was slightly over 60 fps, so I gues we can assume it was off in-game... or the game caps it at ~the vert refresh rate. That would be a bummer, huh?

All it takes is one slow frame rendering to ruin the 'min' result. Happens sometimes.
 

bourbonster

Reputable
Jun 27, 2014
8
0
4,510
GPU load 99% as per GPU-Z monitoring. I kinda suspect in your case that catalyst is using v-sync, dude, with your rig you should be getting higher max fps, me thinks, plus the guy on the youtube video had v-sync on and his max fps also exceeded 60, more exactly 62, altough i don't understand how that would happen.
 

clutchc

Titan
Ambassador


No, v-sync is application controlled:
RqUrTVA.jpg

But since your max is also 62 fps, I suspect the game caps the refresh rate at or near the HZ you set in the options. Try setting yours to 50 HZ and I'll bet the max fps is ~50.