dual core bottlenecking 560ti?

  • Two Factor Authentication is now available on BeyondUnreal Forums. To configure it, visit your Profile and look for the "Two Step Verification" option on the left side. We can send codes via email (may be slower) or you can set up any TOTP Authenticator app on your phone (Authy, Google Authenticator, etc) to deliver codes. It is highly recommended that you configure this to keep your account safe.

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Sounds like a similar problem I was/am having with my computer, and you even said how ****ed up it is.

So you have just upgraded your GPU right? It could be the motherboard if the benchmarks are showing your new card should run the game at much higher FPS.

In cases of mass disappointment, you could always RMA the card. Especially if you've ordered from Newegg. I've done it twice until I figured my motherboard is getting kung fu'd once I get a new pc.

PC hardware is a waste of time and resources the more we go on.

I doubt it is as extreme as yours was. When I got this pc it had a very ****ty quad core in it. When I upgraded to my current dual core, I saw my fps dramatically increase. When I upgraded to the 560ti, I saw my overall fps increase as a result, but the minimum just feels too low. For example when I tried running the Crysis 2 demo at fullHD and extreme settings, it was a slideshow on my 8800gt. 40's fps average with my 560ti is a vast improvement I think.

I think my true real bottlenecks lie in BC2 and Crysis. BC2's fps didn't seem to improve much at all, but it doesn't seem to dip that much either.
 

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Okay something new I just discovered. Decided to play some Borderlands again and try out the beginning section of General Knoxx. It started stuttering like mad when exiting the station in front of Scooters area. I took a screenshot to show the area I am talking about, though its crap that it didn't capture the fps marker. I was getting a flat 39 in that area, the same freaking fps I got with my 8800gt. Even with a bottleneck, the 560ti should eat this freaking game for breakfast, lunch, and dinner right?

Is Knoxx this badly optimized, or should I be worried?
 

Attachments

  • borderlands.jpg
    borderlands.jpg
    579.1 KB · Views: 11

JohnDoe641

Killer Fools Pro
Staff member
Nov 8, 2000
5,330
51
48
42
N.J.
www.zombo.com
That's not a good example, Borderlands is an Unreal engine game which won't raise it's minumum fps with a graphics card boost, what that is is a CPU bottleneck or like you said, a badly optimized DLC.

Borderlands is also capped at 62-63 fps like all Unreal Engine 3 games seem to be, so that may have to do with the lower fps, if there's a way to uncap them, maybe then they'd be higher. I've never understood why they're all capped, maybe it's a result from the lolol consoles or something. :c
 
Last edited:

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
That's not a good example, Borderlands is an Unreal engine game which won't raise it's minumum fps with a graphics card boost, what that is is a CPU bottleneck.

Now my knowledge of bottlenecks is pretty slim, but I lowered all the details and gamed at 1280*800 to make it more cpu heavy. My fps went up which would tell me no bottleneck? I started flicking on all settings but left dynamic shadows off and left the resolution the same. My fps stayed the same. So then I upped it to 1920*1200, fps went down slightly but was still roughly the same. The moment I enabled dynamic shadows, it all went to hell. Could it be possible the shadows themselves are just poorly optimized?
 

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Sorry to keep bumping this but I have been doing more research into this trying to figure out what would be the bottleneck.

PSU - Enough wattage and amperage is going through the system. Have not seen any signs of an ailing PSU.
RAM - Ganged and unganged run in dual channel mode regardless and the speed difference is minimal. Unganged seems to work better for multi threaded tasks since it allows multiple sticks to access a core at a time. At least this is what I got from this thread http://www.overclock.net/5426161-post25.html
CPU and motherboard - Most likely the bottleneck but after doing some searching I found out a few things. The current AM3 cpu's have a hyper transport rate of 2000mhz/4000m/t's, which is what I currently have. My mobo is an AM2+ with a front-side bus of 5200m/t's, which comes out to 2600mhz. I got my information from this site http://www.hardwaresecrets.com/article/The-HyperTransport-Bus-Used-By-AMD-Processors/19/4

1,800 MHz = 3,600 MT/s = 7,200 MB/s
2,000 MHz = 4,000 MT/s = 8,000 MB/s
2,400 MHz = 4,800 MT/s = 9,600 MB/s
2,600 MHz = 5,200 MT/s = 10,400 MB/s

None of the current AM3 cpu's offer 5200m/t bus speeds, therefore I am guessing there is a bit of downclocking going on. This could be a reason for a bottleneck, but it seems to be a purposeful one?
 

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Just wondering, What is your windows rating?

I doubt the Windows rating is that important but here it is broken down

Processor 6.6

Memory 7.4

Graphics 7.9

Gaming Graphics 7.9

Primary Harddisk 5.9

Course it gives me a 5.9 overall because of the lowest score.
 

Kantham

Fool.
Sep 17, 2004
18,034
2
38
Windows rating is terribad because it puts so much judgment on the HDD method/speed as if it was a majorly important factor.
 

NRG

Master Console Hater
Dec 31, 2005
1,727
0
36
35
Windows rating is terribad because it puts so much judgment on the HDD method/speed as if it was a majorly important factor.
I'm pretty sure the logic behind the windows rating is that "your computer is only as fast as the slowest part", which is pretty reasonable because the hard drive is always the slowest part of the computer. If you want to score the responsiveness of the OS in general (instead of games or professional software), the hard drive makes all the difference. Even people who know nothing of computers will notice the better performance with a better hard drive and therefor the overall score will seem to make more sense to the average joe.
 

NRG

Master Console Hater
Dec 31, 2005
1,727
0
36
35
They both probably hit an fps cap or (more likely) hit CPU limitation in the very simple FPS test it runs :p 7.9 is the highest you can get I think.
 
Last edited:

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Alright this will probably be my last post before I totally say screw it. I installed EVGA Precision, which is mainly used for overclocking but has a very nice utility that showcases your gpu temperature, fan speed, core clock, etc.

I tested it with Crysis 2 and let it run for a good 5 minutes before making conclusions.

GPU temp reached roughly 84c, which could be better but my 8800gt ran this hot so I am assuming this is normal.

Core clock and Shader clock never downclocked the whole time I was playing. The reason why I wondered about downclocking was because when I alt tab and stay on the desktop, my gpu will downclock itself to inactive levels. I am assuming its because of the power save feature that this series of cards have.

Checked CPU-Z and it never showed a sign of downclocking either. I even checked task manager and noticed my cpu was only about 50%-60% usage.

Fan speed reached about 55%, would it be safe to manually increase this a tad or do I risk damaging the unit?

The one thing I did notice that may be troublesome is the GPU usage. At the start of the test level it hovered around 95%. When I reached the point where my fps dipped, my usage dropped to about 65%. These aren't random dips by the way. Is that normal? Is the drop in usage causing my fps to drop, and if so would my cpu be to blame for not providing enough power to satisfy the g-card?

To close I just want to thank all you who have provided input. It is much appreciated :)
 

Kantham

Fool.
Sep 17, 2004
18,034
2
38
84C?

What is your resolution? That's very hot considering Precision says my 570 GTX SC runs at a highest of 65C on 1440x900 and 70% speed, aprox 2800 RPM.

EDIT: I read that Nvidia cards, at least most of them can endure temperatures up to 120C. But I don't buy that quite much. I'd be extremely worried at 90C.
 
Last edited:

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
84C?

What is your resolution? That's very hot considering Precision says my 570 GTX SC runs at a highest of 65C on 1440x900 and 70% speed, aprox 2800 RPM.

EDIT: I read that Nvidia cards, at least most of them can endure temperatures up to 120C. But I don't buy that quite much. I'd be extremely worried at 90C.

1920*1080. Also as I said mine runs roughly at 55% fan speed so I wonder if that would make a difference? Then again it could be a tad bit dusty as I play with my case off.
 

SkaarjMaster

enemy of time
Sep 1, 2000
4,872
11
38
Sarasota, FL
What would happen if it ran 100% fan speed all the time like some older cards (a lot older that is)? Would that solve your problem or would it then be too loud?
 

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
What would happen if it ran 100% fan speed all the time like some older cards (a lot older that is)? Would that solve your problem or would it then be too loud?

I upped it to around 70% and it sounded like my 8800gt, which sounded like a jet taking off. I don't mind the noise really, but running it at 100% wouldn't cause an issue later down the road?
 

Sir_Brizz

Administrator
Staff member
Feb 3, 2000
26,020
84
48
Well, it's a fan and the bearings do have a lifespan, so yeah, eventually it would.
 

Capt.Toilet

Good news everyone!
Feb 16, 2004
5,826
3
38
42
Ottawa, KS
Well, it's a fan and the bearings do have a lifespan, so yeah, eventually it would.

Yeah that is understandable. I might just opt for an even 70% fan speed. If my 8800gt can last 4 years running like a fighter jet, I am sure this card can. If not then I will be surprised. Anyway I will report back when I get home to see if the same fps drops occur and if my gpu usage lowers with the increased speed.

EDIT: Found this toms benchmark of Metro 2033 and found this quote rather intriguing.

We used the fastest single-GPU graphics card available in order to expose any platform-oriented bottlenecks in Metro 2033. With that said, it’s hard to imagine anyone buying a GeForce GTX 580 and gaming at 1680x1050. If they did, they’d see performance start to drop off in a noticeable way starting with AMD’s Phenom II X4 970, continuing on through Intel’s dual-core offerings, and ending with an older Core 2 Quad Q9550.

The moral of the story here seems to be that, as you step up to higher-end graphics, a dual-core processor simply isn’t fast enough.

I am assuming they are talking about higher end graphics cards and not just the engine graphics? If so then this might very well be my problem after all. But tbh it really shouldn't since its got a 3.2ghz clock. Are these new g-cards really that cpu hungry?

http://www.tomshardware.com/reviews/sandy-bridge-core-i7-2600k-core-i5-2500k,2833-18.html

Another Edit: I used EVGA precision to set a custom fan profile and had it go only as high as 70% fan speed. I peeked at 75c, so about 10c cooler and it still isn't as loud as my 8800gt.
 
Last edited:

Sir_Brizz

Administrator
Staff member
Feb 3, 2000
26,020
84
48
Well, it is odd. I don't know about dual core processors. I have a q6600 and a radeon 5750 and I can play all of the games you've mentioned at 1600x1200 with some AA and mostly max settings and the game is smooth (I don't really check the framerates, but I don't notice hitching or frame lag).