Vanilla EVGA GTX 670 Testing Anyone?

[toyota]

I'm still playing with both rigs tonight but i'm just happy i hit 10K. +160 core, +200 Mem, 115% power, Stock reference cooling which is shocking. It hated the low temps from the water. Must be a bios thing

grr you have to stop replying with those settings because that does not tell me the actual clocks.

 

[Don Karnage]

grr you have to stop replying with those settings because that does not tell me the actual clocks.

1265 core, 6414 memory

 

[toyota]

1265 core, 6414 memory

how would you get a higher graphics score than me with the same memory speed and 35 mhz lower core?

 

[Don Karnage]

how would you get a higher graphics score than me with the same memory speed and 35 mhz lower core?

Im not sure bro.

 

[toyota]

Im not sure bro.

yeah I get about 130 points less at the same core and memory speeds as you. a different cpu, mobo and ram could still impact the graphics score by just over 100 points or so I guess.

 

[Termie]

Toyota, run 3dMark11 using a afterburner or presicion graph in the background to show temps, boost, power, and voltage.

I guarantee you are not holding 1300 during your entire run.

Repost from my thread:

Hey all - I have another update for you. This one is a closeup view of how 670 overclocking works, along with the highest 3dMark11 score I've achieved so far, running at a theoretical boost of 1110+150 = 1260, and 6600 on the memory:

There are a number of things to look at in this graph. First, note that 3dMark11 has 6 main tests - four graphics test, a physics test (where GPU load is very low), and a combined test. You can see the 4 high peaks, the one small peak, and then the final somewhat high peak on the graph. Here are the other things to look out for:

(1) I had the fan at maximum (80%), to keep the temperature under 70C, which is one key to full boost. The card averaged around 67-68C in this benchmark, and I don't believe it actually went above 69C. This should mean that voltage is maintained at max (1.175v).

(2) The other major limiter on boost is the power limiter. What you can see in the first two tests is that the power limiter is actually exceeded - it actually goes beyond 122% (the maximum), and then voltage is cut. What results is a decrease in core boost by up to 19MHz (I believe). There is nothing I (or anyone) can do to increase the power limiter beyond 122%, as far as I know, but maybe tweaks will come out later.

(3) Given that temps weren't a limiter, and for some reason power didn't exceed 122% in the second two graphics tests or the combined test, the core is at 1260 the entire time.

(4) Also, I've done some more testing on memory overclocks, and as far as I can tell, I get no additional benefits from exceeding 6600. I did this testing in BF3, which is the easiest way I've found to do it (using a singleplayer campaign). The memory keeps going up, but FPS does not (but I haven't seen it go down either), so rather than stress the memory unnecessarily, I'm leaving it at 6600MHz.

 

[toyota]

yes I know it does not stay at 1300 or whatever else its set at for 3DMark as that bench varies the load on your pc.

now you are getting a higher graphics score than I can at those same settings. will you disable HT and run the bench again at those same clocks you just did?

 

[Termie]

yes I know it does not stay at 1300 or whatever else its set at for 3DMark as that bench varies the load on your pc.

now you are getting a higher graphics score than I can at those same settings. will you disable HT and run the bench again at those same clocks you just did?

I'll give it a go this weekend. I know my physics score is increased a lot by HT, but I doubt the graphics score is affected. I guess we'll soon see. Anyway, it seems you're definitely in the ballpark - it could just be my card running at higher boost for a second or two that would lead to a higher score - we're all between 9900 and 10,000.

By the way, Don has Virtu MVP on his Ivy Bridge run, which is known to throw off 3dMark11 scores - in fact it said it right on the score report. That could be the difference right there.

 

[spinejam]

3DMark11 runs w/ higher o/c's than Heaven 3.0. (Asus gtx680 ref.)

Do you guys see the same trend w/ your 670's?

 

[toyota]

I'll give it a go this weekend. I know my physics score is increased a lot by HT, but I doubt the graphics score is affected. I guess we'll soon see. Anyway, it seems you're definitely in the ballpark - it could just be my card running at higher boost for a second or two that would lead to a higher score - we're all between 9900 and 10,000.

By the way, Don has Virtu MVP on his Ivy Bridge run, which is known to throw off 3dMark11 scores - in fact it said it right on the score report. That could be the difference right there.

yeah I am looking at some other gtx670/2500k users and they are getting the same or less than me at the same clocks. as for you, it could simply be HT and your faster memory giving you that 1% higher graphics score.

 

[Termie]

3DMark11 runs w/ higher o/c's than Heaven 3.0. (Asus gtx680 ref.)

Do you guys see the same trend w/ your 670's?

Noticed that on my first 670, which was RMA'd. It would freeze in Heaven at much lower clocks than 3dMark11. Haven't really pushed it on my new card, but it's pretty clear that Heaven is a higher sustained load than 3dMark11 - with 1260 boost, I can't keep temps below 70C in Heaven even with max fan, while I can in 3dMark11.

 

[toyota]

Termie and Don Karnage, were you using stretched(full screen) or centered in 3DMark? I was testing at 1283/7012 and could get 9974 for graphics in centered but then I tried stretched and got 10347 for graphics.

http://3dmark.com/3dm11/3614597

 

[guskline]

Toyota, Termie and Don, are you guys actually playing games on you GTX 670s at these incredibly high settings or just seeing who can get the highest 3D mark11 score?

 

[SickBeast]

Hey Don so what's your GPU situation now? You had a 7850, then I think a 680 and now the 670. What's going to be your ongoing GPU setup? Are you putting them all under water?

 

[cmdrdredd]

Toyota, Termie and Don, are you guys actually playing games on you GTX 670s at these incredibly high settings or just seeing who can get the highest 3D mark11 score?

I'm running my card at 1170/6800 and it will boost to 1250 but never drop below 1230. 24/7

 

[Termie]

Termie and Don Karnage, were you using stretched(full screen) or centered in 3DMark? I was testing at 1283/7012 and could get 9974 for graphics in centered but then I tried stretched and got 10347 for graphics.

http://3dmark.com/3dm11/3614597

I always run stretched - I just figured that would be a more accurate test than windowed, which would have more Aero overhead. Guess you just proved that. Awesome score! You think that memory is scaling now?

Toyota, Termie and Don, are you guys actually playing games on you GTX 670s at these incredibly high settings or just seeing who can get the highest 3D mark11 score?

I actually play at 1245 occasionally, never at 1260, which is what I benched at.

Hey Don so what's your GPU situation now? You had a 7850, then I think a 680 and now the 670. What's going to be your ongoing GPU setup? Are you putting them all under water?

Once he saw how awesome my 670 was he had to get one.

 

[Don Karnage]

3DMark11 runs w/ higher o/c's than Heaven 3.0. (Asus gtx680 ref.)

Do you guys see the same trend w/ your 670's?

Heaven runs with the same exact overclocks as 3dmark

Termie and Don Karnage, were you using stretched(full screen) or centered in 3DMark? I was testing at 1283/7012 and could get 9974 for graphics in centered but then I tried stretched and got 10347 for graphics.

http://3dmark.com/3dm11/3614597

I have always used the default of centered. Stretched gives you higher scores? "facepalm"

Toyota, Termie and Don, are you guys actually playing games on you GTX 670s at these incredibly high settings or just seeing who can get the highest 3D mark11 score?

No we just benchmark. Games are irrelevant

Hey Don so what's your GPU situation now? You had a 7850, then I think a 680 and now the 670. What's going to be your ongoing GPU setup? Are you putting them all under water?

I got a 670 for free from a contact of mine and i just bought another to run SLI. The Asus 680 and 670's i bought i sold without even opening the boxes. My freakin RIVE had issues so i had to send it back to newegg so my plans for TRI are on hold till i decide what i want to do.

 

[Don Karnage]

Highest GPU score the mini me pcb can muster is 10,190

http://3dmark.com/3dm11/3617459

 

[Termie]

Highest GPU score the mini me pcb can muster is 10,190

http://3dmark.com/3dm11/3617459

Offsets?

By the way, have you started benching SLI? My brief tests with SLI showed nearly 100% scaling in 3dMark11.

 

[Don Karnage]

Offsets?

By the way, have you started benching SLI? My brief tests with SLI showed nearly 100% scaling in 3dMark11.

+170 core, +400 Mem, 115% Target Power which translates to 1245 core and 6804 mem. Memory really didn't scale well beyond that and SLI is oh so beautiful especially in BF3. I can't wait to try TRI

 

[Termie]

+170 core, +400 Mem, 115% Target Power which translates to 1245 core and 6804 mem. Memory really didn't scale well beyond that and SLI is oh so beautiful especially in BF3. I can't wait to try TRI

Awesome score with that core speed. So is it true that stretched mode gets higher scores than windowed? Guess I've been cheating this whole time.

 

[Don Karnage]

Awesome score with that core speed. So is it true that stretched mode gets higher scores than windowed? Guess I've been cheating this whole time.

Asshole. Why didn't someone tell me about this sooner?

 

[toyota]

I always run stretched - I just figured that would be a more accurate test than windowed, which would have more Aero overhead. Guess you just proved that. Awesome score! You think that memory is scaling now?



I actually play at 1245 occasionally, never at 1260, which is what I benched at.



Once he saw how awesome my 670 was he had to get one.

default is centered and the other option is stretched to fill the screen but neither of those are windowed.

Awesome score with that core speed. So is it true that stretched mode gets higher scores than windowed? Guess I've been cheating this whole time.

lol, at least that explains what was going on earlier. from now on we should always specify stretched or centered so people dont think something is wrong with their cards. I just happened to remember someone saying stretched would help the score so I tried it and sure enough it does.

 

[Termie]

default is centered and the other option is stretched to fill the screen but neither of those are windowed.

Weird - then I have no idea why stretched would perform differently than centered...well, I'm just glad I taught you both a new trick...

 

[toyota]

Weird - then I have no idea why stretched would perform differently than centered...well, I'm just glad I taught you both a new trick...

because its taking a really widescreen aspect ratio and filling the screen by sort of zooming in. that cuts off a little info on the sides which means slightly less is being rendered.