Overclock.net › Forums › Graphics Cards › NVIDIA › ~~The GTX 670 Overclocking Master-Guide~~
New Posts  All Forums:Forum Nav:

~~The GTX 670 Overclocking Master-Guide~~ - Page 22

post #211 of 3357
Thread Starter 
When people say the voltage is 'locked' they just mean you can't increase it past 1.175v. Increasing the slider definitely has an effect, it seems to force the card to run at the maximum voltage (1.175) when the card is under very heavy load. When the voltage slider is left at default, the voltage just randomly reduces itself to 1.162v and i don't think anyone has really figured out why it does this. By increasing the slider, you can usually net a 10-30MHz increase in overclock. Some people are misinterpreting the locked voltage though. Because the voltage is locked, they go around saying things like, "changing the voltage slider won't do anything because it's locked," which is only half true at best. The voltage slider DOES help stabilize a higher overclock, but it won't let you get past 1.175v.
post #212 of 3357
Quote:
Nope. There might be a way to increase the power limit at some point, but definitely not the voltage without a physical modification of the card. Increasing the power limit without also increasing the max voltage probably won't have a very profound effect. In most cases, the 670's are limited by voltage before the power target.

So +200 is the maximum boost for my card
I was hoping to go a little bit further biggrin.gif

anyways i guess +200 is enough for me
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K ASUS P8P67 Asus GTX670 DC2  G.Skill 8GB (2x4GB) 2133MHz 
Hard DriveOSMonitorKeyboard
For OS: Patriot Pyro SSD 60GB, Storage:2xWD Cav... Windows 7 Ultimate 64-bit Philips 22" Microsoft Sidewinder X4 Keyboard  
PowerCaseMouse
Corsair TX750w Xigmatek Elysium Black Logitech G9x  
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K ASUS P8P67 Asus GTX670 DC2  G.Skill 8GB (2x4GB) 2133MHz 
Hard DriveOSMonitorKeyboard
For OS: Patriot Pyro SSD 60GB, Storage:2xWD Cav... Windows 7 Ultimate 64-bit Philips 22" Microsoft Sidewinder X4 Keyboard  
PowerCaseMouse
Corsair TX750w Xigmatek Elysium Black Logitech G9x  
  hide details  
Reply
post #213 of 3357
iLLLi ... SP answered your first question, the answer to the 2nd is that I've definitely found Heaven to be most effective as a stability tool when run in full screen vs. Windowed. Lots of times I've found that loops complete in windowed (800x600 is what I run) whereas they won't when switched to full screen (1920x1200 for me). Whether that's 'resolution' based, or just full screen vs. windowed based I'm not 100%, but I suspect it's resolution based. It's best to check stability at the highest possible resolution I think.
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
post #214 of 3357
i don't know if i'll get an answer from them, but i think i'll try to contact them and see what they say. its too bad i dont have two monitors with two different max resolutions. that way i could test myself if there is a difference between full screen at different native resolutions vs. windowed at any resolution. i have a un-scientific theory that maybe running at higher native full screen resolutions would push the card harder vs windowed, so it might effect how well the card does. or possibly lower resolutions would inhibit the max o/c since not as much horsepower would be needed for the tests. say a person uses a 22 inch lcd. i wonder if the results would be the same if you used the same card but with a 27inch. or what happens if a person used a 27inch lcd but ran the benchmark windowed or full screen in the resolution of the 22.
i guess i'll just stick to the full screen native res testing for now smile.gif
post #215 of 3357
Quote:
Originally Posted by Gainward View Post

Quote:
Nope. There might be a way to increase the power limit at some point, but definitely not the voltage without a physical modification of the card. Increasing the power limit without also increasing the max voltage probably won't have a very profound effect. In most cases, the 670's are limited by voltage before the power target.
So +200 is the maximum boost for my card
I was hoping to go a little bit further biggrin.gif
anyways i guess +200 is enough for me

+200 is a meaningless value because it means a different clock on everyone's card. What max boost does this offset value actually produce on your card, if you run a full load (like Heaven bench) and observe the clocks in Afterburner or Precision?
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
post #216 of 3357
A great guide,but it actually made me depressed when I saw the results some of you are getting,like those with 1350Mhz frown.gif
Here's my GPU-Z shot:

486

It seems my Kepler boost is very small,since my card's boost can barely hit 1200Mhz under load(the temps are under 70 all the time,ofc)...
I increased the Power Target to 122%(max value) and tried with and without "increasing" the Vcore,but if I push on with my OC,Heaven's rocks start giving me a lightshow.Too bad,was hoping I could at least approach 1250Mhz...
post #217 of 3357
dont feel too bad. 1300+ is pretty rare from what i have read.
1200 range is more common.
post #218 of 3357
Hey - new owner of a MSI 670 gtx here, just threw an Artic Frozr II on in in my new build.

In the midst of OC'ing it, and have a question:

I cannot seem to get the Max recorded power consumption (in GPU-Z) to go above ~110% (or 107% as far as the OSD for Precision X is concerned). I've maxed the power target and Voltage sliders in P-X, and restarted multiple times - same result.

Am I just sitting with an unlucky card that capped my power-percent lower that what I would have wanted?

My voltage monitor shows 1.175 flatline (as one would expect)... So perhaps I'm just screwed. Or, perhaps I'm not screwed: I did buy the 'OC' version of MSI card so it came overclocked to a certain degree... might the reference TDP have been increased on the card, hence reporting a lower 'power percent' even though I'm hitting 1.175?

Current build and results:

2500K @ 4.5 w/1.28v (73 C Max temp)
8GB
ASRock Extreme4 Gen3 on latest BIOS, Intel drivers, etc.
MSI GTX 670 'OC' w/ Frozer Cooler current stable at 1231 gpu load clock (+120) / 3402 memory load clock (+400); load temps 56deg C


Thanks for the advice in advance!

~E~
Edited by Eruethemar - 6/21/12 at 10:15am
post #219 of 3357
Quote:
Originally Posted by Eruethemar View Post

Hey - new owner of a MSI 670 gtx here, just threw an Artic Frozr II on in in my new build.
In the midst of OC'ing it, and have a question:
I cannot seem to get the Max recorded power consumption (in GPU-Z) to go above ~110% (or 107% as far as the OSD for Precision X is concerned). I've maxed the power target and Voltage sliders in P-X, and restarted multiple times - same result.
Am I just sitting with an unlucky card that capped my power-percent lower that what I would have wanted?
My voltage monitor shows 1.175 flatline (as one would expect)... So perhaps I'm just screwed. Or, perhaps I'm not screwed: I did buy the 'OC' version of MSI card so it came overclocked to a certain degree... might the reference TDP have been increased on the card, hence reporting a lower 'power percent' even though I'm hitting 1.175?
Current build and results:
2500K @ 4.5 w/1.28v (73 C Max temp)
8GB
ASRock Extreme4 Gen3 on latest BIOS, Intel drivers, etc.
MSI GTX 670 'OC' w/ Frozer Cooler current stable at 1231 gpu load clock (+120) / 3402 memory load clock (+400); load temps 56deg C
Thanks for the advice in advance!
~E~

Low power consumption is good. It means that you're not going to be throttled from using too much power.
My System
(16 items)
 
  
CPUMotherboardGraphicsRAM
4770K at 4.4 (delidded) Asus Maximus VI Extreme EVGA 980ti SC X2 G.Skill Trident 2400 (2 X 8) 
Hard DriveOptical DriveCoolingOS
Samsung 830 256 GB SSD, Sandisk Extreme Pro 480... LG Blu-Ray Heatkiller CPU Rev4.0 Ni-BL; Aquacomputer Kyrog... Windows 10 Pro 64bit 
MonitorKeyboardPowerCase
Dell U2713HM Filco Majestouch 2 Tenkeyless Corsair AX1200i Case Labs SM8 
MouseMouse PadAudioOther
Zowie ZA12 Artisan Zero Asus Xonar Essence STX APC Smart-UPS SMT1500 
  hide details  
Reply
My System
(16 items)
 
  
CPUMotherboardGraphicsRAM
4770K at 4.4 (delidded) Asus Maximus VI Extreme EVGA 980ti SC X2 G.Skill Trident 2400 (2 X 8) 
Hard DriveOptical DriveCoolingOS
Samsung 830 256 GB SSD, Sandisk Extreme Pro 480... LG Blu-Ray Heatkiller CPU Rev4.0 Ni-BL; Aquacomputer Kyrog... Windows 10 Pro 64bit 
MonitorKeyboardPowerCase
Dell U2713HM Filco Majestouch 2 Tenkeyless Corsair AX1200i Case Labs SM8 
MouseMouse PadAudioOther
Zowie ZA12 Artisan Zero Asus Xonar Essence STX APC Smart-UPS SMT1500 
  hide details  
Reply
post #220 of 3357
Thread Starter 
Quote:
Originally Posted by Eruethemar View Post

Hey - new owner of a MSI 670 gtx here, just threw an Artic Frozr II on in in my new build.
In the midst of OC'ing it, and have a question:
I cannot seem to get the Max recorded power consumption (in GPU-Z) to go above ~110% (or 107% as far as the OSD for Precision X is concerned). I've maxed the power target and Voltage sliders in P-X, and restarted multiple times - same result.
Am I just sitting with an unlucky card that capped my power-percent lower that what I would have wanted?
Two things:
First, Having a low power consumption is a good thing. It means you have a lot more headroom before your card is throttled by TDP.
Second, the power consumption is largely based on what program/app you're using to stress your GPU. I guarantee if you use Furmark you will instantly hit your card's maximum TDP and it will stay there the entire time. So if you really want some piece of mind, and proof that your GPU can go above 110% TDP, then run a minute or two of furmark. Just keep in mind, it will push your TDP so high that the card will be forced to throttle all the way down to the Boost Clock (this is normal with furmark).
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › ~~The GTX 670 Overclocking Master-Guide~~