Overclock.net › Forums › Graphics Cards › NVIDIA › [Official] MSI GTX 680 Lightning Owners Club
New Posts  All Forums:Forum Nav:

[Official] MSI GTX 680 Lightning Owners Club - Page 181

post #1801 of 7827
Hi,

I have a MSI GTX680L now (pics later). When I booted first time the fans spin like crazy and loud continuously. I don't remember if it was the nvidia or mainboard driver, but it works normal and silent now, but during the boot it still spins like insane and slows/quiets down during the process of booting. Is there a way to fix it? Perhaps a BIOS setting. I don' have time to figure it out myself at the moment, so I just ask away, if you don't mind.

Nice card, I only wish the leds were green (nvidia color, as the packaging), but blue is nice, too. Can I put them off at all?
post #1802 of 7827
Have you maybe had a look at the litre of thermal paste MSI has used on your lightnings xoleras? A few people at OCAU have re-pasted and gotten -8 degrees on MSI's effort.

Might get you off that clock wall?

My 24/7 clocks are 1310mhz core on normal bios ~ 1.2v, been 15+ hours stable on BF3 MP.
post #1803 of 7827
Quote:
Originally Posted by resis View Post

Hi,
I have a MSI GTX680L now (pics later). When I booted first time the fans spin like crazy and loud continuously. I don't remember if it was the nvidia or mainboard driver, but it works normal and silent now, but during the boot it still spins like insane and slows/quiets down during the process of booting. Is there a way to fix it? Perhaps a BIOS setting. I don' have time to figure it out myself at the moment, so I just ask away, if you don't mind.
Nice card, I only wish the leds were green (nvidia color, as the packaging), but blue is nice, too. Can I put them off at all?

You can take the GPU Reactor cap off, and the little board underneath it, that's about it. And the fan speed at start-up is an anti-dust technique built into the BIOS, can be gotten rid of by flashing to the...3A BIOS is it? Someone else can jump in on that.

Quote:
Originally Posted by GenoOCAU View Post

Have you maybe had a look at the litre of thermal paste MSI has used on your lightnings xoleras? A few people at OCAU have re-pasted and gotten -8 degrees on MSI's effort.
Might get you off that clock wall?
My 24/7 clocks are 1310mhz core on normal bios ~ 1.2v, been 15+ hours stable on BF3 MP.

1310 on normal BIOS? Damn. You have to try the edit on the LN2 BIOS tongue.gif I tried the non-LN2 BIOS and got crazy downclocking at anything over 1300mhz.
Quote:
Originally Posted by xoleras View Post

DPH, My instability temperature threshold has been low on the lightnings, when my core temp is 70C and i'm overvolting a lot I can feel the side of my card and it is hot. I mean..HOT. Like, my top card in SLI it hurts to touch the heatpipe sometimes. The cooler does great on dispersing heat from the core and VRM but there must be other temp sensitive components i'm thinking. Thats the only explanation as to my results. I can get 1402mhz @ +100mV but maxing everything out will make it unstable at the same settings. I absolutely cannot go in and max out every voltage slider in existence, that ruins my overclocks and honestly I don't think its the best approach to air OC'ing. If anyone else has other opinions on this, please share.
I really think there are more components that suffer when volts get high, although I can't say for sure. I do know that overvolting too much on air will always lessen stability for me, I really think past 1.3V is not meant for air and won't help things on air..... I really feel like more is not always better, you have to find a happy medium.
I don't know, thats what my findings have been. If anyone else has any overclocking particulars or oddities they've found, please share by all means thumb.gif Keep in mind that this is in the context of SLI. Temps are never ever an issue with single card with the 2nd card removed. But in SLI temps are much more sensitive for GPU1. (GPU2 blows hot air directly onto GPU1)
If any other lightning SLI users can share your results and thoughts, i'd love to hear em!

Always appreciate your input thumb.gif +rep

Well, memory lowers my core OC. You mean that with the core voltage at +100mv and the clock at 1402mhz, you can raise the Memory and PLL voltages while leaving the memory clock the same speed and you'll crash? That would mean it's the temp I would assume. If so, what's the temp increase that makes you crash at? Like what's the temp threshold for your 1402mhz OC? Just curious. Trying to figure these out. Because my 570s would not crash no matter what the temp was. 10C or more would make no difference on stability. I know I have to stop comparing to another generation, but it's all I have to go on experience-wise, so if the 600-series has major differences then I have to figure them out.. But yeah if a voltage/clock combo was stable on them, then it was stable no matter what the temp. In the summer before I got my AC, temps would hit high 80Cs and not crash. After the AC, I tried higher OCs on the same volts and it made zero difference.

But yeah, thanks again for the observations thumb.gif
My Only Hobby
(14 items)
 
  
CPUMotherboardGraphicsRAM
i7 3930k Rampage IV Extreme EVGA 980Ti SC 16GB Samsung 30nm 
Hard DriveCoolingOSMonitor
Seagate 3TB 7200rpm H100 Windows 8.1 Acer XB240H 
KeyboardPowerCaseMouse
Razer Lycosa Cooler Master 1200w Silent Pro Gold Cooler Master HAF X Razer DeathAdder 
  hide details  
Reply
My Only Hobby
(14 items)
 
  
CPUMotherboardGraphicsRAM
i7 3930k Rampage IV Extreme EVGA 980Ti SC 16GB Samsung 30nm 
Hard DriveCoolingOSMonitor
Seagate 3TB 7200rpm H100 Windows 8.1 Acer XB240H 
KeyboardPowerCaseMouse
Razer Lycosa Cooler Master 1200w Silent Pro Gold Cooler Master HAF X Razer DeathAdder 
  hide details  
Reply
post #1804 of 7827
Quote:
Originally Posted by dph314 
1310 on normal BIOS? Damn. You have to try the edit on the LN2 BIOS I tried the non-LN2 BIOS and got crazy downclocking at anything over 1300mhz.

I definitely plan to and have a massive text document with all steps! biggrin.gif I couldnt get beyond 1453mhz on LN2 bios with +93 mV (1.36v multi) on air in 3d mark11. Only pulled off those benches due to 12degree ambient, so im pretty much waiting for either the waterblocks to get cannoed out to Australia or another freezing morning to start hexxing.
post #1805 of 7827
Quote:
Originally Posted by GenoOCAU View Post

Have you maybe had a look at the litre of thermal paste MSI has used on your lightnings xoleras? A few people at OCAU have re-pasted and gotten -8 degrees on MSI's effort.
Might get you off that clock wall?
My 24/7 clocks are 1310mhz core on normal bios ~ 1.2v, been 15+ hours stable on BF3 MP.

Temps are not an issue for me really, like I said this is in the context of SLI, GPU2 blows hot air onto GPU1 and makes GPU1 very very sensitive to temps. This is why SLI overclocks are generally m uch worse than single card, in addition to being harder to sync clocks/voltages.

I'm almost 100% sure I can do higher than 1400mhz 24/7 if I so wanted on either of my cards, but GPU2 definitely makes GPU1 very hot to the touch - since GPU2 blows hot air onto GPU1 in SLI.
post #1806 of 7827
Quote:
Originally Posted by xoleras 
Temps are not an issue for me really, like I said this is in the context of SLI, GPU2 blows hot air onto GPU1 and makes GPU1 very very sensitive to temps. This is why SLI overclocks are generally m uch worse than single card, in addition to being harder to sync clocks/voltages.

I'm almost 100% sure I can do higher than 1400mhz 24/7 if I so wanted on either of my cards, but GPU2 definitely makes GPU1 very hot to the touch - since GPU2 blows hot air onto GPU1 in SLI.

Oh fair enough, I guess the best longterm option for high overclocked 24/7 sli would be to find a different board that you can split them further apart. Either that or deal with it I guess.
Oh FYI been playing with undervolting (does actually work @ the multimeter), played for about an hour and a half of MP BF3 on 1300mhz @ -18mV all fine so far. Under volting like over volting (goes up or down in increments of 6) also affects idle volts, currently sitting @ .977 in windows.
Edited by GenoOCAU - 7/30/12 at 7:36pm
post #1807 of 7827
Quote:
Originally Posted by Valenz View Post

.
I really hope they release an extreme because I have a new catleap inc and will need more then 2 gigs to push it. galaxy is my only option being I don't support EVGA anymore. thumb.gif


Hi, I'm new here and I own a msi L. Overclocking it to 1350 / 1386 works on all benches and completes on AFB 2.2.3.
But when playing BF3 I have to LOWER voltages for stability on a 1350 clock and crysis 2 no stability not even on standard clocks. I get lock ups after 5 min.
Tried both bios F8 and 3A and voltages are working. I'm beginning to wonder if the crashes are related to the Catleap monitor since I'm running 2560x1440 res and this card begins to fail on memory
at that point on crysis2 fully loaded on ULTRA settings and extreme TESS?
post #1808 of 7827
Quote:
Originally Posted by Shempio 
crysis 2 no stability not even on standard clocks. I get lock ups after 5 min

Settings in Afterburner then goto monitoring and add 'Memory Usage' to OSD, should be able to monitor (no pun intended) if 1440dpi is stealing all 2gb. Shouldn't be though, if anything you should get lower min fps and stutters as opposed to no stability at standard clocks/crash.
post #1809 of 7827
I have a Catleap, but not those games. I'd back down on the clocks if it's crashing and try that. I did a run of maxed out Heaven at 2560x1440 82Hz and was generally in the 1.4GB mem usage, but had one spike of 1.890.
X-Rig
(14 items)
 
Death Bomb
(9 items)
 
 
CPUMotherboardGraphicsGraphics
I7-5960X Asus Rampage V Extreme 1502 EVGA Titan X EVGA Titan X 
RAMHard DriveHard DriveOptical Drive
G.Skill F4-3000C15Q-32GRK Samsung SM951 M.2 Samsung 850 Pro 512GB Pioneer BDR-2209 
CoolingOSMonitorKeyboard
EKWB Blocks, Fans, and Pumps, HW Labs Black Ice... Windows 10 Pro Retail BenQ BL3201PH Max Keyboard Nighthawk X8 
PowerMouse
EVGA 220-P2-1200-X1 Roccat Kone Pure Optical 
CPUMotherboardGraphicsRAM
I7-7700K De-lidded @ 5GHz Asus Strix Z270i Nvidia Titan Xp G. Skill F4-3600C15D-16GTZ 
Hard DriveCoolingOSPower
2X Samsung 960 Pro 512GB M.2 EK Supremacy EVO CPU. Titan Xp Full Cover w/Ba... Windows 10 Pro 64 Bit SeaSonic X750 
Case
Fractal Nano S 
  hide details  
Reply
X-Rig
(14 items)
 
Death Bomb
(9 items)
 
 
CPUMotherboardGraphicsGraphics
I7-5960X Asus Rampage V Extreme 1502 EVGA Titan X EVGA Titan X 
RAMHard DriveHard DriveOptical Drive
G.Skill F4-3000C15Q-32GRK Samsung SM951 M.2 Samsung 850 Pro 512GB Pioneer BDR-2209 
CoolingOSMonitorKeyboard
EKWB Blocks, Fans, and Pumps, HW Labs Black Ice... Windows 10 Pro Retail BenQ BL3201PH Max Keyboard Nighthawk X8 
PowerMouse
EVGA 220-P2-1200-X1 Roccat Kone Pure Optical 
CPUMotherboardGraphicsRAM
I7-7700K De-lidded @ 5GHz Asus Strix Z270i Nvidia Titan Xp G. Skill F4-3600C15D-16GTZ 
Hard DriveCoolingOSPower
2X Samsung 960 Pro 512GB M.2 EK Supremacy EVO CPU. Titan Xp Full Cover w/Ba... Windows 10 Pro 64 Bit SeaSonic X750 
Case
Fractal Nano S 
  hide details  
Reply
post #1810 of 7827
Quote:
Originally Posted by GenoOCAU View Post

Settings in Afterburner then goto monitoring and add 'Memory Usage' to OSD, should be able to monitor (no pun intended) if 1440dpi is stealing all 2gb. Shouldn't be though, if anything you should get lower min fps and stutters as opposed to no stability at standard clocks/crash.


just checked my log on memory usage in Crysis 2 and it exceeded 2000 to 2017 and then a minute later the game crashed even at stock settings.

BF3 maxed out doesn't even come close in mem usage according to my log files, 1700 is tops.

2 gig seems like NOT enough mem on a 2560x1440 monitor for crysis 2 maxed out.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › [Official] MSI GTX 680 Lightning Owners Club