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 645

post #6441 of 7827

Thank you! i will be adding that soon. I'm using red monsoon fittings throughout. Should i use bitspower blacks to connect on gpu for a all round black look?

thinking 2 of These:

http://www.frozencpu.com/products/10037/ex-tub-606/Bitspower_Ultimate_G_14_Thread_12_ID_x_34_OD_Compression_Fitting_-_Matte_Black_BP-MBCPF-CC5.html?tl=g30c409s1609

Thoughts? or should i just use my red fittings? check build log for pictures
Core X9
(16 items)
 
  
CPUMotherboardGraphicsRAM
3930k ASUS RAMPAGE IV EXTREME GTX980 Gaming G.Skill F3-2400C10D-16GTX 
Hard DriveHard DriveHard DriveCooling
Western Digital Caviar Black Samsung 850Evo Western Digital Green Custom Loop 
MonitorKeyboardPowerCase
QNIX 2710 cooler master quickfire pro EVGA Supernova G2 1300W Thermaltake Core X9 
MouseMouse PadAudioAudio
G502 qck heavy Asus STX DT 990 
  hide details  
Reply
Core X9
(16 items)
 
  
CPUMotherboardGraphicsRAM
3930k ASUS RAMPAGE IV EXTREME GTX980 Gaming G.Skill F3-2400C10D-16GTX 
Hard DriveHard DriveHard DriveCooling
Western Digital Caviar Black Samsung 850Evo Western Digital Green Custom Loop 
MonitorKeyboardPowerCase
QNIX 2710 cooler master quickfire pro EVGA Supernova G2 1300W Thermaltake Core X9 
MouseMouse PadAudioAudio
G502 qck heavy Asus STX DT 990 
  hide details  
Reply
post #6442 of 7827
Quote:
Originally Posted by shakurass View Post


With that profile config I still seem to be locked to +100 mV. It would explain a lot if the core voltages I am choosing aren't being applied, is there a way for me to check this? With the Aux power set to -100 mV I am able to get the Heaven benchmark to run at +150 MHz on the core with no CTD however there are frequent graphical glitches.

AFAIK, +100mv is the limit with AB. Graphical glitches mean that your OC is not actually stable at that voltage. You should back off a bit. You can recover some performance by OCing the memory. Try +120 or +130 and run the mem at +500, and see how that goes. If you get a boost freq of 1300 or a bit more, that's actually very good. Keep it in perspective; the reference card is 1006 base and 1058 boost.

Your voltage is being applied if you have the AB profile edit in place. +100mv just isn't enough to make your chosen core OC stable.

If you want to play with heavy-duty voltage -- which you might if you are water cooling -- then search this thread for "artmoney hack". With the stock air cooler, I wouldn't touch it. But if I had water cooling, I would definitely be tempted.
post #6443 of 7827
Quote:
Originally Posted by SeekerZA View Post

Thank you! i will be adding that soon. I'm using red monsoon fittings throughout. Should i use bitspower blacks to connect on gpu for a all round black look?

thinking 2 of These:

http://www.frozencpu.com/products/10037/ex-tub-606/Bitspower_Ultimate_G_14_Thread_12_ID_x_34_OD_Compression_Fitting_-_Matte_Black_BP-MBCPF-CC5.html?tl=g30c409s1609

Thoughts? or should i just use my red fittings? check build log for pictures

Yes, those would look nice. I use the same black fittings from them too, just rotary.

Your old ones look nice too, it's all preference what you want.
    
CPUMotherboardGraphicsRAM
8700k Apex X TITAN X (Pascal)  Trident Z 4400MHz 
Hard DriveCoolingCoolingOS
750 NVMe 1.2TB EK HF/rx360/rx120 EK FC/GTs/mcp655 Windows 10 Pro  
MonitorPowerCaseAudio
BL3200PT 1600 T2 800d Pro Audio 2.1 + DragonFly Red 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
8700k Apex X TITAN X (Pascal)  Trident Z 4400MHz 
Hard DriveCoolingCoolingOS
750 NVMe 1.2TB EK HF/rx360/rx120 EK FC/GTs/mcp655 Windows 10 Pro  
MonitorPowerCaseAudio
BL3200PT 1600 T2 800d Pro Audio 2.1 + DragonFly Red 
  hide details  
Reply
post #6444 of 7827
Hi

I'm using the unlocked 3A bios on my 680 lightning, and have the core voltage set to +81 in afterburner 2.2.3

However, when afterburner is reset, the core voltage drops by about 10 each time

So, for eg, set to +81 now, close afterburner then open, it has changed to +72, then the next time +61 etc

Sorry for being newb, but why is this?
post #6445 of 7827
Quote:
Originally Posted by RoflcopterZA View Post

Hi

I'm using the unlocked 3A bios on my 680 lightning, and have the core voltage set to +81 in afterburner 2.2.3

However, when afterburner is reset, the core voltage drops by about 10 each time

So, for eg, set to +81 now, close afterburner then open, it has changed to +72, then the next time +61 etc

Sorry for being newb, but why is this?

Everyone has that problem with AB 2.2.3. Try 2.3.1 or later and apply the profile edit mentioned in my post above (post #6438) and on page one of this thread.
Edited by wiredg - 3/24/13 at 12:08pm
post #6446 of 7827
Quote:
Originally Posted by MrTOOSHORT View Post

Yes, those would look nice. I use the same black fittings from them too, just rotary.

Your old ones look nice too, it's all preference what you want.

I was thinking about the roratry's actually. It would make installation on that primochill advance tubing much more easier since that specific tubing is slightly larger than average. Would you Claim Leak Proof ? And confident enough to install all over? Even though i'm just looking for two to use on gpu. Can you post pic of the fitting by gpu? I would love to see what effect the extra height difference looks like compared to the standard non rotary fitting with tubing
Core X9
(16 items)
 
  
CPUMotherboardGraphicsRAM
3930k ASUS RAMPAGE IV EXTREME GTX980 Gaming G.Skill F3-2400C10D-16GTX 
Hard DriveHard DriveHard DriveCooling
Western Digital Caviar Black Samsung 850Evo Western Digital Green Custom Loop 
MonitorKeyboardPowerCase
QNIX 2710 cooler master quickfire pro EVGA Supernova G2 1300W Thermaltake Core X9 
MouseMouse PadAudioAudio
G502 qck heavy Asus STX DT 990 
  hide details  
Reply
Core X9
(16 items)
 
  
CPUMotherboardGraphicsRAM
3930k ASUS RAMPAGE IV EXTREME GTX980 Gaming G.Skill F3-2400C10D-16GTX 
Hard DriveHard DriveHard DriveCooling
Western Digital Caviar Black Samsung 850Evo Western Digital Green Custom Loop 
MonitorKeyboardPowerCase
QNIX 2710 cooler master quickfire pro EVGA Supernova G2 1300W Thermaltake Core X9 
MouseMouse PadAudioAudio
G502 qck heavy Asus STX DT 990 
  hide details  
Reply
post #6447 of 7827
Quote:
Originally Posted by gl0ry View Post

Sounds about right. That's the same type of OC I get. Just know that with those beta drivers your clock speeds will be buggy.

http://www.youtube.com/watch?v=_tgT5TXvDAk

Not sure if nvidia intended for the drivers to do that, but both beta 314.xx drivers do that.

Yeah mine does the same thing however once you reboot its fine and stays that was once you leave AB untouched. This is as high as I could go with a CX600 Power Supply and I refuse to buy another one since this one is pretty good managing this overclock. Right now I've got 1293 on the core and 3557 on memory. I don't want to spend on a better power supply that might not yield better clock speeds...Right now this thing murders anything I throw at it. Go Lighting!!
post #6448 of 7827
Quote:
Originally Posted by wiredg View Post

AFAIK, +100mv is the limit with AB. Graphical glitches mean that your OC is not actually stable at that voltage. You should back off a bit. You can recover some performance by OCing the memory. Try +120 or +130 and run the mem at +500, and see how that goes. If you get a boost freq of 1300 or a bit more, that's actually very good. Keep it in perspective; the reference card is 1006 base and 1058 boost.

Your voltage is being applied if you have the AB profile edit in place. +100mv just isn't enough to make your chosen core OC stable.

If you want to play with heavy-duty voltage -- which you might if you are water cooling -- then search this thread for "artmoney hack". With the stock air cooler, I wouldn't touch it. But if I had water cooling, I would definitely be tempted.

Any time I even touch the memory clock it results in things like this happening:



Point of interest: with memory clock at +0 I have been able to get the benchmarks to run no issues with:

Core Clock: +130
Core Voltage: +100
Aux Voltage: -100
Power Limit: 133
Memory Clock: +0
Big-Beetre
(14 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 3770K Asus P8Z77-V DELUXE MSI GTX 680 Lightning Corsair Vengeance LP 16GB DDR3 2133MHz CL11 Qua... 
Hard DriveHard DriveCoolingOS
Samsung 840 PRO Seagate Barracuda Custom Loop Windows 8.1 Pro 
KeyboardPowerCaseMouse
Ducky Shine 3 DK9008 Corsair AX760i Corsair Obsidian 750D Razer DeathAdder 2013 
Mouse PadAudio
SteelSeries QcK Logitech Z-5500 
  hide details  
Reply
Big-Beetre
(14 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 3770K Asus P8Z77-V DELUXE MSI GTX 680 Lightning Corsair Vengeance LP 16GB DDR3 2133MHz CL11 Qua... 
Hard DriveHard DriveCoolingOS
Samsung 840 PRO Seagate Barracuda Custom Loop Windows 8.1 Pro 
KeyboardPowerCaseMouse
Ducky Shine 3 DK9008 Corsair AX760i Corsair Obsidian 750D Razer DeathAdder 2013 
Mouse PadAudio
SteelSeries QcK Logitech Z-5500 
  hide details  
Reply
post #6449 of 7827
Quote:
Originally Posted by shakurass View Post

Any time I even touch the memory clock it results in things like this happening:



Point of interest: with memory clock at +0 I have been able to get the benchmarks to run no issues with:

Core Clock: +130
Core Voltage: +100
Aux Voltage: -100
Power Limit: 133
Memory Clock: +0

I am familiar with that type of artifact. On my system, it's always caused by too little core voltage for the chosen core freq, or too much core freq for the chosen core voltage (same/same). I get flashing coloured shiny blobs when I push the mem too hard. On air, I've been able to get 1333 boost (+136 core) and 3843 mem (+840 mem) using +100 mv core and the stock mem voltage in Heaven, which I had considered stable, but I get the shiny blobs in Crysis 3, so I no longer consider it stable. Perhaps I just got stupidly lucky with the mem controller or sdram chips in my card. I have no problem with +700 mem in any app using stock mem voltage, or +600 using -50mv mem voltage, which I tend to use now. So I'm kind of prejudiced (or spoilt).

I have no idea why your card is so sensitive to mem OC. My first Lightning was a dud; it crashed on stock core freqs and I returned it. But the mem on that one OC'd decently too (+400 or so if I recall), so long as I underclocked or overvolted the gpu.

Most users here report that Lightning mem OC is good, maybe not as lucky as me with this current card, but all around quite good. Still, you have got a decent core OC. I just can't recall anyone saying that they couldn't OC the memory at all (correct me if I'm mistaken, folks). Unfortunately, I don't think being unable to overclock anything is grounds for RMA.

Now, if the stock mem freq were to cause artifacts at stock core freqs, you'd have a case. Do you think you could make it do that? Maybe on air? smile.gif Heh, OK, I'm slightly evil. But Hell, you paid a lot for your Lightning, and it is marketed as the best 680 out there (which it basically is). You deserve more than the stock mem freq for that cash outlay, imho. MSI would disagree with me here, of course. Overclocking is not guaranteed. But it certainly is implied in the marketing hype.
post #6450 of 7827
So I got my 680 Lightning yesterday in hopes to overclock it at least above 1346Mhz, but the best it does it 1333Mhz and that also after a lot of voltage pumped. Crashes each time on 1346 even at 1.306V. And all the time the temps were touching 60C. And this was all under water.

My other reference 680 does 1333 @ 1.212V without crashing. Have even benched once at 1346Mhz and temps never broke 43C.

Planning to get rid of this lightning.
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