• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

MSI 290X Lightning problem

Jasseh

Jasseh

Jasseh

Associate
Joined
26 Jul 2012
Posts
1,340
Location
England
I installed my new 290X Lightning in my PC yesterday and everything was fine. I decided to try overclock it today, I went for lower settings than I had read it could do and it passed 3DMark Fire Strike fine but after running the test again the screen went black.

I tried restarting the PC (MSI Afterburner wasn't set to apply the overclock on system startup yet) but it fails to load. Within the BIOS I can see graphic problems that weren't there before and when it attempts to boot into Windows it goes to system repair then freezes.

I've tried switching back to my 7850 and it boots fine so it has to be the card. Is there anything I can do to get this working or am I going to have to RMA it?

Settings I tried were 1150 core/1500 memory. +100 mV on core, +100 mV on memory, +50 mV on Aux. Fans on 45%, max temp was 68.
I had tried 1550 on memory but it caused artifacts in 3DMark so I lowered it.

I'm really hoping there is something I can do to get it working again without having to RMA the card, never spent so much on one thing for my PC before...

Anyone able to help?


Edit: I know it doesn't matter but this is the result of the first run through 3DMark Fire Strike on those settings: http://www.3dmark.com/fs/2000491
 
Last edited:
Marine-RX179 said:
What driver are you using? It might worth sticking with the 13.12 WHQL driver if you were having issue with beta drivers.
Click to expand...

I have 14.3 installed, had no issues other than the artifacts in 3DMark before lowering the memory clock from 1550 to 1500.

I don't think changing driver will help as it shows graphical problems before getting into Windows (BIOS and splash screen) with the 290X in. It won't actually boot all the way to Windows with it in :(.

I should also have mentioned, I was increasing the memory clock by 50 between tests, I didn't jump straight to 1550.
 
I tried switching to LN2 mode, no change.

I just made it back into Windows with the 290X in but using on board graphics. The system repair removed Catalyst Control Center.

I then went back into the BIOS and switched back to using the 290X. Rebooted back into the BIOS and the graphical issues are still there. Tried booting into Windows but it is frozen on the splash screen showing "Gigabyte Insist on Ultra Durable" with a small section of artifacting visible to the left of "Insist".

Edit: Just tried the card in another PCIe slot, same problem :(.

Edit again:
Photo of screen in the BIOS:
9y4b.png
 
Last edited:
Thats memory corruption probably due to damaged trace. Why did you set the fan at 45% when over volting. You should have left left it on Auto.

The card needs a reflow so will have to be RMA'd. I stick em in the oven or hit em with a Heat gun but I can't advise that for our customers.
 
8 Pack said:
Thats memory corruption probably due to damaged trace. Why did you set the fan at 45% when over volting. You should have left left it on Auto.

The card needs a reflow so will have to be RMA'd. I stick em in the oven or hit em with a Heat gun but I can't advise that for our customers.
Click to expand...

I set it to 45% after watching a bench/overclock review on the card that did the same and achieved higher clocks than I was trying for :(.

I don't think I will be trying to overclock it again after RMA which probably means I've now wasted £500 (plus I guess RMA postage costs if I caused the problem) :(.

Thanks for the responses.
 
If your temps were maxing out at 68c then it shouldn't be that, also 8pack, a damaged trace would be a physical fault would it not? So unless OP physically damaged the circuit board again it's just a dodgy card from the off?

Jasseh, if it's faulty and replaced within the first 28 days then OCUK will pay for the return postage, just don't forget to remind them in the customer service section of the forum once the RMA has been completed :)
 
Just to calm you a little.
Its not entirely your fault. The 14.X drivers have issues with the fan control. They didn't show the fan RPM on all the cards, but they are potentially dangerous to Lightnings, because those have dual fan control, the two big, and the little yellow fan are controlled separately, and if you use the 14.X drivers, even if its on auto, it will only control the little fan, the two bigger ones will just spin at 30%. causing your card to overheat.

Also, it could just be that you have a faulty memory chip. It could happen to every card.
 
Mtom said:
Just to calm you a little.
Its not entirely your fault. The 14.X drivers have issues with the fan control. They didn't show the fan RPM on all the cards, but they are potentially dangerous to Lightnings, because those have dual fan control, the two big, and the little yellow fan are controlled separately, and if you use the 14.X drivers, even if its on auto, it will only control the little fan, the two bigger ones will just spin at 30%. causing your card to overheat.
Click to expand...

Ouch, is there a way to get around that?
 
Use none Beta driver.

I reviewed on 13.12 and all was 100% fine. Could be faulty mem chip I agree but more likely it overheated and damaged a trace. RMA is the best way with this one.
 
Use none Beta driver.

I reviewed on 13.12 and all was 100% fine. Could be faulty mem chip I agree but more likely it overheated and damaged a trace. RMA is the best way with this one.
 
Space Monkey said:
If your temps were maxing out at 68c then it shouldn't be that, also 8pack, a damaged trace would be a physical fault would it not? So unless OP physically damaged the circuit board again it's just a dodgy card from the off?

Jasseh, if it's faulty and replaced within the first 28 days then OCUK will pay for the return postage, just don't forget to remind them in the customer service section of the forum once the RMA has been completed :)
Click to expand...

I know there is no chance I've physically damaged any part of the card. Way too valuable to risk that. I even dug out an esd wristband to earth myself.

Mtom said:
Just to calm you a little.
Its not entirely your fault. The 14.X drivers have issues with the fan control. They didn't show the fan RPM on all the cards, but they are potentially dangerous to Lightnings, because those have dual fan control, the two big, and the little yellow fan are controlled separately, and if you use the 14.X drivers, even if its on auto, it will only control the little fan, the two bigger ones will just spin at 30%. causing your card to overheat.

Also, it could just be that you have a faulty memory chip. It could happen to every card.
Click to expand...

I didn't know about that problem with 14.x drivers, wouldn't have used them if I'd known :(. I don't think the card overheated though unless CCC, GPU-Z and MSI Afterburner were all reporting incorrect temps, it didn't go over 68.

My last 3 upgrades (including this one) have all needed RMAs now, I should listen to my partner and stop upgrading my PC...

3TB HDD arrived DoA
Z87X-OC had problems
Now this... :(

Thankfully OCUK have good customer service and dealt with the other 2 RMAs quickly. I hope this one will be quick too.
 
Jasseh said:
I know there is no chance I've physically damaged any part of the card. Way too valuable to risk that. I even dug out an esd wristband to earth myself.



I didn't know about that problem with 14.x drivers, wouldn't have used them if I'd known :(. I don't think the card overheated though unless CCC, GPU-Z and MSI Afterburner were all reporting incorrect temps, it didn't go over 68.

My last 3 upgrades (including this one) have all needed RMAs now, I should listen to my partner and stop upgrading my PC...

3TB HDD arrived DoA
Z87X-OC had problems
Now this... :(

Thankfully OCUK have good customer service and dealt with the other 2 RMAs quickly. I hope this one will be quick too.
Click to expand...

Think of this as your third and final everything from here on in will be wonderful :)
 
Just been looking about and it seems that the 14.3 drivers are Lightning killers. They stop the side fans from spinning up when the card gets hot, thus the cards middle fan can't cope. Loads of reports around the world of the same thing.
 
urg, these beta 14x drivers are proving to be seriously concerning ^ :p bettttaaaaa kills

Ironic that the latest leaked betas was just before my fan stopped and I don't even have a lightning! Could just be nothing :p, back to officials I think
 
So is this just a case of dodgy betas or are we likely to encounter this problem in future official drivers, because if it's likely to appear again in the future, it's going to make updating your drivers a bit like Russian Roulette!
 
You must log in or register to reply here.
Back
Top Bottom