Gigabyte G1 Gaming 980Ti - Half Dead??

[Dumb Dude 24/7]

Hello and thanks in advance for any help/advice given in relation to this problem.

As the tittle suggests, my G1 Gaming 980Ti is suddenly failing when gaming.

The screen turned a solid yellow the first few times it happened and from then on out for the rest of my trial and error diagnosing, the screen turned solid white and both of these solid colour screens also froze the pc.

Just like this short vid on youtube. - Not my video by the way, I found it trying to research my card's problem. - https://www.youtube.com/watch?v=TimLeq1cubI

It does this less than 5 minutes into The Witcher 3 & Mad Max (the most recent games I currently have) and sometimes flickers during those few minutes of gameplay, but it's fine when only web browsing and/or during media play-back and other "lighter" display tasks.

I'm as certain as I can be that it is not an overheating issue as I monitor that closely using GPU-Z. When the card was working perfectly I would regularly alt-tab out of both games during game play to see where the cards thermals were and with the in game graphics set to high for both, I've not really seen it go past 64c (maximum). They were often at mid 50's when doing the random alt-tab out. These temps struck me as great given that a lot of people on a lot of dfferent forums have said that The Witcher 3 is typically a heat enducing game up into the 70's & low 80's.

Speaking of alt-tabbing out to check on GPU-Z, when I used to do that to monitor the temps I also got used to seeing nearly all the sensor charts/graphs (GPU Core Clock, GPU Memory Clock, VDDC and others, but not all) maxxed out all the way and stay there as a steady bar moving along as the sensors refreshed and added to the graph. Since My 980Ti has been messing me around though, when I do the same alt-tab check on GPU-Z in the few minutes of game play that I get, most of those same sensor readings go up and down regularly & in a consistent interval looking like an oscilloscope display.

I have a few GPU-z screenshots that I will try to attach to show what I mean a bit clearer than my tired description. UPDATE - Couldn't find an attachment option to upload the screenshots from my pc to this post or site. Will Look again after I get some sleep - been awake all night.

Also, I have never overclocked this card (I've only grabbed MSI Afterburner since my card has been acting up) + I have all update options set to manual so it's got the exact same drivers and bios/vbios it had while it was still fully working + I have tried using a backup (new) PSU I had and that made NO difference. I am currently using a backup GPU I have and though it's a lesser card (R7 370), it plays those games fine for hours.


Apologies for my tired ramblings if I have given too much detail and thanks again for any help on this subject.

 

[Dumb Dude 24/7]

First and foremost, thank you both for taking the time to respond. I greatly appreciate it.

Given that your temperatures are more than fine and you've tried everything up to and including a different PSU, it's likely on its last legs.

That's unfortunate. I had higher hopes/expectations from choosing a higher tier Gigabyte GPU like the G1 980Ti. They are my first choice for motherboards too.

You didn't mention trying purging your drivers completely with Display Driver Uninstaller and reinstalling the latest ones fresh though, so that might be worth a try.

I haven't tried that yet because I had made zero changes to any drivers or bios' before the problem arose, though in my research I have read that a wipe with DDU then a reinstall of the latest drivers has (rarely) solved similar-ish issues for a lucky few out there.

Otherwise your best bet would be to try underclocking it a bit. The memory first and then the core if necessary. Drop them down quite a bit and see if the issue persists. Sometimes it can help to solve this sort of instability, though it would only be a band aid fix and at the cost of some performance.

I did try underclocking it some after I grabbed Afterburner, sadly it changed nothing.

I'm not sure I'd feel comfortable running this card with a band aid level work-around because I forgot to mention that after this first happened I tried using a different display using the DVI output converted to VGA with an adapter block and when the same solid colour screen error happened on that display, it seems to have killed the backlight for that screen so now that display is dead as well.

That's not all actually. Since this whole GPU failure fiasco started, my pc won't fully shut down anymore. Windows shuts down fine, but the fans stay on and the motherboards bug code display stays at 05 causing me to have to turn it off using the power supply switch on the back - a ROYAL PAIN to say the least.

So somehow this GPU failure has resulted in a couple of other problems, the shut down issue (just a pain in the gluteus) and the more costly issue of killing a display.

I'm assuming it's well out of warranty too?

I'm not sure to be honest. It was a gift from my grand parents and they've both since passed .

I could try to ask around the family for who might have they're paperwork/records (thats if my grand parents kept the receipt/invoice let alone if someone bothered to keep & store any of that sort of stuff), but that seems like a looooooong shot n' a half unfortunately.

If it is a few months out, I still thought better of a product from Gigabyte than for it to brick itself not long after the warranty period and take a monitor with it based off of their good reputation for quality (both from personal experience and from research).

Hello!
I am afraid I think that the GPU is on the way out. This is not a memory failing rather it is the GPU itself. You should try a different output socket from the card to the monitor, but don't hold out too much hope. I don't really think it's that either, but it's worth a shot. I am afraid that GPU's do fail a lot more often than CPU's. They run hot and that generally means a shorter life. You could try using Afterburner to underclock the card to extend it's life a bit but I am afraid that once it starts doing things like this, it invariably gets worse until it dies completely.

Thanks pp111.

I did try a different output as detailed above and it killed the display I tried it with.

My card seemed to run cool as 980Ti's go, so I really can't see it being a temperature problem.

I also tried to underclock the card and nothing changed.

These points have been covered in more detail in my response to Aretak's post ^^^^

 

[Dumb Dude 24/7]

Well..... I spent all weekend asking various relatives if they knew and/or could remember anything about this card and a couple of them recall that my grand parents asked them for advice on graphics cards and my uncle is pretty sure that it was bought from Scan, but neither can remember when for sure (one of them thought it could have been around the start of 2016 and another thought it was mid 2016). I really can't remember when I got it either.

I also couldn't find anything when looking through their records that were kept, only personal & sentimental papers were saved really.

I am really surprised that it has killed a display. I would be contacting Gigabyte, even outside of the warranty period about this. I would mention the circumstances and possibly speak to @GIGA-Man the worst you might hear is a no, but you can always try.

Hi Ezyryder,

I thought about contacting Gigabyte given what happened, but with me not finding the proof of purchase and not even being sure when the card was actually bought and not being able to prove that the display worked perfectly fine before the card messed up....I can only foresee them telling me no, as you say. So I kinda think to myself "whats the point" if I'm being honest.

I would mention the circumstances and possibly speak to @GIGA-Man the worst you might hear is a no, but you can always try.

I'm sorry Ezyryder, but I don't know who that is. I've never had to contact Gigabyte directly for anything before.
In fact I've never had to contact any company regarding a failing item (especially one that took out another device with it), so I don't know how to go about it or which department of Gigabyte I should be trying to contact. Just regular customer service??

 

[Dumb Dude 24/7]

Hay.... It does look like the card GPU is failing, we can check the serial number to see if the card is still under warranty. The SN number is normally on a sticker near the PCIe interface. If the card is still in warranty I can try and get it fixed, issue is I wont be able to replace the card if it can't be fixed, not without proof of purchase anyway.

Hi GIGA-Man,

I just wrote out a pm to provide you with the SN number you requested using the conversation function, but once I was finished it would not allow me to send the message. I don't know what happened.

I have the message copied and saved in a txt document, how can I get the pm to you?

 

[Dumb Dude 24/7]

@GIGA-Man is the Gigabyte service rep on here. I'd try contacting gigabyte on the contact below, explaining it's a gift from a deceased relative. It maybe without receipt of payment they won't do anything though so if you could find proof of postage it would help. Because if you found that you could push for a full replacement under sale of goods act.

https://www.gigabyte.com/uk/Support/Warranty

Hi Andrew,

Thanks for the explination and sorry for the late response, but I have been trying everything I can think of over the last 2 weeks to find/get the proof of purchase so that I actually had some news worth updating this with and so far I've found nothing.

Cheers for the link, I will probably give it a shot with Gigabyte if all else fails.

gigabyte rep is very handy on here and hopefully get some word soon.

with fans not turning off, running latest bios and version/build of windows 10?

Hello orbitalwalsh,

The man does seem to be highly recommended by everyone here, cheers for your recomendation of him.

As for my unusaul shut down issue.... I am running Win 7 and again, I changed nothing on the system prior to card playing up.

I only mentioned this issue due to it happening as soon as the card started freaking out and took a display with it.

I haven't gotten around to dealing with the shut down issue yet as I'd like to deal with the card first. If I can't figure out & resolve the problem, then I definatly will post a new thread in the correct section.

What he said.

Hi Ezyryder,

Thanks for the recomendation. I am trying to contact him as I type.

 

[Dumb Dude 24/7]

I think I have a way to show the GPU-Z & Afterburner graph screenshots now.

https://www.dropbox.com/sh/4ee9d7wrfvnctls/AADdKyyDYU3TNp4QsOutdB23a?dl=0

1 & 2 - show the 980Ti's strange behaviour with 2 different power supllies.

3 - Though this is showing my old card that has fewer sensors, the graphs are more what my 980Ti looked like while it was working perfectly & before this yellow/white screen failure. I never took any GPU-Z screenshots of the 980Ti while it was working normally, so this is the best I can do to show the difference.

4 & 5 - These show the 980Ti's same whacky behaviour as 1 & 2, but using Afterburner.

 

[Dumb Dude 24/7]

Send the info to [email protected]

I have sent a copy to that address. Thanks for your time.

 

[Dumb Dude 24/7]

Got it, checking the details and will get back to you.

Hello GIGA-Man,

I have sent a new reply to the marketing email address with an update.

I don't know if it will change anything, but I have gathered about as much info & supporting proof as I can given the unfortunate circumstances so I can only hope at this point.

Cheers.

 

[Dumb Dude 24/7]

Sorry, I have not been about. I have check your mail and all the details, the card is out of the warranty period so I wont be able to replace it. I can check if we can try and repair but it's probably unlikely based on the symptoms you reported. It does sound like the actual GPU has gone faulty.

I'm really sorry I cannot help more.

Hi GIGA-Man,

No need to apologise for the delay, I'm sure you and your team are kept quite busy.

How do I go about getting the card checked to try for a possible repair as I've got nothing left to lose at this point?

 

[Dumb Dude 24/7]

You can send the card back to us under RMA and I will make sure it doesn't get rejected. Make sure you mention this conversation in the e-mail.

[email protected]

Thank you very much for your time in this, I appreciate it greatly and will start that process over the weekend.

Do I email that address to get further instructions/an address to send it to or do I go off and get an RMA address listed somewhere on Gigabytes website and then notify you using that email address that it's been sent off?

 

[Dumb Dude 24/7]

Yes, e-mail that address and start the process. In the e-mail explain you had been talking to me on the OCUK forums.

Hello GIGA-Man,

Just a quick update.

I have contacted the email address provided, filled out the RMA forms & sent the card off via Royal Mail - Special Delivery on Saturday the 18th so it should be there Monday the 20th.

Thank you for your advice and help with this.

 

[Dumb Dude 24/7]

We have the card and it's undertesting, I've asked the RMA guys to contact you directly with any updates. You should hear something from them very soon, they'll also keep me up dated so I can help if needed.

Thank you very much for your assistance along the way with this card, it is greatly appreciated & makes me glad that I choose Gigabyte for motherboards & graphics cards.