[SOLVED] RTX 3070 Ti not detected in BIOS ?

[BooTheFearless]

Hey all,

So, I just recently purchased a Gigabyte RTX 3070 Ti Aorus. I installed it yesterday evening and it was working fine. I went to bed before my wife, and she woke me up saying that the screen just went black. The monitor was reporting "no DP signal available".

Naturally, I check the physical connection of all cables etc. etc. Come to find out, the video card is no longer being recognized in the BIOS. The only video signal we are able to get is from the motherboard port. Now, I've tried a few things: reseating the card, resetting the BIOS, checking for BIOS updates (none available), using two single PCIe 8-pin power cables rather than a single cable w/ a split end, finagling w/ drivers (to no avail because the video card isn't being picked up in the BIOS) etc.

My PSU is a 750w so perhaps it's possible enough power isn't being delivered to the card ? What I don't understand though, is that the card worked perfectly fine for at least 4 hours before no longer being detected by the BIOS. I also forgot to mention that the card is lighting up w/ the fans rotating etc. Any suggestions?



SPECS:

Intel i7 7770k (kaby lake)

750w PSU

Asus Tuf z270 mk.2 mobo

 

[Lutfij]

Welcome to the forums, newcomer!

We're going to need a little more information. 750W is the wattage advertised on the PSU, you will need to mention the make and model of the unit and it's age. As for your specs, please list them like so:
CPU:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:

For the sake of relevance, can you specify the BIOS version at the time of writing?

 

[Feingh69]

Hello,

I have the same exact card and I am having the same exact problem. However, you got further than I ever did. Mine never came up and never registered it was even there in the BIOS. Lights come on , LCD screen works, and fans spin. I got two of these through the Newegg shuffle and both do exactly the same thing. Updated BIOS, reset CMOS, tried all 3 PCIe slots. Can't install drivers because it isn't "there". I wondered how I was so lucky to win back to back. Now I know why. Looks like they are dumping non working cards. I am RMA both cards as we speak. Don't waste your time return it. I spent 3 days trying to get either one to work.

 

[BooTheFearless]

Welcome to the forums, newcomer!

We're going to need a little more information. 750W is the wattage advertised on the PSU, you will need to mention the make and model of the unit and it's age.

For the sake of relevance, can you specify the BIOS version at the time of writing?

Apologies for the late reply! And thank you for the nice welcome And, apologies for the lack of information; I've posted this below

CPU: Intel Core i7-7700K 4.2 GHz Quad-Core Processor
Motherboard: Asus TUF Z270 MARK 2 ATX LGA1151 Motherboard
Ram: Corsair Vengeance LPX 16 GB (2 x 8 GB) DDR4-3200 CL16 Memory & G.Skill Ripjaws V Series 16 GB (2 x 8 GB) DDR4-3200 CL16 Memory
SSD/HDD: Samsung 970 Evo Plus 1 TB M.2-2280 NVME Solid State Drive
GPU: Gigabyte GeForce RTX 3070 8 GB AORUS MASTER Video Card (also have GIGABYTE GTX 1070)
PSU: Thermaltake Toughpower Grand RGB Sync Edition 750 W 80+ Gold Certified Fully Modular ATX Power Supply (PSU is ~2.5 years old)
Chassis: Thermaltake mid-ATX tower
OS: Windows 10 Home 64bit
Monitor: Dell S3219D 32.0" 2560x1440 60 Hz Monitor

For an update on my issue, I eventually RMAd the card back to the manufacturer; turns out the GPU chip itself needed to be replaced (according to GIGABYTE). However, after getting it back yesterday, I've found that the card's core clock speed won't go beyond 240MHz; even while under load with userbenchmark running it doesn't surpass this speed. I should note, as well, that the temps aren't exorbitantly high; I haven't seen the temperature go above 52C with all of the benchmarks that I've ran on the card. Even setting the TDP percentage to 111% in AORUS engine still provides a "power limit reached" message in GPU Shark. I used AORUS Engine to OC the core clock to 1200MHz but obviously the card's base clock is 1500MHz so it should be much faster than this.

GPU-Z reports a performance cap reason as Power; I've also ran GPU shark which reports, while idling, that the card reaches the power limit policy at 68% TDP. Is it possible that my PSU, as originally stated, is insufficient in providing power to the RTX 3070? It provides more than enough power to my older GTX 1070 card and it doesn't have these issues. I'll attach some screenshots later today (I'm posting on my work computer at the moment). I've purchased an 850w and 1000w PSU from amazon; both should be here tomorrow. I'll update my status then as well to see if the PSU was the source of the issue.

One last thing, I did run DDU twice to clean any drivers on my system to ensure that there aren't any driver issues causing the problem. I also installed the Nvidia driver linked on the RTX 3070's manufacturer page to ensure that perhaps it wasn't some odd issue with a new driver causing the card to not go beyond it's idle clock.

 

[BooTheFearless]

Hello,

I have the same exact card and I am having the same exact problem. However, you got further than I ever did. Mine never came up and never registered it was even there in the BIOS. Lights come on , LCD screen works, and fans spin. I got two of these through the Newegg shuffle and both do exactly the same thing. Updated BIOS, reset CMOS, tried all 3 PCIe slots. Can't install drivers because it isn't "there". I wondered how I was so lucky to win back to back. Now I know why. Looks like they are dumping non working cards. I am RMA both cards as we speak. Don't waste your time return it. I spent 3 days trying to get either one to work.

Did you receive both cards back from the manufacturer? As stated in my update above, I got the card back but I'm still having issues; GIGABYTE replaced the GPU chip in the card which is fine. It's detected no issues by my system now. Though, I'm having the previously described issue. At least we aren't the only ones with issues haha; hopfully you'll find yourself a working card. I'll likely end up doing the same if trying a new PSU doesn't help.

 

[RodroX]

Apologies for the late reply! And thank you for the nice welcome And, apologies for the lack of information; I've posted this below

CPU: Intel Core i7-7700K 4.2 GHz Quad-Core Processor
Motherboard: Asus TUF Z270 MARK 2 ATX LGA1151 Motherboard
Ram: Corsair Vengeance LPX 16 GB (2 x 8 GB) DDR4-3200 CL16 Memory & G.Skill Ripjaws V Series 16 GB (2 x 8 GB) DDR4-3200 CL16 Memory
SSD/HDD: Samsung 970 Evo Plus 1 TB M.2-2280 NVME Solid State Drive
GPU: Gigabyte GeForce RTX 3070 8 GB AORUS MASTER Video Card (also have GIGABYTE GTX 1070)
PSU: Thermaltake Toughpower Grand RGB Sync Edition 750 W 80+ Gold Certified Fully Modular ATX Power Supply (PSU is ~2.5 years old)
Chassis: Thermaltake mid-ATX tower
OS: Windows 10 Home 64bit
Monitor: Dell S3219D 32.0" 2560x1440 60 Hz Monitor

For an update on my issue, I eventually RMAd the card back to the manufacturer; turns out the GPU chip itself needed to be replaced (according to GIGABYTE). However, after getting it back yesterday, I've found that the card's core clock speed won't go beyond 240MHz; even while under load with userbenchmark running it doesn't surpass this speed. I should note, as well, that the temps aren't exorbitantly high; I haven't seen the temperature go above 52C with all of the benchmarks that I've ran on the card. Even setting the TDP percentage to 111% in AORUS engine still provides a "power limit reached" message in GPU Shark. I used AORUS Engine to OC the core clock to 1200MHz but obviously the card's base clock is 1500MHz so it should be much faster than this.

GPU-Z reports a performance cap reason as Power; I've also ran GPU shark which reports, while idling, that the card reaches the power limit policy at 68% TDP. Is it possible that my PSU, as originally stated, is insufficient in providing power to the RTX 3070? It provides more than enough power to my older GTX 1070 card and it doesn't have these issues. I'll attach some screenshots later today (I'm posting on my work computer at the moment). I've purchased an 850w and 1000w PSU from amazon; both should be here tomorrow. I'll update my status then as well to see if the PSU was the source of the issue.

One last thing, I did run DDU twice to clean any drivers on my system to ensure that there aren't any driver issues causing the problem. I also installed the Nvidia driver linked on the RTX 3070's manufacturer page to ensure that perhaps it wasn't some odd issue with a new driver causing the card to not go beyond it's idle clock.

Thats a decent PSU.

So the GPU is only boosting to 240MHz?

Can you post screenshot of GPU-Z and GPU Shark?

Also its the RTX 3070 then, not the TI?

Does the card have a BIOS selection switch ?

Im guessing your old GTX 1070 works without issues right now?

 

[BooTheFearless]

Thats a decent PSU.

So the GPU is only boosting to 240MHz?

Can you post screenshot of GPU-Z and GPU Shark?

Also its the RTX 3070 then, not the TI?

Does the card have a BIOS selection switch ?

Im guessing your old GTX 1070 works without issues right now?

I can answer some of those questions; I'll have to grab the screenshots later today when I'm on break. Posting from my work computer atm.

So the GPU is only boosting to 240MHz? Essentially, yes. The GPU sits at 240MHz even under load while running benchmark programs like userbenchmark (I ran furmark once or twice but have seen many people saying to avoid it so I haven't ran it beyond that). As mentioned earlier, temps don't go above ~50C during these benchmarks and the fans run w/out issue.

Also its the RTX 3070 then, not the TI? Correct, it's the RTX 3070 and not the Ti

Does the card have a BIOS selection switch ? Yes, it does have a switch on it; though I think it controls the fan speed. It has "BIOS OC" and "Manual" iirc.

Im guessing your old GTX 1070 works without issues right now? Yessir; 1070 works like a charm and I've never encountered any issues with it.

 

[Phaaze88]

It sounds like Gigabyte sent you back a damaged card.
I've read that the gpu does that if there's a short somewhere, or a bent/broken PCIE port. It refuses to boost to protect itself.
Doesn't help you much, since you can't really use it in that state.

 

[RodroX]

I can answer some of those questions; I'll have to grab the screenshots later today when I'm on break. Posting from my work computer atm.

So the GPU is only boosting to 240MHz? Essentially, yes. The GPU sits at 240MHz even under load while running benchmark programs like userbenchmark (I ran furmark once or twice but have seen many people saying to avoid it so I haven't ran it beyond that). As mentioned earlier, temps don't go above ~50C during these benchmarks and the fans run w/out issue.

Also its the RTX 3070 then, not the TI? Correct, it's the RTX 3070 and not the Ti

Does the card have a BIOS selection switch ? Yes, it does have a switch on it; though I think it controls the fan speed. It has "BIOS OC" and "Manual" iirc.

Im guessing your old GTX 1070 works without issues right now? Yessir; 1070 works like a charm and I've never encountered any issues with it.

I have to agree with Phaaze88 here it seems they send you back a broken card. This is only a guess but perhaps they just changed the switch, saw that the card posted and send it back to you. With or without even checking if it was working properly on windows, boosting the way it should, etc.

 

[BooTheFearless]

It sounds like Gigabyte sent you back a damaged card.
I've read that the gpu does that if there's a short somewhere, or a bent/broken PCIE port. It refuses to boost to protect itself.
Doesn't help you much, since you can't really use it in that state.

I've suspected the same thing at this point. I believe I mentioned earlier that I've purchased an 850w and 1000w PSU to rule that out as a possibility; bought from Amazon for ease of returning in the event that neither of those work for me. Hopefully I'll be able to provide everyone a better idea as to my situation once I can get you all some screenshots of my GPU-Z and GPU Shark readings.

Though, GPU-Z does say that the PerfCap reason is Pwr; do you think that could be caused by a bad/insufficient PSU in my case? And, is it possible that the card is keeping itself at 240MHz to protect itself from damage because of a lack of power?

 

[Phaaze88]

That PerfCap reason: Power, and the stuck at low clocks are 2 of the symptoms of that short/physical damage I mentioned. Almost nothing to do with the psu.

Your ol' 1070 still works, so I figure the possible causes of short/damage aren't on your end, such as a defective monitor cable or PCIe connectors.

 

[BooTheFearless]

Well, thanks Gigabyte haha. That really stinks; I've already reached back out to their support by submitting a ticket but Lord knows when I'll hear back. I'll still try using a new PSU just for the heck of it at this point to see if anything changes; a co-worker of mine was experiencing a similar issue with his RTX 3070 and upgrading his PSU fixed his issue so we'll see. I'll still update the thread later today with my GPU-Z and GPU Shark screenshots for posterity's sake; perhaps there's something that will give folks a better light to my issue than my memory can.

Thanks @Phaaze88 & @RodroX ! I really appreciate both of your input on this. Either of you think it'd be worth sending it back in for RMA or just get my refund at this point.

 

[Phaaze88]

Hmm... On one end, they did get it to come back on, but it's almost like they didn't bother testing it afterwards - I've no way to prove that though.
That, or the gpu was damaged during transit back to you. Would be hard to believe that one, unless they didn't use the original packaging.

I think I'd lean towards a refund.

 

[BooTheFearless]

Hmm... On one end, they did get it to come back on, but it's almost like they didn't bother testing it afterwards - I've no way to prove that though.

Well, I know for a fact that they replaced the GPU chip itself which was the original cause of the card not showing in the BIOS. I was sent a screenshot of their logs by the support representative I was in touch with while the card was undergoing the RMA process. To their credit, they "fixed" it relatively quickly. Ugh, just stinks that you pay good money for a card only for it to not work out of the box, have them repair it, and then it still be broken.

Guess I'm staying away from Gigabyte moving forward; surprisingly I've never had any bad experiences prior to this. All my GPUs from 2011 when I built my first PC to now have been Gigabyte.

 

[RodroX]

Well, I know for a fact that they replaced the GPU chip itself which was the original cause of the card not showing in the BIOS. I was sent a screenshot of their logs by the support representative I was in touch with while the card was undergoing the RMA process. To their credit, they "fixed" it relatively quickly. Ugh, just stinks that you pay good money for a card only for it to not work out of the box, have them repair it, and then it still be broken.

Guess I'm staying away from Gigabyte moving forward; surprisingly I've never had any bad experiences prior to this. All my GPUs from 2011 when I built my first PC to now have been Gigabyte.

Far as I know, I could be wrong, no oem (Asus, Gigabyte, MSI, EVGA, XFX, Zotac, etc) will go as far as to replace "the chip" meaning the nvidia GPU itself. That will probably mean sending the card back to Taiwan or China and so they don't do that.

Perhaps the "chip" they changed was the BIOS. Then again I can not be sure.

 

[BooTheFearless]

Perhaps the "chip" they changed was the BIOS. Then again I can not be sure.

I suspect you're right, more than likely. If you're interested enough, I'll post my screenshots in a bit here when I get off work to give you a better idea of what's going on with the card. Who knows, maybe reinstalling it again, the card will magically fix itself lol.

Like @Phaaze88 mentioned, it's possible there's still a short somewhere in the card that would prevent it from clocking up to base speed. However, if that were the case, wouldn't the temperatures be high, or in general unstable? I don't think I mentioned earlier that when I used AORUS engine to force the card to 1200MHz, it was overall stable based on what I was seeing in GPU-Z.

I gotta say, thanks to both of you a million times over for pitching in and giving your two cents; I really appreciate it and you've both been VERY helpful.

 

[RodroX]

I suspect you're right, more than likely. If you're interested enough, I'll post my screenshots in a bit here when I get off work to give you a better idea of what's going on with the card. Who knows, maybe reinstalling it again, the card will magically fix itself lol.

Like @Phaaze88 mentioned, it's possible there's still a short somewhere in the card that would prevent it from clocking up to base speed. However, if that were the case, wouldn't the temperatures be high, or in general unstable? I don't think I mentioned earlier that when I used AORUS engine to force the card to 1200MHz, it was overall stable based on what I was seeing in GPU-Z.

I gotta say, thanks to both of you a million times over for pitching in and giving your two cents; I really appreciate it and you've both been VERY helpful.

Video cards are soo complicated that sometimes its really hard to pinpoint the right issue from the distance. As for your question about high temps, not necessarily.

In any case I would keep the GPU at stock settings, and if you want try the new PSU.

 

[BooTheFearless]

These are my GPU-Z and GPU Shark specs of the card while idling; does anything here look out of the ordinary to anyone? @RodroX & @Phaaze88? Currently downloading Superposition benchmark as recommended by GIGABYTE technical support; I called them and actually got in touch with someone. I was pretty much told to run this benchmark, contact customer support, and send the card out for another round of RMA repair... gee, thanks Gigabyte lol. Once superposition finishes downloading, I'll run it and nab some screenshots of these specs while the card is under stress.

 

[BooTheFearless]

And here it is under load; both GPU Shark and GPU Z. Will post 1070's results for posterity's sake later.

Hopefully with this information in hand, you all can get a better idea as to what's going on with this video card. As you can see, the clock stays consistent at a whopping 240MHz.

 

[BooTheFearless]

And, here it is boosted to 1200MHz via Aorus Engine OC tool.

 

[RodroX]

These are my GPU-Z and GPU Shark specs of the card while idling; does anything here look out of the ordinary to anyone? @RodroX & @Phaaze88? Currently downloading Superposition benchmark as recommended by GIGABYTE technical support; I called them and actually got in touch with someone. I was pretty much told to run this benchmark, contact customer support, and send the card out for another round of RMA repair... gee, thanks Gigabyte lol. Once superposition finishes downloading, I'll run it and nab some screenshots of these specs while the card is under stress.

So at idle your GPU is runing at 240MHz, VRAM at +1500MHz while consuming +180watts, yes theres something very wrong with that card. Mine (RTX 2070) at idle runs at 300MHz/100MHz (GPU/VRAM), power consumption less than 10watts.

So yes as Phaaze88 wrote already, either send it back for RMA again, or ask for a refund while you still can.

 

[BooTheFearless]

Thanks again for your input @RodroX ! I've already contacted Gigabyte and re-initiated the RMA process. I did get confirmation that they would issue me a refund for the price of the GPU if they're unable to repair, but considering how difficult GPUs are to get, I'm willing to risk their RMA process a second time to see if it can be fully repaired (or at the least have them send me a replacement).

The two PSUs I purchased are here so later today I'll install those to see if anything is different; I doubt it as you've all said the card is likely just caput. But, eh, I've got em so why not. If anything changes, I'll be sure to update the thread with the results of that.

Thanks again to everyone who shared their thoughts!! I really really appreciate it; this forum and it's members ROCK!

 

[RodroX]

Thanks again for your input @RodroX ! I've already contacted Gigabyte and re-initiated the RMA process. I did get confirmation that they would issue me a refund for the price of the GPU if they're unable to repair, but considering how difficult GPUs are to get, I'm willing to risk their RMA process a second time to see if it can be fully repaired (or at the least have them send me a replacement).

The two PSUs I purchased are here so later today I'll install those to see if anything is different; I doubt it as you've all said the card is likely just caput. But, eh, I've got em so why not. If anything changes, I'll be sure to update the thread with the results of that.

Thanks again to everyone who shared their thoughts!! I really really appreciate it; this forum and it's members ROCK!

Best of lucks, and please remember to come by later and share with us what was the resolution.

Cheers