Getting constant problems with 4070

[LAUCHGeorge]

So basically, since over a year I have been getting errors that in some way or another show up as a TDR error thingy in event viewer, that includes nvlddmkm crashing, but just now it also started to show in my PC literally just randomly rebooting, with it showing up in the event viewer as a "The computer has rebooted from a bugcheck.  The bugcheck was: 0x00000116 (0xffffdd8b3d279010, 0xfffff8077cd61a40, 0xffffffffc000009a, 0x0000000000000004).".

I already tried a lot of things, and since the problems have started my PC now has a completely different CPU, MB and RAM. I also already made sure this is not caused by my GPU riser cable, and the temps of my GPU look absolutely fine.

 

The exact model of my GPU is an ASUS RTX 4070 strix oc.

 

This card also already went into ASUS's RMA two times, both times with them sending it back and basically saying "it works for us".

After the second RMA  (and that was like, 3 or 4 months ago) the problems started to magically disappear until today, when my PC randomly rebooted.

 

Back then (either before the first or second RMA) I also ran a stress test in OCCT, where the system completely crashed. After opening OCCT on startup, one of the attached screenshot was the last known status of the GPU prior to the crash. I also attached some more screenshots I found from back then, showing some more stuff in event viewer. (a lot is in german as my windows install back then, was set to german)

 

I have absolutely no idea what to check for anymore, and I don't think that contacting ASUS support yet again will lead to a lot.

 

I have fully reinstalled windows a few times already since then.

 

Sorry if this was not all that clear, I am more than happy to provide more information or context if needed.

[Mark Kaine]

check ur ram.

 

1 stick only, no xmp.

check each slot and stick...

turn of all overclocking. 

buy new ram.

 

to be clear this is 99.99% your ram having issues. 

 

Also, FULL SPECS, please?

[LK_heero]

so just to be sure what steps have you done ?

did you maybe updated the GPU software?

is the motherboard BIOS up to date?

do you have the problem if you put XMP off ? did you do a memtest ? 

the PSU (unlikely i guess) is new ?

 

5 minutes ago, Mark Kaine said:

check ur ram.

 

1 stick only, no xmp.

check each slot and stick...

turn of all overclocking. 

buy new ram.

 

to be clear this is 99.99% your ram having issues. 

 

Also, FULL SPECS, please?

aa damn you where a min urlier

 

PS. i had my 7200 mhz ram sticks work at 7200 untill i bios updated so maybe turn it down untill it does run stable. (6400 mhz) 

[LAUCHGeorge]

Just now, Mark Kaine said:

check ur ram.

 

1 stick only, no xmp.

check each slot and stick...

turn of all overclocking. 

buy new ram.

 

to be clear this is 99.99% your ram having issues. 

 

Also, FULL SPECS, please?

It's late so sorry for not being very clear, but I literally already replaced RAM, CPU and Motherboard since these problems first started appearing. My thought would also have been RAM, but that's ruled out now. Current relevant specs are:

GPU: ROG-STRIX-RTX4070-O12G-GAMING
CPU: 9800x3d
MB: ROG Strix X870e-e Gaming wifi
RAM: 2x32 DDR5-6000 CL30-40-40-96 g.skill trident z5 neo
PSU: bequiet pure power 12m 1000w
CPU-Cooler: NZXT Kraken Elite 360

[LK_heero]

2 minutes ago, LAUCHGeorge said:

It's late so sorry for not being very clear, but I literally already replaced RAM, CPU and Motherboard since these problems first started appearing. My thought would also have been RAM, but that's ruled out now. Current relevant specs are:

GPU: ROG-STRIX-RTX4070-O12G-GAMING
CPU: 9800x3d
MB: ROG Strix X870e-e Gaming wifi
RAM: 2x32 DDR5-6000 CL30-40-40-96 g.skill trident z5 neo
PSU: bequiet pure power 12m 1000w
CPU-Cooler: NZXT Kraken Elite 360

did you run memtest just to be sure because some of the errors do look like ram ..

because its a ROG Strix it should be build in to the bio's

[LAUCHGeorge]

4 minutes ago, LK_heero said:

so just to be sure what steps have you done ?

did you maybe updated the GPU software?

is the motherboard BIOS up to date?

do you have the problem if you put XMP off ? did you do a memtest ? 

the PSU (unlikely i guess) is new ?

 

aa damn you where a min urlier

I'd rule out stuff like mb, ram and cpu, but before I upgraded those anyway, I did run various memory tests to rule that out. I also did do a full clean install of windows multiple times between now and when those issues first started appearing. As described I also want to point out that the "symptoms" rn. are not as bad as they once were, but that GPU is still not acting normal.

I already thought about the PSU, but that's very unlikely

[Mark Kaine]

1 minute ago, LAUCHGeorge said:

It's late so sorry for not being very clear, but I literally already replaced RAM, CPU and Motherboard since these problems first started appearing. My thought would also have been RAM, but that's ruled out now. Current relevant specs are:

GPU: ROG-STRIX-RTX4070-O12G-GAMING
CPU: 9800x3d
MB: ROG Strix X870e-e Gaming wifi
RAM: 2x32 DDR5-6000 CL30-40-40-96 g.skill trident z5 neo
PSU: bequiet pure power 12m 1000w
CPU-Cooler: NZXT Kraken Elite 360

it could still be your ram... the thing is "bugcheck" is typically ram... but its also true it could be anything really...  so you don't have XMP enabled? Ran memtest86, like 6 hours minimum? 

 

another recommendation would be to uninstall, NZXT CAM fully, restart pc...

[LAUCHGeorge]

1 minute ago, LK_heero said:

did you run memtest just to be sure because some of the errors do look like ram ..

because its a ROG Strix it should be build in to the bio's

On that new board I ran occt memory test for a bit, not for too long though

[LK_heero]

4 minutes ago, LAUCHGeorge said:

On that new board I ran occt memory test for a bit, not for too long though

i would say Memtest84 its been the standard for a long time i recommend you run that (takes about 15 min for a run and run it like 4 times) if it says pass its 100% not the ram

[LAUCHGeorge]

Just now, Mark Kaine said:

it could still be your ram... the thing is "bugcheck" is typically ram... but its also true it could be anything really...  so you don't have XMP enabled? Ran memtest86, like 6 hours minimum? 

 

another recommendation would be to uninstall, NZXT CAM fully, restart pc...

I did not run that with my new system yet, but I'd also like to point out that bugcheck 0x00000116 seems to be related to TDR shit, which in turn points at all the issues I had with my GPU. It's also weird that none of this ever happened on any other GPU, and I did use quite a few, especially back then during the 2 RMA's my 4070 went through.

[LAUCHGeorge]

Just now, LK_heero said:

i would say Memtest84 its been the standard for a long time i recommend you run that (takes about 15 min for a run and run it like 4 times) if it says pass its 100% not the ram

I'll look into doing that, can't do that rn. since it's really late but I have sorta already given up on it being a memory issue

[freeagent]

Guaranteed its your memory settings, or your board does not like your ram.

[LK_heero]

9 minutes ago, LAUCHGeorge said:

I did not run that with my new system yet, but I'd also like to point out that bugcheck 0x00000116 seems to be related to TDR shit, which in turn points at all the issues I had with my GPU. It's also weird that none of this ever happened on any other GPU, and I did use quite a few, especially back then during the 2 RMA's my 4070 went through.

top 5 awnsers for 0x00000116

1. Update Your Graphics Driver: Ensure you have the latest drivers for your graphics card. You can download the latest drivers from the manufacturer's website (e.g., Nvidia, AMD) 

2. Check for Overheating: Overheating can cause this error. Make sure your system is properly ventilated and clean any dust from your GPU and other components 

3. Check Hardware: Ensure that your graphics card is properly seated in its slot. Also, check for any hardware issues such as faulty RAM or a failing GPU 

4. Adjust TDR Settings: You can adjust the Timeout Detection and Recovery (TDR) settings in the Windows registry to give your GPU more time to recover from a timeout 

5. System Updates: Make sure your operating system is up to date with the latest patches and updates.

'm not saying the RAM is faulty, but maybe the settings are incorrect?

 

I managed to get my 7200 MHz RAM working with XMP 2 by manually tweaking the voltage. I started with 0.01 V increments, and as the errors decreased, I fine-tuned it with 0.001 V adjustments. It took me a lot of hours to get it working.

 

(Hence why I didn't go through the process again after it stopped working with a BIOS update...)

[LAUCHGeorge]

On 2/4/2025 at 1:40 AM, LK_heero said:

i would say Memtest84 its been the standard for a long time i recommend you run that (takes about 15 min for a run and run it like 4 times) if it says pass its 100% not the ram

I ran 4 passes of memtest86, all passed.

[LK_heero]

i guess you're not like me and maybe have a second PC ( i use my old PC as a server now so ) ? 

or a friend and check if it works in his PC ?

just to make sure it isn't the port in your motherboard

[LAUCHGeorge]

1 hour ago, LK_heero said:

i guess you're not like me and maybe have a second PC ( i use my old PC as a server now so ) ? 

or a friend and check if it works in his PC ?

just to make sure it isn't the port in your motherboard

As I have said multiple times already, since this problem first occured, I already swapped my MB, CPU and RAM all out. Meaning I am quite confident those are not the problem here.

[LK_heero]

23 hours ago, LAUCHGeorge said:

As I have said multiple times already, since this problem first occured, I already swapped my MB, CPU and RAM all out. Meaning I am quite confident those are not the problem here.

Oh, right! I forgot you mentioned you swapped your motherboard. If ASUS says the GPU is 100% okay, then I'm out of ideas.

As your suggested wattage is 700 - 799 Watts and you're PSU is 1000W, you're more than covered.