Pascal Titan X Frequent Code 43 with 372.xx Drivers

Zarathustra[H]

Zarathustra[H]

Extremely [H]
Joined
Oct 29, 2000
Messages
38,865
Hey all,

I'm wondering if anyone has any ideas regarding this. I know a code 43 error is usually indicative of a hardware failure, but I don't believe this to be the case for me, I believe it is related to the drivers.

My setup:

I dual boot Linux and Windows. Linux Mint 18 is rock solid with latest Nvidia Linux Drivers (367.44). I've never had an issue in Linux.

In Windows 10 (Anniversary Edition) I never have any problems if I install the drivers the card shipped with (369.05). The card is perfectly stable. Only downside is I don't have any optimization for newer titles.

I'd stay with the 369.05 drivers, but right now I am playing through Deus Ex: Mankind Divided.

My issues with the 372.xx series have been as follows:

372.54: A shit show of crashes, inability to install drivers, Code 43 errors, and just general awfulness.

372.70 & 372.90: These are mostly equivalent. Work much better than 372.54, providing stable gameplay in heavy titles, including Mankind Divided, but frequently my GPU flips a code 43 on reboot. Maybe one in 5 times.

If I DDU the driver, and reinstall it (losing all my settings, which is a pain in the ass) it comes back, and everything is happy again... until the next time it randomly flips a code 43 during reboot.

I don't think it is a defective GPU, because 369.05 in Windows 10 AE, and 367.44 in Linux both work perfectly and are completely stable. Something about any driver in the 372.xx series seems to be causing the problems.

If I only didn't need 372.xx for Mankind Divided...

Anyone have any ideas?
 
Zarathustra[H] said:
Hey all,

I'm wondering if anyone has any ideas regarding this. I know a code 43 error is usually indicative of a hardware failure, but I don't believe this to be the case for me, I believe it is related to the drivers.

My setup:

I dual boot Linux and Windows. Linux Mint 18 is rock solid with latest Nvidia Linux Drivers (367.44). I've never had an issue in Linux.

In Windows 10 (Anniversary Edition) I never have any problems if I install the drivers the card shipped with (369.05). The card is perfectly stable. Only downside is I don't have any optimization for newer titles.

I'd stay with the 369.05 drivers, but right now I am playing through Deus Ex: Mankind Divided.

My issues with the 372.xx series have been as follows:

372.54: A shit show of crashes, inability to install drivers, Code 43 errors, and just general awfulness.

372.70 & 372.90: These are mostly equivalent. Work much better than 372.54, providing stable gameplay in heavy titles, including Mankind Divided, but frequently my GPU flips a code 43 on reboot. Maybe one in 5 times.

If I DDU the driver, and reinstall it (losing all my settings, which is a pain in the ass) it comes back, and everything is happy again... until the next time it randomly flips a code 43 during reboot.

I don't think it is a defective GPU, because 369.05 in Windows 10 AE, and 367.44 in Linux both work perfectly and are completely stable. Something about any driver in the 372.xx series seems to be causing the problems.

If I only didn't need 372.xx for Mankind Divided...

Anyone have any ideas?
Click to expand...

Try another slot, maybe? Did you change anything else, like BIOS updates?
 
trandoanhung1991 said:
Try another slot, maybe? Did you change anything else, like BIOS updates?
Click to expand...


Nope.

I can switch back and forth between the drivers. Old ones are perfect. New ones are problematic. I dual boot back and forth to Linux too all the time. Linux runs perfectly, Windows does not (when new drivers are installed)

That suggests to me it cant possibly be the slot.
 
Have you tried a clean Windows 10 1607 (AU)? Looks like your issue is WDDM 2.1 drivers.

If a clean install does the same, contact Nvidia.
 
Have you tested your system RAM or is the PCI-E LED lit up when this happens?
 
Any code between 30~50 is usually related to imc/memory.

I expect your ram's not happy when your cpu is overclocked, so only crashes when whole system is stressed. Try putting more voltage to the ram and/or imc portion of the cpu I'd say.
 
Shintai said:
Have you tried a clean Windows 10 1607 (AU)? Looks like your issue is WDDM 2.1 drivers.

If a clean install does the same, contact Nvidia.
Click to expand...


Yeah, I'm leaning in this direction too. I was hoping to avoid this. I traditionally always did clean Windows installs in the past, but I can't keep up with the frequent Windows 10 releases.
 
spine said:
Any code between 30~50 is usually related to imc/memory.

I expect your ram's not happy when your cpu is overclocked, so only crashes when whole system is stressed. Try putting more voltage to the ram and/or imc portion of the cpu I'd say.
Click to expand...


Could be. I don't know why this would only impact the 372.xx line of drivers under Windows, and nothing else, but it is a possibility.

My system passed days of stability testing in Prime95 at 4.8, but I'm going to temporarily drop it to 4.7 and see if that makes a difference. If it does, I'll know to go back up to 4.8 and play with the memory controller voltages. Appreciate the suggestion.
 
Zarathustra[H] said:
Yeah, I'm leaning in this direction too. I was hoping to avoid this. I traditionally always did clean Windows installs in the past, but I can't keep up with the frequent Windows 10 releases.
Click to expand...

I think its worth a try. May be some old DLL screwing things up. Specially if you got no issues with WDDM 2.0, but issues with WDDM 2.1. That sounds like software.
 
Zarathustra[H] said:
Hey all,

I'm wondering if anyone has any ideas regarding this. I know a code 43 error is usually indicative of a hardware failure, but I don't believe this to be the case for me, I believe it is related to the drivers.

My setup:

I dual boot Linux and Windows. Linux Mint 18 is rock solid with latest Nvidia Linux Drivers (367.44). I've never had an issue in Linux.

In Windows 10 (Anniversary Edition) I never have any problems if I install the drivers the card shipped with (369.05). The card is perfectly stable. Only downside is I don't have any optimization for newer titles.

I'd stay with the 369.05 drivers, but right now I am playing through Deus Ex: Mankind Divided.

My issues with the 372.xx series have been as follows:

372.54: A shit show of crashes, inability to install drivers, Code 43 errors, and just general awfulness.

372.70 & 372.90: These are mostly equivalent. Work much better than 372.54, providing stable gameplay in heavy titles, including Mankind Divided, but frequently my GPU flips a code 43 on reboot. Maybe one in 5 times.

If I DDU the driver, and reinstall it (losing all my settings, which is a pain in the ass) it comes back, and everything is happy again... until the next time it randomly flips a code 43 during reboot.

I don't think it is a defective GPU, because 369.05 in Windows 10 AE, and 367.44 in Linux both work perfectly and are completely stable. Something about any driver in the 372.xx series seems to be causing the problems.

If I only didn't need 372.xx for Mankind Divided...

Anyone have any ideas?
Click to expand...

Have your tried this not sure if it'll help, but it might keep you from having to clean install windows...https://forums.geforce.com/default/...or-new-cpu-core-analyzer-updated-04-05-2016-/
 
HubDubDoubleDubDub said:
Have your tried this not sure if it'll help, but it might keep you from having to clean install windows...https://forums.geforce.com/default/...or-new-cpu-core-analyzer-updated-04-05-2016-/
Click to expand...


Which one are you referring to? DDU? Yes, I do that every time it throws a code to get it working again.

It appears as if the way the system works (not sure if it is Windows or Nvidia) is that whenever a problem is detected, it triggers a code. The problem may not be currently occurring, but the code was thrown at some point so the hardware remains in a disabled state. It's really rather annoying.
 
I've had a good deal of problems with windows 10 installs and updates due to overclocks. I'm not sure why because its all totally stable under stress testing and normal use. I always go back to stock due to this.
 
spine said:
Any code between 30~50 is usually related to imc/memory.

I expect your ram's not happy when your cpu is overclocked, so only crashes when whole system is stressed. Try putting more voltage to the ram and/or imc portion of the cpu I'd say.
Click to expand...
Zarathustra[H] said:
Could be. I don't know why this would only impact the 372.xx line of drivers under Windows, and nothing else, but it is a possibility.

My system passed days of stability testing in Prime95 at 4.8, but I'm going to temporarily drop it to 4.7 and see if that makes a difference. If it does, I'll know to go back up to 4.8 and play with the memory controller voltages. Appreciate the suggestion.
Click to expand...

Yeah, so the same thing still happens downclocked. Just booted into Windows and another code 43.

Time to try reinstalling Windows 10.
 
Actually, come to think of it, before I do, I am going to try one more thing. I am currently using the Nvidia registry patch that allows X79 to operate at Gen3 speeds. I wonder if that could the the issue. It's never been an issue before, but who knows with WDDM2.1
 
Zarathustra[H] said:
Actually, come to think of it, before I do, I am going to try one more thing. I am currently using the Nvidia registry patch that allows X79 to operate at Gen3 speeds. I wonder if that could the the issue. It's never been an issue before, but who knows with WDDM2.1
Click to expand...


Well, it turns out that isn't it. It actually resets that registry entry when I reinstall the drivers, so if that were it, the first resintall would have fixed it.


So I've done some more digging, and it looks to me likw hwat is causing it is a Kernel-PNP 441 error as follows:

Code: 
Device PCI\VEN_10DE&DEV_1B00&SUBSYS_119A10DE&REV_A1\4&b16bfe2&0&0010 could not be migrated.

Last Device Instance Id: PCI\VEN_10DE&DEV_1B80&SUBSYS_37021458&REV_A1\4&B16BFE2&0&0010
Class Guid: {4d36e968-e325-11ce-bfc1-08002be10318}
Location Path: PCIROOT(0)#PCI(0200)#PCI(0000)
Migration Rank: 0xF000FFFFE0000122
Present: false
Status: 0xC0000719

At every boot when I find the GPU in Code 43 mode, there is a log entry with this identical error. I'm not sure what "could not be migrated" means, but I googled the status code "0xC0000719" and have found suggestions that it might just require a reset of the BIOS to reset the PCIe bus addresses.

I am going to try that and see if it helps, but not tonight.

I need to go through and write down all my overclock settings first, or I'll lose them when I reset it.

Not sure why this would effect only 3762.xx drivers though. That part just doesn't make sense.


PS.

It is amazing when you google these error codes, how many fake Microsoft support sites you come across urging you to install their tool to fix the problem (which is probably some sort of malware)
 
Zarathustra[H] said:
Well, it turns out that isn't it. It actually resets that registry entry when I reinstall the drivers, so if that were it, the first resintall would have fixed it.


So I've done some more digging, and it looks to me likw hwat is causing it is a Kernel-PNP 441 error as follows:

Code: 
Device PCI\VEN_10DE&DEV_1B00&SUBSYS_119A10DE&REV_A1\4&b16bfe2&0&0010 could not be migrated.

Last Device Instance Id: PCI\VEN_10DE&DEV_1B80&SUBSYS_37021458&REV_A1\4&B16BFE2&0&0010
Class Guid: {4d36e968-e325-11ce-bfc1-08002be10318}
Location Path: PCIROOT(0)#PCI(0200)#PCI(0000)
Migration Rank: 0xF000FFFFE0000122
Present: false
Status: 0xC0000719

At every boot when I find the GPU in Code 43 mode, there is a log entry with this identical error. I'm not sure what "could not be migrated" means, but I googled the status code "0xC0000719" and have found suggestions that it might just require a reset of the BIOS to reset the PCIe bus addresses.

I am going to try that and see if it helps, but not tonight.

I need to go through and write down all my overclock settings first, or I'll lose them when I reset it.

Not sure why this would effect only 3762.xx drivers though. That part just doesn't make sense.


PS.

It is amazing when you google these error codes, how many fake Microsoft support sites you come across urging you to install their tool to fix the problem (which is probably some sort of malware)
Click to expand...


Alright, to test to see if this "resetting the PCI adresses" thing was bs. or not, and to see if overclocking had any impact, saved my BIOS settings using Asus OC profile, and reset them to optimized default, and decided to do a few reboots to see if I could tease the issue into coming back.

First boot, great, it worked, and came back up, No Code 43 error. Promising.

Second reboot. Damn. Code 43 error again. Alright, so it's neither the PCIe address thing or my OC settings causing this issue...

Rebooted into BIOS to restore my OC using Asus OC profile, and booted back up again, and then someting really weird happened. The GPU is back. It booted back up without a Code 43 error. This is really weird. Usually once the Code 43 error hits, I need to DDU it and reinstall drivers for it to come back.

I'm not counting on this being permanent though. If things go the way they have been going, next reboot or the reboot after, I'll be in Code 43 again.

When I muster up the energy to do so, I guess I'm going to have to do a fresh reinstall of Windows 10 AE. Which sucks. I feel like I only just reinstalled Windows not that long ago. It was before the AE update though.

PS.

Learned the hard way that Aus OC profile does not save my boot CSM settings. Now I have to figure that BS out again to get it to point towards my grub entry, instead of just gooting into Windows 10 directly.
 
Zarathustra[H] said:
Alright, to test to see if this "resetting the PCI adresses" thing was bs. or not, and to see if overclocking had any impact, saved my BIOS settings using Asus OC profile, and reset them to optimized default, and decided to do a few reboots to see if I could tease the issue into coming back.

First boot, great, it worked, and came back up, No Code 43 error. Promising.

Second reboot. Damn. Code 43 error again. Alright, so it's neither the PCIe address thing or my OC settings causing this issue...

Rebooted into BIOS to restore my OC using Asus OC profile, and booted back up again, and then someting really weird happened. The GPU is back. It booted back up without a Code 43 error. This is really weird. Usually once the Code 43 error hits, I need to DDU it and reinstall drivers for it to come back.

I'm not counting on this being permanent though. If things go the way they have been going, next reboot or the reboot after, I'll be in Code 43 again.

When I muster up the energy to do so, I guess I'm going to have to do a fresh reinstall of Windows 10 AE. Which sucks. I feel like I only just reinstalled Windows not that long ago. It was before the AE update though.

PS.

Learned the hard way that Aus OC profile does not save my boot CSM settings. Now I have to figure that BS out again to get it to point towards my grub entry, instead of just gooting into Windows 10 directly.
Click to expand...


As I suspected. I rebooted to BIOS again to fix my UEFI boot settings, and once fixed, when I booted into Windows 10 again, the Code 43 error reappeared.

On a whim, I decided to shut the system completely off and boot it back up again to see what would happen. Surprise surprise, the GPU is back. So at the very least I have learned that it is somehow intermittent, and I don't need to DDU the driver every time it happens.

I'm still going to reinstall Windows when I have enough time, but it will probably be a while. I have surgery on Wednesday (hip arthroscopy), and I don't know how long it will be until I can sit comfortably in a chair afterwards. Supposedly up to a 6 week recovery...
 
Zarathustra[H] said:
What do you mean by the PCIe LED?
Click to expand...
Sorry thought you ment this was a QCode, realizing now it's not.
Seeing a lot of user reports with Lucid VirtuMVP installed (software that usually comes with motherboards that I never install). If it's not this I would assume it is a driver issue. I would just go back to 369.05 or whatever was stable for you until you're well enough to tinker around with this after your surgery. Is there performance problems with 369.05 on DeusEx MKD? The 372.xx drivers have all been crap for us TitanXP users, I tested all of them and they were all worse than 369.05 in most situations (with the exception of NoMansSky). My issue with 372.xx is artifacting in photoshop and occasionaly in the Windows UI.
 
I'm just gunna share this, for what it's worth...


As an avid overclocker, I've found that my Titan XP is much more sensitive to BCLK overclocking than previous card(s) (Asus 980 and 780ti in same mobo), when PCIe Gen 3 is enabled. Forcing Gen2 immediatelty allows greater overclock of the BCLK.

I say this, because where it relates to you is that if I push my BCLK too far, everything will be fine, but when that crash happens 3 hours into a game or whatever, on reboot it fails to POST with an error code. I've never bothered to look at it mind, but it's a triple beep so memory related.

The thing is, on every boot the motherboard 'trains' the memory and ultra tweaks a whole bunch of ultra fine timings never exposed to the user. Temperature affects everything, so this training has to be done on every boot. ON soft reset, it doesn't bother, and so what happens is that the machine has gotten warmed up so that the original training is now out of whack, so on reboot it fails. Full power off and on again, and thus force a full re-train, and it'll POST just fine.


That's basically what I believe is happening to you.


I'd say you need to boost voltage to the SA portion of the CPU. Why now and not before? Couldn't say, but hope this helps in some way.
 
Zarathustra[H] said:
Now I have to figure that BS out again to get it to point towards my grub entry, instead of just gooting into Windows 10 directly.
Click to expand...

Do you have the problem if you don't boot from GRUB? (I.e., using the motherboard boot device chooser. F10 at boot or whatever it is.)
 
spine said:
I'm just gunna share this, for what it's worth...


As an avid overclocker, I've found that my Titan XP is much more sensitive to BCLK overclocking than previous card(s) (Asus 980 and 780ti in same mobo), when PCIe Gen 3 is enabled. Forcing Gen2 immediatelty allows greater overclock of the BCLK.

I say this, because where it relates to you is that if I push my BCLK too far, everything will be fine, but when that crash happens 3 hours into a game or whatever, on reboot it fails to POST with an error code. I've never bothered to look at it mind, but it's a triple beep so memory related.

The thing is, on every boot the motherboard 'trains' the memory and ultra tweaks a whole bunch of ultra fine timings never exposed to the user. Temperature affects everything, so this training has to be done on every boot. ON soft reset, it doesn't bother, and so what happens is that the machine has gotten warmed up so that the original training is now out of whack, so on reboot it fails. Full power off and on again, and thus force a full re-train, and it'll POST just fine.


That's basically what I believe is happening to you.


I'd say you need to boost voltage to the SA portion of the CPU. Why now and not before? Couldn't say, but hope this helps in some way.
Click to expand...


This is very good information.

Not sure it applies to me though, but I will check it out.

Firstly, when I overclock I do it 100% with multipliers, so no BCLK to worry about.

Secondly I tried resetting the BIOS to optimized defaults (IE no overclocking) and it still happened.

GPU loads in GEN2 mode due to Nvidia's restriction on X79. I can force it into GEN3 mode by editing the registry, but I've run it both ways and it doesn't appear to have an impact on this issue. The issue DOES happen during boot though, so it is possible it temporarily loads as a gen3 device, before the Nvidia driver is loaded, and that's when it errors. I will try to drop the BIOS setting down to GEN2 and see if it persists.

So, I don't think my overclocking settings are the issue, but you make a good point regarding the voltage. It could be that the optimized defaults on my motherboard aren't quite right for stable SA. I can't remember which settings SA are (I'm usually an overclock once, then set it and forget it kind of guy, so I have to relearn overclocking every time I buy a new motherboard/CPU) but I will look at that.

Then again, it could just be that after 5 years of running at 4.8Ghz my CPU or Motherboard VRM's are starting to get tired.

This is the first time I've ever kept a CPU/motherboard combo long enough that it might actually be failing. It's also the first time I've had a CPU/Motherboard combo that has still been viable performance-wise 5 years later. How the enthusiast desktop hobby has changed...

Thanks for the suggestions!
 
Last edited:
jwcalla said:
Do you have the problem if you don't boot from GRUB? (I.e., using the motherboard boot device chooser. F10 at boot or whatever it is.)
Click to expand...

I only have very limited testing without grub, but it did happen once without it, yes.

How do you think this could be related?

All Grub does is acts as a menu that points to the respective boot loader for each OS. Windows should boot the same way with or without grub. Grub just inserts itself before the windows boot loader.

I'd be curious what your theory is, as I ahve never heard of issues like this due to grub before.
 
Zarathustra[H] said:
I only have very limited testing without grub, but it did happen once without it, yes.

How do you think this could be related?

All Grub does is acts as a menu that points to the respective boot loader for each OS. Windows should boot the same way with or without grub. Grub just inserts itself before the windows boot loader.

I'd be curious what your theory is, as I ahve never heard of issues like this due to grub before.
Click to expand...

Only that there is something different with the way GRUB boots Windows. I forget the details because it was a long time ago. But, e.g., on my Windows 7 box, I can not run Windows backup at all when booting from GRUB. I have to boot from the actual Windows disk in order to run backup (on 7 at least). It's probably not related at all to your problem, but since I saw you were using it I'd figure why not ask if you've tried without it.
 
jwcalla said:
Only that there is something different with the way GRUB boots Windows. I forget the details because it was a long time ago. But, e.g., on my Windows 7 box, I can not run Windows backup at all when booting from GRUB. I have to boot from the actual Windows disk in order to run backup (on 7 at least). It's probably not related at all to your problem, but since I saw you were using it I'd figure why not ask if you've tried without it.
Click to expand...


That's really interesting. Is that with EFI boot or traditional bios boot?

It really shouldn't make a difference in either. I wonder why it does. With EFI all grub does is has its own EFI entry with a menu that either loads the Linix kernel or points to the Windows EFI entry. Windoes shouldn't know the difference either way.

Same with traditional boot methods. Grub just owns the MBR so instead of the MBR pointing directly to the Windows partition it loads grub which either loads the Linux kernel or just forwards things on to the same partition it otherwise would.

This is very strange.
 
spine said:
I'm just gunna share this, for what it's worth...


As an avid overclocker, I've found that my Titan XP is much more sensitive to BCLK overclocking than previous card(s) (Asus 980 and 780ti in same mobo), when PCIe Gen 3 is enabled. Forcing Gen2 immediatelty allows greater overclock of the BCLK.

I say this, because where it relates to you is that if I push my BCLK too far, everything will be fine, but when that crash happens 3 hours into a game or whatever, on reboot it fails to POST with an error code. I've never bothered to look at it mind, but it's a triple beep so memory related.

The thing is, on every boot the motherboard 'trains' the memory and ultra tweaks a whole bunch of ultra fine timings never exposed to the user. Temperature affects everything, so this training has to be done on every boot. ON soft reset, it doesn't bother, and so what happens is that the machine has gotten warmed up so that the original training is now out of whack, so on reboot it fails. Full power off and on again, and thus force a full re-train, and it'll POST just fine.


That's basically what I believe is happening to you.


I'd say you need to boost voltage to the SA portion of the CPU. Why now and not before? Couldn't say, but hope this helps in some way.
Click to expand...


So I just had the opposite experience (weird).

I tried if doing a full shut down and power on (instead of just warm rebooting) would help.

First time: Code 43.

Hmm What if I disconnect power for a few seconds and then boot?

Second time: Code 43.

Grr.

Retried four more times shutting down and powering up. Code 43.

Then, on a whim, I just did a warm reboot to see what would happen, and it came right up. (??)

Something really weird is going on here. I certainly hope reinstalling Windows helps, otherwise I am going to have to take it out of my loop for an RMA which will be a major pain in the ass. I still can't help but think its something wrong with the Nvidia drivers / WDDM2.1 though because this only happens with the 372.xx drivers.

I'm thinking maybe an incompatibility between 372.xx drivers and the x79 platform, or something like that.
 
Zarathustra[H] said:
So I just had the opposite experience (weird).

I tried if doing a full shut down and power on (instead of just warm rebooting) would help.

First time: Code 43.

Hmm What if I disconnect power for a few seconds and then boot?

Second time: Code 43.

Grr.

Retried four more times shutting down and powering up. Code 43.

Then, on a whim, I just did a warm reboot to see what would happen, and it came right up. (??)
Click to expand...

Actually, never mind, I take this back this time. I think it might have just been a loose monitor cable tricking me into thinking it had gone into code 43.

I did unplug and reattach the HDMI cable right before the last attempt.

There are no event errors for today, like there were yesterday or the days before. Maybe - just maybe - the cold boot trick worked.
 
spine said:
I'm just gunna share this, for what it's worth...


As an avid overclocker, I've found that my Titan XP is much more sensitive to BCLK overclocking than previous card(s) (Asus 980 and 780ti in same mobo), when PCIe Gen 3 is enabled. Forcing Gen2 immediatelty allows greater overclock of the BCLK.

I say this, because where it relates to you is that if I push my BCLK too far, everything will be fine, but when that crash happens 3 hours into a game or whatever, on reboot it fails to POST with an error code. I've never bothered to look at it mind, but it's a triple beep so memory related.

The thing is, on every boot the motherboard 'trains' the memory and ultra tweaks a whole bunch of ultra fine timings never exposed to the user. Temperature affects everything, so this training has to be done on every boot. ON soft reset, it doesn't bother, and so what happens is that the machine has gotten warmed up so that the original training is now out of whack, so on reboot it fails. Full power off and on again, and thus force a full re-train, and it'll POST just fine.


That's basically what I believe is happening to you.


I'd say you need to boost voltage to the SA portion of the CPU. Why now and not before? Couldn't say, but hope this helps in some way.
Click to expand...



So, the problem persists with the 373.xx drivers, but miraculously disappears if I DDU and reinstall 369.xx drivers.

The doing a col reboot thing seems to help, MOST of the time, but not every time.

I am starting to suspect (but I am not 100% certain yet) that it might be related to my headphone amp.

I know it sounds odd, but I could swear that the problem is less likely to occurr if I wait to turn on my headphone amp until after the computer is booted. It is odd, I know.

I've wanted to get a dedicated external DAC for some time now, so maybe I'll just scrap the Titanium HD in favor of a Schiit DAC with a toslink connector. Even if there is a ground fault in my amp, it won't reach the computer via toslink!

Still odd that this would only occur with 372.xx and newer drivers though.
 
Zarathustra[H] said:
So I just had the opposite experience (weird).

I tried if doing a full shut down and power on (instead of just warm rebooting) would help.

First time: Code 43.

Hmm What if I disconnect power for a few seconds and then boot?

Second time: Code 43.

Grr.

Retried four more times shutting down and powering up. Code 43.

Then, on a whim, I just did a warm reboot to see what would happen, and it came right up. (??)

Something really weird is going on here. I certainly hope reinstalling Windows helps, otherwise I am going to have to take it out of my loop for an RMA which will be a major pain in the ass. I still can't help but think its something wrong with the Nvidia drivers / WDDM2.1 though because this only happens with the 372.xx drivers.

I'm thinking maybe an incompatibility between 372.xx drivers and the x79 platform, or something like that.
Click to expand...

you're beating yourself up about this, but if nobody else has the problem, then it's probably not the driver
 
You must log in or register to reply here.
Back
Top