PC Crash while using DAZ Studio
mbug90
Posts: 53
Today I was using DAZ Studio and was posing a model when my PC decided to crash. It was the first time my new PC had a BSOD at all. The code was INTERRUPT_EXCEPTION_NOT_HANDLED
I'm on Windows 10.
Dump file here: https://1drv.ms/u/s!AhpXMWQ8tVpLfiElHb0KVHJ4KSU?e=SEH0qK
Post edited by mbug90 on
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
If its a one off, its a one off. You just cant read enough into it.
That said, Daz does usually find a way of exposing PCs that are not quite 100% stable more so than browsing and games I've found.
In the search bar type "reliability history" and you'll see a graphic showing all the important events that have occured on your PC in recent days/weeks, including driver installs, hardware problems, software crashes, etc.
Even though Windows is the OS we love to hate, it does have some very nice troubleshooting and monitoring features.
I went to that reliability history and found this. Don't know what it means though.
Source
Windows
Summary
Hardware error
Date
5/30/2021 4:11 PM
Status
Ready for upload
Description
A problem with your hardware caused Windows to stop working correctly.
Problem signature
Problem Event Name: LiveKernelEvent
Code: 193
Parameter 1: 804
Parameter 2: ffffffffc0000001
Parameter 3: 108
Parameter 4: fffff8057abdb770
OS version: 10_0_19042
Service Pack: 0_0
Product: 768_1
OS Version: 10.0.19042.2.0.0.768.101
Locale ID: 1033
Files that help describe the problem
WATCHDOG-20210501-1151.dmp
sysdata.xml
WERInternalMetadata.xml
WERInternalRequest.xml
memory.csv
sysinfo.txt
WERDataCollectionStatus.txt
Extra information about the problem
Server information: 909ad8fc-9638-4323-96b2-43df3d6b3e21
Also, here's an analysis of one of the dmp files:
Are you using a GPU? If so, were there any recent NIVDIA driver updates shown in Reliaibility History? My first thought with an error like that is a GPU driver problem or even a hardware problem. But since we know absolutely nothing about your system all we can do is speculate. Since it's a "new" computer, did you build it? Has it worked before with DAZ?
Device name WINDOWS-P35T52V
Processor Intel(R) Core(TM) i9-10900K CPU @ 3.70GHz 3.70 GHz
Installed RAM 64.0 GB (63.9 GB usable)
Device ID A94FDB29-EBBC-4A49-A20F-907948BD5789
Product ID 00326-00873-87071-AAOEM
System type 64-bit operating system, x64-based processor
Pen and touch Pen support
Edition Windows 10 Home
Version 20H2
Installed on 4/30/2021
OS build 19042.985
Experience Windows Feature Experience Pack 120.2212.2020.0
-------------
I am using a GPU. It's an nvidia GeForce GTX 1650. No driver updates were made recently and other users seem to be having problems with the latest version of Game Ready Drivers (see: https://www.nvidia.com/en-us/geforce/forums/game-ready-drivers/13/453674/geforce-46647-grd-feedback-thread-released-51821/). Another company built the PC for me. It has worked before with DAZ until that day.
Yeah, there's ALWAYS people having trouble with every NVIDIA driver release since the beginning of time. Apparently GPU drivers are the most difficult software on the planet to get right. You might try rolling back to an earlier version. Most new releases have almost nothing to do with most of us, and are designed only for the latest and greatest hardware and/or software. I generally wait 6 months or a year or more before updating GPU drivers, until the bugs get worked out.
I agree with jmtbank, if it only happened once I wouldn't even worry about it, let alone go searching for a problem. No computer is 100% reliable all the time.
A brand new computer that had a BSOD, and he shouldn't try to find out what went wrong? And if the cause of the problem was an error by the company that built it for him, or a hardware failure, he shouldn't pursue it, and he should just accept it? How many "one - offs" should he wait for?
How many? Maybe when it becomes a persistent problem. I'm not arguing that it's a concern but one-off glitches are part and parcel of computing. DS crashed for me yesterday on my new rig for the first time using dForce and it could be one of a thousand things gone askew, hardware or software. Just restarted and the simulation worked perfectly.
Just keep in mind that often the BSOD that the OP described is generally caused by either driver problems, or more likely, an actual hardware failure (as the error message described). The last time I had a BSOD on W10, years ago (in my experience BSOD's are extremely rare), it was due to a failing system hard drive, not a "one-off glitch". And especially if the hardware is under warranty I think it makes sense to pursue stuff like this, but that's just me.
But what in my PC could be failing? It's been running fine since then. Though I haven't tried using Daz Studio again yet.
Any hardware can fail at any time. As I said, maybe your system hard drive is failing, or maybe your GPU is having problems, or maybe your drivers are causing problems and need to be changed.
If it was me, I'd change my NVIDIA drivers to an earlier version, and run a disk check on my hard drive(s). And also try disabling your GPU and see if DAZ works okay without it (as in, no crashes when using only CPU, which would point to a GPU problem).
I'd also go thru the Reliability History and see what's been going on, and check the DAZ log file to see if there are any hints, and also check Task Manager while rendering to see what's going on with the GPU VRAM and what resources DAZ is using and so on.
Unfortunately, computers are a lot more complex than people give them credit for, and the only way you can find the culprit is look at the data.
I checked my system hard drive for errors (Properties -> Tools -> Error Checking). The test came back clean.
I've also attached the log file for DAZ though I don't see anything that looks remotely like a problem.
Also, one more thing... I wasn't rendering anything at all when it happened.
So the last log entries looked like it had been going along fine loading character textures, then presumably that's when the PC had the BSOD? So it was actually a blue screen of death, W10 crash, and not a DAZ crash?
My hunch is maybe it was trying to read another texture and encountered a hard drive failure. When I had my BSOD years ago due to a hard drive failure, I threw every disk test I could think of at it and they all said there was no problem, but on a hunch I just replaced the drive with a new one ( I recall the bad one was like 6+ years old or something) and that fixed it.
BTW, were you in Iray preview mode when you were posing the character? I've been having occasional DAZ crashes (not W10 BSOD's) when not rendering, but just working on the scene, but yours sounds like a system issue.
Maybe somebody knows of a reliable disk tester out there.
I've only had the PC for a few months now. It's inconcievable that my SSD hard drive would fail so quickly. I don't use Iray.
And how could it be a hard drive failure if Error Checking says it's fine?
Have you ran MemTest on it yet to see if you have any problems with the RAM?
https://www.techpowerup.com/memtest64/
No install needed.It's a stand alone diagnostic tool that runs on Windows.
If it's a faulty RAM card you may be able to get it replaced for free if you still have time on your warranty.
This is actually the first thing I run on a new PC build.I test each module separately if I have more than one so that it's easier to locate the faulty ones.
Non-ECC consumer grade RAM isn't as reliable as many would think.I've gotten plenty that seemed to run fine until a certain area was used then Windows would BSOD.
also, never assume a new electronic device is in perfect running condition.It's not uncommon to get RAM, hard drives, and just about anything else DOA(Dead on Arrival) from the factory.
Yeah, all electronic components since the beginning of time have a probability of failure over time that's defined by a U-shaped "bathtub curve". It basically says that any electronic component is far more likely to fail when it's brand new, or when it's really old. But once it's been in service for a while ("mid-life") it's less likely to fail, since it has proven it has no defects and is therefore fairly reliable.
Keep in mind that electronics are real complex and real tiny and require some insanely tight tolerances, so it' guaranteed that a certain percentage of new components will fail due to manufacturing errors, bad components/materials, installation errors, etc. So it's a coin toss if you'll be the lucky recipient of one of those failures. And depending on the manufacturer, the likelihoods could be much higher or much lower. Also, user error often creeps in and can cause unexpected failures.
IT HAPPENED AGAIN! I was posing a figure in DAZ Studio when my PC locked up and then BSOD'd!
Well I guess it ain't a one off. I never trust log files, but its flat out blaming the CPU. As you were posing a character, I probably also wouldnt assume it was the graphics card. Edit: Unless you were in Iray mode? If so, disregard everything else in this post and indeed do start by testing your gfx card - or power supply.
Normally I run a raft of other programs on my computers, so already have a feeling as to whether the machine is stable and its just the software. Usually a full machine lock up that happens frequently is a hardware fault and it could be worth running some game benchmarking or stress testing software to try to recreate a crash to learn more. I don't find memory testing sofware particularly thorough enough to find a 'once every month or two' type crash but are generally easy for people to use.
These people have written an article on PC stress testing. The bit you would consider starting with is the "Prime95 With AVX & Blend" : "Using the more conservative Blend test results in less load on the cores and more load on the memory."
https://www.tomshardware.com/uk/reviews/stress-test-cpu-pc-guide,5461-2.html
If you feel its just Daz and not hardware, then I'd reinstall it first.
How long should I run the test for?
The one they recommend sucks a lot of power, so I'm always lothe to let prime 95 and its like run and run. You hear people on tech forums claim to do 24hrs. I used to only run it for an hour or so.
We go back to my original comment. If you've only had 2 crashes in 4 months and it was hardware instability, you might never track it down if its such a rare occurance.
I had a crypto mining cpu that only developed memory instability on the hottest days this summer. I had assumed its underclock was fully stable up till that point.
Okay, so it's been a little over an hour and a half that I've been running the blend test and the CPU is holding.
Now what?
Ok, Try running Windows Memory Diagnostic just type it in your little search bar and run it. If it's your memory is bad you find out fast enough. It takes a while to run but let run. It will tell you while running you have errors. if it doesn't memory is good.
Windows Memory Diagnostic test came back clean
I also ran the Prime95 Blend Test for a little over 8 hours today with no problems.
This has been used to find problems when most of the stress tests have failed to find them https://www.ocbase.com/
One thing to check is, how much RAM do you have and how big is your swap file. There is a pattern of users with 8GB RAM to run out of usable memory which has caused crashes.