Recommended Posts

Hello everyone,

Don't really want to flood the forum or anything, but I seem to have a problem with my PC. Just as the title says, my computer crashes while rendering.

I've done some 3D renders of low- to medium-poly models without really complicated geometry in the past and everything has been working properly. However I've recently finished modelling a high-poly object which I really want to render. But the PC starts working on it, then crashes into a blue screen with a crash dump.

I've tried several different rendering programs and it happens with all of them. Sometimes it even crashes in pre-render phase, while I am setting the lighting options and such.

I'm guessing the problem lies in the hardware, maybe in RAM or CPU. But I don't have enough knowledge to confirm/deny that, which is why I came here. I don't think the problem could be heat-related since the chrashes happen 1-5mins from the start of rendering.

The PC specs are as follows:

CPU: AMD Phenom 9850 Quad-Core 2.5GHz (poorly ventilated, not overclocked)

Motherboard: ASUS M2N68-AM

PSU: Aero Cool E80-700 700W

GPU: ATI Radeon HD 7850 (overclocked, dual fans)

RAM: 4GB (can my motherboard support more?)

OS: Windows 7 Ultimate 64bit

I think that pretty much covers my problem. Any help would be appreciated.

Link to comment
https://www.neowin.net/forum/topic/1129444-pc-crashes-while-rendering-in-3d/
Share on other sites

Hello everyone,

Don't really want to flood the forum or anything, but I seem to have a problem with my PC. Just as the title says, my computer crashes while rendering.

I've done some 3D renders of low- to medium-poly models without really complicated geometry in the past and everything has been working properly. However I've recently finished modelling a high-poly object which I really want to render. But the PC starts working on it, then crashes into a blue screen with a crash dump.

I've tried several different rendering programs and it happens with all of them. Sometimes it even crashes in pre-render phase, while I am setting the lighting options and such.

I'm guessing the problem lies in the hardware, maybe in RAM or CPU. But I don't have enough knowledge to confirm/deny that, which is why I came here. I don't think the problem could be heat-related since the chrashes happen 1-5mins from the start of rendering.

The PC specs are as follows:

CPU: AMD Phenom 9850 Quad-Core 2.5GHz (poorly ventilated, not overclocked)

Motherboard: ASUS M2N68-AM

PSU: Aero Cool E80-700 700W

GPU: ATI Radeon HD 7850 (overclocked, dual fans)

RAM: 4GB (can my motherboard support more?)

OS: Windows 7 Ultimate 64bit

I think that pretty much covers my problem. Any help would be appreciated.

You need to remove your GPU OC before you can continue troubleshooting, anything OC'd can cause bluescreens

Also install something like coretemp to monitor your temps while you render

CPU: AMD Phenom 9850 Quad-Core 2.5GHz (poorly ventilated, not overclocked)

NEXT!

People need to understand, if you don't get good air flow around the CPU your going to kill the machine.....try wrapping your head in 3 or 4 scalfs then go to the gym...you're gonna collapse.

I would look at the CPU and GPU temps. Also try running memtest to make sure the ram isnt taking a dump on you. Try running a benchmark also like 3d mark just to rule out the OC on the GPU as well.

Try and post a screen shot of the BSOD if you can. That will also help.

remove the GPU and attempt to render the 3D again

What do you mean? Just to remove the GPU entirely from the case and try to run the render that way? Will it even work?

CPU: AMD Phenom 9850 Quad-Core 2.5GHz (poorly ventilated, not overclocked)

NEXT!

Even though it's poorly ventilated as I mentioned, it has never suffered from excess heat and the temperature readings have always been low enough. The CPU has a little fan on it, not sure about the effectiveness of it.

What do you mean? Just to remove the GPU entirely from the case and try to run the render that way? Will it even work?

Even though it's poorly ventilated as I mentioned, it has never suffered from excess heat and the temperature readings have always been low enough. The CPU has a little fan on it, not sure about the effectiveness of it.

Problem is now you are stressing the CPU more than likely almost to it's limit....you are rendering 3D we are not talking about burning a few dvd's 3D rendering is crazy on the CPU

What do you mean? Just to remove the GPU entirely from the case and try to run the render that way? Will it even work?

Even though it's poorly ventilated as I mentioned, it has never suffered from excess heat and the temperature readings have always been low enough. The CPU has a little fan on it, not sure about the effectiveness of it.

How do you know your GPU is overclocked if you don't know how to remove the OC ?

You should also stick a fan at the front of your case sucking in, and a fan at the back blowing out, that way you always have a 1 directional airflow moving hot air out and cool air in

The GPU is over-clocked by ASUS and I can manually tweak the clock speeds with the software provided, so I assume it's a legit OC (it even says it on the packaging). I can max out the GPU Clock at 1050MHz, GPU Voltage at 1165mV and Memory Clock at 5800MHz.

Ever since the last Windows reinstall in autumn I haven't been monitoring the temperatures. Can you recommend the best tool for the job?

I would look at the CPU and GPU temps. Also try running memtest to make sure the ram isnt taking a dump on you. Try running a benchmark also like 3d mark just to rule out the OC on the GPU as well. Try and post a screen shot of the BSOD if you can. That will also help.

Please tell me how to do the memtest and benchmark, I've never done it before. Also, what's a BSOD?

Firstly the GPU is unlikely to be the culprit if he's rendering, he hasn't specifically mentioned what render suite(s) he's using, but they all crash, and most render suites won't use the GPU for rendering, unless specifically set to it and most can't even then.

HE probably hit on the problem in his first post. bad memory. Rendering especially high poly rendering uses a LOT of memory, in fact it's probably the first time he's ever been close to using all of his computers memory.

DO you have 1 4GB stick or 2 2GB sticks ? if it's 2 sticks, remove one at a time and check.

also what resolution are you rendering the output at ?

The GPU is over-clocked by ASUS and I can manually tweak the clock speeds with the software provided, so I assume it's a legit OC (it even says it on the packaging). I can max out the GPU Clock at 1050MHz, GPU Voltage at 1165mV and Memory Clock at 5800MHz.

If you have not moved the sliders for the OC, then it is a factory OCd card from ASUS, if you have moved the sliders, then it is a Custom OC by you

Put the sliders to where ever they were when you got the card, or disable the overdrive check box and hit apply

BSOD = Blue Screen Of Death

For monitoring

GPUz - GPU temps

CPU-z or coretemp - CPU temps

Firstly the GPU is unlikely to be the culprit if he's rendering, he hasn't specifically mentioned what render suite(s) he's using, but they all crash, and most render suites won't use the GPU for rendering, unless specifically set to it and most can't even then.

HE probably hit on the problem in his first post. bad memory. Rendering especially high poly rendering uses a LOT of memory, in fact it's probably the first time he's ever been close to using all of his computers memory.

DO you have 1 4GB stick or 2 2GB sticks ? if it's 2 sticks, remove one at a time and check.

also what resolution are you rendering the output at ?

My main renderer with which I work with is Keyshot 2, but Vray, Indigo and others failed too. I have 2 sticks of 2GB of RAM. But how does removing one stick at a time help me? Do you think one of them could be corrupt or something?

If you have not moved the sliders for the OC, then it is a factory OCd card from ASUS, if you have moved the sliders, then it is a Custom OC by you

Put the sliders to where ever they were when you got the card, or disable the overdrive check box and hit apply

BSOD = Blue Screen Of Death

For monitoring

GPUz - GPU temps

CPU-z or coretemp - CPU temps

I have tried rendering with various clock settings and it's still the same, although maybe I didn't try rendering with stock clock speeds.

I have tried rendering with various clock settings and it's still the same, although maybe I didn't try rendering with stock clock speeds.

Well my last 6850 would crash either the display driver or the machine if I even OCd 1MHz

Again, you can not troubleshoot an OCd machine, I`m not saying it is 100% the GPU, but you need to have everything @ stock

Well my last 6850 would crash either the display driver or the machine if I even OCd 1MHz

Again, you can not troubleshoot an OCd machine, I`m not saying it is 100% the GPU, but you need to have everything @ stock

Would setting the clock speeds to stock card speeds (from google) do the job?

EDIT: Well, as it turns out, I cannot go below Memory Clock speed 3880 MHz and google says the stock speed is 1200MHz, so...yeah

Edited by Redix

Go download memtest86 and burn a cd from it. Pop the CD in and reboot the computer and boot to the CD. Run the test with 1 stick at a time them pop them all back in if they pass and run a test with them all in. That will rule out the Ram for sure.

A screenshot of the blue screen would be a big help also. But rule out he ram first. Trust me.

Would setting the clock speeds to stock card speeds (from google) do the job?

EDIT: Well, as it turns out, I cannot go below Memory Clock speed 3880 MHz and google says the stock speed is 1200MHz, so...yeah

Just uncheck the Overdrive option in CCC and hit apply, the card will drop to its factory clock speeds

Capture.PNG

So, I was surprised when it reached 15% of the rendering process...but then it crashed again, so removing the OC doesn't seem to help. Also, the BSOD.

Stop 0?00000050: PAGE_FAULT_IN_NONPAGED_AREA ? This Windows stop error indicates that information that the operating system was looking for was not found in memory (ram). This occurs when the system looks for information in the ram that doesnt exist. This is usually caused by defective RAM (once again, including Ram on your motherboard, L2 cache and video card ram).

Get your ram testing on ;)

Stop 0?00000050: PAGE_FAULT_IN_NONPAGED_AREA ? This Windows stop error indicates that information that the operating system was looking for was not found in memory (ram). This occurs when the system looks for information in the ram that doesnt exist. This is usually caused by defective RAM (once again, including Ram on your motherboard, L2 cache and video card ram).

Get your ram testing on ;)

I'd have thoroughly tested RAM today, but it turns out I have no empty CDs left which I can use and the nearby stores have none. I hate putting off such tasks but I just have no options right now. So tomorrow I'll go to a bigger store further away, find & buy those CDs and start working.

Hopefully the problem is one of my regular RAM units and not motherboard RAM or anything like that. That would be an excuse to buy more RAM.

Though what I did today was I opened the case to provide better ventilation for any components that could be overheating. I started rendering once again, and checked if there was any excessive heat coming from somewhere. No, there wasn't. And the air coming from the fan on the CPU is just mildly warm under full rendering. So maybe we can abandon the overheating theory.

Hopefully the problem is one of my regular RAM units and not motherboard RAM or anything like that. That would be an excuse to buy more RAM. Though what I did today was I opened the case to provide better ventilation for any components that could be overheating. I started rendering once again, and checked if there was any excessive heat coming from somewhere. No, there wasn't. And the air coming from the fan on the CPU is just mildly warm under full rendering. So maybe we can abandon the overheating theory.
Using your PC with the side panel off will only disrupt the normal airflow of the PC and make it not as cool as it could be; it's a very poor way of measuring temperature.

And I don't know what you mean by regular RAM and motherboard RAM...there's only one type of RAM inside your PC, and those are the two 2GB sticks in their slots...

This topic is now closed to further replies.
  • Posts

    • YouTube has finally brought back its DMs feature, but only in these countries by David Uzondu Late last year, YouTube started testing a "new" way to share videos directly with friends, without having to leave the app. Now, the video giant has announced that is now rolling out a revamped direct messaging inbox, which lets you share videos, Shorts, and live streams and have conversations about them, directly on YouTube. The platform limits this feature to 18+ users who are signed in to a verified channel and use the latest mobile app version. Direct messaging on YouTube first became a thing back in 2017 inside the mobile app (later renamed to "Messages"), where users could chat one-on-one and share clips directly, but all that came to an end on September 18, 2019, when Google decided to shut it down after giving users a month to download a .zip file archive of their past chats. No one really knows why YouTube killed the feature, but users were encouraged to migrate to the public Comments section, on Community tab posts, and via YouTube Stories. The previous incarnation suffered from moderation challenges, prompting Google to implement stricter safety guidelines and age verifications for this new iteration. Here's a list of the countries where the re-launched feature is currently available, though note that Brand Accounts do not have access to it, at least for now: Countries American Samoa Austria Belgium Brazil Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Guam Hungary Iceland Ireland Italy Latvia Liechtenstein Lithuania Luxembourg Malta Netherlands Northern Mariana Islands Norway Poland Portugal Puerto Rico Romania Singapore Slovakia Slovenia Spain Sweden Switzerland U.S. Virgin Islands United Kingdom United States Before you can use the feature, you first have to send an invite link to your contact. Invite links expire exactly seven days after you create them. If the person on the other end accepts the invite, you can exchange videos directly and text back and forth inside the app. To delete a message, just long-press on the message and tap unsend to remove it for both users. You can also delete entire conversations by long-pressing the thread and selecting delete, but the other person will continue to see the chat history on their end. To make sure everything remains safe, YouTube monitors these messages to ensure they follow Community Guidelines.
    • The problem of course is simply that government does not always know best. My point is that agency is taken away from the EU consumer in these cases. I'm sorry, but I do not believe that governments (politicians) are inherently good, and "looking out for me." Primarily they look to themselves and their own personal desires first, foremost, and always. When the EU or the DOJ fines these companies, claiming to "represent the welfare of the consumer," how much of these billion-dollar judgments are handed to the consumers they claim to represent? Not even a dollar, as I've seen. Yet the EUC lawyers who are paid to sit around and dream up these suits make huge commissions on the fines the EUC adjudicates, which is an ironclad fact I hope everyone is aware of. It's also rank corruption, of course, but that's another topic. Last, when the EU inflicts these judgments, or the DOJ, take your pick, the costs are bundled right along in the cost of the goods and services these companies provide the consumers they are "looking out for." If you are someone who believes his government is his savior then you have my condolences. I think Apple is right here, because the whole scheme of consumer choice is that consumers pick and choose among the products companies offer. Microsoft Windows is more compatible with third party software and hardware than any desktop OS on Earth, which is my sole reason for choosing it. Just because the EUC forces companies do certain things it knows the companies do not want to do, "or else", has no bearing on consumer benefit. This Siri thing is almost idiotic it's so infantile. But this is what the EUC does when the EU in Brussels becomes cash-strapped and needs a big infusion of cash. Some people get upset by "big companies" but it's the opposite when governments dwarf the size and scope of these companies, which is so obvious it hurts.... I mean you can't honestly believe that forcing Apple to do things with Siri it has its own reasons to decline is something that "opens up" Apple, do you? Say it aint' so...
    • Looks like many years since the request was made, a directory tree view finally may be added. https://github.com/files-community/Files/pull/18537
    • Is it still super slow or has it improved on that area?
    • There's this from last year https://gist.github.com/threat...364659a8887841aa43deca4efd9 but nothing about a buffer overflow that MS somehow can't code against. No matter what, it makes sense to take a "protected by default" approach.
  • Recent Achievements

    • One Month Later
      sjbousquet earned a badge
      One Month Later
    • Week One Done
      sjbousquet earned a badge
      Week One Done
    • First Post
      DragonOfMercy earned a badge
      First Post
    • First Post
      bella52 earned a badge
      First Post
    • Reacting Well
      Techinmay earned a badge
      Reacting Well
  • Popular Contributors

    1. 1
      +primortal
      501
    2. 2
      PsYcHoKiLLa
      213
    3. 3
      +Edouard
      156
    4. 4
      Steven P.
      84
    5. 5
      FloatingFatMan
      72
  • Tell a friend

    Love Neowin? Tell a friend!