|
Problem description: Recently bought a new MSI 970 GTX. It came with a Ubisoft game coupon, picked Far Cry 4. My poor game choices aside, it also makes my computer crash. Hardware BSoD, which I had never seen before now. I wouldn't have thought the game crashing would cause these, but I am unsure to blame the card or the game, and most definitely want to rule out the graphics card. Crashes seem to happen rarely, often only after an hour or five of gameplay. These are as mentioned rare, only five events over a week, but all have produced a Hardware BSoD (although they do mention recent software changes too, but that's a stock line) and four of the five produced associated minidumps. When the problem first arose, last weekend, I experienced three of these unexpected crashes and restarts in an eighteen hour period. After some other checking I also verified FC4's files, which did have some issues apparently, and was crash free for the next few days. Then it crashed again. That was a few days ago, but I haven't had much time for either gaming or investigating my issues with it. Recent changes: The new MSI 970 GTX as mentioned, also purchased 32gb of RAM, (only 24 of which is installed currently) and a new PSU, which is currently also not installed. The RAM is XMP enabled, and running well at 1.5V, replacing 6GB previously at 1.65V. Both (old and new) are 1600/CL9. Attempted fixes: At first, checked for similar issues, checked video drivers (347.09) and GPU Bios (which is apparently a thing to check now) were up to date and not known for being problematic. Thoroughly tested the RAM with Memtest86+ V5.01 (at least four complete error free passes, over two sessions totalling about 12hrs) so am confident it's not a problem. My inexpert analysis of the minidumps generated point towards error checking routines in my CPU, some blame farcry4.exe some point towards windows driver faults. Windows reliability monitor mentions 'Video hardware errors' in most events. I am somewhat unsure what to make of all this, hence the posting here. Haven't been able to make any other game crash but have not tried much, only a few hours of The Talos Principle. (Which runs better and with more stability, being built for PC.) Current investigations/fixes: Have disabled NVidia gameworks graphic settings in FC4, also reducing graphics quality and trying to see if this ameliorates things. The fact that crashes are rare and not so easily provoked makes this honestly a slightly tedious prospect. Anyone have a recommendation for a benchmarking program that would help test my GPU? Honestly I don't care at all if Far Cry 4 is a buggy POS or not, but I would certainly like to know if I should return/replace my graphics card. As I have not experienced any other issues with windows and that the card is currently running my three displays (of which I only run games on one, @1920x1200) I am assuming the card is good? Maybe? It's possible further errors persist in FC4's files, despite further verifying completing successfully. Redownloading the 30gb of game is not a pleasant prospect. Also an uncertain one, as interruptions in the first download I suspect were to blame for the initial errors. System specs: Here's a snapshot of my system, courtesy of PC Parts Picker: OS: Win7 Ultimate, SP1 CPU: Intel Core i7-920 2.66Ghz Quad-Core Processor Motherboard: Asus P6T ATX LGA1366 Motherboard Motherboard and CPU are old (2009 old) and next to be replaced. CPU is running at stock clock of 2.66ghz, lots of thermal headroom on the CPU (see cooler below) to overclock and get a few more months out of it but haven't done so. Video Card: MSI GeForce GTX 970 4GB Twin Frozr V Video Card Memory: A-Data XPG V1.0 16GB (2 x 8GB) DDR3-1600 Memory X2 A note: only three of these four sticks are currently installed. Storage: Samsung 850 Pro Series 1TB 2.5" Solid State Drive OS and relevant game both installed on this drive. My PSU, Corsair HX620W also from 2009. Have the replacement (SeaSonic 860W) (absurdly over-powered) but still boxed. These parts are somewhat irrelevant to this issue but also installed: CPU Cooler: be quiet! Dark Rock Pro 3 93.3 CFM Fluid Dynamic Bearing CPU Cooler Storage: Western Digital AV-GP 3TB 3.5" 5400RPM Internal Hard Drive Used to be an external, now internal. Nothing special but it was a gift. Storage: Seagate Barracuda 500GB 3.5" 7200RPM Internal Hard Drive X2 These two drives are particularly old (older than 2009 even) and will be given away (with caveats and warnings) soon. Storage: Western Digital Caviar Black 1TB 3.5" 7200RPM Internal Hard Drive X2 One of these was formerly my OS drive, now general storage. Location: Canadaland I have Googled and read the FAQ: Where do you think I got this template from? Also yes. Other notes: In short I would like to know if I have a problem with my GPU, and should I return/exchange it, and the best way to determine this. My Steam gift inventory is available to help recompense particularly helpful posters, should things be particularly involved. Pastebins of WinDbg's take on this minidumps: http://pastebin.com/JkfMM23x http://pastebin.com/uqSL8D6g http://pastebin.com/RryJabMw http://pastebin.com/CEz20muv Teledahn fucked around with this message at 21:33 on Jan 10, 2015 |
#
?
Jan 10, 2015 21:28
|
|
|
|
| # ? Apr 24, 2024 21:23 |
|
|
Have you experienced the issue with any other games? While your initial verification check on Far Cry 4 indicates the issue could be your hard drive/RAM, it could also be a fluke. There are two things I can think of to try. 1) Move Far Cry 4 to one of your other hard drives, and see if crashes stop. If they stop, probably the HD is the issue. 2) Try other games and see if crashes persist in a similar manner. If they persist, could still be anything (GFX, HD, RAM), but if they don't, probably something more specific to Far Cry 4
|
|
#
?
Jan 11, 2015 15:13
|
|
|
Honestly, I perhaps over-reacted a touch. Initial benchmarking and other tests aren't bringing up any problems. I am more than willing to accept that this bloody console port is far more error prone than my GPU. That said, Hardware BSoDs are hardly what someone wants to see after installing an expensive new piece of equipment. Going to keep on with the testing.
|
|
#
?
Jan 12, 2015 07:59
|
|
|
If you have any further issues, make sure you have the latest motherboard BIOS. The latest BIOS is often required for newer hardware to work properly.
|
|
#
?
Jan 12, 2015 18:11
|
|
|
So I should have believed the minidumps when they said 'probable cause: genuine_intel'. Prime95 can bring down my system when benchmarking, haven't tried their torture tests. Should have looked harder at the CPU far sooner. I was about to say 'glad to have found the problem' but then realised I may have found a second. Think this is related possibly to running my ram at its rated 1600mhz, as my venerable i7 920 prefers 800/1033. We'll see.
|
|
#
?
Jan 24, 2015 20:58
|
|
|
|
| # ? Apr 24, 2024 21:23 |
|
|
The problem is almost certainly not the CPU, unless it is overclocked too far. Do verify the CPU isn't overheating via RealTemp, and see if the system can complete at least one full pass of Memtest86+. Memory issues wouldn't be too shocking given memory has age-related failures and you're running three sticks.
|
|
#
?
Jan 24, 2015 22:31
|
|