|
I just had blocks like this start jumping around on the screen, plus grey triangles extending from the center of the screen out to the edges flickering while playing Fallout New Vegas. Does it look like my graphics card is going bad? It's a ~3 year old Radeon HD 5770 and has been overclocked by about 10% using the AMD Overdrive utility. If it is going, can underclocking it and turning down graphics settings prolong its life for at least another few weeks?
|
#
¿
Feb 24, 2014 00:32
|
|
|
|
| # ¿ Apr 16, 2024 09:28 |
|
|
Haha, I know how buggy New Vegas is, that's why I was asking. KSP and Minecraft seemed OK even though the card sounds like a jet taking off on all 3. I underclocked it by about 10% and ran the FurMark burn-in test for 10 minutes keeping it at 99°C and there were no problems, and I played another hour of Fallout with the settings turned down a few notches without problems. I'm crossing my fingers that it was just a one-time glitch or a problem with the replacement textures I have installed, but one of my friends gave me his old GTX 260 in case my card dies.
|
|
#
¿
Feb 24, 2014 19:12
|
|
|
Zero VGS posted:Yeah, you might take a look and see if you need to take a can of duster to it and/or redo the paste. Duster and paste are both really cheap so you're not out much and you might keep the thing alive longer.  The aluminum fins on the heatsink were so clogged with dust that I had to go at them from both sides 3 or 4 times just to get the dust bunnies out and I went through half a can of air cleaning the whole thing. The thermal paste was pitted and brittle, I'm not sure what the stuff looks like when it overheats but I'll bet that's what happened. I was getting 12fps at 99°C on the FurMark burn-in test before, I'm getting 19fps@69°C now. I'm going to be a lot more proactive about cleaning my computer now and at least check the heat sink I can't see into with a small mirror next time.
|
|
#
¿
Mar 6, 2014 04:53
|
|
|
If the newest game I'm playing is Fallout 4 at 1080p, a GTX 1060 6GB should be fine, right?
|
|
#
¿
Apr 27, 2017 18:59
|
|
|
PC is an ASrock B550 Pro, Ryzen 5 3600 with 16GB of RAM. I have an RX 5700 XT and everything seems fine except for junction temperature, which tends to spike to 110° and cause a crash to desktop maybe 20% of the time and a hard restart the other 80% I made sure fans and heat sinks are clean, and the card's fans and my case fans are keeping everything else to a max of 90° with the occasional spike to 96. It's done it with Detroit: Become Human, Deus Ex: Mankind Divided , Control (I know it's really resource intensive), and Ghost Recon: Wildlands. I'm running one monitor at 2160p30 and one at 1080p60m plus whatever model the original Oculus Rift is. It's happened across the past 5 or 6 driver updates, so I don't think that's the issue. Any ideas?
|
|
#
¿
Oct 15, 2021 03:16
|
|
|
This is the case I have. https://www.newegg.com/black-corsair-275r-airflow-atx-mid-tower/p/N82E16811139147 I'm going to run Ghost Recon for a while and take a screenshot of what other temps are to make sure I'm remembering right, I'm not really familiar with what proper temperatures are and I could be remembering wrong. To clarify, 90C is when I'm running games, this is where I'm at with just a browser and background apps (1Password, Razer Synapse, 3DConnexion driver/software, Steam, ExpressVPN, qBittorrent, Greenshot). Does that seem normal or is it far off enough that I should just cut my losses, backup everything, and start from scratch?  edit: I popped open the case to double check and everything looks normal. I vacuumed it since there was a small amount of visible dust on fans, but that was all. Room temperature is 23C GWBBQ fucked around with this message at 04:33 on Oct 15, 2021 |
|
#
¿
Oct 15, 2021 04:16
|
|
|
I just kept it to the case fans and filters, not the electronics. I quickly grabbed a few screenshots while playing Control and trying to get as much poo poo on screen as possible. I was definitely remembering wrong about other temperatures. The only thing more frustrating than trying to figure this out is the fact that I can't reliably reproduce it.     
|
|
#
¿
Oct 15, 2021 04:49
|
|
|
Dr. Video Games 0031 posted:The other potential cause is a bad mount on your GPU cooler. Fixing this would require taking your graphics card apart and re-applying thermal paste, if that's something you'd be comfortable with. Comedy option: just strip the fans and panels off of the case and put in in a 40 gallon fish tank of mineral oil.
|
|
#
¿
Oct 15, 2021 05:35
|
|
|
Dr. Video Games 0031 posted:Thermal Grizzly Kryonaut is the current high-end thermal paste of choice among enthusiasts. It's popular to the point where finding a legitimate source for it that's guaranteed not to be cheap counterfeit poo poo is a bit of an issue. If someone knows a source for it that is known to be reliable (and not just "I bought it and it was fine:") then I'd recommend that. Else, Noctua's NT-H2 is pretty good as well: https://www.amazon.com/Noctua-NT-H2-Thermal-Cleaning-Performance/dp/B07MZ45X9G/ref=sr_1_12?dchild=1&qid=1634274756&sr=8-12 Thanks a lot for walking me through this, I'm really unfamiliar with much of anything other than the work side of things where I spend 15 minutes with a Dell support rep and the next day we hang out with one of two techs from our immediate area while they fix whatever isn't working. GWBBQ fucked around with this message at 06:51 on Oct 15, 2021 |
|
#
¿
Oct 15, 2021 06:48
|
|
|
|
| # ¿ Apr 16, 2024 09:28 |
|
|
PittTheElder posted:I'm not an expert by any means but that idle temp of >60 degrees seem crazy high. Maybe by EVGA 1080 just is different, but it usually idles at like 35-40 degrees. What does your idle CPU temp look like? Are your case fans exhausting properly? 
|
|
#
¿
Oct 16, 2021 02:08
|
|