Scared of Blender

(Roken) #1

Probably not scared - but I’ve had a bad week. I bought a shiny, new 32" QHD monitor 3 months ago, which died last week. Return and replaced under warranty. On the day the replacement arrived, my gfx card died in a puff of smoke (literally). Also under warranty.

Now since these are both display, I have to wonder, coincidence, or related?

Everything seems to be working now, but Blender scares me at the moment. Should I simply stop GPU rendering, and leave CPU going all night, or chalk it up to an unhappy coincidence and carry on as normal?

0 Likes

(SterlingRoth) #2

It’s probably all a coincidence.

Especially the monitor, there’s no way I know that a program displaying things could damage a monitor.

Though, if the gpu was malfunctioning, that’s a more likely candidate for damaging the monitor.

Most likely they were just defective units and you happened to get lucky twice in a row.

I wouldn’t worry about blender, or any other program, at all.

0 Likes

(Avery) #3

Their can be some unfortunate circumstances that come up causing a lot of headaches when it comes to computers and 3d art. But you really just have to keep persisting through it! It can be hard, but if you don’t keep going just because your worried about your computer breaking down, you’ll never be able to keep working and getting better at your work.

0 Likes

(Renzatic) #4

Yeah, it’s either a coincidence, or you had a flaky GPU.

Blender is just a desktop app. It runs at the same resolution and refresh rate as anything else that’d be open on your desktop, like, say, your browser, or Spotify. There’s nothing specific about it that’d burn out your monitor.

Now, as has been mentioned above, a malfunctioning GPU can kill a monitor. It’s rare, but it happens. Blender is a pretty GPU intensive program, so if your GPU is just two hairs away from flaking out entirely, and you start getting hot and heavy with some Blender action, it can spaz out, and do some collateral damage before offing itself.

Course this isn’t Blender’s fault directly. Running any modern graphics intensive game could do the same amount of damage for much the same reasons. With everything said and done, I’m about 99.999997% sure it was faulty hardware, and it just happened to be Blender that broke your particular camel’s back.

0 Likes

(Roken) #5

As it happens. when it died (the GPU) I wasn’t doing anything heavy at all - in fact, I was replying to a forum post, and the system shutdown without any warning. Pulling the side panel off and trying to power up again was when i spotted the smoke. A couple more tests confirmed, and removing the gfx card started my system with a successful POST beep, followed a little while later with the “No VGA found” beep error code.

My worry is, has stressing the GPU previously killed it, or was it already dying. I hope the latter, but I guess I’ll find out. Early stress testing has shown no issues.

0 Likes

(Renzatic) #6

There are a number of things that could’ve caused Blender to break your GPU. Was it an old card? Was it a budget model? Maybe it had a small manufacturing defect that went undetected until just that moment? How’s the cooling setup in your case?

Blender isn’t singularly hard on your hardware. That said, it is stressful enough that it could expose problems that you would never notice if you just used your machine to watch movies, or read emails. Like, for instance, your old card had a not quite properly seated heatsink, and it wasn’t dissipating heat efficiently. In that setup, a few multi-hour long GPU cycles renders could end up adding months or years of wear and tear to the card over a much shorter amount of time because you’re keeping up that stress.

…but if your card has all its bits and pieces in working order, then Blender’s fine. It isn’t any worse than running Doom 2016 at 4k for 2-3 hours.

0 Likes

#7

Stressing the gpu? If you made it operate above it’s recommended temperature, then yes, this might have damaged it. You might also investigate to see if it might have gotten a tiny bit of water from someplace. This too could have damaged the monitor.

Was the monitor old? Maybe the malfunction of the monitor damaged the gpu.

0 Likes

(sundialsvc4) #8

I would also very strongly(!) recommend that you have a qualified electrician immediately check the wiring of your house or office. What else is on the same circuit as that computer? Even if the circuit has “three prong plugs,” does it actually have a strong earth ground? (Could current be leaking into that “ground” wire from somewhere?)

An “uninterruptible power supply (UPS)” with an appropriately-generous rating is cheap insurance against power problems – provided that it is known to be attached to a good earth ground.

The services of a licensed electrician – someone who really knows what (s)he’s looking at and how to fix any problems that are found – is well worth the hour-or-so service call. You might discover things about your home’s wiring … :astonished: … that you never suspected. :fire:

0 Likes

(colkai) #9

I’d agree with this, things going up in smoke is rarely a good sign.
Even if it is coincidence, could be the GPU died because the heatsink wasn’t mounted correctly, or had shifted. The monitor going as well could indicate a spike, at least investing in a surge protector may be worth while.
We moved house and had a ceiling light which blew bulbs with disconcerting regularity. When I checked the light switch, I found one wire was just pushed home, either badly installed, or somehow, had loosened over time. I’m lucky, I was trained in telecomms and electronics so am pretty comfortable around electricity, but for the layman, I’d advise leaving it to a qualified person.

0 Likes

#10

It could have been your psu too, there might be something wrong with it, sending surges into your computer. And then there is that one, a power surge might have struck you, from somewhere, maybe a storm?

0 Likes

(Roken) #11

Thanks, all. In no particular order:

I’ve ruled out PSU. It’s still quite new, and more than capable of handling the load. In any event, I would have expected collateral damage if it was a surge from the PSU before it got to the gfx card. Similarly, the wiring (and there’s plenty on the circuit more fragile). No storms at the time (or even recently).

I’m pretty sure the monitor was a fault with the monitor. Not a complete failure, but a single red line running straight down the display.

When I say stressing the GPU, I don’t mean past tolerance, I simply mean consistent heavy load.

I’m currently leaning towards a GPU fault - possibly manufacturing, possibly a cap dead, or died. There’s been no moisture got to it, and this seems likely.

Five days in with the replacement (same model, different manufacturer - EVGA old, MSI new), and no problems so far (though the last worked for 8 months, so who knows). The key difference is the old used 8 pin supply, whilst the new uses the reference 6 pin supply, so maybe that made a difference (and yes, reference design is 6 pin supply).

Guess I’ll just have to keep an eye on things and keep my fingers crossed.

0 Likes