• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Is it safe to buy AMD again?

A few years ago I built a 3700x with a 5700XT and, while I've been pleased with the performance for the price (the gpu was about 60% of the price of anything equivalent from nvidia at the time) it was only after I built it that I found all the posts and reddit threads etc about the instability of the GPU.

I get great benchmark results for the spec, can't complain but wow do I get a lot of crashes. Despite all the driver updates since, it's still unstable. Certain things really trigger it, like cutscenes in games, or if I pause a game and then the screensaver comes on (I know, right?). Actual performance while I'm playing is great.

But, I swore I'd never buy a radeon again. Is it safe to? Are the new models stable?

With all those issues I'd say you either have a faulty card or something else is the issue in your system
 
my card the amd 5700 has been through 2 mainboards and 3 different cpu's and no problems except for the random rare crash that any pc can have.

tuCtRwP.png


installed a new psu earlier so been running benchmarks a bit to make sure its stable :)
 
Last edited:
@Stubbsy - It's a good idea to check for a new BIOS for the card, with tools like EVGAs precision X1 (or whatever brand your card is), then let it auto update.

Check the card isn't overheating, e.g. GPU and hotspot.

Check there's nothing odd going on with the GPU clocks and memory clocks under load, they might be boosting too high. You can use AMDs software to force lower clocks.

Update your motherboard's chipset drivers.

A faulty or unreliable power supply is quite likely too. If you have another PC to test the card in with a decent power supply, I would try that.

There's a chance it could be a BIOS setting causing issues. Can be hard to track down though...

It seems a bit silly to write off a company because not everything went smoothly for you (this can happen whatever company you buy from), these days I would say you don't have the luxury of choice with the market the way it is, unless you have a lot of spare cash.
 
Last edited:
If buying AMD by Sapphire.

You'll be fine they are IMO the best GPU manufacturer, and have been for a long time, I would say EVGA (Nvidia) were on par but they are out of the game now.
 
I was surprised this morning to see so many replies to this thread thanks.

I recognise that perhaps I've been more unlucky than some, but I'm sure I'm not the only person with issues with this card given the amount of threads I found around the time (of course it's all legacy now).

The specific card is the PowerColor Radeon RX 5700 XT Red Dragon 8GB. I haven't changed any of the default setting in Wattman. Of course I did once, but as it resets every time the machine crashes, I cba to change any more.

The PSU is Seasonic Core Gold GC-500 500W 80+ Gold Power Supply.

MB is MSI B450-A PRO MAX and I'll check the bios on it, and the card today, thank you for the suggestions.

I don't think it's getting too hot, but what are the best tools to log temperature, please?

This will make you laugh, the Roblox windows app makes the machine crash 100% of the time. It's got to be a driver issue, as that game isn't very taxing at all, compared to other things I play...
 
I like GPU-Z to monitor cards.

You could try rolling back just the display driver using device manager. Or, just install download an old driver package.

Try a really early one.

I personally think it's probably not driver related.

I wonder if it could be some odd incompatibility with the monitor, display resolution or refresh rate settings.

You could try disabling any G-sync or FreeSync settings on your monitor, then set different refresh rates (like 30, 59.9, 60 etc).

Or, just try a different monitor.

In the end though, I think you may need to test the GPU on a different PC.
 
Last edited:
there are people reporting similar issues then it starts looking like a pattern

maybe its got something to do with specific hardware combinations and may not impact you
 
Stubbsy I'm with you on the 5700XT. Worst card I've ever owned for random crashes and I've owned 100+ cards going back to the 4200ti. It did heavily seem to be a driver fault. After the first extremely poor six months the drivers got a lot better but occasionally AMD would release a new driver and it would act up like it used to before. Even then, when it was good there were too many driver crashes for me on graphically demanding games (RDR2 or MSFS2020). I also used that card on 2 builds all fresh upto date bios's, drivers and such.

These cards were turds. Good for all the "works on my machine" people but they were the lucky ones who got the polished turds.
 
Isn't there usually a way of checking if there's a driver problem?

Like BSODs etc. Or, it might show up in the Windows event log

It's very quick to try different display drivers by 'updating' the display driver from device manager.

I would uninstall the AMD software, to rule out an issue there.
 
Last edited:
I cannot fault the performance of the 6900XT for the price but the issues I was having with driver time outs and other oddities on desktop like black screens etc made me give up. A shame tbh.
 
The PSU is Seasonic Core Gold GC-500 500W 80+ Gold Power Supply.

Seasonic in general make excellent psu but this particular part of their range was replaced when Nvs Ampere was released. The newer gpu architecture has viscous power spikes that would cause the over power protection to kick in and the pc would shut down. It is a low powered unit but should be adequate for your needs.

I know you are saying you have driver issues but a driver can fail if the gpu is not getting power supplied correctly. If you could swap it out for a day to test it would be helpful.
 
Black screen system crash I’ve had this before on a very early motherboard bios. Usually while I was surfing or watching YouTube. From memory during the early days of the Ryzen 3000.
 
I had 5700xt Nitro+ for a few years in intel system with 6700k than in AMD system with 5900x and never had issues on either

you havent listed down rest of the spec , PSU etc , what kind of temps is the GPU getting for GPU, hotspot, vram and if running CPU and RAM at stock ?
 
Last edited:
I genuinely don't think I've had a single crash on my 6800xt rig, mind you I never had any issues with my radeon 7 either. I'm either super lucky or it's pretty safe :)
 
My way of thinking is: Get a good PSU, use one cable for each 8/6 pin plug and don’t install crapware, set the drives to default, don’t use overlays and other utils and things should just work. I’ve had AMD GPUs from 970 Pro, 980 Pro, 2*6990M crossfire in laptop, 290X, Fury X, 480XT, 5700XT, 6900XT and they all mostly worked without issues. Almost all the crashes I have had happened when I was using a util to check temp/frequency’s. The worst bug was with the Fury X, the drivers would not install, I had to use device manager to update drivers for around a year! it stll worked fine though.
 
Last edited:
My way of thinking is: Get a good PSU, use one cable for each 8/6 pin plug and don’t install crapware, set the drives to default, don’t use overlays and other utils and things should just work. I’ve had AMD GPUs from 970 Pro, 980 Pro, 2*6990M crossfire in laptop, 290X, Fury X, 480XT, 5700XT, 6900XT and they all mostly worked without issues. Almost all the crashes I have had happened when I was using a util to check temp/frequency’s. The worst bug was with the Fury X, the drivers would not install, I had to use device manager to update drivers for around a year! it stll worked fine though.


So basically it's not fit for purpose :)
 
Last edited:
Back
Top Bottom