• Notice regarding image hosts and attachments. See here.

  • 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.

The AMD Driver Thread

humbug said:
What's really annoying..

I don't know if this is Windows, AMD's drivers, Faulty PSU, Faulty GPU, i doubt its the CPU or RAM as i hard-stressed them overnight to rule them out, they passed with 0 errors, maybe the Motherboard? I bought the same board during covid and it arrived semi DOA, the No 1 PCIe X16 didn't work at all, i got sent a replacement, same board which has been trouble free for 6 years but whose to say the reason that PCIe slot was not working on the original is because its a weak point on this board?
I still have the original DOA board, rainforest refused to take it back. because of the lurgy?

Without having a spare PC or access to one its hard to diagnose, made even harder given that after swapping PSU cables after suffering back screen crashes it was perfectly fine again for a week. Its so random and intermittent.
Click to expand...

I watch some of those guys on Youtube that repairs electronics, and they've said that a lot of graphics cards with huge heatsinks just bow under their own weight and eventually crack the BGA solder under the GPU chips giving all kinds of different issues, many of which can be intermittent because of power load and thermal cycling/expansion.

There's a reason why a lot of modern motherboards have massively reinforced 16x PCIE slots and the GPUs come with substantial support brackets. They are like Frankenstein animals that have grown too big for their bones to support their weight.
 
Steampunk said:
I watch some of those guys on Youtube that repairs electronics, and they've said that a lot of graphics cards with huge heatsinks just bow under their own weight and eventually crack the BGA solder under the GPU chips giving all kinds of different issues, many of which can be intermittent because of power load and thermal cycling/expansion.

There's a reason why a lot of modern motherboards have massively reinforced 16x PCIE slots and the GPUs come with substantial support brackets. They are like Frankenstein animals that have grown too big for their bones to support their weight.
Click to expand...
www.techpowerup.com

Sapphire PULSE RX 7800 XT Specs

AMD Navi 32, 2430 MHz, 3840 Cores, 240 TMUs, 96 ROPs, 16384 MB GDDR6, 2438 MHz, 256 bit
www.techpowerup.com

That might be it but its not a huge card, its quite small by modern standards, it doesn't look like its sagging and the PCB looks straight, if i wiggle the card around a little while its running it doesn't do anything, it just keeps running...
I would have thought the first thing to go with that would be memory errors, that manifests as Error Code 43, i have never had that.

280 mm (L) × 128.75 mm (W) × 52.57 mm (H)
Weight is 1090 Grams

Installing some sort of support is tricky, i have an Antec C7 ARGB case, 3 fans under the GPU, on top of the PSU shroud, i might still get one as i think even if this is not the cause in time it could be if its not supported. Ill have to find one that doesn't need to use the PSU shroud to stand on.

I'm going to buy a new PSU next week, if the GPU is still playing up after that i'll RMA it, i have a few months left on the warranty.
 
I had same issue, in that I had zero space on the bottom of case for a stand thing, so in the end I just mounted vertically to be safe.
 
humbug said:
www.techpowerup.com

Sapphire PULSE RX 7800 XT Specs

AMD Navi 32, 2430 MHz, 3840 Cores, 240 TMUs, 96 ROPs, 16384 MB GDDR6, 2438 MHz, 256 bit
www.techpowerup.com

That might be it but its not a huge card, its quite small by modern standards, it doesn't look like its sagging and the PCB looks straight, if i wiggle the card around a little while its running it doesn't do anything, it just keeps running...
I would have thought the first thing to go with that would be memory errors, that manifests as Error Code 43, i have never had that.

280 mm (L) × 128.75 mm (W) × 52.57 mm (H)
Weight is 1090 Grams

Installing some sort of support is tricky, i have an Antec C7 ARGB case, 3 fans under the GPU, on top of the PSU shroud, i might still get one as i think even if this is not the cause in time it could be if its not supported. Ill have to find one that doesn't need to use the PSU shroud to stand on.

I'm going to buy a new PSU next week, if the GPU is still playing up after that i'll RMA it, i have a few months left on the warranty.
Click to expand...
Do you still have any of your old GPUs? might be able to use one of them to narrow things down
 
humbug said:
Installing some sort of support is tricky, i have an Antec C7 ARGB case, 3 fans under the GPU, on top of the PSU shroud, i might still get one as i think even if this is not the cause in time it could be if its not supported. Ill have to find one that doesn't need to use the PSU shroud to stand on.
Click to expand...
How about this?

My basket at OcUK:
Total: GBP20.98 (includes delivery: GBP11.99)​
 
Dicehunter said:
I know you can't say much due to corpo rules but realistically could we expect any work done for bringing FSR4 to Vulkan ? Several games, Namely Star Citizen, Use Vulkan and right now AMD users have access to either FSR2 which is plain horrible or the engines own upscaler which falls short of DLSS that is available to Nvidia users.

Even Star Citizens own devs have previously said they would love to implement FSR4 but due to AMD they cannot so AMD users are left with inferior upscaling options while Nvidia users get their own upscaler that provides superior image quality.
Click to expand...
i'm not commenting on that because i genuinely don't know or care all that much about fsr. i'm sure it's in progress, we'll just have to wait and see.
 
AMD_Vik said:
i don't know what to tell you, different people have different opinions and priorities. i don't tend to play games that run bad by default.
Click to expand...

I've never seen an Nvidia rep say they don't care about one of their companies popular technologies, It's not a good look.

And it's not about a game running badly, AI based upscaling can do a better job than a games AA implementation.
 
Last edited:
Dicehunter said:
I've never seen an Nvidia rep say they don't care about one of their companies popular technologies, It's not a good look.

And it's not about a game running badly, AI based upscaling can do a better job than a games AA implementation.
Click to expand...

antialiasing is a fair point though i'd argue it's mostly forced because of the prevalence of bad implementations of TAA in mordern games. This isn't something we historically needed to throw ML at. in any case, i'm entirely uninvolved with fsr, there's nothing about it i could or even would talk about.

the things i focus on are stability, reliability, experience, baseline performance, overall software quality etc.
 
Last edited:
Dicehunter said:
I've never seen an Nvidia rep say they don't care about one of their companies popular technologies, It's not a good look.

And it's not about a game running badly, AI based upscaling can do a better job than a games AA implementation.
Click to expand...

Meh I like the personalised take tbh over boilerplate corporate responses. There is a human behind the avatar after all.
 
Last edited:
honestly, I just want people to demand more performance and reliability form hw venors like us, and tighter optimisation from game developers, rather than clamouring for the latest concessions.

like, yeah I'm quite opinionated about it all. I just hoped the ecosystem as a whole would be in a better spot.
 
GoogalyMoogaly said:
Do you still have any of your old GPUs? might be able to use one of them to narrow things down
Click to expand...

It is the PSU....

I ran OCCT GPU stress test, passed no issues.... then i ran the power stress test, the moment both the GPU and CPU loaded up it crashed.

Its always the same, i can stress the GPU on its own for hours on end and its perfectly stable, or the CPU on its own, the moment i stress them both together it crashes
 
humbug said:
It is the PSU....

I ran OCCT GPU stress test, passed no issues.... then i ran the power stress test, the moment both the GPU and CPU loaded up it crashed.

Its always the same, i can stress the GPU on its own for hours on end and its perfectly stable, or the CPU on its own, the moment i stress them both together it crashes
Click to expand...
Ah well, at least you've (hopefully) figured it out.
You said you've got a new PSU arriving soon didn't you? Hope that fixes it.

I've been struggling with an issue with my Ryzen 5500 + B550 motherboard + 7900XT where the USB devices intermittently disconnect for a second (works fine with a 4060Ti). So annoying trying to troubleshoot intermittent issues. I've decided it's something about the power draw or something of the 7900XT that messes up the IO controller, which is tricky to fix.
 
I've just seen QSSR is out for the base PS5. Hopefully now Radeon can pivot back to some of the PC features they promised.

Edit. Probs FSR4 on 6xxx GPUs, it would fit with QSSR on PS5
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom