AMD issues.

Chris Pratt · September 3, 2021

13 minutes ago, Wictorian said:

But higher distrubution doesn't effect the rate.

It would have been better to say "apparent failure rate", I suppose. That was actually my point, it's not an actual failure rate. Failure rates are generalized and honestly kind of meaningless because of that. It's based on RMAs or some other similar factor, but tells you nothing about the cause. As a contrived example, let's say there was some PSU that fried some particular CPU every time. The average user is more apt to RMA the CPU alone, contributing to it's failure rate, but really the problem was with the PSU. It's never so cut and dry of course, but I'm trying to illustrate what is being talked about here.

If more AMD CPUs are paired with more cheap components that cause issues, then AMD CPUs end up with a higher failure rate irrespective of their actual quality or longetivity. That's a correlation. From that, people may make assumptions that AMD CPUs are bad and fail all the time (causation), but you can't say that actually with the available data.

Wictorian · September 3, 2021

1 hour ago, Chris Pratt said:

It would have been better to say "apparent failure rate", I suppose. That was actually my point, it's not an actual failure rate. Failure rates are generalized and honestly kind of meaningless because of that. It's based on RMAs or some other similar factor, but tells you nothing about the cause. As a contrived example, let's say there was some PSU that fried some particular CPU every time. The average user is more apt to RMA the CPU alone, contributing to it's failure rate, but really the problem was with the PSU. It's never so cut and dry of course, but I'm trying to illustrate what is being talked about here.

If more AMD CPUs are paired with more cheap components that cause issues, then AMD CPUs end up with a higher failure rate irrespective of their actual quality or longetivity. That's a correlation. From that, people may make assumptions that AMD CPUs are bad and fail all the time (causation), but you can't say that actually with the available data.

Ok I get your point but still we can safely say AMD caused more issues in the past

Chris Pratt · September 3, 2021

5 minutes ago, Wictorian said:

Ok I get your point but still we can safely say AMD caused more issues in the past

Well, sure, but the same can be said of Intel at various points in its history. That's irrelevant at this point.

Dzaxis · September 3, 2021

I do think AMD has more problems but i wouldnt count them to a certain component. just more a sytem as a whole. AMD issues with USB, thunderbolt, memory, random shutdowns and such all seem a lot more common than on intel systems.

the components for a build can all be great, but for some reason just not work together. a coulple of my friends swapped grappics cards a while back because ones new build was shutting off within 20 minutes of playing a game. no thermal issues or anything just suddenly the power would go... swapped cards with another friend just to test and both had no issues. so he knew his card was fine as it worked perfectly in his mates system. and new his system was fine as it worked perfectly with his mates card.. but when they swapped back the random shut offs came back straight away.. so in they end they swapped permanantly and since then everything has been fine. both 3060 cards just different models. baffling problem really.

then my company has over 400 computers. 80 of them were new / upgraded in the last 18 months to AMD systems (many in staff homes due to the pandemic and people working from home) and you definitely notice when 2/3 of your IT tickets are for those same systems time and time again.

its just quirky i think. there isnt one issue that stands out above all others as common or even notorious just a whole bunch of little issues that more often seems to be amd systems.