G

[Xysto]

.

[Jurrunio]

You need a much higher sample size than that to be conclusive, but it is a good way to remind yourself who to buy from next time

[TahoeDust]

No EVGA listed because when they do fail, they actually replace them.

[Howitz]

then theres sapphire that sells 4x more than xfx who sells 5 times more than diamon and so on... get a percentage out of total units sold instead.

[Phentos]

I used to be a PC builder in my area while in school, so I've heard of my fair share of dead GPUs. Out of the AMD GPUs I've put in systems over probably the last 12 months before I stopped back in June (having gotten an actual decent job), I remember 3 failed GPUs. Two were Powercolor (an RX 480 and a used R9 290X) and one Gigabyte RX 570.

 

I also did not process the RMAs, that was on the users. All 3 failures were not caused by me or the users.

 

Personally I've had one GPU kind of flake out on me and that was a Powercolor PCS R9 390. Replaced that with a Sapphire Nitro R9 Fury, because I was too poor for a 1070 or better at the time. 

Edited October 19, 2018 by Phentos
Confused some statistics a bit.

[imreloadin]

12 minutes ago, Phentos said:

All 3 failures were not caused by me or the users.

It's always the users...you can just never trust them!

[Phentos]

1 minute ago, imreloadin said:

It's always the users...you can just never trust them!

[Enderman]

This is basically impossible.

Maybe like 1% of actual GPU failures will be posted online in forums or reviews.

You also have no idea about the amount of GPUs that the manufacturer has sold.

Then there's also a million other variables like operating conditions, hours of runtime, environmental factors, etc etc etc.

[Xysto]

4 hours ago, TahoeDust said:

No EVGA listed because when they do fail, they actually replace them.

,

 

4 hours ago, Jurrunio said:

You need a much higher sample size than that to be conclusive, but it is a good way to remind yourself who to buy from next time

 

[Jurrunio]

2 minutes ago, Xysto said:

Ya, but I'm sure the sample size is already impossibly massive.

 

 

massive? It's nothing compared to how many units they sell per day

[Xysto]

.

 

 

 

 

 

 

On 10/19/2018 at 5:00 PM, imreloadin said:

It's always the users...you can just never trust them!

LTT

[Xysto]

,

[Reznik]

Whatever numbers you come up with won't be accurate so what value is it to you or anyone?   It will be a compiled list of inaccurate, potentially (very likely) misleading information.  

[imreloadin]

26 minutes ago, Xysto said:

Exactly. Do you thinkSapphire, XFX, ASUS et al would share this data?  I'm guessing no.

 

You must be a former Mike's customer.  My sympathies. jk ish.

Mikes? I live in the states so I'm not sure if that's a Canadian business or something

[Xysto]

.

 

[Xysto]

2 hours ago, imreloadin said:

Mikes? I live in the states so I'm not sure if that's a Canadian business or something

 

,

[Enderman]

1 hour ago, Xysto said:

Yes impossible, but when does anyone have a comprehensive data available to make a decision?  You look at the information that's available thru research and reviews to arrive at a conclusion.  Anecdotal information is still useful in an imperfect world.

The manufacturers have the exact information on how many cards they sell and how many fail.

They do not make this information public.

[Xysto]

.

[Xysto]

2 hours ago, Phentos said:

I used to be a PC builder in my area while in school, so I've heard of my fair share of dead GPUs. Out of the AMD GPUs I've put in systems over probably the last 12 months before I stopped back in June (having gotten an actual decent job), I remember 3 failed GPUs. Two were Powercolor (an RX 480 and a used R9 290X) and one Gigabyte RX 570.

 

I also did not process the RMAs, that was on the users. All 3 failures were not caused by me or the users.

 

Personally I've had one GPU kind of flake out on me and that was a Powercolor PCS R9 390. Replaced that with a Sapphire Nitro R9 Fury, because I was too poor for a 1070 or better at the time. 

thanks for that thoughtful reply.  It's what I hoped for when I originally posted to LTT.  We never have access to perfect and complete information ever, so we take a guess, maybe just buy what's on sale. As a community tho, I think there is more to learned from our collective experiences.  Why bother otherwise?  thanks agin.