> >Hmm, maybe that is why my EVGA GeForce 7950 GT KO (512 MB)'s fan stopped
> >working. At least EVGA's unlimited warranty rocks since I got a GeForce
> >8800 GT (512 MB) as a replacement, but its warranty is only for a year.
> Problem with 7900GT was that the ram on many cards that companies like
> EVGA were overclcoking it to just could not do. Many people had issues
> with 7900GT.
What type of issues from OC speed? Like lock ups, blue screens, fan
deaths, etc.?
--
[Laser pulsing] "Bah. It's as easy as crushing an ant! You know, the..."
[grunting] "Wh-wh-whoa! Hey, take my wallet and leave me alone!" --Mr.
Burns from The Simpsons (Fraudcast News; FABF16/FABF18 episode)
/\___/\
/ /\ /\ \ Ant @ http://antfarm.ma.cx (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Please nuke ANT if replying by e-mail.
( )
>> For example, PNY hardly is the brand choosen by gamers that want to
>> overclock. Those that show the high return rates on the other sides are
>> brands that aim on the "enthusiast" market (gamers) with
>> factory-overclocked products. Such products are probably overclocked even
>> more, and often also modified (i.e. by replacing the cooler).
>
> That is one school of thought. The other school of thought says the Golden
> Samples are purchased by "armchair" enthusiasts who like the idea of
> overclocking, but prefer to let the factory do it for them under warranty.
This is probably very rare. Either someone is interested in overclocking
(in which case he wants to push the card to the limit) or he isn't (in
which case he wouldn't pay the premium for a bit of overclocking).
> The return rates more likely indicate that despite Gainward's best efforts
> with stress screening, the factory overclocking did result in slightly more
> failures... which may be okay for Gainward, because its profit margins on
> the overclocked cards are higher.
Maybe, but it is even more likely that the cards have been returned
because the buyer realized that the high premium for factory
overclocking was a waste of money.
However, without more details (why the cards have been returned) we can
only guess. And the lack of details makes these statistics even more
worthless.
>> The chipset statistics is equally useless. There very likely is no
>> relaibility difference between a GTX 260 or GTX 280 GPU itself,
>
> The GTX 260 and 280 share the same GT200 core, same cooler, but the GTX 280
> generates more heat even at reference clocks. On what basis do you think
> there is no reliability difference between the two GPU tiers?
First, simply because as you say the GPU itself is identical, and the
differences in heat dissipation and power consumption are not
groundbreaking. And second, my last employer was Nvidia customer with
access to many details that only board partners do have, and no-one is
aware of any general reliability differences between GTX 260 and GTX 280.
>> At the end of the day it is just what it is, a return statistics of a
>> certain french retailer. It however says nothing about the reliability
>> of a specific gfx card brand or chipset.
>
> On the contrary, the return statistics take a real-world cross section of
> current products. If you are shopping for a card, you'll trip over all kinds
> of overclocked cards. Hell, AMD claims to have lowered the reference clocks
> on some cards (e.g. HD4890), just so board vendors can artificially create
> "overclocked" SKUs. With a sufficiently large sample size (as it was in this
> case), the average return rate per brand or chipset is representative of
> what a buyer may see.
Maybe. Maybe not. Who knows. Fact is, with a survey created by a single
retailer which doesn't tell you side parameters you don't know ****. For
example, you don't know why cards have been returned. In the EU, online
retailers are required to take back an item within a certain period of
time after purchase if the customer doesn't like it anymore, so it isn't
known how much of the cards actually had defects and haven't been given
back because the customer was disappointed by the performance, regreted
the buy because the next day the cheaper/faster/cooler successor came
out, or because he found a better deal elsewhere. You also don't know if
the statistics only cover retail products or also bulk/OEM versions, or
if it includes grey market versions that were not made for the European
market. You don't know if the same return rates are seen by other "large
French e-retailers", or retailers in other countries, so you don't know
if this statistic from a single retailer really reflects average return
rates, or for some reason shows quite unusual return rates, i.e. maybe
because the shop tries to sell products that have been returned on to
other customers to avoid having to return them to his distributor
(something which isn't rare in the retail market), which means that a
single defective card would be counted multiple times as "return".
This survey contains so little information that no sane and logic
conclusions can be drawn from it. Of course, you could tell yourself
that this shows the defect rates of various products, but that is
exactly as worthless as this survey.
>What type of issues from OC speed? Like lock ups, blue screens, fan
>deaths, etc.?
Video graphics corruption. Was easy to test for by running 3DMark 06
Deep Freeze test. I now always run that bench everytime I buy a new
video card because I know if the card makes it through Deep Freeze a
few times it is good to go. I was able to fix one of the cards by
simply downclocking the ram in the vid bios but sent it back to EVGA
anyway after setting the bios back to default ram speed.
--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."
"Beckett" <beckett@work.invlaid> wrote in message
news:hpmsd55ldbpgsp6uuhulfltj5ek3io08sv@4ax.com...
> On Tue, 20 Oct 2009 15:43:31 -0500, ANTant@zimage.com (Ant) wrote:
>
>
>>What type of issues from OC speed? Like lock ups, blue screens, fan
>>deaths, etc.?
>
> Video graphics corruption. Was easy to test for by running 3DMark 06
> Deep Freeze test. I now always run that bench everytime I buy a new
> video card because I know if the card makes it through Deep Freeze a
> few times it is good to go. I was able to fix one of the cards by
> simply downclocking the ram in the vid bios but sent it back to EVGA
> anyway after setting the bios back to default ram speed.
The only time I saw corruptions was in DOS boot disk's Norton Ghost 2003
(VGA mode?) after a few minutes of making a big image and seeing garbage
on my screen. That was when I discovered the video card's fan was not
moving and I couldn't turn it myself (stuck).
--
"When you can't fight on and drop to die; you're just a big tasty feast
for the crows, ants, buzzards and flies." --unknown
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.ma.cx (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Nuke ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
>> What type of issues from OC speed? Like lock ups, blue screens, fan
>> deaths, etc.?
>
> Video graphics corruption. Was easy to test for by running 3DMark 06
> Deep Freeze test. I now always run that bench everytime I buy a new
> video card because I know if the card makes it through Deep Freeze a
> few times it is good to go. I was able to fix one of the cards by
> simply downclocking the ram in the vid bios but sent it back to EVGA
> anyway after setting the bios back to default ram speed.
Any known issues with ATI Radeon 4870 (512 MB) video cards? I bought a
MSI R4870-T2D512 OC Radeon HD 4870 512 MB in December 2008. Haven't seen
any serious issues except ATI's buggy DVXA video driver (had to use
v8.10's ativva* files) to avoid blue screens with DVB Viewer Pro v4.
--
"I used to own an ant farm but had to give it up. I couldn't find
tractors small enough to fit it." --Steven Wright
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.ma.cx (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Nuke ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
On Tue, 20 Oct 2009 22:16:36 -0700, Ant <ant@zimage.comANT> wrote:
>Any known issues with ATI Radeon 4870 (512 MB) video cards? I bought a
>MSI R4870-T2D512 OC Radeon HD 4870 512 MB in December 2008. Haven't seen
>any serious issues except ATI's buggy DVXA video driver (had to use
>v8.10's ativva* files) to avoid blue screens with DVB Viewer Pro v4.
No, I have 4870 and no issues. Thought I had an issue with my new 5870
when I got a freeze in a game a couplemof times but it turned out it
was ATI beta driver they released for the card. Turned off catalyst AI
and it never froze again.
Re: Graphics Card Failure Rates 
Linear Mode
