• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Hardware failure rates 4Q 2012 (numbers inside!)

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
In the past I have posted some links with Hardware failure rates from an overseas company. The latest data below is from 11/12 (new one should be out soon).


Source: http://www.behardware.com/articles/881-1/components-returns-rates-7.html


Motherboards


- ASRock 1.67% (against 1.90%)
- Gigabyte 1.77% (against 2.17%)
- MSI 2.24% (against 2.11%)
- ASUS 2.34% (against 2.66%)

In comparison to the previous period, ASRock has improved its score and keeps top spot in the rankings. Gigabyte has moved into second position at the expense of MSI while ASUS has closed the gap on MSI with a notable reduction in returns rates.

If we look more particularly at the rates observed on motherboards based on the Intel P67 or Z68 Express chipsets, we get the following numbers:

- Asrock 1.41%
- MSI 2.44%
- Gigabyte 2.47%
- ASUS 2.96%

In terms of the big picture, only four models have a returns rate of over 5%:

- 6.40% for the MSI Z68A-GD80 G3
- 6.19% for the ASUS P8P67 Pro
- 5.38% for the ASUS Rampage IV Extreme
- 5.33% for the MSI-P67A-C43


Power Supplies



- Fortron / FSP Group 0.42% (against 0.81%)
- Cooler Master 1.01% (against 1.25%)
- Antec 1.17% (against 0.80%)
- Seasonic 2.20% (against 1.92%)
- Corsair 2.30% (against 2.20%)
- Thermaltake 2.36% (N/A)

Antec has lost top spot to FSP Group, which posted a record this time. Cooler Master has also overtaken Antec and pushed it into third place. Corsair has benefitted from the arrival of Thermaltake, which is now the worst performer though with still reasonable return rates.

Just one model has a rate of over 5%, the Corsair V2, at 5.6%. Here are the scores for the 400-450 Watt models:

- 3.54% for the Cooler Master GX 450W
- 2.11% for the Corsair CX430 V2
- 1.40% for the Antec Neo ECO 450C
- 1.06% for the Antec High Current Gamer 400
- 1.01% for the Antec Neo ECO 400C
- 0.90% for the FSP (Fortron) HEXA 400
- 0.28% for the Cooler Master Elite Power 400W
- 0.00% for the FSP (Fortron) AURUM 400

And for the 500-550 Watt models:

- 2.10% for the Corsair CX500 V2
- 2.04% for the Antec High Current Gamer 500
- 2.03% for the Seasonic S12II-520
- 1.94% for the Cooler Master Silent Pro M500
- 1.80% for the Cooler Master GX 550W
- 0.96% for the Seasonic M12II-520
- 0.56% for the FSP (Fortron) AURUM 500
- 0.53% for the FSP (Fortron) HEXA 500
- 0.46% for the Antec High Current Gamer 520M
- 0.27% for the Antec Neo ECO 520C


RAM


- Kingston 0.27% (against 0.40%)
- Crucial 0.30% (against 0.23%)
- G.Skill 1.01% (against 1.10%)
- Corsair 1.06% (against 1.44%)

Kingston takes over from Crucial at the top of the pile here but both brands have maintained a very low returns rate. G.Skill and Corsair, which produce more high-end models, have both improved on their scores, especially Corsair which has closed the gap on G.Skill.

Only one kit has a returns rate higher than 5%, the G.Skill RipJaws X Series 16 GB (4x 4GB) F3-12800CL9Q-16GBX with a rate of 5.45%. Corsair’s worst score is for a model at 4.08% compared to Kingston’s worst at 2.14% and Crucial’s at 1.03%.


GPUs


- Sapphire 1.32% (against 1.20%)
- ASUS 1.53% (against 1.55%)
- PNY 1.56% (against 1.62%)
- MSI 1.69% (against 2.25%)
- Gigabyte 1.82% (against 2.18%)
- Gainward 2.05% (against 2.43%)

Sapphire retains its position at the head of the field. Several manufacturers see their returns rate drop significantly: MSI, Gigabyte and Gainward. 6 cards have a higher than 5% rate:

- 16.89% for the Gainward GeForce GTX 580 "Phantom" 1.5 GB
- 7.87% for the Gainward GeForce GTX 580 "Phantom" 3 GB
- 6.19% for the Sapphire Radeon HD 7970 3 GB (21197-00-40G)
- 5.69% for the ASUS EAH6950 DCII/2DI4S/2GD5 2 GB
- 5.56% for the Gigabyte GeForce GTX 560 Ti OC 1 GB

The GTX 580 Phantoms have been problematic indeed, especially the 1.5 GB model. Sapphire, though with the best average score, still has a high rate on its first 7970, just like ASUS and Gigabyte on other GPUs.

Here are the figures by GPU:

- Radeon HD 6870: 2.89% (against 2.00%)
- Radeon HD 6950: 2.04% (against 4.08%)
- Radeon HD 6970: 4.03% (against 5.85%)
- Radeon HD 7950: 4.44% (N/A)
- Radeon HD 7970: 5.56% (N/A)
- GeForce GTX 560 Ti: 1.87% (against 3.77%)
- GeForce GTX 570: 2.58% (against 2.49%)
- GeForce GTX 580: 7.56% (against 5.68%)

Compared to the previous period, the Radeon HD 6950 and GeForce GTX 560 Ti see a significant improvement in reliability. The higher you go up the range, the higher the returns rate, whether for the Radeon HD 7970s or the GeForce GTX 580s. If we exclude the two Gainward Phantom’s from the figures for the GTX 580, its figures drop back to 5.27%.


HDDs


- Western 1.48% (against 1.63%)
- Samsung 1.65% (against 1.23%)
- Seagate 1.70% (against 1.89%)
- Hitachi 3.77% (against 3.95%)

With an improved score, Western has taken over from Samsung at the top. Samsung’s returns rate has also got worse. Seagate is hot on the heels of Samsung and Hitachi, as was the case in our last report, brings up the rear with quite a high rate. It is to be hoped that Toshiba, which has taken over Hitachi’s 3.5" hard drive production, will be able to remedy this problem.

4 hard drives have a higher than 5% rate:

- 7.65% for the Hitachi Deskstar 7K3000 3 TB
- 6.91% for the Seagate Barracuda XT 2 To
- 5.78% for the Seagate Barracuda 7200.12 160 GB
- 5.30% for the Samsung SpinPoint F3 (HD253GJ) 250 GB

Here now are the stats for the 3 TB drives sold over the period:

- 7.65% for the Hitachi Deskstar 7K3000 3 TB
- 4.86% for the Seagate Barracuda XT 3 TB
- 3.36% for the Seagate Barracuda 7200.14 3 TB
- 3.12% for the WD Caviar Green 3 TB

And the 2 TB drives:

- 4.60% for the WD Caviar RE4 2 TB
- 3.91% for the WD Caviar Black 2 TB
- 3.32% for the WD Caviar Green 2 TB WD20EARS
- 2.28% for the Seagate Barracuda 7200.14 2 TB
- 1.94% for the WD AV-GP 2 TB
- 1.46% for the WD Caviar Green 2 TB WD20EARX
- 1.03% for the WD RE4-GP 2 TB


SSDs


- Intel 0.45% (against 1.73%)
- Samsung 0.48% (N/A)
- Corsair 1.05% (against 2.93%)
- Crucial 1.11% (against 0.82%)
- OCZ 5.02% (against 7.03%)

There have been a lot of changes in the rankings here since our last half year figures! First of all, Crucial, previously top dog, has dropped to fourth place. Although with a slightly higher score, its returns rate remains low but it has been overtaken by Corsair, with a much lower rate, Samsung, which is now 2nd according to our stats, and Intel, which now has the best score, returning to the sort of form we have seen from it in the past.

The OCZ SSDs are in last place overall but with a returns rate that is finally starting to drop. In fact, behind this rate there are some big disparities between ranges. The most popular ranges, namely the Vertex 3s and Agility 3s, do relatively well with returns of 1.51% and 2.03% respectively. Some models however have catastrophic scores and OCZ monopolises the group of models with returns of over 5%:

- 40.00% for the OCZ Petrol 64 GB
- 39.42% for the OCZ Petrol 128 GB
- 30.85% for the OCZ Octane 128 GB SATA II
- 29.46% for the OCZ Octane 64 GB SATA II
- 9.73% for the OCZ Vertex 2 120 GB 3.5"
- 9.59% for the OCZ Vertex 2 120 GB
- 6.73% for the OCZ Vertex 2 60 GB
- 5.43% for the OCZ Agility 3 240 GB
- 5.12% for the OCZ Vertex Plus 128 GB

The Future


What does the future hold? Of course, no one can predict this but over the next period, for sales between 1st April 2012 and 1st October 2012 for returns before October 2012 (namely between 0 and 6 months of use), OCZ continues to have a high rate of 5.65%... once again impacted by the Petrol and Octane SATA II models. Among the most popular models, things are much better, with the exception of the Agility 4: 0.93% on the Vertex 4s, 1.22% on the Vertex 3s, 2.59% on the Agility 3s and 5.60% on the Agility 4s.

For Hitachi hard drives, the current rate is around 1.76% over this period. It remains to be seen how things will evolve. Here are the products with the highest returns sold over this period, five by category (minimum sample of 100):

Motherboards:
4.28% for the ASUS Rampage IV Extreme
4.20% for the ASRock H77 Pro4/MVP
2.92% for the ASUS P8Z68-V PRO GEN 3
2.81% for the ASUS P8Z77-M PRO
2.47% for the MSI Z77A-G45

Power supplies:
2.84% for the Akasa Venom Power 550
2.43% for the Corsair Gaming Series GS600
2.12% for the Corsair CX500 V2
2.09% for the Corsair Gaming Series GS800
2.00% for the Cooler Master Silent Pro M500

RAM:
3.48% for the Corsair Vengeance LP Blue 16 GB (4x 4 GB) DDR3-1600 CL9 CML16GX3M4A1600C9B
3.31% for the Corsair XMS3 16 GB (2x 8 GB) DDR3-1600 CL11 CMX16GX3M2A1600C11
3.28% for the Corsair Vengeance 16 GB Blue (4x 4 GB) DDR3-1600 CL9 CMZ16GX3M4A1600C9B
2.94% for the Corsair XMS3 8 GB (2x 4 GB) DDR3-1600 CL9 CMX8GX3M2B1600C9
2.83% for the G.Skill RL RipJaws 8 GB (2x 4GB) DDR3-1066 F3-8500CL7D-8GBRL

Graphics cards:
11.88% for the Sapphire Radeon HD 6770 1 GB (11189-10-20G)
7.62% for the Sapphire Radeon HD 7970 OC Edition 3 GB (11197-01-40G)
6.75% for the Sapphire Radeon HD 7870 OC Edition 2 GB (11199-03-20G)
6.06% for the ASUS ENGT520 SL/DI/1GD3/V2(LP)
5.07% for the Sapphire Radeon HD 7970 3 GB (21197-00-40G)

Hard drives:
4.36% for the Hitachi Deskstar 7K1000.C 1 TB
4.20% for the Western Digital Caviar Black 1.5 TB (WD1002FAEX)
4.11% for the Hitachi Deskstar 7K3000 2 TB
3.57% for the Seagate Barracuda 7200.12 320 GB
2.39% for the Western Digital Caviar Blue 500 GB (WD5000AAKS)

SSDs:
37.19% for the OCZ Octane 128 GB SATA II
29.52% for the OCZ Octane 64 GB SATA II
28.10% for the OCZ Petrol 128 GB
22.48% for the OCZ Petrol 64 GB
8.52% for the OCZ Agility 4 256 GB

Several products already have a higher than 5% returns rate: the OCZ SSDs (no surprises here!) and an entry level ASUS graphics card as well as four Sapphire cards, including a Radeon HD 6770, two Radeon HD 7970s and a 7870 OC.
 

PolePosition

Member
Joined
Jul 15, 2012
Location
Louisiana
I imagine the 5.38 RIVE return rate is primarily due to human abuse, which is probably a greater key role in these returns along with human error as opposed to manufacturer defects, where most products are tested before they leave the factory. They will honor most RMAs though to maintain public image and repuatation even when such abuse is difficult to detect and determine or has been determined in some cases.
 
OP
EarthDog

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
THe entire article is in a link in the first post. I copy/pasted it all. Cliff's - Nope..


I imagine the 5.38 RIVE return rate is primarily due to human abuse, which is probably a greater key role in these returns along with human error as opposed to manufacturer defects, where most products are tested before they leave the factory. They will honor most RMAs though to maintain public image and repuatation even when such abuse is difficult to detect and determine or has been determined in some cases.
While you are correct, all models suffer this same inflated rate. We cant extract that out of any of the metrics given from any maker. They are what they are, but to say, like you did in PM, that more XXXX boards had human error on them than YYYY and that is why YYYY has better numbers, has literally no merit with the information given. Perhaps its true, perhaps not, nobody knows and to extrapolate and speculate that is premature at best.

All models are also tested before they leave the factory as well... :confused:
 
Last edited:

Bobnova

Senior Member
Joined
May 10, 2009
I imagine the 5.38 RIVE return rate is primarily due to human abuse, which is probably a greater key role in these returns along with human error as opposed to manufacturer defects, where most products are tested before they leave the factory. They will honor most RMAs though to maintain public image and repuatation even when such abuse is difficult to detect and determine or has been determined in some cases.

Barely honor, heh. Have you read the myriad of horror stories about Asus warranty service? :D
 

PolePosition

Member
Joined
Jul 15, 2012
Location
Louisiana
Yeah, I've read and heard about them, and no doubt ASUS is the worse for tech support, so you had best be a tech guru and have the supreme ability to resolve your own issues if you're going to buy ASUS. Fortunately, there is rarely ever the need for an RMA except those of a small percentage.
 

SMOKEU

Member
Joined
Nov 7, 2010
Location
NZ
I don't believe the motherboard failure rates at all. Out of the 10 or so motherboards I've had, only about 1/3 haven't failed. Most are either DOA or fail within 6 months. I've only had Asus and Gigabyte boards.

My mom had a HP desktop, and that had 2 motherboard failures (100% failure rate).
 

Bobnova

Senior Member
Joined
May 10, 2009
Let's see, out of 8 motherboards I've used extensively in the last 4 years, zero have failed. If I include the 6 in other people's computers that I be built, none have failed. If I extend our back further, put off the ~45 motherboards I've used it installed, none have failed. Some of this dates back to the 90s,they still work.
 

trekky

Member
Joined
Oct 3, 2011
Let's see, out of 8 motherboards I've used extensively in the last 4 years, zero have failed. If I include the 6 in other people's computers that I be built, none have failed. If I extend our back further, put off the ~45 motherboards I've used it installed, none have failed. Some of this dates back to the 90s,they still work.

ya my next build probly will be asus
asrock 2 mobo in 2 differ rigs give me trouble all the time
one board sound chip broke 2 times just went in 3 months time!

XFX seems like a garbage brand also history my XFX exp
late 2011 got a 6950 came broken (took me few month if find out cause 7300LE to 6950 ya.... only ran @500mhz core)
last month got 2 XFX DD 7950
one broken 39 days after I ordered it! it was just running idle also (not a lick of load when it went out)
I guess windows updates do kill :eek:

as for OCZ I have Agility 3 and runs great
so does my friends Agility 3

but A-Bit boards seems pretty good (well for old lasting wise)
got a P2 rig and a P3 rig all A-bit!


but nice copy n paste earthdog :thup:
 

macklin01

Computational Oncologist / Biomathematician / Mode
Joined
Apr 3, 2002
Location
Bloomington, IN
Some interesting data there ... thanks!

It'll be interesting to see how the TLC-based SSDs fare compared to MLC down the road. I feel a bit skittish on that for now, but I'd imagine they will come out fine with sufficient overprovisioning ...
 

madhatter256

Special Member
Joined
Jul 5, 2008
Location
CFL
Do these figures factor in the amount of product volume?

Asus makes a ton of motherboards compared to the others, so even though they have a relatively higher failure rate, considering the volume of production they output, it may be very small. but then my experience with Asus is there will always be a model or specific board that just won't last and fails too early - case in point P5RD2-VM series... poor design leading to mosfet/VRM failure. About 2/3 of that model we sold failed within the year.
 
OP
EarthDog

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Do these figures factor in the amount of product volume?

Asus makes a ton of motherboards compared to the others, so even though they have a relatively higher failure rate, considering the volume of production they output, it may be very small. but then my experience with Asus is there will always be a model or specific board that just won't last and fails too early - case in point P5RD2-VM series... poor design leading to mosfet/VRM failure. About 2/3 of that model we sold failed within the year.

Its a percentage. Simply failed/produced for everything.
Exactly. Its a %, that isnt affected by one mobo selling 10k while another sells 3k. % is % is %. They gave minimum values and data for each. :thup:
 

macklin01

Computational Oncologist / Biomathematician / Mode
Joined
Apr 3, 2002
Location
Bloomington, IN
It seems like the most relevant metric for us, anyway. We're not concerned with sales volumes or RMA volumes. We're concerned with estimating the chance that we're purchasing a piece of junk, based upon somebody's helpful statistics. :)
 
OP
EarthDog

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Agreed. While this clearly isn't a large enough data set to sit on conclusively, it gives readers the best idea we have of return rates as its about the only thing published that I can find.

I'm hoping their new version comes out here very soon.