Huge list of failure rates on PC components (French, but I translated nearly everything)

[Askew]

I bought an OCZ SSD once, it failed.

I bought an OCZ PSU once, it also failed.

 

Don't know if that's coincidence, but those are the only 2 things I have purchased from them.

[CM Rajiv]

I can say we've made major improvements to our PSU line (especially with the V series and upcoming VSM) to really return the confidence into the product line with enthusiasts. There are still the lower end lineups that are intended for non-OC, basic use but we're primarily focused on keeping people happy when they buy a Cooler Master product. 

[ttam]

Your source please.

I got stuck in the RMA department once at my new job.

A lot of it is stupid. A lot of it is the normal it just failed. A lot of it is reasons no one could stop. But there a lot of stupid.

Not the reasoning he had. Just mine haha

[RicY]

Can contribute on the Asrock 970 ext3 s**tboard, first one came DOA, second one is working, for now. Good thing that the RMA took only ~7days.

 

So this list seems to be quite belivable.

[Leonard]

I got stuck in the RMA department once at my new job.

A lot of it is stupid. A lot of it is the normal it just failed. A lot of it is reasons no one could stop. But there a lot of stupid.

Not the reasoning he had. Just mine haha

Ok, from a business perspective though, most of these things can not be helped due to the sensitive nature of the products. There are any number of combinations that can cause these problem and most of it is man made( either they make mistakes or they lack the necessary checks and balances for the products. I haven't gotten any of my PC parts DOA as yet and i hope to keep this trend.  . 

[TheRuke]

I have a Sapphire 7870 

[applebook]

Looks like Corsair is consistently among the worst. Sort of surprising.

[Kiwiandapple]

I can say we've made major improvements to our PSU line (especially with the V series and upcoming VSM) to really return the confidence into the product line with enthusiasts. There are still the lower end lineups that are intended for non-OC, basic use but we're primarily focused on keeping people happy when they buy a Cooler Master product. 

 

My next Keyboard will very likely be a CM one. Only heard good things about them. Especially the QuickFire line-up. Sadly I don't think my current Das one will have to be.. removed anytime soon... But if you keep making those amazing keyboards, i'm watching you! (please get an RGB or a soft yellow backlite mechanical keyboard, then I will have to sell my Das and buy that!)

[hawaiims]

Looks like Corsair is consistently among the worst. Sort of surprising.

Then again, all memory kits have extremely low failure rates, even in the highest rate ones (which were all corsair in this case)

As for the PSUs, it's not really surprised to see models from the CX series as they are as far as I know the only "name brand" who sell very cheap 80+ bronze PSUs (you often find cx500 PSUs for 30$)

My next Keyboard will very likely be a CM one. Only heard good things about them. Especially the QuickFire line-up. Sadly I don't think my current Das one will have to be.. removed anytime soon... But if you keep making those amazing keyboards, i'm watching you! (please get an RGB or a soft yellow backlite mechanical keyboard, then I will have to sell my Das and buy that!)

CM keyboards are amazing. I bought a brand new simple little quickfire rapid with mx blues about a year ago for only 59$! Apparently these ten keyless rapids are identical in build to the much more expensive and loved filch keyboards. Both companies just rebrand the same keyboard with their logos etc.. I did however had one problem with the removable micro USB cord for the keyboard failing, but thankfully they managed to send me a new USB cable quite fast.

[LogicalDrm]

As for the PSUs, it's not really surprised to see models from the CX series as they are as far as I know the only "name brand" who sell very cheap 80+ bronze PSUs (you often find cx500 PSUs for 30$)

 

I checked that in our price range aswell and seems you are correct. Only brand that I would get PSU from and had lower pricing than Corsair was Silverstone. You could get their 500W 80+ Silver with 43€ when Corsairs CX430 was at 55€.

[Kiwiandapple]

@hawaiims

 

Awesome. I just hope they make in a year time an RGB or yellow backlit keyboard! Then I am all over it ^^ 

[Osmium]

Very interesting. Not all of these numbers are unexpected but a few were surprising to say the least. 7870s having a whopping 12% failure rate.

[powerrock]

Hi all.  New to this forum, my friend sent me the link and after reading, I will add my 2 cents from what I know.

This data is from Hardware.fr a publication in EU with financial link to LDLC, one of the biggest online webshop in France, "unnamed" in this thread.

Every year this data is published, at first, people really cared about it but now it is not very relevant because the data is known to be inaccurate.

  This data does not take into account people making returns and switching to another brand or model due to change of mind.

  Nor does it take into account any returns which are not true failures such as compatibility or other factors.

  Also this does not take into account lower cost items which fail but are not returned and thrown out instead.

  There are many factors involved not accounted for and I cannot imagine for example Sapphire at such high rates.  At these rates, they can close their business.

  Also this data does not take into account from products not returned to LDLC but direct to the manufacturer.

[That Norwegian Guy]

I am glad to say that the only WD HDD I've bought is the WD Red 3TB. Also that all my SSDs are Samsung, all my RAM is Crucial, in Power Supplies I swear to FSP and Cougar (the latter only GX series, not mentioned here, the only Cougar PSU with a fail rate to speak of is their lowest end one)

 

Some things to point out though. While Gigabyte have a low return rate, almost all of their Z77 and AM3+ boards had a fundamentally broken RAID controller. However, since practically nobody uses onboard RAID, it hasn't affected their score much.

 

As for this:

PNY jumps from second to first place, while Sapphire keeps it's last position. We have to however clarify that PNY sales are not often for higher end cards, which are more subject to failure.

 

I'd call that a very liberal interpretation of statistics. What is more likely is that people who spend more on graphics card are less likely to tolerate small faults and problems than those who cheap out.

 

As a footnote, I can be very pleased that I was right all along in not recommending Corsair Power Supplies or RAM, contrary to the rest of the bandwagoneering PC crowd. Also funny to see the high fail rate of Seasonic PSUs, most often recommended by elitists in the PSU forums. (HA!!!)

 

LEPA, Cougar, Antec, Enermax > Seasonic

 

Your source please.

 

Not really arguing his statement, but at least when it comes to 3TB and up HDDs, they are often returned because the end user doesn't know how to create a working partition with them. AKA, stupidity.

 

I only heard good about FSP Aurum PSUs. So, there's also a prove now, but sadly people here only know & buy in 90% cases Corsair.

 

Bought my Fortron Source Epsilon 600W PSU in 2008, it has survived three builds and is now sitting in my Prodigy gaming rig, one of my best purchases. I'm pretty sure it also delivers more than 600W as I once tried to overload it with components to feed and it still ran (quietly)

[LAwLz]

Every year this data is published, at first, people really cared about it but now it is not very relevant because the data is known to be inaccurate.

[Citation needed]

 

 

This data does not take into account people making returns and switching to another brand or model due to change of mind.

How many people does that? Buy one part, and use it, and then return it because you changed your mind. I am willing to bet that next to no people does that, and even if they do then it will probably affect most brands equally as much.

 

 

Nor does it take into account any returns which are not true failures such as compatibility or other factors.

Again, it should affect all brands pretty much equally.

 

 

Also this does not take into account lower cost items which fail but are not returned and thrown out instead.

I agree with that one. Most of these things are not low cost items though. If your motherboard for 100 dollars breaks and the warranty is still valid then most people will return it. Why wouldn't they? They need the part so it would be stupid to buy it again if you can get it for free.

 

 

There are many factors involved not accounted for and I cannot imagine for example Sapphire at such high rates.  At these rates, they can close their business.

You don't think you can stay in business with a 2.06% failure rate? Really? For some reason their 7870s are unreliable but that's it. Remove those two cards and their failure rates are just 2.06%, which is just ~0.5% higher than Asus, MSI and Gigabyte. Maybe they messed up their stock overclock settings on those cards? Anyway bottom line is that they don't have high failure rates overall, it's just one specific GPU that has it. Also check the stats from previous year. 1.32% failure rates. They don't have to close their business just because they get slightly higher return rates during 1 year.

 

 

Also this data does not take into account from products not returned to LDLC but direct to the manufacturer.

I agree with that one as well, but the only company I can think of that encourages customers to send it back to them instead of the retailer is Corsair. I am willing to bet that 95% of all RMAs (other than for Corsair products) go to the retailer and not the manufacturer, so it shouldn't skew the results too much.

Of course these are not super accurate and you should blindly trust them 100%, it's a very good guideline if you ask me.

 

 

 

As a footnote, I can be very pleased that I was right all along in not recommending Corsair Power Supplies or RAM, contrary to the rest of the bandwagoneering PC crowd. Also funny to see the high fail rate of Seasonic PSUs, most often recommended by elitists in the PSU forums. (HA!!!)

 

LEPA, Cougar, Antec >>>> Seasonic

First of all, the difference is very small. It's like a 1% difference. So out of each 100 PSU sold, 1 is defect (and will be replaced). Statistically speaking, you will have to buy 100 PSUs before you would notice a difference in reliability.

However, you are forgetting that not breaking does not mean the PSU is good. The reason why Seasonic units (including rebranded ones such as most of Corsair units, all XFX units and some Antec ones) are often recommended is because they are quiet, got properly balanced rails, don't lie about the wattage or efficiency, have very good failure mechanisms (so if it fails it doesn't kill your CPU, GPU, motherboard etc as well), low ripple, and many more things.

It's not just about the risk of failure you have to take into consideration.

[hawaiims]

 

The higher end corsair psus are of decent quality, HX, TX series etc.. the problem is their CX series which is very budget oriented and popular, but actually has quite high failure rates. FSP indeed has pretty reliable power supplies, and they manufacture for other brands as well. If you haven't seen it already, here's a list of all brands and the actual manufacturers that make PSUs for them (interestingly enough brands such as corsair use 4 different OEMs depending on their models if my memory is right) 

 

http://www.tomshardware.com/reviews/power-supply-oem-manufacturer,2913.html

[That Norwegian Guy]

The higher end corsair psus are of decent quality, HX, TX series etc.. the problem is their CX series which is very budget oriented and popular, but actually has quite high failure rates. FSP indeed has pretty reliable power supplies, and they manufacture for other brands as well. If you haven't seen it already, here's a list of all brands and the actual manufacturers that make PSUs for them (interestingly enough brands such as corsair use 4 different OEMs depending on their models if my memory is right) 

 

http://www.tomshardware.com/reviews/power-supply-oem-manufacturer,2913.html

 

Yeah you're absolutely right, and I thank you for your great work on translating this into a post. IMO the best post on Linus Tech Tips Forums since it was created!

[powerrock]

Well, I would take the data with a very big grain of salt...

The data are not audited, not by any authority or even the manufacturers.  Are these numbers possible?  OCZ aside...

 

- 9,64% Seagate Constellation ES ST2000NM0011

- 4,87% Hitachi 7K3000 (HDS723020BLA642)

- 5,04% WD Caviar Black 1,5 TB (WD1502FAEX)

- 12,67% Sapphire Radeon HD 7850

- 7,44% Sapphire Radeon HD 7870 OC V2
- 7,41% Sapphire Radeon HD 7870 OC V1
- 7,02% Sapphire HD 7950 With Boost (11196-16)

-15,76% Sapphire Radeon HD 7870 OC Edition 2 GB

- 14,29% Sapphire Radeon HD 7970 OC Edition 3 GB
- 11,88% Sapphire Radeon HD 6770 1 GB
- 11,82% Sapphire Radeon HD 7870 GHz Edition 2 GB

 

It would be very interesting if there was someone from Sapphire or Seagate or WD on this forum to either agree or disagree to the data.

As well, can these numbers be "tweaked" because some companies give more SPIFF or MDF than others?

Maybe I haven't read enough, but I don't recall ever seeing any other webshop presenting RMA figures nor have I ever read about RMA rates on any magazine or PC related sites.

Sorry, I am just suspicious....

 

 

[Andi]

I checked that in our price range aswell and seems you are correct. Only brand that I would get PSU from and had lower pricing than Corsair was Silverstone. You could get their 500W 80+ Silver with 43€ when Corsairs CX430 was at 55€.

be quiet! System Power 7 400W 38€

vs

Corsair Builder Series CX430 37€

 

in germany bequiet! wins in every pricepoint. 5 years till world domination The next mid price psu's will be sold in the us too

[Andi]

 

Well, I would take the data with a very big grain of salt...

The data are not audited, not by any authority or even the manufacturers.  Are these numbers possible?  OCZ aside...

 

- 9,64% Seagate Constellation ES ST2000NM0011

- 4,87% Hitachi 7K3000 (HDS723020BLA642)

- 5,04% WD Caviar Black 1,5 TB (WD1502FAEX)

- 12,67% Sapphire Radeon HD 7850

- 7,44% Sapphire Radeon HD 7870 OC V2

- 7,41% Sapphire Radeon HD 7870 OC V1

- 7,02% Sapphire HD 7950 With Boost (11196-16)

-15,76% Sapphire Radeon HD 7870 OC Edition 2 GB

- 14,29% Sapphire Radeon HD 7970 OC Edition 3 GB

- 11,88% Sapphire Radeon HD 6770 1 GB

- 11,82% Sapphire Radeon HD 7870 GHz Edition 2 GB

 

It would be very interesting if there was someone from Sapphire or Seagate or WD on this forum to either agree or disagree to the data.

As well, can these numbers be "tweaked" because some companies give more SPIFF or MDF than others?

Maybe I haven't read enough, but I don't recall ever seeing any other webshop presenting RMA figures nor have I ever read about RMA rates on any magazine or PC related sites.

Sorry, I am just suspicious....

 

i worked at a big hardware onlineshop and Sapphire is well known for their stupid high rma rates...but 10% seems a bit too much.

[hawaiims]

i worked at a big hardware onlineshop and Sapphire is well known for their stupid high rma rates...but 10% seems a bit too much.

Their 10% rate is only for their 7870... If you removed those cards from their lineup, according to the article if you removed the 7970 and 7870 from the equation, Sapphire's RMA rates would be 5.47%. It's still way higher than the other brands, but it's definitely not that horrible. 

 

The seagate RMA rates seem to also be quite accurate, because apparently that whole backblaze rma thing was complete BS. 

[JAW96]

Their 10% rate is only for their 7870... If you removed those cards from their lineup, according to the article if you removed the 7970 and 7870 from the equation, Sapphire's RMA rates would be 5.47%. It's still way higher than the other brands, but it's definitely not that horrible. 

 

The seagate RMA rates seem to also be quite accurate, because apparently that whole backblaze rma thing was complete BS. 

backblaze thing was a ton of bs

[ccaian]

http://www.newegg.com/Product/Product.aspx?Item=N82E16820227776

 

52% failure rate. How is that even possible.

[Goddard]

Are there failure rates higher when using non-NAS drives for NAS?

[topsu]

Keep in mind these count in all rmas like coil white etc...