Puget Systems print logo
Read this article at https://www.pugetsystems.com/guides/749
Article Thumbnail

Most Reliable PC Hardware of 2015

Written on December 28, 2015 by Matt Bach


At Puget Systems, we record a huge amount of data for each and every system we sell including benchmarks, BIOS screenshots, thermal images, and system photos. In fact, much of this data is published on our website and can be accessed through our part information pages. Simply view a part information page, scroll down to the Links section, and you can view BIOS Screenshots, Thermal Images and Product Photography from recent systems we built.

While this data is nice, one of the most important things we track is something we don't have publicly available: the failure rates of individual components. Reliability is one of our primary values, so this data is invaluable for tracking both individual component, product line, and overall brand failure rates. With 2015 coming to a close, we thought we would run some reports and share what hardware we found to be the most reliable in 2015.

One thing we want to make clear is that we can only comment on products that we have sold significant quantities of. There may be even more reliable products out there, but either we have not sold it or we have not sold it in large enough quantities for us to have enough data to call it reliable or not. With that said, let's get started!


A reliable motherboard is essential in a high quality computer. Not only is a motherboard very difficult to swap out, but the effects of a poor quality motherboard can be far reaching and difficult to troubleshoot. This is complicated by the fact that motherboards are one of the most complex components in a computer. There are SATA, USB, fan, and network controllers as well as the physical ports, audio chips, and everything else that is needed to inter-connect every component in your system. This is a huge number of small parts that have to work perfectly together and any one of these could potentially have a problem. If there is a single dead USB port, slight static over the audio, or the voltage levels are measured outside of norm, it does not meet our standards and is considered to have failed.

Because of this, motherboards have one of the highest overall failure rate of any core component with about 1 out of every 20 motherboards failing for one reason or another. This may seem like a high failure rate, but the silver lining is that nearly all of these failures we catch in-house before the system is shipped to the customer. In fact, motherboards as a whole only have a ~.7% failure rate (or less than one out of every 100) when you only look at post-shipping failures.

However, in 2015 there were four motherboards that had an above-average history of reliability:

Asus Z170-A
(2.34% failure rate)

Asus H97I-PLUS
(2.65% failure rate)

Asus Z97-A/USB 3.1
(2.7% failure rate)

Asus Sabertooth Z97 Mark 2
(2.72% failure rate)

CPU (Processor)

2015 was again an absolutely great year for CPUs in terms of failure rates. Note that in 2015 we did not sell enough AMD CPU/APUs for us to make a call on their reliability. For Intel, however, even though more and more technology is moving from motherboards to CPUs (memory controller, voltage regulator, etc.), the overall failure rate for Intel CPUs has not increased since last year. For 2015, we saw the failure rate of Intel CPUs to be:

Intel CPUs Failure Rate
Overall .33%
Intel Core i3/i5/i7 .38%
Intel Xeon E3/E5 .2%

One interesting thing to point out is that we saw roughly half the failure rate with Intel Xeon CPUs versus the Intel Core i3/i5/i7 CPUs. Because of this, instead of listing all the different CPU models that are extremely reliable, we are simply going to say that every Intel CPU made in 2015 (especially Xeon) is incredibly reliable.



RAM (Memory)

In the past, Puget Systems has almost exclusively used Kingston RAM in our computers. However, Kingston was a bit slow to release DDR4 modules so for 2015 we also offered Crucial and some Samsung modules in addition to Kingston. 

Desktop DDR4 RAM Failure Rate
Crucial DDR4-2133 4GB (CT4G4DFS8213) 0%
Kingston DDR4-2133 8GB (KVR21N15D8/8) 0%
Crucial DDR4-2133 16GB (CT16G4DFD8213) 0%

Since it's launch last year, DDR4 desktop RAM has had an overall very good failure rate. In fact, the only desktop DDR4 module we sold a significant quantity of in 2015 that did not have a 0% failure rather was the Crucial DDR4-2133 8GB (CT8G4DFD8213) and even that model has less than a 1% failure rate - all of which we found in house, so no customers saw a problem.

Server/Workstation DDR4 RAM Failure Rate
Crucial DDR4-2133 4GB ECC Reg. (CT4G4RFS8213) 0%
Crucial DDR4-2133 8GB ECC Reg. (CT8G4RFS4213) .09% (single stick)
Crucial DDR4-2133 16GB ECC Reg. (CT16G4RFD4213) .14%

ECC RAM is specifically designed to be reliable, but we surprisingly had a higher failure rate with ECC Reg. RAM than we did with the standard desktop RAM. To be fair, ECC Reg. modules are often used in very large configurations (up to 1TB or larger) which may increase the load on each individual stick. The good news is that every single failure we saw (even for the models not listed above) was due to the RAM preventing the system from POSTing. There was not a single instance of memory errors or system instability so even with the higher failure rater, Reg. ECC RAM is still better for ensuring data integrity than normal RAM.

Hard Drive

Just like RAM, we typically only use a few brands of SSDs and hard drives that we historically know to be extremely reliable. In 2015 we sold primarily Samsung and Intel SSDs and the reliability for both brands was excellent. In fact, the failure rates were so low that instead of listing individual models we are simply going to list the failure rates for each of the product lines we used in 2015.

Desktop DDR4 RAM Failure Rate
Samsung 850 EVO (128GB-2TB) - SATA .06% (1 drive)
Samsung 850 Pro (128GB-2TB) - SATA .09% (1 drive)
Intel DC S3500/S3700 (160GB-1.6TB) - SATA 0%
Intel 750 (400GB-1.2TB) - PCI-E .5% (1 drive)

Each one of these product lines had incredibly good failure rates with at most only a single drive having a problem. Traditional platter drives - with all their moving parts - did not fare as well in 2015, but there was one product line that was better than the others:


Western Digital Red (1TB-6TB)
(.74% failure rate)



The good news is that of the other WD drive lines we carry (mostly Blue, Black, and RE), only a single drive from those lines failed in the field. All of the other problems were caught in-house before they made it to our customers.

Video Card

Like motherboards, video cards tend to have a bit higher failure rate than other hardware components. In addition, with the wide range of video cards we offer (including a mix of different brands), naming the most reliable model is a bit tough as we don't tend to sell incredibly large quantities of any one card. However, of the cards we sold enough of to have a good feel for their reliability, there are a few that stand out with a low failure rate. In no particular order, these cards are:

Asus GeForce GT 610 1GB Silent
(0% failure rate)

Asus GeForce GTX Titan X 12GB
(.86% failure rate)

EVGA GeForce GTX 960 2GB SC
(0% failure rate)


EVGA GeForce GTX 750 Ti 2GB
(.83% failure rate)

EVGA GeForce GTX 970 4GB​
(0% failure rate)


This list may look very NVIDIA heavy, and that is simply due to the fact that we sell considerably more NVIDIA cards than AMD cards so there are more NVIDIA models that we have sold enough of to be confident in our data. On the other hand, NVIDIA GeForce cards on a whole have proven to be much more reliable than AMD Radeon cards in recent years. In fact, in 2015 NVIDIA GeForce cards only had an overall failure rate of 1.64% (down from 2.34% last year) versus AMD Radeon cards which had an overall failure rate of 10.2% (which is actually much better than the 17.9% we saw last year). Keep in mind, however, that with the lower number of Radeon cards we sell this failure rate may be skewed a bit simply due to the fact that a single failed card has a much greater impact when the total number of cards is lower.

One last interesting thing we want to mention is why there are no NVIDIA Quadro video cards in the list above even though we sell a decent amount of them. Simply put, Quadro as a whole actually has a higher failure rate than GeForce (about 2.5% versus 1.64%). - although luckily the majority of the issues were found in our initial testing so customers themselves did not see many failures. One reason for the higher failure rate may be that Quadro cards tend to be used more in heavy load environments with the GPU put under heavy load for longer than most GeForce cards.

Power Supply

Out of all the power supplies we sold an appreciable amount of in 2015, there were actually only two models that stand out as being extra reliable:


EVGA SuperNOVA 850W P2
(.76% failure rate - 1 PSU)

EVGA SuperNOVA 1000W P2
(.6% failure rate - 1 PSU)


Power supplies as a whole had a 2.6% failure rate in 2015, but both of these models were well below the average. We especially want to point out the EVGA SuperNOVA 1000W P2 as this is the second year in a row this model made it onto our most reliable list.


So there you have it: the most reliable hardware we sold in 2015. While there is too much data to make many broad generalizations, there are a few trends we want to point out:

  • Samsung and Intel SSDs are actually one of the most reliable components within our systems with a combined failure rate of just .19%
  • Intel CPUs are a close second at .33% total or just .2% for Xeon E3/E5 CPUs alone
  • Even though DDR4 is relatively new, reliability so far has been excellent. Currently, DDR4 has about half the failure rate of DDR3 (.46% vs .9%)
  • Although they are fairly new to the PSU game, EVGA power supplies are the most reliable brand we offered in 2015 with a total failure rate of just 1.63% across all EVGA models (average for all power supplies was 2.6%)

Keep in mind that if you purchase a system from us that includes parts not on this list, that does not mean you are at a significantly higher risk of your machine failing. Our internal testing catches the majority of problems before we ship the system so if anything this list is more about the parts that make the production process smoother for us than parts that our customers should specifically use in their systems

This wraps up the most reliable hardware of 2015. Overall, it was a great year for reliability at Puget Systems. We saw some small increases in failure rates compared to 2014 (such as across all video cards and power supplies), but the majority of those failures were found in-house. As a whole, hardware appears to be continuing the trend of becoming more and more reliable.

Tags: reliability, reliable, hardware, 2015
Prince Chèn

I got a GTX 970 from EVGA, and it failed. My first GPU to ever fail at all, and worse than that: upon arrival . I see reviews on Newegg saying to not get EVGA ones.

Posted on 2016-03-03 01:18:37

Your reply means nothing, that happens with literally EVERY manufacture. So your one experience doesn't mean anything.

Posted on 2016-11-24 04:23:42

Nice post, hope to see more like this. I am so happy with my Asus SABERTOOTH x58, it never failed. I do remember PCs that always had problems of some kind, it was AMD, moving to Intel never had any issue. Worth paying little extra for quality components.

I will refer back for updated list when building new PC.

Posted on 2016-03-04 07:34:03

Thanks! We've been doing one of these at the end of each year for about four years now (list of articles is at https://www.pugetsystems.co.... We're also trying to be a bit better about highlighting specific products that are extra reliable, so watch for "Product Spotlight" articles in the near future! Just to give you a sneak peak, I'm working on an article right now about Samsung SATA SSDs because they are by far the most reliable hardware component we have ever sold. In fact, since the start of 2015 only a single Samsung SATA drive has failed - and that may have actually been due to user error rather than an actual hardware fault!

Posted on 2016-03-04 19:05:54

Any opinion on the huge annual hard drive reliability review by Backblaze? As a cloud storage provider, they had a pool of over 50,000 disks to draw from, which is a pretty sizable data set. They found that HGST (aka Hitachi's storage division, now a wholly owned subsidiary of WD) have the lowest failure rates by a fair margin, followed by Toshiba, WD in third, and Seagate the worst. In particular, they found that 3TB drives generally fared worse than either 2TB or 4TB models, and 6TB were more reliable than any of the latter - although not as great a value in price per GB. Pretty interesting article and worth a read.
I was inches away from pulling the trigger on a WD Red 3TB for my new NAS until I stumbled on the review. Now I'm getting an HGST Ultrastar, which looks very well reviewed and lower cost.

I am aware that a massive cloud provider has quite different needs than an average consumer, so the findings should be taken with a grain of salt. For example, they buy enterprise grade NAS drives almost exclusively, which operate a bit differently and can be quite a bit pricier than consumer desktop drives. They jam 45 drives per server rack, each of which are operating 24/7, so they're taking a serious beating, which means failure rates will be naturally higher. Further, since all the drives have multiple levels of redundancy in that environment, value outweighs absolute reliability. That's why they buy way more cheap Seagate drives, known to have awful reliability, than trustier and pricier HGST models. When one starts to fail, they just pull it and buy more.

Based on the needs of an average person, I'd much prefer a couple of quality drives that will last many years without worry.

Posted on 2016-03-19 04:38:56

I've seen that reliability review and I don't see any reason to doubt any of their findings. We use WD for our platter drives for a number of reasons but they have been very reliable for us. Right now we actually tend to use their NAS (Red) and enterprise (RE) drives for the better reliability but we in the past we never really had any major concerns when we were using the consumer drives.

Really, I think for a typical user HGST, Toshiba, WD, or Seagate would work great. Each brand has it's own little pros and cons but really unless you get unlucky you shouldn't have too much trouble with the drive.

Posted on 2016-03-19 04:46:50

Matt, I've been using Seagate for many years and need to replace several drives due to needing more capacity. I was going to not only get larger drives, but also upgrade to their Enterprise Capacity line. We don't RAID so no need to get their Enterprise NAS HDD. Plus the Enterprise line features 5 vs 3 yr warranty. I got to this forum from a MacRumours Forum thread on WD vs Seagate. I've found the articles and comments amazingly helpful, thanks. The fact that you don't mention Seagate other than in passing now and again, vs your analysis, use and commitment to WD, plus godisafairytale's comment above: "cheap Seagate drives, known to have awful reliability" - is WD really that much more reliable than Seagate? And since we run our 4 bay server (non-RAID) 24/7 - our goal is to get new drives that vibrate the least and are designed to run 24/7, that we should seriously re-consider getting WD RE vs Seagate Enterprise Capacity drives? I'd also love to hear your thoughts on your reply about "I think for a typical user HGST, Toshiba, WD, or Seagate would work great." which follows godisafairytale's comment about Seagate being known to be unreliable. godisafairytale, I'd like to hear your reply too and if your Seagate comment is based on experience, reports, or hearsay. Thanks!

Posted on 2016-08-13 17:31:03

That is really hard to say since we don't have a ton of experience with Seagate here at Puget Systems. So I unfortunately can't really say how much more reliable Western Digital would be over Seagate.

However, what I can say is that while almost any drive manufacturer is very likely to be fine for a normal user, for a 24/7 server like you are doing, I would stick to either HGST or WD. Those are the two brands I hear the most good things about. HGST a bit more on the server side, WD a bit more on the workstation side.

Posted on 2016-08-15 19:05:36

I'd be very curious to see the reliability on X99 boards, G.Skill RAM, the newer Samsung 950 Pro NVMe SSDs, and the newer EVGA 80 Plus Titanium PSUs, if you have any data on those...

Posted on 2016-03-19 04:42:46

We don't use G.skill RAM so I can't comment at all on that. Same with the EVGA Titanium PSUs (we did look at them actually but the cost for the small amount of extra efficiency didn't make sense for us). The P2 and PS EVGA power supplies we do carry, however, and they have been some of the most reliable units we have ever offered. So I would expect the Titanium models to follow suite.

The 950 Pro we do sell, although right now we don't sell a ton of M.2 drives so the sample size is a bit small. From what we have sold, however, they are at 2.5% failure rate (2 drives total failed). That is a relatively high failure rate for a storage drive (especially compared to Samsung's SATA drives which have am amazing track record for us https://www.pugetsystems.co..., but given the small sample size and how much they are pushing performance it isn't too bad at all. Especially since we saw about twice that failure rate (~5%) with the older XP941 that is a great improvement- although again the sample size was pretty small.

As for X99 in general, that is really hard to say. We primarily use ASUS and most of our sales are with either the X99 Deluxe or X99-E WS boards so I can really only speak for those particular models. We tend to go for the models that we think will be the most reliable, however, so I'm pretty sure our failure rates should be relatively accurate for X99 as a whole. With that said, we see about twice the failure rate with X99 boards compared to something like Z170 (5.34% vs 2.64%). However, we catch almost all of those failures in-house so to our customers the failure rate for X99 is only .38% (compared to Z170's .53%). So for us, X99 has more DOA problems but if anything is more reliable in the long run.

Just for reference, for X79 we saw a 11.75% failure rate so X99 is definitely much more reliable than the previous chipset.

Posted on 2016-03-19 05:10:47

Wow, thank you so much for the prompt and thoughtful reply Matt!
Do you have any insight or experience with MSI or EVGA boards?
When you quote failure rate, over what period of time is that measured?
Thanks again.

Posted on 2016-03-20 05:23:38

Those failure rates were all since the 950 pro or the chipsets were launched so it should be time independent. As for MSI and EVGA motherboards, we really haven't used either too much so I can't give much insight. MSI especially we don't use much beyond their video cards every one in a while. EVGA we do use their video cards and power supplies though and have had an overall very positive experience so I would expect that to continue to their motherboards.

Really, we've looked at MSI and EVGA boards in the past but they never perfectly fit our product line like Asus tends to. There was never really a major issue, however, so I don't think there is really inherently bad or wrong with either MSI or EVGA boards.

Posted on 2016-03-20 15:36:38
Ayumilove Ayumi Love

Hi Matt, based from your comment, does it mean that the failure rate posted in this article is mostly on the user side? It would make more sense to know the failure rate of the hardware caught in-house as it would represent better estimation to simulate the failure rate of a user who is purchasing it directly from shop to create their PC themselves.

Posted on 2016-03-23 22:36:35

Do you mean the difference between the failure rate for the Z170 board in the article and what I mentioned in our conversation? Most of that difference is because we have carried more boards than what we listed in the article - the article is simply covering which boards were the most reliable out of all the ones we carried. Since those are the most reliable boards, the average failure rate for that chipset across all the motherboards we have offered is going to be higher.

How many failures we see in-house and how many happen in the field depends quite a bit on the exact model. In general I believe we catch about 9/10 failures in-house and most of the ones we don't catch usually occur after the system has been with the customer for a significant amount of time. Very little makes it past our testing unless it is a failure that doesn't happen until the machine has been in the field for a while.

Posted on 2016-03-23 22:58:34

I got a gt610 from Gigabyte, it was faulty...

Posted on 2016-08-16 18:22:28
Dolorem ipsum

Thank you for releasing this information.

could you clarify one thing? "EVGA power supplies are the most reliable brand we offered in 2015"
did you offer other brands on PSU in 2015?

Posted on 2017-02-01 19:23:16

I believe that year the only other brands we sold a decent amount of were Seasonic and Silverstone. Neither were really bad or anything like that (we wouldn't have sold them if they were), it is just that year EVGA was even better.

Posted on 2017-02-01 19:30:27