Very surprising PSU failure rate chart

whats up with all the reviews of the enermax galaxy 1000w psus? they say something is dodgey on the 3.3v rails?
for a psu of such a high end ranking it should be damn perfect. :confused:
 
Could be the results were done before Corsair came to the market.

They are apparently made by Seasonic but with higher quality components inside.

Thats the problem with power supply units, too many re-stickered OEM units to really tell what you actually have in your rig.

Not sure about Enermax, always refused to use them as they don't work well with DFI boards (which I think is due to their 5v rails - sure I read this some where in the past).
 
i ordered a 620w corsair psu since i needed a new psu for a new machine im putting together, and corsair was the cheapest comming in at 73 quid for the 620w one.



someone should create a poll to see what psu would be best to get in certain price brackets.
 
Hiper didn't do so well if you read the rest of the page :

'Hiper's rates oscillate from 0.4% and 3.8%. The two models above 2% are the modular R 580 watts.'
 
Keep in mind that those figures are based on reported issues.

As antecs are commonplace and included in many cases, they would have larger volumes and therefore be subject to greater reports of failure.

I've always known of Antecs to be quality PSUs, never had one die on me, or heard of anyone have one do so.

Seasonic all the way though! :D
 
krooton said:
Keep in mind that those figures are based on reported issues.

As antecs are commonplace and included in many cases, they would have larger volumes and therefore be subject to greater reports of failure.

I've always known of Antecs to be quality PSUs, never had one die on me, or heard of anyone have one do so.

Seasonic all the way though! :D

But I think having percentage failure rates eliminates the issue of more sales=more failures.
 
Interesting but it doesn't really seem to tally with the experiences of the users here so I'm unsure how much weight I'd put on it given the lack of raw data. I'm surprised at the Antec percentages but not altogether convinced of them, it also doesn't take into account the various uses that people put them to e.g. how loaded the system is so people may be attempting to run completely unsuitable systems on any of the PSUs or may be underestimating others.

The English utilised in the article is also somewhat questionable "For Fortron, the figures are a bit more disparate from 0% for the most reliable model to 1.5%" - disparate does not mean constrained within a fairly narrow band, particularly when it follows that sentence with "Hiper's rates oscillate from 0.4% and 3.8%" which would surely be the more logical place to use a word like disparate or indeed about the Antec reported figures where they really are disparate varying as they do from 2.6% to 21.6%.

It isn't likely to change any of my recommendations thus far.
 
I think its a percentage of reported failures, and not sales? Thus Antec would receive more complaints by the virtue that they sell more PSUs.
 
ive been saying this all along, hipers are probably better than most of the psu's sold here, just that some people jumped on the lets flame hiper bandwagon, mine has been going strong for months and im running a beast of a rig
 
Would direct RMAs not also figure in to this somewhere?

e.g. Tagan like customers to deal direct with them (and OcUK have a notice on the site saying contact Tagan if RMA required), so OcUK shouldn't get as many complaints about Tagan PSUs if they aren't dealing with the RMAs at all.

There are so many flaws in the way the data was collected that it renders the results practically useless.
 
Phil99 said:
Would direct RMAs not also figure in to this somewhere?

e.g. Tagan like customers to deal direct with them (and OcUK have a notice on the site saying contact Tagan if RMA required), so OcUK shouldn't get as many complaints about Tagan PSUs if they aren't dealing with the RMAs at all.

There are so many flaws in the way the data was collected that it renders the results practically useless.

but hipers had the 2nd lowest failure rate, surely the would have been placed a lot higher, but the fact is hipers dont blow, theres just a lot of hype going on, sure the data aint the best but it does give a realistic view, it would be impossible to truely know the real figures because of the numbers sold by different countries and companies
 
thing with the hipers if the failures are all centered around the 580w models, and do remember when people buy psu's they probably get less powerful ones since they don;t need all that power. do remember though that hiper do do 525w, 425w, 480w and 530w psu's too and those seem to have a very low failure rate. dunno about the new 670w and 730w ones though, they have very good specs but not seen any in reviews yet.
 
Psycho Sonny said:
but hipers had the 2nd lowest failure rate, surely the would have been placed a lot higher, but the fact is hipers dont blow, theres just a lot of hype going on, sure the data aint the best but it does give a realistic view, it would be impossible to truely know the real figures because of the numbers sold by different countries and companies

How do you know that it provides a realistic view (any more than I know it doesn't)? Do take note however that if you believe the data there the Type-Rs are the ones with the highest failure rates of the Hipers which are the ones that people are (or rather should be instead of just a blanket warning against Hiper) more wary of.

As I've said a few times, Hiper aren't all bad but from reading threads about them dying on here it does seem like the Type-Rs are more likely to have problems than would be expected. It could be unlucky or it could be a more fundamental flaw, I just don't know.

The figures also of course don't take account of people who change their PSU before it dies, I know of quite a few people on here who have replaced their Hiper after advice (whether good or bad) so who can say if they would subsequently have catastrophically died. The point being that there are too many variables and unknowns about that data to make it particularly useful. :)
 
millsy888 said:
antec at 9.5% :eek:

Antec used to be my first choice when it came to PSU's - I've owned 3-4 over the past few years and they've all been perfect. However, the past 12-18 months they've really gone down the drain. I wouldn't even consider one now. I hope they get their act back together :(
 
When i got my Antec Neopower 480W about 2 years back it was one of the early modular psu's, the main reason why i went for it really.

Cost me £90 back then but am extremely happy with it, been no problems and now that i have a PCi-E gfx card am thankful for the dedicated PCi-E power cable it came with.

Quite surprised to see the high failure rate, but as said above they seem to be commonplace with Cases that come with PSU's.
 
I guess the reason that the Hiper's (especially the 580w modular) get so many complaints is because a lot of people on this forum and other forums have actually bought them. They are very affordable, offer good performance on paper and have been used without problem by a lot of people (like me). It's no great surprise that there are a good number of complaints.... they are a disproportionately popular PSU in that price and power bracket.
 
man_from_uncle said:
I guess the reason that the Hiper's (especially the 580w modular) get so many complaints is because a lot of people on this forum and other forums have actually bought them. They are very affordable, offer good performance on paper and have been used without problem by a lot of people (like me). It's no great surprise that there are a good number of complaints.... they are a disproportionately popular PSU in that price and power bracket.

I think even taking that into account there were a lot of quality issues. Fans clicking and ticking, fans fitted the wrong way around, strange noises. I had a few friends with Hipers and saw 2 blow up, and one get extremely noisy.

With all the contradicting information around all you can go on is personal experience and recommendations.
 
Back
Top Bottom