If you don't want to read the whole post, skip to the next bold section.
Can you tell the difference between a good connector and a bad connector?
Can you tell the difference between a PSU that will run a PC for a week then kill it and a PSU that will run a PC for five years and shut down peacefully?
Those are the differences real reviews are for. The majority of the truly crap PSUs on Newegg will turn a PC on. They may even run it for a while. But all you have to do is look at the $60 1000w PSU reviews, there are a lot of positive reviews for PSUs that blow up at the 450w mark.
Not useful.
As a note, I
was one of the people being sent PSUs, GPUs, SSDs, motherboards, etc. in return for 2000-5000 words.
http://www.overclockers.com/author/bobnova
Newegg "reviews" can't even
be 3000 words, they're length limited to a hell of a lot shorter than that. As a note, never ever call it "free". Reviewing is a lot of work. The average
real PSU review took me between 5 and 8 hours, on test equipment that cost a decent chunk of money and took another dozen hours to make (or, if you buy premade, $2000 to $14,000). That's the same five to eight hours for a unit that I could sell for $200 (really, really, really, really rare), as a unit I could sell for $90 (more common), as a unit I could sell for $40 (most common), and as for a unit I could not sell at all.
Before I got involved in reviewing, I shared your viewpoint on reviewers. It was my opinion that the reviewers that complained or weren't completely enthusiastic were a bunch of whiny *******. Now however, I am one of them, and I understand. It is a complete pain in the rear. Enough of one that
I told all the sites and manufacturers that I was working for and with to stop sending me stuff. Think about that for a moment.
That, or go spend 45 minutes taking and resizing pictures, an hour and a half sitting in a cold room and/or with
really noisy fans taking voltage readings with a multimeter and looking at a scope picture, an hour inserting and captioning the pictures you took and resized earlier, another hour taking apart a PSU that will almost certainly cut you
somewhere and identifying all the parts and taking more pictures and trying not to do any damage to it, and two hours writing about something you've already written about a dozen times (this is my average PSU review. If the PSU did something interesting, it was longer). Then spend another 15 minutes (or an hour, if it's a be quiet! Dark Power Pro) putting the PSU back together, another 30 minutes re-testing it to make sure you didn't kill it, ten minutes making a classified/etc. listing, and a month or so waiting for someone to buy it for a profit after shipping of $50.
Then let me know how "free" the review items for real reviewers are.
If the average reviews were actually useful reviews, which a decent number of people in this thread go to some effort to make (which I appreciate!) it'd be one thing, but the average reviews are not useful.
Some of them, such as positive reviews for fire hazard PSUs, are downright unhelpful bordering on dangerous.
TL-DR section on PSU reviews:
"Overall perceived quality" makes for a crap review. "Actual quality" makes for a good review. Without proper tools or at the
very least significant knowledge and pulling the PSU apart (a hazardous endeavor) it is simply not possible to gauge the difference between a genuinely good PSU and a well disguised crap PSU.
I, at this point, could do a look-at-it-and-turn-it-on review and tell you if a PSU is crap. I
cannot do a look-at-it-and-turn-it-on review and tell you it is good. I can tell you that it might be good, based on X Y and Z, but I have
sitting right next to me three PSUs that
look great, have great components from good companies, and are on a totally reasonable platform, that have ripple over
ten times the amount allowed by the ATX specifications in certain conditions. Hit one of those conditions with one of these units powering your PC and your PC is
going to die.
How do I know? Dedicated test equipment and careful testing. There is no other way.
So yes, I do think that the program should be shut down, and now you know why.
