I hate to be a critic, but when your highest load measured was 380W and that's with SLI and a power hungry motherboard, measuring the efficiency at anything more than 400W is useless. The most telling page on this review is the one that showed the actual power usage. I think those should be emphasized more. The Zalman 750W registered just 351W at full load. That's under 50%. Why bother with high-wattage PSU's? What a waste of money. Anyone who paid more for anything over 450W is wasting their money. Could there maybe do a review that shows power supplies that are not excessive for people? Otherewise they are appropriating their money incorrectly when it comes to buying a computer. $60 buys you a 550W Antec with all the cabling most people would need. Why would someone pay $130 for something that honestly isn't going to provide you with anything better? $80 will put you a good way forward upgrading from a 22" to a 24" monitor which will last you alot longer than a PSU.

EDIT: Don't get me wrong. I thought the review was very well done. It just doesn't apply to me nor would I necessarily recommend any of these PSU's to people looking to build. There are PSU's out that there that are just as good and are cheaper because they aren't excessive.

Why leave out Corsair ??

Does anyone know what the differences are between the three Silencer 750 Quad's currently on the market, besides color? The three sell at different prices and have inconsistent manufacturer designations. Are there mechanical reasons I should be willing to pay more for the newer blue model over the older red one?

Good review, however realistically why would anybody run their PSUs at near 100% load? Running the power supply so close to its rated capacity is just asking for eminent failure. Apart from an academic curiosity, I wouldn't pick a power supply with the highest efficiency at 100% load but maybe at 75% or 50% load. I think the BFG unit is an overall better choice.

Where did you find a 5:4 CRT monitor to run at 1280x1024 @ 85hz?

Just put together the new comp today, which i got a Corsair HX 520w for. It will be just fine. But i probably wont be able to run any Crossfire on it. Not with another 4870 anyway as they are around 174watts each. Had i gotten a 620w i could have.

The funny thing was in the manual, where asus recommends at least a PSU on 600w for their mobo in a large notice. Of course, when one continues to read, they have then compare 3 packeges, the largest with a tricked out rig using 3 way 8800 SLI and 7 drives so...

A small comp using a high-end cpu, high end graphics and a single drive can easily use a 500w without screaming. The worst thing is the second graphics card. Thats really the one that hits you bad.

As for wires, my card came with a 2 x molex -> 6pin adapter. Two of those together with the two normal PSU leads would give give you 4 6pin leads to run SLI even though the CPU doesnt have enough PCIe-leads.

Just one thing that I'm curious about is the power consumed by these PSU's at 0% load or at the minimal load of a PC on standby.

Considering that a PC might spend 3/4 of its day on standby or turned off with the power still connected, these figures could have another measurable impact on total power consumption.

It's too bad PSUs fail more than anything IME. I'd take a long warranty over any other features if I were in the market now, assuming this much similarity.

Did I miss a table that mentioned that the Silencer 750 is not an ATX power supply, but instead follows the EPS12V form factor, making it too deep for use in many cases?

I'll also mention from personal experience that the 24-pin cable on the Silencer 750 is too short to use in the Antec P182 case, so you'll have to use an inefficient extension.

I would have cared for the off-load.
my new PSU takes 40W if the machine is off.
That is 70 Euros for me per year. Unacceptable!
I use a powersocketadapter with a powerswitch to adress this problem.
Also I'm against the gigantomanie with PSUs, a 400W PSU is good enough for 99% of PCs.

The one thing that's always missing from these PSU reviews (not just TR, but everyone) is how well they handle ripple in the AC input. Lots of people I know live in older neighborhoods and have to deal with a lot of variance on their AC lines. It would be really nice to see how well these PSUs handle AC inputs from 105V through to 125V.

Nice review! I was quite surprised to see the shape of the efficiency curves for some of the PSUs - I've never seen on that starts high and then decreases before, usually they're slightly bell shaped (low at low loads, rising to a peak ~50%, then dropping off slightly at full load). Any ideas? I thought it might be something to do with how the loads are configured, but it seems similarly balanced to other review sites such as SPCR.

I have a question...when will TR review some more 'mainstream' PSUs? Something less than 600W, preferably less then 500W even? It's not that I dislike these, it's just that my personal build...uhh, tastes, tend to lend themselves to more lower-power systems (i.e. 1 graphics card on the lower edge of the midrange, low-end processor, etc.).

Enermax for $99.99 here. Enjoy.

http://www.microcenter.com/single_product_results.phtml?product_id=...

An easy analogy to the excessive PSU phenomenon would be this:

Buy a car with 1,000 horsepower, but by design, the pedal cannot go down further than about 25% of the way to the floor. So you have all of this "capability" that simply cannot be accessed. It is easier on the engine and other parts to not drive beyond 25%, but there are better, cheaper options that fit more closely to what is needed.

I think a lot of it has to do with e-peen stuff. "Well *MY* system needs a fricken KILLOTWATT PSU just to run !!!"

I'm with Usacomp2k3 on this one. Liked the review and interesting points, so I'm not being critical of the subject matter (and it would be immensely useful if I was spending $2500 on a new computer right now, which I do every 3-5 years), but these PSUs *ARE* for super-high-end SLI systems or people that just want to say, "Well *I* have a 750 watt PSU! HAH!".

In the next PSU review, I have a few suggestions to make that would bring up some interesting discussion points on the forums:

1) Try to throw an "odd-ball" PSU into the mix. i.e. a really crappy generic brand or a super-high end PSU in with a bunch of budget-PSUs. These scenarios come up in the forums a lot and it would be interesting to have some results of what happens when you waste $140 on a PSU you don't need or put a $15 PSU in a system that should have a $40-$60 PSU. For example, in the latest review I'd have been very interested to see how, say, a Seasonic S-12 450 PSU would have done with that test system. On the flipside, it would be neat to see a 1.1KW PSU lose in an idle efficiency test on a low-end computer.

2) Have a special "load" category for Folding to see what kind of impact the PSU would have on annual bills for a folding machine.

3) Make a chart that shows Cost Per Year in avg electricity to run at idle or load 24x7. Some PSUs pay for themselves within a year of usage.

Something i would also like to see just for the kicks is of the maximum power ability of today. How far can they really go before crankning out, shutting down or just explode. ;) Of course, i can see why you might not want to do it, or actually, what it actually gives to do it as they all seem to clear their respective ratings. Could still be fun though :P

Although maybe that was just in the old that it was more critical when just everything was "generic" in one sense or another. Back in the old days i used a 250w Fortron source sparkle PSU that went quite abit higher than its rating. This at a time where some other PSU's barely managed their rating. Mine came with the highly recommended fulltower Inwin Q500 and its removable motherboard tray. A favorite among overclockers and builders at the time.

On a more serious note. Have you thought about testing the rated 100% load for each individual voltage rail to see that it works as specified as long as the total maxium wattage rating of the PSU isnt exceded?

In this review accourding to the beast-settings table at 100% you barely got up to half the individual load on the 3.3v and 5v lines, compared what is specified. Although that would of course require lowering the 12v line abit for that test.

Still, i havent read PSU-design, so it might not be something that is worth checking out. Or for that case, have an impact since nobody would ever skew the balance as much?

Id also like to see reviewed a few more budget psu'S that many many more people are likely to buy.

For instance, the OCZ StealthXStream OCZ600SXS 600W goes for $39 after promo and rebate.
http://www.newegg.com/Product/Product.aspx?Item=N82E16817341010&nm_...EMC-IGNEFL080508&cm_mmc=EMC-IGNEFL080508-_-Power...[.com]

As well as the RAIDMAX HYBRID 2 RX-530SS 530W which sells for merely $25 after promo and rebate!
http://www.newegg.com/Product/Product.aspx?Item=N82E16817152028&nm_...EMC-IGNEFL080508&cm_mmc=EMC-IGNEFL080508-_-Power...[.com]

I mean the prices in the ones mentioned in the above review aren't appealing at all and are unlikely candidates for most shoppers who read these reviews

e/ this was meant to reply to #3

Good review Geoff! Found a typo in here:

http://techreport.com/articles.x/15208/10

"Efficiency isn't everything, though, and the Purepower other tricks up its sleeve."

when it should be

"Efficiency isn't everything, though and the Purepower has other tricks up its sleeve."

Maybe I missed it (if so, my apologies, feel free to point it out), but I have a question: What temperature were these power supplies operating at when tested for efficiency under load?

How warm a PSU gets can have a great effect on its efficiency. Some PSU's can't handle the heat over 40C, some high-end models are engineered to 50C, and some cheapies fall off prior to 40C. A well-ventilated case, or a poor one can make all the difference in how one of these units might hold up, so it's a good thing to know.

Nice review guys, thanks.

It is great to see other new brands doing well, but I am glad my old favs are still hanging strong.

I have been using OCZ units for a while, since TR recomeneded them, a while back, and they have been great.

I just picked up a 600W OCZ model as a spare, and because it cost me a a TOTAL OF $40 after rebate. OCZ rebates have been great, and this one came in after 6 weeks or so.

How can you turn down a SLI certified OCZ PS for $40?
I couldnt.

Deal Edit:
The deal is back at Newegg today, what luck.

http://bensbargains.net/deal/73678/

NewEgg.com has the OCZ OCZ600SXS StealthXStream 600W Power Supply for $85 - $11 code EMCAJAFBE - $35 rebate [Exp 9/2] = $39 with free shipping.

Model info:
http://www.ocztechnology.com/products/power_management/ocz_600w_ste...

Thanks, great review

Boy am I 1sty...

I didn't know BFG made power supplies, so I guess it's a good thing they made editor's choice.

Btw, Geoff, Bellingham WA is teeming with cute, hippy baristas. You should take note if you stop by.

Also, I think you guys were a little hard on the 600w PurePower RX by ThermalTake. Modular PSUs are WAYY more desirable than fixed, imo, and it was offering that feature at a fairly affordable price. Yeah, so it lacks 8-pin. That just means you won't be able to throw in some huge, overpriced monolithic nVidia board ;)

great review!

I have a question though, you keep pointing out how some cables are sheathed more so than other cables... and im still trying to figure out why it matters. Is it just for looks or what?

Keep in mind that non-sheathed cables are much more flexible. For example, if you are hiding the cables behind the mobo tray and you are trying to bring the cable out from underneath and plug it in directly to the mobo a full 260 degrees, it would be easier to do with the cables that left a few inches unsheathed. Anyways, I dont think its particularly important at all that they be sheathed in the first place.

Also, I agree with ssidbroadcast, I think modular cabling is very appealing when looking into a new psu. I think its more important than cable sheathing or cable length or the design/color of the psu....

You mentioned that the BFG Tech ES-800 800W lost the editors choice award just because of cable reach, which I think is sort of a separate issue, if an issue at all. But i've never owned a full tower case so im speaking out of my experience box here. 18.5 inches and above seems like it should be plenty to me...but I duno I could be totally wrong.