Silverstone Zeus 750w ST75ZF Quad Rail PSU
Testing
Published: 10th May 2006 | Source: Silverstone | Price: |
In order for the results from all current and future PSU reviews to remain fair and comparable, Overclock3D uses a custom built Power Supply load stress tester.
The tester is capable of placing loads on the following rails:
+3.3v - 20a Load
+5.0v - 20a Load
+12v1 - 10a Load
+12v2 - 10a Load
+12v3 - 10a Load
+12v4 - 10a Load
(or 40a on a single +12v rail)
The results are collected from a Mastech MAS-345 Multimeter which logs its readings via RS232 to a PC.
None of the 4 rails dropped below 12.00v and most only experienced a drop of 0.02v with a 10a load applied to each of the rails - that is simply AMAZING.
Silverstone have clearly spent a lot of time and effort ensuring that the Zeus is of the highest quality, and in my opinion, this places the unit directly in fierce competition with the likes of PCP&C and Zippy.
Noise Testing
At present Overclock3D doesn't have the correct equipment to reproduce accurate stats on the dBA output, so for the moment you will have to rely on my ear :)
The Zeus is not a quiet Power Supply - I will say that now. From what I could tell the installed fan appears to run at 100% even when idle. As stated on the previous pages, the Zeus uses a Sanyo Denki San Cooler 80mm fan that pushes 42cfm at 34dBA.
Having listened to various fans, I would say that the noise output is closer to 30dBA. The installed fan does not make any mechanical or high pitched noises, just the noise of air turbulence which is to be expected.
Check back on this review at a later date for professional noise level testing results.
Most Recent Comments
Jesus - that is rated with twice as much power as my current PSU!
I'm liking the current idea of only needing to plug in the connections you want - leaves a nice tidy box (as the actress said to the bishop).
Yes, i'm a big fan of Modular too (my last 3 PSU's have been).
Coming up soon is a Silverstone Strider 600w Modular review - should be a nice bit of kit :)
It also worries me slightly that 18a might not even be enough to power the latest generation of ATI cards (x1900xtx) especially when used in crossfire mode as the current draw from these cards is quite unbelievable.
120.6w according to xbitlabs if yoi wanted to know specific numbers :)
http://www.xbitlabs.com/articles/video/display/gpu-consumption2006_4.html
but all in all a good review but a way to pricy psu for my taste.
120.6w according to xbitlabs if yoi wanted to know specific numbers :)
http://www.xbitlabs.com/articles/video/display/gpu-consumption2006_4.html
but all in all a good review but a way to pricy psu for my taste.
So thats a good 10a for a single card (despite hearing some cards drawing more).
However, now that Silverstone have confirmed that they are re-arranging the PSU rails, this problem doesn't really apply.
Cheers for the linkage :)
So thats a good 10a for a single card (despite hearing some cards drawing more).
this is the methode xbitlabs uses and sounds to me like it could be accurate.
We utilize a special modified mainboard that has special shunts attached to PCI Express x16 power lanes as well as we attach the shunt into additional 12V power connectors that some of the graphics cards have.
but wouldnt surprise me at all if it differers between manufactures too, some like to raise the speed and/or use faster memories for example.
do you have any thoughts of starting to measure power consumption like this ?
at present time i have no reason to believe xbitlabs measurements are off, i looked at them before i bought my 7800gt and come to the conclusion my psu could manage the extra load and it has, flawless.
As for OC3D measuring power consumption, this is not something we are planning on performing at present, but if there is a need for it then we'll certainly look into getting the equipment.
However, on the PSU side of things, we will very shortly be measuring: Power Consumption (for Efficiency), Noise and Heat. The key thing for us is making sure that all of these tests are performed under the same conditions, so results are comparable.
As for OC3D measuring power consumption, this is not something we are planning on performing at present, but if there is a need for it then we'll certainly look into getting the equipment.
their isnt directly a need for it but from my point of view it would be interesting as its one part i look at when researching a card.
also interesting to know if a current psu can manage the load or not, not everyone can just go out and buy a stronger psu if problem should arise.
one simple way but not so accurate is to only measure the pcie power cable, remember some site measuring that way back in the agp days.
However, on the PSU side of things, we will very shortly be measuring: Power Consumption (for Efficiency), Noise and Heat. The key thing for us is making sure that all of these tests are performed under the same conditions, so results are comparable.
for psu's efficiency and noise is the 2 most interesting aspects so will be interesting to se it comming :)
if you havent already guessed it i chose components to minimize power usage but still be powerfull for what i do.
CSS Valid
I'm liking the current idea of only needing to plug in the connections you want - leaves a nice tidy box (as the actress said to the bishop).