facebook rss twitter

Review: Akasa vs FSP - 500W PSU shootout

by Josh Blodwell on 2 April 2007, 08:48

Tags: Akasa, FSP Group (TPE:3015)

Quick Link: HEXUS.net/qah72

Add to My Vault: x

Thoughts, HEXUS.awards, HEXUS.where2buy,HEXUS.right2reply



Both the FSP Blue Storm II 500W and the Akasa GreenPower 500W managed to pass all of our standard load tests. They both lived up to their 500W labelling, and, in fact, their specs were so similar that you can barely tell them apart by looking at the test results.

Both are more than capable of handling a fairly powerful system, according to our test results, then.

We did, however, come across some shortcomings in both of the supplies.

We'll cover the FSP first, as we found fewer significant problems with it. The first problem was its claim of 85 per cent efficiency: it was 'only' 81.4 per cent efficient in our tests. It's close, but not technically accurate. It would be better if it was labelled 80 per cent-plus efficiency, like the Akasa is.

The only other fault we could find was the rippling on the PEG connector, attached to the primary 12V rail. While this was within the specifications, it caused us some concern as we would have liked to see a nice, clean output.

While the Akasa did manage to pass our tests, we came across a significant problem. Both PEG connectors are attached to the primary 12V rail. This means you may come up against it if you attempt to run a pair of high-powered graphics cards in SLI or CrossFire mode.

Further, one-half of the EPS connector (pins 5-8, CPU2) was also sharing the primary 12V rail, too.

Our tech. hound, James Smith, chased Akasa for a formal statement about the wiring on the GreenPower, and here is what company spokesman Adrian Young had to say on the matter:

"The AK-P050FG8_BKUKV1 has two PEG connectors on V1, this is fine for a mid-range SLI configuration as a potential hi-efficiency PSU upgrade. We tested this PSU using 2 x XFX nVidia GeForce 7800GT. The system was running stress tests and 3DMark06 all day and performed faultlessly. However due to the growing public realisation that a high quality 500w PSU can drive a high end SLI system we have decided to change the 500w 80+ (to AK-P050FG8_BKUKV2) to accommodate two PEG connectors on different rails (V1 & V2) to support high peak demands This new version should be available imminently."

As was mentioned previously, we tested the supply with an Intel Core 2 Extreme X6800 system, running a pair of GeForce 7900GTX cards in SLI mode and the supply worked fine. However it may make sense to wait for the second revision of the supply, which should be available very soon.

If Akasa does alter the wiring in its second revision of the supply, and it remains the same price, then it would be difficult to choose between the FSP Blue Storm II and the Akasa GreenPower. Purely on the basis of this test, though, we'd have to recommend the FSP over the Akasa. It's a little better and cheaper.

HEXUS Awards



Essentials Recommended
FSP Blue Storm II 500W


HEXUS Where2Buy

The Akasa GreenPower 80+ 500W is available for £62.15 here, and the FSP Blue Storm II 500W retails at £54.99 here

HEXUS Right2Reply

At HEXUS, we invite the companies whose products we test to comment on our articles. If any of the company representatives for the products reviewed here choose to do so, we'll publish their commentary here verbatim.

HEXUS.related reading

HEXUS.net - HEXUS.reviews ::400-500Watt PSU shootout
HEXUS.net - HEXUS.reviews ::600-700W PSU shootout
HEXUS.net - HEXUS.reviews ::HEXUS PSU (Power Supply Unit) Roundup - Taoyuan 2005 - thirty-four tested
HEXUS.net - HEXUS.reviews ::Zalman ZM400A-APF PSU


HEXUS Forums :: 7 Comments

Login with Forum Account

Don't have an account? Register today!
I am a bit confused about the testing - I must be missing something :confused:

The same power enters both units

The Akasa is more efficient (i.e. more power is ‘throughput’ than with the FSP)

…but when you compare the temperatures before/after load…

…the Akasa gets almost 4 degrees hotter

My imagination - or this does not (quite) make sense ?


If only there were a highly qualified accountant in the house - so that we could get a certified tally on the Watts ;)
It just means it's more efficient at the maximum draw. If efficiency at mid-draw is worse then more heat could be produced.

Interesting comments about the load sharing on the rails - it seems to have taken some manufacturers a while to get away from the idea of super power hungry CPUs. But then again there is the arguement for ‘clean power’ and by keeping everything on a seperate rail from the CPU they might be gaining stability - could explain the lack of ripple compared to the unit which puts the second PEG on the same rail as the CPU.
Probably the most important component in your system but often neglected

Hehe I thought we had seen the last of that phrase at the start of PSU reviews… I think everyone who is involved in computers now is well aware of their PSU choice.

I found you statement about Radeon X1950XTX a little alarming. You meant single card rather than Crossfire yes? I can't imagine hooking up two expensive cards as well as my entire system to a 500W PSU.
DeSean
Hehe I thought we had seen the last of that phrase at the start of PSU reviews… I think everyone who is involved in computers now is well aware of their PSU choice..
Looking in the Hexus.Hardstuff section, you may be surprised at how many are anything but aware of it :)
DeSean
Hehe I thought we had seen the last of that phrase at the start of PSU reviews… I think everyone who is involved in computers now is well aware of their PSU choice.

I found you statement about Radeon X1950XTX a little alarming. You meant single card rather than Crossfire yes? I can't imagine hooking up two expensive cards as well as my entire system to a 500W PSU.

I think you'd be surprised what can be run by a high quality PSU.

Alot of the power requirements are over exaggerated by the PSU companies - the inquirer ran a quad core, 8800gtx SLI system on a 500W psu easily without problems

http://www.theinquirer.net/default.aspx?article=36066