Article updated 2 May 2005

SkyHawk Power One GM620SC and The Emperor's New Clothes






When I originally Googled SkyHawk Power One the entire first page listed sites which in turn linked back to the original Legit Reviews article. Fortunately that article has now been restored. The issue at hand is what SkyHawk has done since the original article to repair their reputation and more importantly the PowerOne itself. As the evidence above shows this company did not hesitate to own up to discrepancies found and immediately set about rectifying the issue. You can you imagine if more auto-makers did this? If Politicians did this? If people did this? The world would be much different, and perhaps we could drop the term litigation; from our Lexicons.



I've chosen to tear down my unit for you the reader, to give a clear, concise view of the original wiring found on first revision units. The yellow wires are those responsible for all 12V current.



The photo below exemplifies all lines originating from their perspective sources on the mainboard. To the far right are the yellow 12V lines. The 12V silk-screen can just be made out.



A closer view of the 12V origin from the board.



Now flipping the PSU on it's back we see where and how these lines originate from a single source, where they should be separated if they exceed 18A and/or 240VA.



Finally a macro of the area originally revealed through the actions of a concerned Legit Reviews forum member and their article which followed.



I've read all 33 pages of the BTX White Paper, which from what I understand is the brainchild of Intel. The very same issue which has the Power One under scrutiny are the EPS12V Spec 2.1 found at SSIForum.org, which stipulates multiple 12V rails must be employed where greater then 18A are present and potential exposure to current greater then 240VA exists. This is in accordance with Safety Standards for Information Technology Equipment, IEC 60950-1. Below we find the section referred to in the original article EPS V2.1 Sec.6-1 (there's also a more recent EPS 12V 2.8 for 700 ~ 800W).



Onto the ATX power design aspect of these standards which are outlined in the Formfactor.org document ATX 2.01 which is itself 44-pgs in length. Section 1.2.1 (pg.7). Specifically the section below are those Sky Hawk failed to meet, yet claimed their PSU had.




Unbiased Journalism:

Insofar as the original article I wondered (although I intuitively knew) why it wasn't left in-tact with an Update so potential consumers of the Power One could understand how and why it failed to meet certain standards. We now know Legit Reviews was threatened with legal repercussions for their publishing of that article. After having a few discussions with Nathan at Legit Reviews, I understand the disagreement was their wording in the article. It's likely had one End-user in the Legit Forums not brought this issue to light, none of us would have known about the solder-points on the 12V rails?

What the ramifications of a failure to meet the EPS/ATX guidelines we can only surmise. To date I don't know if anyone has claimed these units caused specific damage to their systems? It seems the original investigation while revealing the PowerOne's failure to meet specific standards, did not show that was the reason behind that person having difficulty. Nonetheless, we should credit Legit Reviews with their original publication.

One reason I initially criticized them was their retraction of the article, however; understanding why led me to fry bigger fish. Truth is, the world’s fish populations are in short supply. Do to over zealous profiteers a feeding frenzy can often result with no concern for the future. While I find investigative journalism is an integral aspect of our purpose, it shouldn't be the encompassing or driving force. There are other venues for that type of writing. We are here to provide our readers with accurate information pertaining to al products which pass through our Labs. Additionally it's my duty to seek out products of special interest, albeit cutting edge, or those which offer superior performance ideally for low cost. Even with the PowerOne's misgiving's I thought it had potential.


As a product tester I take the time to perform scientifically accurate, repeatable, verifiable, tests backed by thorough research. It’s these tests and accompanying benchmark utilities themselves which are the common unbiased denominator. That a writer distance themselves emotionally from the product they evaluate is not nearly important as the skill it takes to maintain objectivism in a test environment. Reducing the number of variables and/or any condition which might give one product an unfair advantage over another is paramount to allowing yourself to "like" one product more then another. It's a given you’re not going to create a "favorable" test environment for a product you "like." In fact for the most part I write manufacturer's whom I believe are building a better mousetrap, because I'm already using my experience for the good of my readers in seeking out the most exciting products. What separates mediocrity from an exceptional review is the ability to communicate esoteric material so it's comprehendible by the layperson as well as the technically inclined. However in the writing of the writing of this article I've learned valuable lesson, while Hardware Review sites may accept advertising fees from the very same industry they scrutinize, it is not necessarily the case there will remain an inherent conflict of interest although it is a thin red line. It is up to us to point out the fact the The Emperor has no clothes.