The Inquirer-Home

Gigabyte claims stitch-up over EMI tests

Blames shady competitors for excess radiation readings
Wed Sep 26 2007, 19:03

AFTER OUR NEWSDESK got hold of a leaked ETC file - which, frankly, isn't exactly a big deal for us who handle non-certified early testing samples every other day - a separate very detailed story, appeared at some place called Dailytech.

It was only a matter of hours before Gigabyte itself would respond.

Here's the firm's response, in its entirety:

"GIGABYTE Response to Inquirer September 25 article: GIGABYTE P35 Board Fails EMI Tests

All GIGABYTE motherboards undergo rigorous and thorough testing (including EMI testing) before being brought to market. Attached is GIGABYTE official certification for the GA-P35-DS3, where it clearly shows the GIGABYTE GA-P35-DS3 passing all EMI requirements.

It seems a little strange that one of our top selling boards just so happened to have a negative report from the Taiwan ETC several months after the launch of the product. Any motherboard manufacturer or anyone else for that matter can pay to have EMI testing done from ETC. In this case, GIGABYTE has contacted ETC and they have confirmed to us that one of our competitors is actually the one who paid for the testing mentioned in the Inquirer article.

Not to impugn the integrity of the ETC, but any time you have research paid for by a competitor of the study's focus, it does beg the question of how accurate the "report" is. In all fairness, the fact that one of our competitor motherboard manufacturers paid for this report should have been mentioned in the article so readers can make up their own minds as to the accuracy. For example, as mentioned in the article, there is no information in the report about where the motherboard tested came from. Was this a sample purchased from a retail outlet? Were there any modifications made to the board before testing?

Clearly, this is a malicious act by a competitor to negatively impact GIGABYTE motherboard sales, and they have used the Inquirer to do so. Hopefully we can get a retraction or at the very least a clarification as the article has been picked up by several other websites and can have a serious negative impact on our motherboard sales."

Now, we're talking about ETC Electronic Testing Centre in Taiwan, one of well known institutions in this field. We did a check, which confirmed that MSI and Asus boards were also in the same test batch, supposedly passing all fine.

If it was a targeted paid effort to twist results by either MSI or Asus, the party in charge would, presumably, arrange with ETC to "fix" the results of both competitors, not just one of them. And, if the ETC did fix it, and leak it into the public domain - as with most things in Ole Taipei, we imagine a seedy karaoke bar with a drink too many next to a willowy local beauty or two - it would be the end of the road for them.

As a CE (or FCC, or UL) certificate means far more for a complete system box, rather than a board alone, there are many factors which can affect a board level test - for instance, if it sits in a good quality enclosure, the emission measurements would be superior to the same hardware in a lousy casing, not to mention open environment. The audio output, USB and Firewire connections also can play a part.

Then, as readers on various forums have also mentioned, you can reduce " spread spectrum" in the BIOS - it reduces the overclocking margin, but also dampens the noise somewhat. These things are usually done at the final system assembly point.

Truly complete EMI tests are always every expensive, both in money and time spent. And, after all, it is only done by humans with all their weaknesses, including the karaoke one above. This mainboard episode could be just a tip of the iceberg for many vendors, and one never knows what was, in haste or ignorance, pushed under the - hopefully non static - test lab carpet. So, it is advisable, if you care about a particular product a lot, to do a fair bit of testing on your own, based on the parameters that, in the end use, concern YOU most.

Feel free to compare the two studies of these boards: one by ETC and one supplied by Gigabyte from its affiliated test setup - the judgment is your own. µ

The original ETC pdf is here.

Gigabyte's test pdf is here.


Share this:

blog comments powered by Disqus
Subscribe to INQ newsletters

Sign up for INQbot – a weekly roundup of the best from the INQ

INQ Poll

Heartbleed bug discovered in OpenSSL

Have you reacted to Heartbleed?