Tim Higgins of SmallNetBuilder fame has detailed his thorough experiments on the affects that products with the Atheros Super-G chip have on nearby WLANs: If you have any interest in the issue, you should really read this because he does a great job of explaining his methods and results.
He also goes the extra mile by realizing that it might be a good idea to test the Atheros products against non-Broadcom gear. Turns out that was a great idea because products with Broadcom chips realize greater degradation then others.
With his own studies in hand, Higgins draws some great conclusions. For one, he notes that Atheros' response to Broadcom's claims was essentially, "Is Not!" and that response isn't appropriate. While Atheros was forthcoming with Glenn Fleishman when he and I wrote a story about this situation for PC World, the company hasn't gone out of its way to offer any solid data that might disprove Broadcom's claims.
Higgins suspects that Atheros knew it should test the Super-G product more before shipping but likely was just anxious to get to market with a product that doubles the speed of other gear. We all know it's a competitive market out there, but introducing products that could have serious affects on other nearby products isn't the way to grow a market.
This whole situation contributes to recent whispers in the industry pointing to the potential for real trouble with the lack of interoperability among 802.11 gear. The more competitive the market the more vendors will be tempted to add a special sauce that can differentiate their products. The trouble is that everybody loses if that special sauce isn't compatible with everything else. What's the point of banding behind an industry group like Wi-Fi Alliance if vendors don't get their products tested or if the group doesn't do stringent testing?