the way the shop printed the charts was assinine. I was very unhappy with the article when it was finished.
I'll give you the short version.
All the headers with the exception of the GT Spec Ver.2 made power.
The average increase in spooltime was about 200rpm on the first cold pulls, then non-existant on the second pull. (all headers were wrapped w/ DEI wrap)
The Cusco Header made the most power by 1hp (I consider that Dyno error margin)
The GP Moto header was second making about 17hp more across the pull.
The Fujitsubo came in 3rd with about 12hp across the pull.
The Borla, Gruppe-S, and the rest came in about the same with 5-10hp
Please keep in mind that all headers that did not come with an up-pipe were tested with the Helix up-pipe to be fair against the one piece units.
I'm not mentioning the GT Spec Gen.3 as it has been discontinued and had a catastrophic failure shortly after this test (on RiftsWRX's car).
As for fitment, the Cusco, Borla, Gruppe-s, and Fujitsubo fit the best and were the easiest to install.
The GT Spec, GP Moto were the hardest.
If you have any further questions feel free to pm me.
btw- I hae since tested a set of extrude honed/ cnc port matched, jet hotted stock manifolds. The results were interesting. No loss of spool (cast iron holds heat better), and a gain of about 5hp across the band.
For the money spent ($700) to do the port job right (by hand is impossible to remove as much material as extrud hone) and the cross over pipe and the jet hot coating, I would rather spend the $699 on the GP Moto header, $60 on header wrap and make about 10 more whp.
For all those who complain about losing the "boxer rumble", remember it's teh sound of inneffecency.