And I want to pick on the flow rate argument a little bit....

K20A.org .:. The K Series Source . Honda / Acura K20a k24a Engine Forum - View Single Post - Flowbench data k20a2 head Here's a stock K20A2 head. I don't even think it's possible to get those numbers out of any type of modified LSx head. Not to mention they strap a 300+ duration cam to it. Yeah it's a 4cyl, but that leads up to my next point....

Flow rates are measured with the valve held static. I've got a 2.5" OD pipe that flows 550cfm. So what. Actual flow into the engine depends on the dynamics of the cam and the effect on VE of the surrounding components. THERE's where you make power. By pumping air. So lets recap some air pumping volumes versus specific output: BMW makes 500 on 5.0L versus LS7's 500 on 7.0L. Audi makes 420hp on 4.2L versus LS3's 430 on 6.2L. Hmmm. So maybe Germany makes better headers than the US and that's where the everything goes wrong? Not likely. So maybe its the combustion chamber design? Oh yeah, LSx has 2 big coasters versus a 4-valve cloverleaf. Yeah, but we're still talking small gains. Hmmm... maybe it's the shit flow through the head. You just can't hammer the valves open with a pushrod like you can with a direct acting valvetrain. I mean, at least they're up to 1990 with variable valve timing, albeit you can't vary intake and exhaust indendently or change the overlap. Engines are air pumps. If you aren't making enough power, you're not moving enough air. Period.

So back to it: Yes the LSx is a great motor.... in spite of the pushrods, not because of them.

So, uh, no Northstar fans in here?