Howdy back All:
Your choice Nate. Like spankmeister says, "You may never know the difference".
Spankmeister- When I spoke of an "interference" fit, i meant hand-fitted with a hand file and rotary grinder to the tune of about 2 hours of work. It fit so tight that once in place, it was difficult to remove. Once fitted to the port and welded in place, the part protruding from the port was ground off with a slight swell or bow ourward. When the header was attached it compressed against the swell. Even if the welds broke loose, it could not rattle. It's been in there for going on 6 years now with no sign of problem.
Also please not that I referred to "a properly tuned engine" and to "Higher rpms".
I also am not aware of Dyno proof, but that would be a tough one. Some one would have to put on a header, run the dyno, disassemble the head to only install a properly fitted port divider, reassemble and run another dyno test. To prove what? That with a stock head and cam the difference is slight if any. Probably the same would be true for changing only the cam or even for the header.
I have given myself proof that the port divider matters on a flow bench. While flow bench numbers do not directly relate to HP. The numbers were there.
The whole story is that while we were getting a base line on my modified D8 head, we noted a big difference in flow from 1, 2, 5, & 6 to 3 & 4. We mocked up a port divider with some modeling clay to seperate an mimic the other ports and were able to bring the flow up to equal the other ports. That was with 1.46" exhaust valves in an otherwise stock port with a three angle valve job with the valve set at .400" lift.
Your head and specs may be different from mine and you may not get the same results.
From my point of view, when paying $300 plus for headers, the time and effort to properly fit and install a $20 port divider, to equalize flow and to optimize the potential of the header, is time and money well spent.
Again, that's my two cents, for what it is worth.
Adios, David