One thing I just thought of. If you do venture into using ball powders at some point, since they're so common at the burn rate you're looking for, I think temp sensitivity will be far less if you move up to large rifle primers. I personally haven't validated that, but the physics/chemistry behind that should support that, since ball powders typically need a good flame to ignite consistently. Furthermore, the faster burn rate profile that you need lends itself better to spherical shapes than if the powder were of a slower burning nature. They should, in theory, prove much closer to the consistency of an extruded powder at that faster burn rate than at the slower burn rate of the H4350 class of powders. It's theory, but sound theory, and got me thinking, putting myself in your shoes and tinkering. If you went that route, I don't think you'll struggle to hit 3100 fps, and possibly more, with an accurate load profile with several different powders. At that point, in reference to trajectory, wind drift, and effective range, your 308 Win is superior to a 270 Win shooting streamlined 130gr bullets like the Barnes TTSX (I've radared that bullet at a G7 of .209, compared to the 135gr you're shooting at .216). Not to mention you get the benefit of a "30 cal hit" on an animal with the superior terminal performance of the Afterburner. Kind of makes the good old 308 Win look even better than it normally does! Hard hitting AND flat shooting! Below are a couple quick references to look at in your testing. I trust the Hodgdon/Western information more than the Barnes info (having the pressures listed is very helpful and will help answer any QuickLoad formulated pressure/velocity discrepancies), but the Barnes is still a great reference. You may be able to get away with some of the loads below that show a compressed charge, since the Afterburner is 5gr heavier and may require a little less powder at the top. Hard to say, but I would try and consider something not showing as compressed. Benchmark, one of your preferred I believe in extruded form, is showing as a viable option (QuickLoad shows it compressed pretty substantially, but the Barnes info does not).