The 1619 in low percentage U-L with E-Linear drivers still seems like the best idea. I know it worked with mediocre OPTx, and with a good one, it ought to be stellar. The 1624 does have the glass envelope advantage...

I am still trying to decide if a 40W core is better than a 20W core design. Or if a tertiary winding rig might be best. I guess it depends on how much you are willing to risk on the build.
regards,
Douglas