wiedzmin,

Ralph Lambrecht is well respected here, and you may get a chance to correspond with him here or at DPUG.

If your flash is 1/1000th second and assuming it is valid to compensate for reciprocity failure by adding 10% to development time, guess how that will affect your test results? Your test results will tell you to develop 10% longer than otherwise. I don't consider 10% that significant, because I could just as well be off 15% for other reasons. But reciprocity is a point to consider. Maybe you can find a studio flash with a powerful, 1/100th second flash.

As far as light source issues, everything is a compromise. Electronic flash compared to Daylight is considered a compromise because its spectrum is "discontinuous". But some lab-grade sensitometers ARE made with electronic flash because as a compromise, it is not that bad. Or maybe you can use Tungsten as a compromise, with 80B daylight filtration.

I find Film Speed tests more difficult to nail down. I "develop to the ASA triangle" and then take the point where that curve crosses 0.1 density, as the rated speed of fresh film in standard developer. Once I have that benchmark, I can try a different developer or use expired film stored in unknown conditions, and know by comparison what speed I am getting. This to me, is a great value of film speed testing.

The other great value of film testing is knowing when you are starting to get "out of control".