Fog is not harmless.(posted 8333 days ago)Ed, I always believed that the fog is a constant density over the image and therefore it can be "printed through" with no harm to the image. But recently I found an opposite note in James & Mees "Theory of photographic process" (4th edition): the fog level does depend on exposure, unexposed area has more fog, exposed -- less fog. Thus the effect of this variable level of fog is equivalent to presence of a slight positive image, which of course eats shadow details.
A sort of "good news" is that if we process a test roll and then use the same processing for important negatives, we have the same characteristic curve for test and for real roll, the same shadows etc, thus the contrast as it could be without the fog, may seem to be an imaginary benefit: we cannot achieve it. But it in no way follows that other developer/time/agitation combination cannot give less fog with the given film speed.
Measuring the fog: you have to compare a clear area of developed and fixed piece of film (film base plus fog density) and a piece of only fixed film (film base only). At least if we speak about the chemical fog the film receives in developer.
I consider the fog level as normal if film base plus fog together have no more density than 1 stop. Most my films fit in the range 0.7- 1. The pleasant exception is pulled TechPan: ca .3 stops.
Unfortunately I cannot tell how to decrease the fog. Adding inhibitors (KBr, benzortriazole) decrease the fog at expense of decreasing film speed also.
Please let us know if you find something. I wish you luck.