In article , etclau@hkusua.hku.hkW says...
>I find a similar variation among F3 and F90 as well. I have ask some
>experts ... They told me that if the variation is within one stop, it >is acceptable.

HUH!!!!!!! I find that 1/4 stop error in most color slide materials
shifts the exposure mighty close to the "Opps, guess THAT one goes
into the circular file!" catagory. Having owned many Nikon bodies,
I find that virtually all their meters are either right-on, or
produce 1/3 stop overexposure (easily compensated by setting one
ASA number higher, or using the compensation ring) - this is, of
course, with manual exposure and matrix metering defeated (let's
get serious here about accurate exposure!). Here is a bit of a
run-down (several samples of each):
F2 - (adjustable to 1/6 stop in increments, or infinitely adjustable with a small screw driver)
F3 - 0, +1/3
F3T - 0
N2000 - 0
8008 - 0, +1/6
FG - 0, +1/3
FE-2 - +1/3
FA - +1/3
Hope this helps.