> I recently got the 550EX for my EOS30 and I was also surprised to find > several slides coming out a bit darker than expected. > > If it is true that ETTL generally underexposes by 2/3 stops, them this > must be a widely known fact in this group, I suppose. > > So my question is if anybody else has observed the same underexposure with ETTL? > I think what most people don't get is that E-TTL uses the same metering system used for the 'usual' though the lens metering of reflected light. The catch is that E-TTL only uses the center focusing poing or the selected FP (based on Custom Function 8) and bases the exposure on the subject placed behind it (*).
As a metering system for reflected light, it's heavily dependent on the subject and it needs the corresponding exposure compensation (FEC in this case). The proof is that when using a grey card to meter off (using FEL), the results are invariably consistent. A underexposure of 2/3 of a stop can be due to a pale model who's skin is just 2/3 stops lighter than 18% grey. Overexposure can be due to a black cat (giving you bad luck ;-) ;-) ;-) So, when working with E-TTL, think as you think when using partial metering (*) and apply the corresponding compensation. ETTL is no wonder solution, but it's very useful. Much more that TTL systems where the flash (or camera) "decides" when to stop instead of deciding "how much (light) you actually need". Here I'm not defending the particular implementation of the D-30's ETTL metering which has been reported as *very* inconsistent many times. (**) Regards, Gerard. (*) my experience is based on cameras with partial metering sensors. On cameras where a spot meter is available (-3, -1V) I don't know what is the metering rule. (**) I haven't used a D-30 (yet :-) * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
