>Yes, I agree, but only with the first part of your argument. All lenses, >even 35mm ones, produce a circular image field.
Well. some lenses with flare cutting diaphragms, project a almost square image field. Moreover a lens with a petal shaped hood can also project a square image field. But its true that the image field is "usually" circular ;-) >The image captured on film >is merely a "bite" out of it. In the case of 35mm the circle of capture is >barely larger than the film frame, but it exists nonetheless. And it's very important that the image field is not much greater that the film format because the non used part of the image field contribute to flare. >A CCD/CMOS is smaller than a 35mm frame, so the photo made is a smaller >section of the image coming through the lens; however, the lens is producing >exactly the same size image circle regardless of the format behind it. The >angle of view of the lens does not change (how can it?), but the smaller >sectional area of the image circle used to form the picture changes the >effective AoV as WE perceive it. It requires some lateral thinking to get >ones head round this concept! Ok. >So, when using a flash we need only to cover that part of the subject area >that will be captured on our digital chip. Any extra is waste. Not only is a waste in GN and/or power it also contributes to flare! >This I'll >cheerfully admit is theory only, the proof of the pudding is in the eating! >I shall await Lars' test results eagerly :o) I thik it's more that theory. If you set in a flash designed for 35mm a focal lenght of a smaller format body, you will get an excess of flash coverage. You will loose power and increment flare. And, if you use in a D30 (or in a aps body) the matched hood for a given lens, you are getting less that optimal protection. Moreover all 35mm format lenses, if are used on lower format bodies (film or digital), will give more flare than an optimal lens for this format, as the image circle (or square) is quite greater than it should be. I suspect that Canon won't give us optimal hoods to be used on digital bodies. Best regards Vicente * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
