Limit image circle to only usable area?

[Darin Boville]

This thread

http://www.largeformatphotography.info/forum/showthread.php?t=56270

is talking about the visible image circle vs. the real limitations due to sharpness, etc.

So a thought occurs to me: Is there any way to limit the full image circle to only that which is truly usable? Some sort of rear lens shade or...? Seems like having image are that you really don't want to use can be a problem...

--Darin

[mikec]

You would lose some of the movement capability. However, I was wondering if there might not be another disadvantage with large image circles. For instance, using a long process lense on a 4x5. There is a lot of extra light sloshing about in the bellows that could cause fogging due to intenal reflection all be it minimal. Has anyone experienced that?

[Joanna Carter]

Using a compendium shade is a good way to help avoid internal and bellows flare due to large image circles.

[GPS]

Of course you can do it. With a DIY lens shade built in the proper way you can eliminate all the image circle except for the image itself. A great advantage of this is a superb gain in contrast and color saturation. And yes, movements are then prohibited. But not all lens shades limiting the image circle to the same degree are the same! The longer the shade is, the more you avoid the external light causing flare - with the same limitation of the image circle area. Go wonder...

[Brian Ellis]

"Truly usable" is to some extent a matter of personal standards of quality. It's not like image circles are usable up to a precise point and beyond that point immediately become totally useless.

[dave_whatever]

Another thing to remember is if say you're using a lot of upward rise to take in the top of a building or mountain and you're running into the part of the image circle where the performance is a bit poor - chances are that part of your image that may be softer as a result may contain only sky, which isn't generally going to look bad in the final image if the sky with no detail is a bit soft.

[rdenney]

The light rays approaching and departing from the lens form basically the same cone on either side of it. Thus, a compendium shade on the front can be adjusted to accomplish this goal probably more effectively than a short shade on the rear. That said, some Canon lenses (and, I assume, other brands as well) improve their MTF by including a rectangular baffle on the back of the lens. I'm not sure it improves images to any noticeable extent, but they compete on the basis of reported MTF to so they probably design for good MTF numbers.

In terms of image quality, I judge that with a loupe on the ground glass. A 4x loupe gives me about the same sense of image quality as I would get closely inspecting a 4x enlargement without a magnifying glass.

I'm not too worried about light inside the camera. For one thing, the inside of the camera bellows are the same construction as any shade I would put on the rear of the lens. The only thing I worry about is the image reflection from the shiny emulsion lighting a shiny (even if black) lens board and bouncing back onto the film. That's why I cover the back of my off-brand shiny lensboards with flocking paper.

Most modern wide-angle lenses seem to maintain good resolution even as they are fading to black at their edges, at least at small apertures. I suspect MTF drops as much because of the falloff as because of fuzziness. I've recently been working with an image made on 6x12 with a 47mm f/5.6 Super Angulon, which is slightly beyond its published limits on that format. I had even used slight movements. The corner falloff was pretty extreme, but the highlights that poked through the black corners were quite sharp. That is not the case with older designs, but if it's sharp on the ground glass (viewed at whatever magnification is the target), it will be just as sharp on the negative, unless something else (uncorrectable by limiting coverage in the way suggested) intervenes.

Rick "in the market for a tilting 6x loupe to better make these assessments" Denney

[aduncanson]

From deep inside of here: http://www.globalmatter-lab.net/catalog.htm

I find "Its considered good optical design practice to vignette the illumination circle of a lens bewond (sic) the point of good performance, (see _Modern Lens Design_ by Warren Smith or Kingslake's design books)"

So I have assumed doing so is pretty much standard practice for newer designs. (And I bought a 240mm convertible Symmar for 8x10 use because it reportedly had a larger image circle than newer designs.)