BirdForum is the net's largest birding community dedicated to wild birds and birding, and is absolutely FREE!

Register for an account to take part in lively discussions in the forum, post your pictures in the gallery and more.

Reply to thread

Message: <blockquote data-quote="David Ellsworth" data-source="post: 1862960" data-attributes="member: 40206">Barrel play has a huge effectGiven my experience with my own 100-400mm, I would say that it's quite likely that there's "soft copies" and "sharp copies" of this lens. It's probably due to barrel play.My 100-400 when I first got it had very good sharpness across the entire APS-C frame on a 450D. Not as sharp as my 100mm USM Macro, but close. There was some slight softness at f/5.6, but so little as to be visible only when shooting very contrasty subjects, such as a sunlit Snowy Egret against dark water. Stopping down to f/6.3 was enough to make it fully sharp (stopping down any further made no difference*). Corner sharpness was so good that I had the freedom to compose a subject into a corner if I wanted to. Strangely, the upper-left corner had some visible transverse chromatic aberration (though not much), whereas the other corners and edges had none (i.e. so little I can't see it at 100%). It took teleconverters very well too, having satisfying sharpness across the whole frame even with 2x and 1.4x stacked (assuming a very steady mount, perfect focusing, remote triggering, low atmospheric turbulence and most importantly, the subject stays sufficiently still) — i.e., diffraction-limited performance.I dropped this lens, and when it came back from Canon repair it wasn't the same. Corner sharpness suffered from astigmatism, and the lens needed to be stopped down to f/8 to reach full center sharpness. However, when I nudge the zoom operation ring up+right, all of its wonderful sharpness comes back, corner to corner edge to edge (and even that strange CA-only-in-the-upper-left is back). I can even tighten the smooth-tight ring to lock in that nudge. But this is not very convenient, and I'll probably send it back to Canon for another go at repair (hopefully they'll get it right this time — I hate being without this lens).There's so little barrel play on this 100-400 that I can barely feel it with my hands, and yet such a tiny nudge has a huge effect on image quality. My guess, regarding "soft copy" versus "sharp copy", is that Canon can collimate a 100-400 perfectly in the disassembled state, but then when they reassemble it there's a chance that the two barrels won't mate precisely the way they should, and collimation may be compromised.If you have a 100-400 that is soft — needs to be stopped down more than f/6.3 to reach full sharpness, and/or is not as sharp in the corners as it should be — then you may be able to make it sharper by nudging the zoom operation ring in a particular direction. Figuring out what direction you need to nudge it is made easy by focusing on a point source of light (an artificial star is best), then going a bit out of focus and looking at the nested circles in the defocused image. The circles should be perfectly centered; if they're not, that will tell you what direction in which to nudge the zoom ring. It's much like collimating a refractor telescope (google it).* Well, it's more complicated than that. With the 100-400mm collimated as well as I currently can (which still isn't perfect), stopping down to f/7.1 gets rid of a little bit of remaining softness, but it takes spectral highlights or stacked teleconverters to see this. Stopping down further, to f/8 and beyond, does not help fine sharpness at all (it only hurts it due to diffraction).</blockquote>

Verification