I just tried a quick comparison of various 300mm options I currently have. (I'm trying to decide which to keep and which to sell). This is not a particularly rigorous test - the light levels were varying slightly, for example. A tripod was not used, but the camera was steadied against a solid object and I'm happy that camera shake was not an issue. These are all 100% crops from near the centre of the image and, in each case, the best of several attempts. All images are from a 400D on 'Standard' setting with no sharpening applied. If they don't look quite as sharp as you might expect, bear in mind that:
- the target will really show up any deficiencies
- this was a dull overcast afternoon; I've found the images always seem much sharper in sunlight.
Here goes:
Canon 70-200 f/4 IS with Sigma 1.4x teleconverter (at 280mm f/5.6, i.e. wide open):
 | HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
Sigma 100-300 f/4 EX (at 300mm f/4)
| HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
Sigma 100-300 f/4 EX (at 300mm f/5.6)
| HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
Canon 300mm f/4 (non-IS) (at 300mm f/4)
| HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
I also tried the Kenko Pro teleconverter in place of the Sigma one, but results were nearly indistinguishable. I tried stopping down the Canon + TC shots to f/8, but couldn't see any improvement - I may need to try that again at some point. My next test will be the 300mm lenses with 1.4x TC vs the Canon 70-200 with 2x. I expect the 70-200 to lose, but you never know
My view is that the Canon with TC is sharper than the Sigma at f/4, but the Sigma is better at f/5.6 (i.e. the same aperture as the Canon+TC). All of them are beaten by the 300mm prime (I didn't bother stopping that down to f/5.6 as it was already so good at f/4). It's hard to tell from these pics, as the light was varying a little, but I'd say from what I've seen so far that the results for contrast pretty much follow the same order as the sharpness results. The Sigma definitely loses a little sharpness/contrast at f/4, but it's still pretty good.
To get some perspective, all of these shots can look pretty damn sharp with a little sharpness apllied in post processing. Every combination tested is good enough that I'll have no qualms about using them wide open.
In terms of focussing, my thoughts at the moment are that they are all pretty good, with the prime having an edge for consistency/accuracy and speed. The Sigma can hunt a little in poor light, but I don't think the 70-200 with TC is any better (although it may be better without the TC).
Of course, this doesn't make my choices any easier
The best option for quality would be to keep the 2 Canon lenses and sell the Sigma. I really do, however, prefer using a zoom - on a recent trip to the zoo I just used the Sigma (most of the pics here are with the Sigma: Marwell zoo
). With the Canon pairing I'd have been carrying more weight in total and constantly swapping lenses. I don't like the weight of the Sigma, and the IS on the 70-200 would be useful, but I hate swapping lenses too often. From previous testing the Canon 300 takes a teleconverter better than the Sigma. Decisions, decisions...Also, here are a couple of 200mm shots, although I didn't spend any time on this:
Canon 70-200 f/4 IS (at 200mm f/4)
| HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
Sigma 100-300 f/4 EX (at 200mm f/4)
| HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO |
The Canon here is a little better than the Sigma, but I'm sure they would be equal by f/5.6. Maybe I'll do some more testing if I get the time.
