24mm World Cup: Group C

Canon 24mm f2.8 v Olympus 24mm f2.8 v Sigma 24mm f2.8 v Yashica ML 24mm f2.8

Corner Performance at f2.8 (100% Crops)

The Canon 24mm f2.8 has a distinctive, almost Sigma-like, yellow cast and is arguably the best corner performer of the bunch. Certainly it retains the best contrast out in the corners, but the resolution isn’t quite as good as the Olympus 24mm f2.8. To be fair, none of these are stunning at this aperture. The Sigma and Yashica ML fall distinctly into second and third place behind the class-leading Olympus and Canons.

Centre Performance at f2.8 (100% Crops)

Centre frame, the Olympus just outresolves the Sigma and Yashica. The Sigma is a bit unlucky to walk away with just two points from this comparison because in reality it’s very close to the Olympus. The clincher is its weakness for chromatic aberration. The Yashica looks well behaved in this regard. Following its nicely rendered corners, the Canon looks much less attractive centre frame compared with the more neutral competition.

[nextpage title=”Resolution (f5.6)”]

Canon 24mm f2.8 v Olympus 24mm f2.8 v Sigma 24mm f2.8 v Yashica ML 24mm f2.8:
Corner Performance at f5.6 (100% Crops)

Once again, the Canon turns in an evenly controlled corner performance, just squeezed out of the top spot by the Olympus, with the Sigma and Yashica trailing in joint last place.  By now you’ll probably have noticed that the Yashica is showing the age of its coatings: compared to the modern lenses, it has a distinctive low-contrast ‘glow’, yet resolves reasonably strongly across the frames. Makes you want to track down a longer version for portraits.

Centre Performance at f5.6 (100% Crops)

Here again, the gulf is a little wider than the points suggest: the Olympus and Sigma are very close, but both are better than the Yashica or the Canon by some margin. This particular sample of the Canon 24mm shows some evidence of de-centering: the left side of the image seems slightly better resolved than the right. Just the sort of problem you’d expect to encounter with an adapted lens . . . .

[nextpage title=”Resolution (f11)”]

Canon 24mm f2.8 v Olympus 24mm f2.8 v Sigma 24mm f2.8 v Yashica ML 24mm f2.8:
Extreme Corner Performance at f11: Far Field (100% crops from top left)

The Olympus has overturned the Canon’s early lead convincingly from f5.6 on. And the Sigma has put in a strong showing, too. However, turning to the question of near-field performance, we see a surprising problem in these 100% crops from the extreme cornermost pixels of the bottom right of the frame, which has been present at all apertures.

Extreme Corner Performance at f11: Near Field (100% crops from bottom right)

The lack close-range correction or floating element in the Canon 24mm really lets it down at close range. By contrast, the Sigma and Olympus excel.

Centre Performance at f11 (100% Crops)

The Olympus is really flying at f11, with its tiny front and rear elements, it suffers far less than the larger lenses from diffraction at this aperture, and the corners have sharpened up very nicely. All in all an outstanding performance from the tiny Zuiko.

[nextpage title=”Distortion & Chromatic Aberration”]

Canon 24mm f2.8 v Olympus 24mm f2.8 v Sigma 24mm f2.8 v Yashica ML 24mm f2.8:
Geometric Distortion

Here each frame has been reduced to 2000 pixels, and a sample taken from the extreme upper left corner, demonstrating part of the distortion pattern in the brickwork.

So that’s four points to the Yashica, three each to the Olympus and Canon (the Zuiko’s distortion is slightly more pronounced, but more regular and easier to fix without degrading the capture), and two for the Sigma. At this focal length, it seems that the greater the resolution, the worse the distortion.

Chromatic Aberration (see also f2.8 Performance Page)

As indicated by the wide open performance, there is a touch of chromatic aberration in the Sigma and Canon captures; the Olympus and Yashica results are largely aberration-free.

[nextpage title=”Conclusion”]

Out of a possible maximum score of 36, here’s how they scored:

Canon 24mm f2.8	Olympus 24mm f2.8	Sigma 24mm f2.8	Yashica ML 24mm f2.8
20 points	30 points	21 points	18 points

With such a clear lead, there will only be one finalist going forward from this group: the cracking little Zuiko.

Generally, it’s clear to see that the venerable Yashica was out of its depth in this company. Long in the tooth coatings, poor corners and generally sub-par resolution were revealed in this full frame test. Had it not redeemed points for its excellent CA and distortion control, the gulf would have been even greater. Again, to its credit, it is the nicest lens mechanically in the group, with smooth focus and confidence-inspiring heft, but in reality, it’s not a lens you’re going to want to use on your Canon body unless you’re deliberately aiming for a nostalgic look.

I wasn’t expecting the Canon 24mm to give such strong corner performance on a full frame sensor: if wide open corner sharpness is critical to your shooting style, it might appeal – as long as you keep the subject a long way away: near field performance is much weaker. And it suffers from a deeply nasty case of jaundice.

The Sigma, too, was something of a surprise: as a late-night haunter of photodo, I was secretly hoping that it would pull something special out of the hat to justify its 4.2 rating . . . of course, I can’t guarantee this was the world’s best sample, and it was being asked to do its thing out of a relatively unproven EOS adaptor, but it’s a fine lens on balance – very sharp at working apertures (particularly centre frame), and without conspicuous flaws. It’s just that, in comparison with the Olympus 24mm, it has no conspicuous strengths, either.

On balance, the only one of the bunch to offer a persuasive reason to throw away your convenient autofocus on-brand lens and put up with stop down metering and manual focus is the Zuiko. Positively toy-like in appearance, it delivers in ways that no other lens among this group can dream of. No pain, no gain.

But how will it fare against the best of the rest? We’ll find out in the final . . . .