A macro lens comparison - 65mm, 70mm and 150mm

[Overread]

This is a test done between the Canon MPE65mm macro, Sigma 150mm macro and the Sigma 70mm macro. The test was done in a similar setup to that of my previous test comparing the magnifaction factors of the MPE 65mm macro and the effects of diffration and aperture upon that setup.
Test setup:
Canon 400D camera body and 580EX2 with a lumiquest softbox and an offcamera flash cord for lighting. A tripod and focusing rail setup were used to ensure that shots were taken with the minimal level of shake, with mirror lockup and a remote release being used to further reduce any shake effects.
ISO and shutter speed were set to the same value for each shot (ISO 100, 1/200sec) and whilst the flash was freehand held attempts were made to keep the angle of lighting as similar as possible for each of the shots (though as outlined in the test there is some resulting difference in contrast in the final results - this could be from differences in flash angle or could be the natural contrasty nature of the lens).

Results from the test can be found here: this contains fullsized versions suitable for comparison
Test images
the title of each image notes the lens used and the aperture the shot was taken at.

Thoughts on the results:
Sigma 150mm macro: wide open at f2.8 is clearly not it sharpest, but that is common for many lenses; stopped down to f4 and things make a dramatic improvement. From f4 through to f8 things appear quite solid, then there is a slight degradation around f10, but the real drop starts to occur at f13 where things take a far more noticeable dip in overall sharpness.

Sigma 70mm macro: starting to understand where this lens gets its title as one of sigma's sharpest, even wide open at f2.8 it gives impressive results. Stopped down to f4 and things become even sharper still. Rather like the 150mm from f4 through to f10 the performance is very similar, however between f10 and f13 there is again a noticeable drop in overall performance of the lens sharpness.

Canon 65mm macro: wide open at f2.8 we start to see why this lens is a pure macro only, sharpness at this aperture is far ahead of the other two and would be more comparable to their performances when stopped down to around f4. Again we see the same pattern as with the other glass, prime optimal performance before f10, slight degradation at f10 followed by a little more loss at f13.

Overall views:
f2.8 - wide open - the 65mm is clearly standing out as the strongest here and is far ahead of the other regular macro lenses.

f4-f10 - quite similar overall performance from each of the lenses in this range. Peek optimal performance, followed by a slight dip at f10. f10 performance overall seems quite stable, though the 70mm appears a little sharper, though the flash angle and contrast increase as a result might be affecting this result.

f13 - The 70mm and 65mm appear to be leading the way at this setting, with the 150mm dropping a little behind (though again I repeat that contrast differences might be giving a slight bias toward the 70mm over the 150mm in this respect).


Overall - the MPE65mm is clearly standing up as the best overall macro performer, from wide open through to even closed down to f13. It really goes to show the level of performance that can come from a purely macro dedicated lens, I just wish (again) that it could get down to 1:2 macro it would then be an easy lens to sit with all day for a day spent doing macro work.
Coming next the Sigma 70mm is earning its name as one of sigma's best performers and just manages to beat the 150mm optically speaking when at the more extreme ends of its aperture range. In the middle areas performance is very similar.
And finally the 150mm - the longest focal range by more than double that of the other two, but whilst its not standing quite as high its performance, even down to f13 is still very impressive and certainly usable.

I remind readers that this test is done with no sharpening applied to any of the shots so this is a pure test of the lenses performance rather than showing any differences in sharpening amounts. Further there might be some bias toward the 70mm in the latter test areas because of the angle/power of the flash and the slight contrast increase these shots appear to have over those produced from the 150mm.

Any thoughts on the test or its results are most welcome

[Malcolm_Stewart]

Thanks for all your hard work, and for posting the results.

I have the Sigma 150, and Canon's EF 100 f2.8 USM & EF 50 f2.5 compact macro. One of my test objects is a bird's feather, and using my EF 100 f2.8 I can resolve some very fine inner detail, which I've not resolved using the EF 50 f2.5 - it's nearly there, but not quite. (I don't think I've checked my Sigma yet.) It worried me initially that what I was seeing might be digital processing artefacts, so I checked what I was seeing using a good quality microscope - the very fine details were real! (I'm a member of the Quekett Microscopical Club so should know a little about the risks of false optical images.)

I would imagine that in normal macro usage, f8 would be used quite often simply to get a semblance of depth of field, but it's interesting to see what happens at full aperture, where all microscope objectives work, but in this case the effective aperture is normally slightly reduced by adjusting the illumination cone diameter for optimum contrast.

[Overread]

f8 is, in my experince for most people, where 1:2 macro sort of starts and once down to around 1:1 most shooters are using f13. The scale then turns over as diffraction becomse a bigger problem with increased magnification and by 5:1 f6.3 is about the most many are willing to use due to the loss of sharpness.
Certainly being able to see the difference in wide open use is very interesting, something also that I feel quite valid in just testing out because its the sort of thing I would generaly not shoot so much (due to the tiny depth of field).

Intersting to hear about the resolution of finer details that you've seen a difference in, this is one area where I've read that the camera body used can also be a key factor for. I belive DaftBiker noticed improvement between his rebel series camera body and his 40D and certainly a difference (more resolution again) between hi 40D and 7D.

[daft_biker]

The feather test is quite interesting, especially if you haven't seen how the structure of a feather is woven together. You get a clear view of it through the MP-E but then there's a whole load more detail that you could see through microscope lenses.

Sometimes I like wide open on the MP-E as it's a bokeh paradise. It has it's uses but I find f/9 at 5x looks fine printed/web versioned. Normally I shoot at a bit less but it depends on the subject and the kind of detail I'm after - the other day I was using f/3.5 - f/7.1 at 5x. At 1x I'll try to keep it below f/11.

I have done a detail test between the 40D and 7D but I'm seeing loads of things more clearly on the 7D. I did a test with the 300D and 40D and the difference was clear enough that I haven't bothered this time.