100mm Canon Macro - 100mm Leica R macro comparison

[darcy_lorimer]

I just got my new Canon 20D body and 100mm f2.8 EF USM Canon lens.

One thing I was curious about was the focus issue with R lenses, so I

took some macro comparison shots with my Leica 100mm F4 Macro Elmar R

using a Photodiox adapter on the 20D body.

 

I fully expected, before the tests, no to see any apprecable

difference between these lenses. In fact, I guess I expected to the

Leica lens underperform in comparison to the faster, more modern

Canon lens, which is optimized for the 20D body, yadda yadda yadda.

 

Boy was I surprised. First thing I found out was that the Canon lens

gets slower at close focus distances. But the F-stop reading in the

finder does not change from f2.8! What a con! This is mentioned in

the manual, and I confirmed it by placing a neutral grey card in

front of the lens, and adjusting the focus from infinity to the

closest focus point, with manual exposure mode on the 20D. If the

shutter speed is fixed for a proper exposure at infinity, focusing to

close focus drops the exposure by 2 stops. This makes this lens

effectively f5.6 at near focus points. The Leica R lens, of course,

does no such thing. F4 at infinity, f4 at the close focus endpoint,

and R lens is now faster for macro work than the Canon.

 

I tested the focus accuracy (meaning image sharpness) at macro levels

of about 1:4 with the Leica lens. With the stock screen in the Canon,

I had no trouble focussing, even stopped down 2 or 3 stops. The

digital images manually focused (R lens) were as accurate as ones

made with the Canon lens auto-focussed. I tried a number of other R

lenses from 180mm F2.8 to 24mm, and I had no focus problems

whatsoever.

 

I made a couple comparison shots at 1:4 with both macro lenses under

flourecsent light, ISO 1600. The Canon lens produced a slightly

orange tinted image (large, fine) with no white balance corrections

applied (AWB engaged). The Leica R lens, produced white whites with

no tint. All body settings were identical for both lenses. Finally, I

tried to compare sharpness of both images. I did this by magnifying

sections of both images via the viewer LCD on the back of the 20D. I

think you can get a 10x magnification this way. Not super critical,

but I still found the R lens somewhat sharper. One of the circuit

components I was shooting (which is about 0.2" x 0.3") has ID numbers

laser etched on one side. Blowing these up to the maximum

magnification showed the Leica to be sharper, cleaner, with better

contrast. And remember, this is the manually focussed image being

compared to the auto-focussed Canon image.

 

These tests clarified a number of things for me. Comments I received

in an earlier post about R lenses being impossible to focus on the

20D were incorrect. My faith in R optics is restored, and feedback I

have read about superior Canon optics (particularly telephoto lenses)

is not the case for the 100 mm Macros.

[nasmformyzombie]

One could certainly make a good argument that macro focus should be done manually, regardless of the system used. But why be defensive about Leica R glass? It is exceptional---if you can live with the exorbitant prices, considerable heft and bodies that have limited appeal. As for Canon glass, Canon's best glass is long telephoto, and this has been the case for many years.

[ilkka]

For a 100mm lens to focus to 1:1, it needs to be extended an additional 100 mm from the film plane. This causes the 2 stop reduction in F/speed. If Leica lens does not extend this way, how does it achieve the 1:1 reproduction ratio? Or does its aperture get wider by two stops or focal length shorter to compensate? This is simple physics, nothing to do with relative merits of optical design in Germany or Japan.

[bobatkins]

Wow! The Leica lens cures Canon white balance problems just by being near the camera. I can hardly believe it.

<p>

Bet you're pretty pissed you wasted your money on the Canon junk.

<p>

I hear the Leica 100/2.8 APO Macro makes the old 100/4 Elmar macro look pretty sad, so compared to <b>it</b> the Canon must be a doorstop.

<p>

As for "how does it achive a 1:1 ratio", as far as I know it doesn't. I think the best it can do is 1:3.

<p>

[gluteal cleft]

It would be interesting to see how the EF-S 60mm fared in the mix. I haven't tried one myself, but my friends who use them speak very highly of them. The MTF chart looks noticeably better than that of the 100, but of course, MTF isn't everything.

 

steve

[colin_elliott]

The 100mm Apo-Macro Elmarit-R f/2.8 focuses down to a 1:2 repro. ratio.

It achieves 1:1 by adding the Elpro 1:2-1:1 near focusing attachment. At 1:1, there is a free working distance of 7.2 cm.

It actually gets to 1.1:1, giving an object field of 22x33 mm.

[oceanphysics]

What a joke.

 

All lenses lose speed when focused close. If you don't understand why then you really don't have any business this sort of nonsense. If you didn't see it with the Leica "macro" it's because you aren't very observant, and it's not really a macro because it doesn't really focus that close. Put an extension tube on it to bring it to 1:1 and the laws of physics dictate that it will lose two stops.

 

But being a Leica owner, you're going to see what you want to see. That's how people get to be Leica owners in the first place.

 

The only thing that's possibly more hilarious is the white balance drivel. I don't even want to get started on that.

[kevin m.]

"My faith... is restored..."

 

If it's faith, why bother testing it? If it's science, why such a shoddy test?

[panos_voudouris]

"And remember, this is the manually focussed image being compared to the auto-focussed Canon image."

 

Who said that you use AF in macro? Read a book and learn why you shouldn't use AF.

 

"This makes this lens effectively f5.6 at near focus points. The Leica R lens, of course, does no such thing. F4 at infinity, f4 at the close focus endpoint, and R lens is now faster for macro work than the Canon. "

 

Maybe you should compare both at 1:3 which is the max the Leica can do. Is the Canon then slower?

 

Praise the lord, R optics are good.

[jack_myovski]

I seem to recall that the Nikkor 1:2.8 105mm AF Micro functions somewhat as a zoomlens, which rearranges internal elements and loses a bit of focal length rather than by extension losing speed. The Leica 1:2.8 100mm APO Macro as I recall requires the purchase of a quite dear auxiliary diopter to reach life-size reproduction scale, which attaches to the front of the lens and thus has no effect on speed. I can understand why someone who specifically reserves the lens for static macro subjects might prefer the adapted Leica lens on Canon body, however if the lens is used for moving subjects such as insects then I would imagine the continual running of the aperture to and fro manually between wide-open (for focussing) and perhaps f/16 or f/22 (for depth)would become rather tedious in short order.

[oceanphysics]

<i>I seem to recall that the Nikkor 1:2.8 105mm AF Micro functions somewhat as a zoomlens, which rearranges internal elements and loses a bit of focal length rather than by extension losing speed.

</i>

<p>

Many do. However, it's a relatively small amount that reduces but doesn't nearly eliminate the light loss effect. And if the Canon managed to shorten its focal length all the way down to 25mm in order to remain at a constant effective aperture, I guess that would be something to complain about as well.

<p>

You can put diopters on Canon lenses too, as far as I know!

[nemeng]

FWIW the Leica 100-APO-Macro also slightly changes focal length as you focus closer. From memory I think it's only 100mm at infinity and 90mm lens at min. focus.

[nels]

D'Arcy,

 

Where may I see your images where you put this superior Leica 100mm lens to use?

[al_kaplan1]

When you extend a lens away from the film to focus closer you change the angle of view. It becomes narrower. If you shorten the focal length as you focus closer you gain the close focus while preserving the same angular coverage. A standard 100mm macro at 1:1 has the angular coverage of a 200mm lens at infinity. The variable focal length lenses keep pretty close to the same angle throughout the range.

[darcy_lorimer]

In response:

 

The Canon lens cost me about the same as the Leica lens. Of course the Leica was used, but in mint condition. So I do not consider the Leica exorbitantly priced.

 

I stand corrected on the speed loss. I went back and measured the speed loss of both lenses at the 1:3 magnification (which is the limit for the Leica), and both lenses loose 1 stop.

 

I don't consider the Canon lens to be "junk". Its just not quite as sharp as the Leica.

 

K. Mendenhall-- Ever hear of a colloquialism? Perhaps I should have said "My confidence was restored..." Not every test has be carried out in an optics lab. If the lenses were so close in performance, I should not have seen any difference in the images. Same is true if the test was not capable in resolving any difference. The images were taken back to back and compared with the LCD viewer (10x magnified) on the camera. I admit its not a critical test, but it should not produce a bias toward the Leica lens either.

 

P. Voudouris -- Maybe you should read the manual that came with the Canon lens:

Closeup Photography

[Framing Priority]

"While looking through the viewfinder and framing the subject, focus with the AF or MF mode"

Where in the Canon literature does it say that there is a close focus limit to AF?

The impression I got from previous posts below was that the only way to effectively focus with the Canon 20D was in using the autofocus, and that there was inherent advantage in using AF Canon lenses with this body as opposed to manually focussed R lenses. Now your saying this is not true for Macro work?

[djl251]

Wow - a guy makes a contribution and gets his head chopped. And from the Canon folks no

less. I have Canon - I like it. I have Leica and I like it too. It's really quite simple - Leica glass

beats Canon glass. Always did- always will.