Leica R 21-35/3.5-4.0 asph

FM Forums \| Alternative Gear & Lenses \| Join Upload & Sell	Search Used
1 2 end

Archive 2009 · Leica R 21-35/3.5-4.0 asph

Paul Yi
Offline
Dedicated FM
Upload & Sell: On

p.1 #1 · Leica R 21-35/3.5-4.0 asph

Does anyone have experience with this lens?
How does it compare to Canon 16-35/2.8 or Contax N17-35/2.8?

Feb 28, 2009 at 03:18 AM

mMontag
Offline
Upload & Sell: Off

p.1 #2 · Leica R 21-35/3.5-4.0 asph

I'm also interested in information on the Leica R 21-35.

Does anyone know if there is a non- aspherical & also an aspherical version. Peebles lists this lens as "iffy on a 5D. - i.e. modifications to lens & still may not work. I was watching this one on ebay for a while (before the economy jumped in the dumpster) USD $1,850.00 to $2,250.00 for bids.

You may be aware of this popular reference: http://www.16-9.net/lens_test/
I may be over looking but I don't see reference to aspherical on 16-9.

Thanks for starting this thread.

Feb 28, 2009 at 05:20 AM

weekh
Offline
Upload & Sell: Off

p.1 #3 · Leica R 21-35/3.5-4.0 asph

I've tested it side by side with the 17-40L.
In a blind test done in this forum some years back, many preferred the Canon over the Leica in terms of colour. The sharpness of Canon is also better.

Feb 28, 2009 at 05:58 AM

Manfred W. fEU
Offline
Upload & Sell: Off

p.1 #4 · Leica R 21-35/3.5-4.0 asph

The Vario Elmar 21-35 is asph only one version exists. I find mine very sharp using the lens on my 5DII ,you will need to remove the rear shroud you may only use it up to 24mm at larger apertures the mirror will not return. I do not have the 17-40 but the Leica is every bid as sharp as my 24mm f1.4 L first version ( setting is F3.5 )

Feb 28, 2009 at 11:50 AM

phuang3
Offline
Upload & Sell: Off

p.1 #5 · Leica R 21-35/3.5-4.0 asph

I have both lenses. I think 17-40L is no match to 21-35 ASPH in terms of micro-contrast and color. The later has the IQ of modern Leica lenses, it produces contrasy and vivid color images. Its sharpness is great in the center, but not at the edge - a sign of image curvature. It performs best from 28mm to 35mm. Here are some of my shots you may see them before.

This image is copyrighted by the owner

Feb 28, 2009 at 12:21 PM

weekh
Offline
Upload & Sell: Off

p.1 #6 · Leica R 21-35/3.5-4.0 asph

At around 21mm, in the corners, the Leica is no different from a lensbaby. Period.

Tested on several copies of the lens. Tried several branded / high-end adapters. Used different ff bodies.

Don't waste time with the Leica.

For those who rave about this lens, please don't do it with words. Show the pics taken at the wide end, wide aperture, full res image.

Mar 01, 2009 at 06:18 AM

Paul Yi
Offline
Dedicated FM
Upload & Sell: On

p.1 #7 · Leica R 21-35/3.5-4.0 asph

It is bit on the expensive side.
right now, I'll stay with Contax 18/4, and wait for perhaps Contax 21/2.8 ZE

Mar 01, 2009 at 01:09 PM

phuang3
Offline
Upload & Sell: Off

p.1 #8 · Leica R 21-35/3.5-4.0 asph

Paul,
There will always be trade-offs and compromises between lens designs. If you like center-to-edge sharpness and don't care about the moustache distortion, you can go for a zeiss 21/2.8. There must be someone who always look at a distortion table before purchase. These people will definitely like the hologon 15mm if they can get a RF camera for that, but it also comes with a price of 2.5 to 3 stops on vignetting. If we got full res images on these two lenses, can you tell me which is better? I prefer the color and contrast of a lens, and I showed it, didn't I?

Mar 01, 2009 at 03:40 PM

Conner999
Offline
Dedicated FM
Upload & Sell: On

p.1 #9 · Leica R 21-35/3.5-4.0 asph

There is also the upcoming CV 20/3.5 SLII - might be worth waiting to see how that pans out (sometime this month IIRC)

Mar 01, 2009 at 03:58 PM

robsteve
Offline
Upload & Sell: Off

p.1 #10 · Leica R 21-35/3.5-4.0 asph

weekh wrote:
At around 21mm, in the corners, the Leica is no different from a lensbaby. Period.

Tested on several copies of the lens. Tried several branded / high-end adapters. Used different ff bodies.

Don't waste time with the Leica.

For those who rave about this lens, please don't do it with words. Show the pics taken at the wide end, wide aperture, full res image.

It may be that the 21-35mm is not suited to the 5D. The 5d is missing the offset microlenses to help the edges of the images when using wide angles that have rear elements closer to the film plane and a very obtuse angle for the light hitting the edges of the sensor.

One thing I have noted with the 21-35mm is that the focus is extremely critical and you cannot rely on DOF. The image quality drops dramatically especially on infinity images if you do not actually focus on the subject. On all but the most distant scenes, this best focus can actually be a fraction off the infinity setting of the lens.

On film and and on the Leica digital, the 21-35mm is behind the Leica 19mm in image quality, but you would have to be very picky and print large prints to see it. In the 28mm range it is nearly as good as the Leica 28mm f2.8, and at 35mm it is better than all the Leica 35mm prime lenses.

For the convenience, the Leica 21-35mm mated with a Leica 35-70mm can't be beat. I seldom use my 19mm now except when I need optimum image quality.

Robert

Mar 01, 2009 at 05:02 PM

mawz
Offline
Dedicated FM
Upload & Sell: Off

p.1 #11 · Leica R 21-35/3.5-4.0 asph

robsteve wrote:

It may be that the 21-35mm is not suited to the 5D. The 5d is missing the offset microlenses to help the edges of the images when using wide angles that have rear elements closer to the film plane and a very obtuse angle for the light hitting the edges of the sensor.

Robert

The rear element of any SLR lens that's compatible with the 5D mirror will be far enough from the sensor plane that offset microlenses or their lack will not make a significant difference on IQ. It's only an a non-SLR camera like the M8 where rear elements can get within a few mm of the sensor plane where this becomes an issue.

A lack of telecentricity may remain an issue with IQ on SLR lenses, but will be primarily a vignetting source rather than a source of softness since most retrofocus designs are relatively telecentric by design.

Mar 01, 2009 at 05:10 PM

robsteve
Offline
Upload & Sell: Off

p.1 #12 · Leica R 21-35/3.5-4.0 asph

mawz wrote:

The rear element of any SLR lens that's compatible with the 5D mirror will be far enough from the sensor plane that offset microlenses or their lack will not make a significant difference on IQ. It's only an a non-SLR camera like the M8 where rear elements can get within a few mm of the sensor plane where this becomes an issue.

A lack of telecentricity may remain an issue with IQ on SLR lenses, but will be primarily a vignetting source rather than a source of softness since most retrofocus designs are relatively telecentric by design.

Then why does the DMR also have offset microlenses? Also, why are these problematic lenses on the 5D fine on the DMR, (though a 1.3 crop) or slide film?

Robert

Mar 01, 2009 at 07:07 PM

mawz
Offline
Dedicated FM
Upload & Sell: Off

p.1 #13 · Leica R 21-35/3.5-4.0 asph

robsteve wrote:
mawz wrote:

The rear element of any SLR lens that's compatible with the 5D mirror will be far enough from the sensor plane that offset microlenses or their lack will not make a significant difference on IQ. It's only an a non-SLR camera like the M8 where rear elements can get within a few mm of the sensor plane where this becomes an issue.

A lack of telecentricity may remain an issue with IQ on SLR lenses, but will be primarily a vignetting source rather than a source of softness since most retrofocus designs are relatively telecentric by design.

Then why does the DMR also have offset microlenses? Also, why are these problematic lenses on the 5D fine on the DMR, (though a 1.3 crop) or slide film?

Robert

IIRC the DMR uses microlenses, but they are not offset. This is supported by all the information I've been able to find on it. The M8 introduced offset microlenses to compensate for the problems inherent in using non-retrofocus 35mm format wide-angles with sensors larger than DX crop. Anyways the DMR is not anything close to FF so the issue weekh describes would not necessarily be visible on it.

As to slide, well frankly that comes down to 2 things. First off, film is slightly less sensitive to plane of focus placement (Since it has an appreciable thickness and an R series body will not hold the film truly flat anyways as that requires a vacuum back). Secondly, are you sure the results are that good? You'd need a high-res scan and some 100% inspection to really compare results from film to digital and that rarely happens in a truly comparable fashion (Witness Michael Reichmann's hilariously poor film vs digital comparisons over at L-L).

If what you suggest was truly the case the Leica R 19mm would have worse issues than the ones weekh is claiming for the 21-35 on a 5D, and the 21 Distagon would have similar problems. The fact that these supposed disconnects in performance are limited to a single lens design indicates an issue with the lens design, not the camera.

Edited on Mar 02, 2009 at 02:11 AM · View previous versions

Mar 01, 2009 at 07:51 PM

robsteve
Offline
Upload & Sell: Off

p.1 #14 · Leica R 21-35/3.5-4.0 asph

mawz wrote:

IIRC the DMR uses microlenses, but they are not offset. This is supported by all the information I've been able to find on it. .

The DMR has offset microlenses. If might be hard to find the specs now, but it was either in the Leica literature or the Kodak literature/specs for the sensor.

This is the sensor part number and description. you may be able to Google a spec sheet.

KAF-10010-CXA-CE-AA-OFFSET
Color (Bayer RGB), Offset Microlens, CERDIP Package (sidebrazed), Clear Cover Glass with AR coating (external) and IR coating (internal), Standard Grade

Robert

Mar 01, 2009 at 11:53 PM

mawz
Offline
Dedicated FM
Upload & Sell: Off

p.1 #15 · Leica R 21-35/3.5-4.0 asph

robsteve wrote:
mawz wrote:

IIRC the DMR uses microlenses, but they are not offset. This is supported by all the information I've been able to find on it. .

The DMR has offset microlenses. If might be hard to find the specs now, but it was either in the Leica literature or the Kodak literature/specs for the sensor.

Robert

Got a source for that? It's at odds with everything I've read on the DMR and the M8 (Which touted the introduction of offset microlenses as a solution for the wide issue). And given that the offset microlens design is all over the M8 literature I'd expect the same for the DMR.

In fact I tracked down the PR for the DMR and it doesn't mention offset microlenses at all, but rather an optimized microlens structure (which is not the same thing)

From: http://www.dcviews.com/press/Leica-Digital-Module-R.htm

Image quality: When developing the LEICA DIGITAL-MODUL-R, Leica's camera designers applied the same high standards of imaging performance, mechanical precision and handling as those used for the cameras themselves. Constructive advantages of the Leica R bayonet mount, the large image circle of the lens and the almost vertical incident angle of the light rays - not to forget the extremely tight tolerances of the cameras and lenses in the Leica R system - make already existing system components particularly suitable for digital photography. There is no need for specific modifications for the use of an image sensor instead of a film.

Sensor: The Kodak sensor integrated in the LEICA DIGITAL-MODUL-R is based on the proven 6.8 µm pixel architecture already used in digital backs for medium-format cameras. Kodak has optimized the design of the sensor especially for the LEICA DIGITAL-MODUL-R. For example, the surface of the sensor is protected by a particularly thin layer of glass and an optimized microlens structure is used. It was decided not to implement an anti-alias filter, which reduces image definition. Moiré effects of the sensor can be prevented by the software for critical subjects, while the quality of non-critical subjects is not impaired by a permanently built-in filter.

Given the fact that they're touting the high telecentricity of the R lenses, that seriously argues against the use of offset microlenses (Which would be self-defeating with that combination).

Mar 02, 2009 at 12:09 AM

robsteve
Offline
Upload & Sell: Off

p.1 #16 · Leica R 21-35/3.5-4.0 asph

From the press release you linked

"Kodak has optimized the design of the sensor especially for the LEICA DIGITAL-MODUL-R. For example, the surface of the sensor is protected by a particularly thin layer of glass and an optimized microlens structure is used."

This was the part number gleaned from a posting at the Leica User Forum. Note it says offset microlenses.

"KAF-10010-CXA-CE-AA-OFFSET
Color (Bayer RGB), Offset Microlens, CERDIP Package (sidebrazed), Clear Cover Glass with AR coating (external) and IR coating (internal), Standard Grade"

There was also a post on Dpreview mentioning that the DMR offset was 6 degrees, versus 8 degrees for the M8. I don't know what the author used for his source. A lot of the source press releases and info on the Kodak site is now gone.

Robert

Mar 02, 2009 at 12:45 AM

mawz
Offline
Dedicated FM
Upload & Sell: Off

p.1 #17 · Leica R 21-35/3.5-4.0 asph

robsteve wrote:
From the press release you linked

"Kodak has optimized the design of the sensor especially for the LEICA DIGITAL-MODUL-R. For example, the surface of the sensor is protected by a particularly thin layer of glass and an optimized microlens structure is used."

This was the part number gleaned from a posting at the Leica User Forum. Note it says offset microlenses.

"KAF-10010-CXA-CE-AA-OFFSET
Color (Bayer RGB), Offset Microlens, CERDIP Package (sidebrazed), Clear Cover Glass with AR coating (external) and IR coating (internal), Standard Grade"

There was also a post on Dpreview mentioning that the DMR offset was 6 degrees, versus 8 degrees for the M8. I don't know what the author used for his source. A lot of the source press releases and info on the Kodak site is now gone.

Robert

OK, I did trackdown the product sheet for the KAF10010, and it does indicate use of offset microlenses in all variants of that sensor. I stand corrected on that.

http://www.kodak.com/global/plugins/acrobat/en/business/ISS/productsummary/FullFrame/KAF-10010ProductSummary.pdf

Although it would have been nice if you had actually provided a link to prove your assertion, rather than forcing me to do your research for you. (unlinked forum posts that are straight assertion with actual indication of the source of the info don't count as proof and optimized does not necessarily mean offset)

Dunno why they'd bother though, there's no good reason to use them on a 1.37 sensor for 35mm SLR lenses. Even lenses which aren't terribly telecentric aren't going to need offset microlenses on a sensor with that crop due to the large distance between the rear element and the sensor plane (40+mm minimum for the R's) and the fact the problem areas of the image circle are going to be off the centre anyways. There will be no performance gain from that use of microlenses.

It also doesn't make your argument about the performance of the 21-35 any stronger. Performance on a crop body cannot indicate what edge performance will be seen on FF.

Mar 02, 2009 at 01:29 AM

telyt
Offline
Upload & Sell: Off

p.1 #18 · Leica R 21-35/3.5-4.0 asph

mawz wrote:...(unlinked forum posts that are straight assertion with actual indication of the source of the info don't count as proof ...

I haven't seen any photos demonstrating the claimed weak corner performance of the Leica 21-35. Got proof?

Mar 02, 2009 at 01:43 AM

robsteve
Offline
Upload & Sell: Off

p.1 #19 · Leica R 21-35/3.5-4.0 asph

mawz wrote:

OK, I did trackdown the product sheet for the KAF10010, and it does indicate use of offset microlenses in all variants of that sensor. I stand corrected on that.

http://www.kodak.com/global/plugins/acrobat/en/business/ISS/productsummary/FullFrame/KAF-10010ProductSummary.pdf

Although it would have been nice if you had actually provided a link to prove your assertion, rather than forcing me to do your research for you. (unlinked forum posts that are straight assertion with actual indication of the source of the info don't count as proof and optimized does not necessarily mean offset).

I tried finding the link but had no luck.

Why are we arguing about the DMR? The point to my original post was if the 21-35mm was not any good on the 5D, don't use it, but don't put the blame on the lens for corner performance.

Robert

Mar 02, 2009 at 02:12 AM

mawz
Offline
Dedicated FM
Upload & Sell: Off

p.1 #20 · Leica R 21-35/3.5-4.0 asph

telyt wrote:
mawz wrote:...(unlinked forum posts that are straight assertion with actual indication of the source of the info don't count as proof ...

I haven't seen any photos demonstrating the claimed weak corner performance of the Leica 21-35. Got proof?

I'm not the one claiming that, weekh is (and I've edited the one post upthread where it looked like I was doing so rather than merely comparingon weekh's assertions to other similar lenses known to perform well on the 5D). My involvement on this thread is merely to dispute the idea that crop-sensor or film performance of a lens can accurately indicate 35mm FF digital performance of the same lens.

Mar 02, 2009 at 02:13 AM

1 2 end
FM Forums \| Alternative Gear & Lenses \| Join Upload & Sell 1 2 end