Testing the Leica M 90 Apo 2.0

MP Guy

Just another face in the crowd
Staff member
Local time
6:12 AM
Joined
Jul 28, 2003
Messages
2,770
Location
PNW
I have lately been facinated with the Leica 90 2.0 Apo. Not only is it a great portrait lens, its just as good on the field.

The following shots where taken with an MP and the 90mm Apo 2.0 using Velvia 100. Aperature was 5.6 shutter speed was around 500th of a second.

Scanning was done using the Mionlta 5400 II at 5400 DPI with 8 passes using VueScan software and then the default noiseninja settings.

The resulting image has a size of 7137 x 4774 pixels. This is equivelent to a 34 megapixel camera. at 300 dpi, this would be aproximately 16 x 24 native size. The Canon 1DSII has a native resolution of 4992 x 3328 or 16.6 megapixels and a native print size at 300 DPI of approximately 11 x 16. Obviously the scanned slide is substatiatlly larger.

The first image is the complete shot reduced to 700 pixels in width.

The images after are 100% crops of the original scan.

90mmApoM.jpg



Left side

90mmApoM-left-1.jpg


Here is the same image after being reduced to the native size of a 1DSII and cropped 100%


90apo1ds2sizeleft.jpg


Bottom Left

90mmApoM-bottomleft-1.jpg



Top left

90mmApoM-topleft-1.jpg


Center.

In this shot, look at the bottom right side. On the big gray steel beem, there is a small block with some dots on it with a verticle orientation. If you look to the right of this small dot you can make out parallel lines which seem to be the shadow of those dots.

90mmApoM-center-1.jpg



Here is the same image after being reduced to the native size of a 1DSII and cropped 100%

90apo1ds2sizecenter.jpg


Another Center

90mmApoM-center-2.jpg



Here is the same image after being reduced to the native size of a 1DSII and cropped 100%

90apo1ds2sizeanothercenter.jpg


Right

90mmApoM-right-1.jpg



Bottom right

90mmApoM-bottomright-1.jpg



Top right

90mmApoM-topright-1.jpg
 
Fantastic dissection, and a great shot for this sort of analysis, very busy with many aspects covered.

The process (or workflow) is one I'd consider high-end. It's probably not what you intended to discuss, but I'm curious about the specs of your PC, and to what degree you think it cuts-the-mustard.
 
Cool. For fun, take the same shot with the DMR and interpolate the file to a similar size.
I`ll bet it will do well....
While your scanner gives you a "34MP" scan, in reality it`s closer to a 10-12MP cameras output because of the grain.
Filmscan pixles and CCD/CMOS pixles aren`t the same..

That 90mm really shines, have you tried the new 75mm yet. It seems people are selling their 90mm`s to get the 75mm over at the Leicaforum...
 
Jorge, nice detail, but under the same circumstances with a non AA Summicron I suspect image detail would be about the same as your lens. It would be more interesting to see a comparison of the same subject shot at f/2 or f/2.8 to see if the expense of that lens pays off in maintaining similar sharpness to f/5.6. To demonstrate how sharp the older 90 Summicrons were, I'm enclosing images taken years ago on Kodachome 64 comparing the 6 element (48E) to the 5 element (55E) Summicron both shot @ f/2.8. The image order is full frame, center crop of 48E, center crop of 55E, right end crop of 48E and lastly, right end crop of 55E.
 
I cant wait to see how the 50 Asph Summilux does with a similar test.
 
When I get the M8. This will be one of the first scenes I will photograph. I plan on an apples to apples with this industrial scene.
 
I'm not sure I understand the point of the comparison Jorge. Were you simply showing what a crop looks like for one resolution vs another or were you trying to say something about the quality of a scan vs a digital camera?
 
Jorge, I will be looking forward to your film/digital comparisons with this lens. I recently sold my 90/2 after getting the 90/2 AA at a very nice price from a fellow RFFer. My impression on FILM is that there is no significant difference in lp/mm resolution with the exception of slightly better color saturation and edge performance. I suspect image differences will be even more apparent with a flat digital image receptor vs film with it's tendency to not lay perfectly flat.
 
Back
Top Bottom