Shootout at 4-6x

[ray_parkhurst]

I did a quick comparative shootout (well, it took a few hours to do the stacks...) of 14 objectives I had laying around from various projects and purchases. I did NOT do a good job of ensuring a consistent magnification, so you'll see anywhere from ~4x to ~6x or possibly higher. The two 20mm objectives I tested were tough to get below 6x on my setup so that's where I tested. them. Most of these are fairly modest in price and quality and I wanted to see which of them had acceptable results in this magnification range, which is perfect for showing die variety details on coins. I used a 1956-D Lincoln Cent as the subject and shot around the mintmark area. The stacks required were modest, from 4-7 images depending on the lens. Camera was a Canon XS 10MP APS-C, which was (one of) Canon's first to incorporate EFSC in Live View. Here is a list of the objectives tested, in no particular order:

Canon 35mm f2.8-f22 Macrophoto
Parco 4 0.15
Made In Germany 5x 0.1
Technical Instruments 4x 0.13
Nikon 4 0.1
Olympus 4 0.1
Nikon 4Plan 0.1
Olympus M5 0.1
Bausch & Lomb 48mm 0.08
Bausch & Lomb 48mm f4.5-22 Micro Tessar
Canon 20mm f3.5-22 Macrophoto
Nikon 5x Measuring Microscope (M26 mount)
Zeiss 20mm f3.2-f32 Makrotar
Nikon Plan 4 0.13

Canon 35mm f2.8-22 Macrophoto at f2.8


Canon 35mm f2.8-22 Macrophoto at f4


Parco 4x 0.15


Made in Germany 5x 0.1


Technical Instruments 4x 0.13


Nikon 4x 0.1


Olympus 4x 0.1


Nikon 4Plan 0.1


Olympus M5 0.1


Bausch & Lomb 48mm 0.08


Bausch & Lomb 48mm f4.5-22 Micro Tessar at f4.5


Canon 20mm f3.5-22 Macrophoto at f3.5


Nikon 5x Measuring Microscope


Zeiss 20mm f3.2-f32 Makrotar


Nikon Plan 4 0.13


Edited to correct that the XS was not the first but one of the first Canons with EFSC

[rjlittlefield]

Any chance we can get a picture of the objectives themselves? "Nikon 4 0.1" could describe several lenses with quite different characteristics.

--Rik

[Charles Krebs]

Ray,

That's a lot of work. Thanks for posting this. Can you give a brief summary of your impressions of winners and losers.

Camera was a Canon XS 10MP APS-C, which was Canon's first to incorporate EFSC in Live View

Nitpicking here, but:
Canon 40D Sept 2007
Canon Rebel XSi (450D) April 2008
Canon Rebel T1i (500D) May 2008
Canon Rebel XS (1000D) Aug 2008

[ray_parkhurst]

Any chance we can get a picture of the objectives themselves? "Nikon 4 0.1" could describe several lenses with quite different characteristics.

I knew you guys would ask for a pic. Here you go:



Front row L-R:
Parco 4 0.15
Bausch & Lomb 48mm 0.08
Olympus M5 0.1
Technical Instruments 4 0.13
Olympus 4 0.1
Nikon 4 Plan 0.1
Nikon 4 0.1
Made In Germany 5/0.1

Back row L-R:
Canon 20mm f3.5-22 Macrophoto
Nikon Plan 4 0.13
Zeiss 20mm f3.2-f32 Makrotar
Nikon 5x Measuring Microscope (M26 mount)
Bausch & Lomb 48mm f4.5-22 Micro Tessar
Canon 35mm f2.8-f22 Macrophoto

Ray,

That's a lot of work. Thanks for posting this. Can you give a brief summary of your impressions of winners and losers.

Several of the lenses could not even cover APS-C. I'd call these losers. So here are the losers, in order of worst to less bad:

Parco 4 0.15
Tech Instruments 4 0.13
Nikon 4 0.1 (which also vignettes a bit)
German 5 0.1
B&L 48mm 0.08 (which also had flare/hot spot issue)
B&L 48mm Micro Tessar (similar issue as above)
Olympus M5 0.1

The rest are able to at least cover APS-C, so I'd broadly call them winners. Here they are in my order of preference, decent to best:

Olympus 4 0.1
Canon MP35 (had to stop to f4 to get sharp image)
Nikon 4Plan 0.1
Zeiss Makrotar
Nikon 5x MM
Canon MP20
Nikon Plan 4 0.13

The most flexible lens was the Nikon 5x MM, which also has the interesting property of being effectively telecentric within the fairly shallow stack for coin imaging. There is no perceptible change in image magnification, and the stacker I used (CZP, still haven't graduated to Zerene) did not do any scaling compensation. Perhaps this is a quality of objectives for measuring purposes? The objective has a very long working distance, so it may be simply that there is very little perspective change over a distance that is short relative to the working distance? Whatever the reason, it was very convenient to do the stacks and had a completely different feel compared with the other objectives. The Nikon Plan 4 0.13 also has a large working distance, and had a somewhat similar characteristic.

[naturephoto1]

Ray,

Thanks. Glad to see the Canon 20mm Macrophoto lens performs so well. I will/would have to see how mine compares with my Leitz 25mm f2.5 Photar.

Rich

[ray_parkhurst]

Ray,

Thanks. Glad to see the Canon 20mm Macrophoto lens performs so well. I will/would have to see how mine compares with my Leitz 25mm f2.5 Photar.

Rich

I actually had the Canon 20mm for sale on eBay when I decided to do the shootout. It was one of the last lenses I tested, and about 5 minutes after I got it on the bellows I ended the auction early. It was a joy to use and gave superb results. I just need to remember that so I don't try to sell it again...Ray

[naturephoto1]

Ray,

Thanks. Glad to see the Canon 20mm Macrophoto lens performs so well. I will/would have to see how mine compares with my Leitz 25mm f2.5 Photar.

Rich

I actually had the Canon 20mm for sale on eBay when I decided to do the shootout. It was one of the last lenses I tested, and about 5 minutes after I got it on the bellows I ended the auction early. It was a joy to use and gave superb results. I just need to remember that so I don't try to sell it again...Ray

Ray,

I remember right after you got the lens and you tested it and you were shocked by its performance; it was that good. So. don't sell it!!!

Rich

[rjlittlefield]

Thanks, Ray. (One glitch: at present the image is so big, 2000 pixels wide, that it messes up the page formatting and makes long text impossible to read without scrolling.)

The most flexible lens was the Nikon 5x MM, which also has the interesting property of being effectively telecentric within the fairly shallow stack for coin imaging. There is no perceptible change in image magnification, and the stacker I used (CZP, still haven't graduated to Zerene) did not do any scaling compensation. Perhaps this is a quality of objectives for measuring purposes? The objective has a very long working distance, so it may be simply that there is very little perspective change over a distance that is short relative to the working distance?

Yes, for measurement purposes anything closer to telecentric is better because it gives an accurate measurement even if the focus is slightly off. (This presumes that the user correctly identifies the center of the blur.) Not all telecentric lenses are LWD, but LWD objectives usually act more telecentric than normal ones. I haven't checked recently, but based on earlier tests I would expect CZP's estimate to be accurate.

--Rik

[ray_parkhurst]

Thanks, Ray. (One glitch: at present the image is so big, 2000 pixels wide, that it messes up the page formatting and makes long text impossible to read without scrolling.)

OK, fixed it. Now it's 1000 pixels wide.

[rjlittlefield]

Thanks, Ray. Now that I can see the whole group at a glance, I realize that the Nikon 5X MM is also an oddball in terms of its overall size and shape. I'd guess that big lump in the middle contains some optics that specifically make the thing telecentric.

--Rik

[Chris S.]

The Nikon MM 5x/0.13 is an oddball in its working distance, as well as shape: 64mm! As Ray says, it is a real pleasure to use--it feels almost like cheating. However, I don't use it often, as my Nikon Planapo 4x/.020 has higher resolving power, as its larger NA would suggest.

Nice test, Ray. Thanks for doing all this work!

--Chris

[ray_parkhurst]

Camera was a Canon XS 10MP APS-C, which was Canon's first to incorporate EFSC in Live View

Nitpicking here, but:
Canon 40D Sept 2007
Canon Rebel XSi (450D) April 2008
Canon Rebel T1i (500D) May 2008
Canon Rebel XS (1000D) Aug 2008

OK, I edited original text to say "one of the first". Thanks for the correction...Ray

[Fredlab]

Hello

I don't know if my post is helpfull but...

I'm not sure if a coin is a good subject for a test.
Maybe wing of butterfly will be better.

For the "unknown Nikon MM", i find one example in "ebay.de"
Link
(this seller has one 5x and one 3x to - unfortunately, i have no money to buy them)

Ray... many thanks for this job.
As a crazy tester (but I much fewer objectives to try), i appreciate.

[ray_parkhurst]

Hello

I don't know if my post is helpfull but...

I'm not sure if a coin is a good subject for a test.
Maybe wing of butterfly will be better.

Perhaps, but my hobby is numismatics, not entomology. What is it about butterfly wings that makes photographing them more useful in determining lens characteristics than photographing coins?

[Fredlab]

Hello

Perhaps, but my hobby is numismatics, not entomology. What is it about butterfly wings that makes photographing them more useful in determining lens characteristics than photographing coins?

This kind of microdetails ?


(crop @ 100% - Canon EOS 500D - raw shot)

(same lenses, same position, but this 100% crop came from a Nikon D700)

OK, with money, you have micro-scratches, but with metal, there is no "transparency".
I think - but i'm not an expert - wing scales of butterfly are more discriminative.