Planning for my next lens tests

[raid]

I am considering to perform another testing of my 50mm lenses, and this time I may also include 35mm lenses to widen the coverage of the test. I will use relatively slow B&W film and get it scanned directly after developing from the negatives. Most likely, I will do both a controlled portion of tests where I will use a tripod and cable release and will focus on a fixed object plus a portion where I focus on my daughter's face where I will hand hold the camera. The first part will most likely be considered "maybe OK" and the second part will not be considered a valid "test'. This may be an argument that is fine if the purpsoe of the tests is a general test of lenses so that other users of such lenses benefit from the results. I also want to benefit myself from seeing my own photography results as I would usually take photos. I want to see how a human face is rendered by these lenses, and having some dead object as my "model" will not help me with this goal. Frank will be not be agreeing with my reasoning here. I know.

Now for the fun part; which lenses could I use in my tests:


50mm focal length:

FSU Lenses:

1. 50/1.4 J-3
2. 50/2.8 Industrar 61

Canon Lenses:

3. 50mm/1.2
4. 50mm/ 1.8
5. 50mm 2.8
6. 35mm/1.8



Leitz Lenses:

7. 50mm rigid Summicron (first version) ... now being CLA's by DAG
8. 40mm Summicron C
9. 35mm Summicron (8 elements) early model


Zeiss Lens:

10. 50mm/2 Sonnar ... now being again looked at by DAG


All lenses above are tried lenses and all showed excellent performance in the recent past.


I am glad to see that other people here are in the process of doing tests of their 50mm lenses. maybe we can pool our results eventually and someone could summarize in a detailed write-up the finding.


My earlier flare test helped me identify that the Summicron definitely needed internal cleaning and that the Sonnar needed another look at its optics.

Any suggestions?

 

[raid]

Raid,

I have been a regular reader of your tests, thanks for the work !

Would love to see the test that you describe, in particular since I don't do a lot of B+W portrait work but want to.

Maybe clustering 50/1.2-1.5 together and 35-40's ? Canon, Leica, Jupiter and CV would be most interesting.

Best,

Roland.

Hello Roland,
I mentioned in an earlier thread my wish to somehow group the lenses by design and not by brand. Some people here are more knowledgeable than me about the optical designs. If groued by design, we could maybe see subtle differences for the same design and see larger differences for the different designs.

As for using B&W film, it basically is new to me, considering that over the past 25 years or more I have been using only very slow color slide film. When I came to use rangefinder cameras, I found out that in most posrings of photos the photos were taken in B&W. My wife is not impressed. She prefers color.

 

[FrankS]

Raid, I agree with the use of a human face in lens testing. It has specular highlights in the eyes, fine detail in the eye lashes, a 3 dimensional roundness, skin tones with slight imperfections, and to top it off, the human face is a universally emotionally attractive image (to other humans). That or a nude adult female model. 🙂

 

[raid]

Raid, I agree with the use of a human face in lens testing. It has specular highlights in the eyes, fine detail in the eye lashes, a 3 dimensional roundness, skin tones with slight imperfections, and to top it off, the human face is a universally emotionally attractive image (to other humans). That or a nude adult female model. 🙂

Frank,
I don't want any problems with my wife, so a nude model is out :bang:
Seriously, I have a wonderful little daughter who enjoys being photographed, so she will have to do fo the tests. I also am looking to see whether a 3-dimensional roundness, as you put it, can be seen and to which degree for each lens. I am still uncertain about the lighting to be used. I would love to use only natural light (no flash) but know that with a slow film with ASA25-50, the window light with the regular ceiling artificial light will only allow very slow shutter speeds (with ASA 200 I had 1/60 @ f2 last tests). This is OK for the tripod steadied camera shots, but handholding will be not good for tests at slow speeds. Maybe I can use bounce flash or have some directional light from a light source.

 

[FrankS]

Raid, I don't think that slow film is all that necessary, since the limiting factor for resolution will be our computer monitors.

 

[ZeissFan]

Agreed. Use a fast film. That will allow you to use a faster shutter speed.

 

[VinceC]

I agree that slow film isn't necessary. 200-400 is quite useful. It's also what most of us shoot.

Tripods can be a pain in the neck when you're using a rangefinder camera to take pictures of people (focusing is a real issue).

It seems nearly everyone here is looking for image character, contrast, out-of-focus highlights, flare, etc. Most forum members seldom ever nitpick about sharpness, which is essentially a given except with some very exotic older lenses.

 

[raid]

I will then use XP2 like in my last test and have it scanned directly without bothering with having a true B&W film hand developed and then scanned somewhere else. I will try to address both audiences; those who insist on technical perfection ina test by using a tripod, and also those who want to recreate real life situations. The latter case needs more work since I need to do my very best to weed out any external factors that may influence the results and findings. Maybe having replicates will make things more trustworthy. I wonder whether a true B&W film shows up better or worse than XP2 on our monitors? Anyone has evdience of this or the opposite or is it irrelevant which type of B&W film is used?

 

[FrankS]

Well, real B+W film is sharper I believe, because the image is formed by discrete silver particles whereas the C41 stuff uses dye clouds like reg colour film. (this is my understanding of the difference) If you had a choice, I would use the good stuff. 🙂

 

[raid]

OK Frank. I have the choice here, so I will use Tri-X or Agfa 400 B&W film. This is not a problem, and I agree with you that the C-41 B&W films do not show the details you get from a silver based B&W film. I did not know if scanning makes things any different here.

 

[FrankS]

Raid, when I do my testing of lenses, I will set up so that there will be a good deal of depth/space behind my subject, so that OOF areas will be present. In those areas behind the subject there will be small light sources (I'm thinking Christmas lights) and objects of detail (perhaps leafy plants) in order to display each lense's rendering of OOF detail. If you think these are good ideas, feel free to use them in your tests too.

 

[raid]

Frank,
I have been thinking about such a factor, and I just was looking for the right place in our home so that there will be a window and some space left behind the subject photographed. I will find something suitable.

 

[harry01562]

I'm looking forward to the results of your test, as well. A good 200/400 speed film will give decent results, better than most monitors will display. Very few of us have our monitors adjusted for proper color balance, anyway. It's time consuming, and not easy to really do it right.

Using your daughter is an excellent idea. I think you posted a pic taken outdoors, with great highlights, and excellent bokeh. That sort of pic would be very interesting with different lenses. The biggest consideration is a shot wide open, and another stopped down to about f8, IMO. Probably flare, contrast and bokeh will be the main things to look for at f8. Practically any lens should be good there.

Good luck with the project, and keep your daughter supplied with cookies or ice cream, to keep the smile wide.

Harry

 

[FrankS]

Testing 9 lenses (of various max. f-stops) at different f-stops can get muddled. Here's what I'm planning:

- 2 lenses (with max aperture 1.5) at f1.5
- 7 lenses (including the 2 above) at f2
- all 9 lenses at f4 (this includes for the first time the f3.5 Elmar and Fed)
- Elmar vs Fed at f5.6

That's 20 frames altogether and that's enough for me. Stopping almost any lens down to f5.6 or smaller will result in very good results and not much difference between lenses to compare. That's my rational anyway.

Sound reasonable?

 

[FrankS]

I have 3 other lens tests to do as well.

the COntax/Kiev 50's: I have a J8, Helios, Zeiss prewar 1.5, postwar 1.5, and postwar f2

35mm lenses: J12, Canon f2.8, Summaron (hopefully soon) and 40mm Rokkor

long lenses: Canon 85 f2, 100 f3.5, Leitz 90 f4 Elmar, and 135 f4.5 Hektor

Some folks may say that lens testing is a waste of time, just go out and shoot, and I was of that mind years ago too. Now that I"ve slowed down a bit, I think tests like these are useful to me to learn the signatures of my vintage lenses. (Testing modern lenses I still think is a waste, because they are all sharp and contrasty. It's the vintage fingerprints that are interesting to me.) And I think that others find lens comparisons interesting too. I really liked Raid's 50 tests, and Bill M's tests of his long lenses.

 

[raid]

I'm looking forward to the results of your test, as well. A good 200/400 speed film will give decent results, better than most monitors will display. Very few of us have our monitors adjusted for proper color balance, anyway. It's time consuming, and not easy to really do it right.

Using your daughter is an excellent idea. I think you posted a pic taken outdoors, with great highlights, and excellent bokeh. That sort of pic would be very interesting with different lenses. The biggest consideration is a shot wide open, and another stopped down to about f8, IMO. Probably flare, contrast and bokeh will be the main things to look for at f8. Practically any lens should be good there.

Good luck with the project, and keep your daughter supplied with cookies or ice cream, to keep the smile wide.

Harry

Harry,
I have given the idea of taking some of the shots outdoors some thought. I amy use two rolls of film to tget the job done. One roll for indoors with a tripod and one for outdoors handheld shots.

 

[raid]

I have 3 other lens tests to do as well.

the COntax/Kiev 50's: I have a J8, Helios, Zeiss prewar 1.5, postwar 1.5, and postwar f2

35mm lenses: J12, Canon f2.8, Summaron (hopefully soon) and 40mm Rokkor

long lenses: Canon 85 f2, 100 f3.5, Leitz 90 f4 Elmar, and 135 f4.5 Hektor

Some folks may say that lens testing is a waste of time, just go out and shoot, and I was of that mind years ago too. Now that I"ve slowed down a bit, I think tests like these are useful to me to learn the signatures of my vintage lenses. (Testing modern lenses I still think is a waste, because they are all sharp and contrasty. It's the vintage fingerprints that are interesting to me.) And I think that others find lens comparisons interesting too. I really liked Raid's 50 tests, and Bill M's tests of his long lenses.

Frank: It is better to have several tests than one huge test involving all lenses. Else, things become unclear. My last test showed the need to have two lenses inspected and maybe cleaned, so the testing was not a waste. Plus, I used my family a smy model so I ended up with several nice portraits.

 

[rbiemer]

Hmmm, my Zorki and its Jupiter-8 are off being serviced; when they get back to me I'm going to be doing some testing of my own to decide which of the 50s I own I like best. I have 5 lenses and will be putting each of them on the same camera and shooting the same shot and then ranking them. I'm not going to be counting line pairs or shooting test targets and so this discussion is helpful to me in thinking about what and how.
I'll want to get all the shots on the same roll of film so I'm trying to work out 4 photos from each lens but it looks like 6 is a minimum for what I want to do:
1-Close focus and wide open
2-Far focus and wide open
3-Close focus and same fstop
4-Far focus and same fstop
5-Close focus and min. fstop
6-Far focus and min. fstop
1,2,5, and 6 will be what ever the max and min is for each lens--f3.5, f2, f2.5, and f2.8 for the wide open shots and f16 or f22 for the stopped down.
The middling shots will likely be f8.
The biggest problem, I think will be to get this done in consistent light. The close focus shot would be fairly easy to do indoors but I don't know where I could get anything like infinity focus inside. That's something I need to figure out.
The other thing is that I suppose I should do some side by side shots to see what the color cast/bias of these lenses might be. But to keep the light the same for those I think I'll need to buy a flash. Except that none of them has a noticeable or objectionable color signature and I don't really care.
I'm also thinking that since I'm only looking to see which "look" I prefer, this may be over-kill? And maybe the way to do this would be to shoot 3 shots with each lens: a landscape, an abstract, and a portrait? And let the tech stuff fall where it will?
What do you folks think? Simple or rigorous tests?
Rob

 

[raid]

I agree with Rob that keeping the same lens mounted on the camera and changing aperture settings works best. This way, focusing is not changed and also you don't have to worry too much about wear and tear compared to changing the lens after each shot is taken.