Test Procedure for Enlarging Lenses using Vlad's Test Target

Some of you may have read my previous blog post about comparing the performance of five 50mm enlarging lenses. I'm proud of the work I put into that post, but I wanted to develop a more standardized way of testing enlarging lenses without necessarily having multiple lenses of the same focal length.

In comes Vlad's Test Target (link to his website), a fine-grain mounted negative designed to calibrate camera-scanning rigs. I stumbled upon his product online, and immediately saw how useful it could be in the darkroom to test lens resolution and center-to-corner performance. The test slide is made with Adox CMS Pro II film and has an official resolution of 110 line pairs per millimeter (lp/mm). Details this fine rival the resolution achievable on fine-grain films such as Kodak TMax100¹ and Ilford Delta 100. Adox CMS Pro II is incredibly fine-grained and is capable of resolving up to 800 lines per millimeter (l/mm), so the limitation of the film stock itself will have no effect on this experiment. The design and principle behind the test pattern can be found here. As of now I only have the 35mm version of this slide, which I don't plan to use to test lenses longer than 80mm (Vlad makes test targets in larger sizes).

Vlad graciously provided me with this high-quality scan from his test slide, which shows the test lines alongside their corresponding resolution in lp/mm. Note how small this area is on the slide itself (1.5 x 1.3mm); it can be viewed in some of my test prints below. The slide was scanned by Dominique Ventzke at High End Scans.

To prevent this post from getting too lengthy, I focused on testing only two lenses that in my opinion are on opposite ends of the quality spectrum: the EL-Nikkor 50mm f/2.8 and the EL-Omegar 50mm f/3.5. My hope in testing these particular lenses was that I would see similar results in center sharpness but noticeably different results in the corners. The EL-Nikkor was tested at print sizes of 5x7, 8x10, and 16x20. The EL-Omegar was only tested at 16x20, to most clearly illustrate the difference between the two lenses. I did not print full sheets at each of these sizes (only for the 5x7), but rather adjusted the enlarger to the proper size, and printed a small area of the test slide that contained the area I wanted to analyze. I made all prints at f/8, both to simplify the procedure and to give the lenses their shot at the most "ideal" printing aperture.

Each of the resulting prints was scanned at 1200dpi on a flatbed scanner. At this resolution, the test target and the lenses being used are the limiting factor rather than the resolving capability of the scanner.

5x7 print of the test slide using EL-Nikkor 50mm at f/8.

The 5x7 print was, as expected, very sharp and the details were all excellently resolved. I wasn't expecting to push any limits at this size, just needed a straight print. The corner details appear to be exactly as sharp as the details in the center. When I zoom into the scanned print, I can see slight differences in sharpness, but the difference is indiscernible to the naked eye.

    

Enlarged now to 8x10, much finer details can be seen. The smallest lp/mm are more discernible, but not yet fully resolved.

 

At 16x20, my enlarger has reached its maximum height without having to adjust it to project onto the floor. The very smallest details are visible now, and the resolution limit of the test target has been reached. The corners are quite obviously softer than the center, but still very suitable for printing.

And here are the results from the EL-Omegar. In my last discussion on this lens, it mainly showed inconsistencies between aperture settings and overall poor sharpness at all but the most ideal apertures. These prints reveal another interesting characteristic of this lens, particularly in the corners. The center is arguably as sharp as that of the EL-Nikkor, especially at f/8, which is expected to be the sharpest. What was interesting was the distortion that can be seen in the corners. The test lines that are tangential to the lens appear decent, but significant blurring can be seen in the radial lines.

Cropped in from the previous photo; the difference
between radial and tangential lines is very obvious.

I'm no optical engineer, but I can only assume the effect we see here is a factor of lens design, and not of anything about my setup. I would always expect a lens to lose some sharpness in the corners, but this is almost beyond usable. The distortion blurs details in the 15-17 lp/mm range, which would considerably reduce resolution when printing even the grainiest films.

In conclusion, I'm very satisfied with the results I was able to obtain with Vlad's Test Target. It is a quality product, perfectly suited for darkroom use, and highly recommended by me. I believe the results I obtained are very clear and the differences between these two lenses thoroughly demonstrated. I always welcome feedback of any kind, and hope that this post is of some use to those reading it!

Thank you.

1. The official datasheet for TMax100 lists a resolution of 200 lines/mm. Converting to line pairs per mm gives 100 lp/mm, barely less than the resolution limit of the test target.

Enlarging Lenses & Ideal Printing Apertures?

General advice in various darkroom-focused groups on the internet tells us to only use the "middle" aperture of an enlarging lens, or alternately 3 stops down from wide open. It certainly is good advice, but how rooted in empiricism is it? What exactly do we sacrifice when we adjust up or down one f-stop? What about adjusting 2 f-stops? My goal in this post is to provide some discrete examples of those effects, as well as to compare the overall performance of some well-known enlarging lenses.

I compared 5 enlarging lenses, all of which have a 50mm focal length. They differ somewhat in aperture ranges, but I will describe them further below and explain my testing process. The lenses I tested are:

1. Schneider-Kreuznach Componar f/3.5
2. Schneider-Kreuznach Componon f/4 
3. Nippon Kogaku EL-Nikkor f/2.8
4. Fuji Fujinar-E f/4.5
5. Omega EL-Omegar f/3.5

Some of these lenses have impractically open apertures when used wide-open (looking at the EL-Nikkor f/2.8 in particular), therefore I tested the 5 smallest apertures on each lens, since these are the ones that will likely see the most use.

The negative I decided to enlarge for this test was a recent shot of a toucan at the zoo. I was pushing Kodak 5222 Double-X to 1600 just to see what kind of results I could get. The photo is deliciously grainy, a quality I was looking for in this series of tests because the sharpness of the individual grains will be more evident at different f-stops than the sharpness of the subject or overall image. I made a cropped 5x7 test print of the negative; the exposure was 32 seconds at f/11 using the Fujinar-E lens listed above. The projected image was approx. 10"x12", and I used a #2 Ilford contrast filter.

Testing Process

For my test strips, I cropped even further to include just the area around the toucan's eye and a bit of the blurred background. I focused all lenses at their most open aperture, and stopped down to make 5 different exposures per test strip, one exposure per aperture. I've labeled the apertures on each of the scanned test strips below, and I'll discuss the performance of each lens below the test strip.

Exposure times were extrapolated from the 32 seconds required for the test print and were used for all lenses:

f/4 - 4 seconds

f/5.6 - 8 seconds

f/8 - 16 seconds

f/11 - 32 seconds

f/16 - 64 seconds

f/22 - 128 seconds

Schneider enlarging lenses have a good reputation, and for very good reason. Each successive stop down, and therefore doubling of exposure time, is very consistent along the tested range. f/4 is a little bit soft, but the grains are sharper at f/5.6 and fully sharp at f/8 up to the smallest aperture of f/16.

Similar to the Componar above, the Componon performed excellently. Brilliant consistency across the aperture range, and very sharp from f/5.6-f/16. Not much else to say, it's quite obvious why Componon lenses are some of the most recommended.

The EL-Nikkor performed just about as well as the Componon. Very consistent, and also very sharp from f/5.6 and smaller. Perhaps at f/16 it appears slightly less sharp.

I made a small mistake with this test strip. I forgot to stop down to f/22 for the last frame, and with that requiring an additional 2-minute exposure, I decided to forgo making an entirely fresh test strip. Similar to the performance of the lenses above, the Fujinar is quite consistent in exposure time as it is stopped down, with a bit of softness wide-open. Apertures f/8-16 all seem useable.

Perhaps not surprisingly, the EL-Omegar performed quite poorly. The exposures on the test strip varied considerably, meaning that adjusting this lens up or down a stop does not give an equal exposure by halving or doubling the exposure time, respectively. The lens is at least reasonably sharp around the middle apertures, f/11 is certainly an acceptable exposure if this is the only lens available.

Conclusion

In conclusion, it would seem that the extreme apertures of most enlarging lenses are either sub-optimal in terms of image quality, or impractical in terms of required exposure time. The middle apertures of all the lenses produced acceptable results, and I think are very difficult to differentiate when analyzing prints. Choosing quality lenses will make adjusting exposure when stopping up or down easier, but it seems we have a lot more room to play with than initially thought!