Some resolution tests and multicoatings - just for fun, some picutes, lots of words

[matthew_brain]

For everyones interest as I have not seen results for a biogon on he

net yet, here are the lens resolutions I got with Johns biogon and my

wide angles. The biogon is very good across all the apertures.

 

Although my SA90 results are similar to kerry thalmans at

http://www.hevanet.com/cperez/testing.html

my tests are best intrepretted relative to each other.

 

Note these are the minimum resolutions, some of the lenses had a

marked transverse radial discrepancy (in keeping with standard

estimates for lenses eg

http://www.butzi.net/rodenstock/grandagon/65mm.htm)

 

The results are in lens name Aperture-Tested Center resolution Middle

and Edge

 

Shot Lens Aperture Center Middle Edge

1 Super Angulon 90mm f8 f11 64.8 57.7 28.8

2 Super Angulon 90mm f8 f22 57.7 51.4 40.7

3 Super Angulon 90mm f8 f32 40.7 36.3 36.3

 

 

4 Grandagon 90mm f4.5 f11 64.7 64.7 45.8

5 Grandagon 90mm f4.5 f22 57.7 57.7 51.4

6 Grandagon 90mm f4.5 f32 51.4 45.8 36.3

 

2 Biogon 75mm f4.5 f4.5 43.6 43.6 43.6

3 Biogon 75mm f4.5 f11 43.6 43.6 38.8

4 Biogon 75mm f4.5 f22 48.8 38.8 34.5

5 Biogon 75mm f4.5 f32 38.8 34.5 27.5

 

1 Planar 135mm f3.5 f11 50.8 45.3 40.3

2 Planar 135mm f3.5 f22 40.3 45.3 28.5

3 Planar 135mm f3.5 f32 32.0 28.5 25.4

 

4 Fujinon W 180mm f5.6 f5.6 53.4 47.6 30.0

5 Fujinon W 180mm f5.6 f22 47.6 47.6 42.5

6 Fujinon W 180mm f5.6 f32 42.5 42.5 37.8

 

1 Sironar 180mm f.6 f5.6 53.3 47.6 47.6

2 Sironar 180mm f.6 f22 42.4 42.4 42.4

 

 

3 Tele-Arton 250mm f5.6 f5.6 54.3 48.1 43.0

4 Tele-Arton 250mm f5.6 f22 43.0 48.1 43.0

 

Note the sironar has some coating problems.

 

I'll say again, these results are relative to each other. Done using

a postscript version of the usaf test chart in bright sunlight.

 

A bit of fun. It doesn't make much difference in the field, where not

all the world is on one plane and compromises must be chosen - and

shooting speed maintained.

 

Also for interest is a comparison between a single coated lens

(grandagon 90mm) and a multicoated lens (super angulon 90mm). The

flare is warmer in the single coated lens On the positive and even

more noticable on black and white negatives, shadow detail holds

better contrast with the multicoated lens. Again this is not

scientific and I have not had both shutters tested (to ensure one is

not letting more light on the shadows) but it is interesting.

 

Matt<div></div>

[matthew_brain]

Note the above attached picture looks closest to the film when my monitor is set to full contrast and brightness just down from full where black is black.

 

No contrast or curves have been altered in photoshop, so that no diffence to the flare would be caused by optimizing the scan for the screen.

[neil_poulsen1]

This is interesting data. What strikes me is that at f22, the two modern wide angles demonstrate better performance than the Planar, the Fujinon W, and the Sironar. I would have expected this to be the other way 'round. It would be interesting to have results at f11 for these three lenses, to compare the wide-angles with these three lenses at an optimum aperture.

 

As to multi-coated vs. single-coated lenses, the warmer results for the single-coated could be due to the brand difference. I've heard that Rodenstock lenses are a tad warmer than Schneider lenses.

 

What would be neat is to have data on multi-coating vs. single-coating for the same lens brand and focal length for lenses made in the same time-frame with multiple examples. How much difference does multi-coating really make, and what's the nature of those differences?

[david_goldfarb]

The comparison between a single-coated Grandagon 90/4.5 and a multicoated 90/8.0 Super Angulon isn't really a test of the coatings, since a faster lens will generally be more flare prone anyway, though for your own purposes, this test confirms that you would want to use the S.A. in flare-inducing situations. A better test of the effect of the coating alone would be to compare a single-coated Super Angulon 90/8.0 with a later multicoated one.

[david_goldfarb]

I have a late, pristine, single-coated 90/8.0 Super Angulon, so if there is someone who has a multicoated version of the same lens who lives in the New York area and wants to do that side-by-side test, I'm game.

[james_driscoll2]

I have also done similar quick and dirty tests- I compared a 180mm

symmar vs a 180mm apo symmar, and a 121mm super angulon vs a 120mm

super angulon.

 

I have noticed there to be a slightly warmer cast to the chromes taken

with the single coated lenses, but the diff. between them was minimal

compared to when I used to have a Nikon 35mm outfit that had all newer

lenses except for my 50mm 1.4 which was single coated.

 

With the Nikon, you could tell right away which shots where taken with

the singel coated lens-there was a big diffrence in contrast and

color.

 

At first, I was surprised when I did the symmar test-because after the

nikon experience I expected a night and day diffrence.

 

On the last book I did (1000 paris buildings-due out in the fall), I

used a 72mm SA, 90mm SA (single coated), 120mm SA, 180mm (single

coated), 240mm Nikkor, and a 300mm symmar (single coated, w/slight

separation).

 

I noticed almost no performance diff. with the 90mm compared to the

others, but did notice a slight reduction in contrast with the 180mm.

The 300mm also showed almost no diff. from the "new" lenses.

 

When I also did the test of 121mm SA vs 120mm SA, I also noticed very

little of a change between the "look" of the lenses. When put side by

side on a light box, the 121mm only had the slightest amount of lower

contrast, but I have to admit it was like splitting hairs.

 

I also used an uncoated 240mm Plaubel lens, which I purchased to give

me soft and off-color effects. Of course when you buy said lens to

give you that, it turns out to be sharp as a tack and pretty on with

its color!!! I was able to count the lines in a sweater on a

polaroid, and I only shot the damn lens a f11!!!

 

Not bad for a $40 optic.

 

I think it really seems to be hit or miss, once the 70's hit, I think

the production tolerances got much better so even the single coated

lenses perform almost as well as there multi-coated versions. Every

"late" single coated lens I tested was fine-every pre 1970 one would

give diffrent results.

 

It harks back to the old days, when I heard that professionals would

"borrow" 5 versions of the same lens from the dealer to test and would

keep the best performer. I think if you did that today, you notice

almost no (or little) diffrence between them.

 

th

[matthew_brain]

David,

 

I completely agree, lens design is far more important in flare handling. My comparison in this case is to show the effect the lens coating has on the flare - note the colour difference of the sunrays. The SA90 f8 has a blue hue to the coating. Testing the same lens designs would yield better results of the actual image enhancing effect of the coating.

 

Neil,

 

My main suprise was that the biogon and planar didn't acheive higher resolutions although their performance at wide apertures was very impressive compared to the modern designs.

 

The sironar has a coating problem and the fujinon and tele-arton have a very slight film on one of the elements when looking at a light. This could well be the cause of the lower results. A good service could change these.

 

Some time soon I am planning to repeat the tests.

[jason_greenberg_motamedi]

Matt,

 

Thanks for posting this. I have, for some time, been curious about the resolution of lenses used wide open, and have been slowly testing a number of fast lenses. My results are posted here:

 

http://astro.temple.edu/~sanford/LF_Info/Fast_Lens.html

 

As you write, the Biogon doesn't achieve stellar resolution stopped down, however I must say that at f/4.5 it is higher than any lens I have seen, and higher than many lenses stopped down to f/11 or f/16--very very impressive. I would be curious to see what the resolution of the Planar is open to f/3.5.

 

Thanks again,

[wieslaw1]

And I could not care less about your resolution charts and counts.

[arne_croell]

With respect to the f/3.5 135mm Planar (single coated), I can relate the following qualitative impressions from a series of test shots (sorry, no numbers): f/3.5 is rather soft, but it gets much better already at f/4! Big change for just 1/3rd of a stop. It had the best performance at f/8.

[j.w.]

These results are useful for those of us contemplating the purchase of LF lenses. Agreed, lens testing in itself gives only a certain amount of information on the overall performance in actual real-world applications, where there's plenty of variables. But such data is a good starting point for an informed decision.

 

Although they're oft times criticize for their over-reliance on lens testing numbers, Pop Photo's method is still useful for comparison between similar lenses, all else being equal.

 

In an ideal world, there would be a LF publication that would do such lens testing for us, and publish the results periodically. I'd even like to see ratings of lens flare/coloration and bokeh.

 

In the meantime, its great that people like Matt take the time to do this. One thing I noted of interest is that for the Super Angulon and Grandagon, edge sharpness improves as the lens is stopped down from f/11 to f/22, while center sharpness decreases a bit. That adds another degree of complexity to the assumptions pertaining to optimum aperture for sharpness.

[michael_briggs2]

<p>

Matt, thanks for the lens test results. I've duplicated them in html table

form for easier reading:

</p>

 

<p>

<TABLE>

 

<TR>

<TH>Shot</TH><TH>Lens</TH><TH>Aperture</TH><TH>Center</TH>

<TH>Middle</TH><TH>Edge</TH>

</TR>

<TR>

<TD>1</TD><TD>Super Angulon 90mm f8</TD>

<TD>f11</TD><TD>64.8</TD><TD>57.7</TD><TD>28.8</TD>

</TR>

<TR>

<TD>2</TD><TD>Super Angulon 90mm f8</TD>

<TD>f22</TD><TD>57.7</TD><TD>51.4</TD><TD>40.7</TD>

</TR>

<TR>

<TD>3</TD><TD>Super Angulon 90mm f8</TD>

<TD>f32</TD><TD>40.7</TD><TD>36.3</TD><TD>36.3</TD>

</TR>

<TR>

<TD>4</TD><TD>Grandagon 90mm f4.5</TD>

<TD>f11</TD><TD>64.7</TD><TD>64.7</TD><TD>45.8</TD>

</TR>

<TR>

<TD>5</TD><TD>Grandagon 90mm f4.5</TD>

<TD>f22</TD><TD>57.7</TD><TD>57.7</TD><TD>51.4</TD>

</TR>

<TR>

<TD>6</TD><TD>Grandagon 90mm f4.5</TD>

<TD>f32</TD><TD>51.4</TD><TD>45.8</TD><TD>36.3</TD>

</TR>

<TR>

<TD>2</TD><TD>Biogon 75mm f4.5</TD>

<TD>f4.5</TD><TD>43.6</TD><TD>43.6</TD><TD>43.6</TD>

</TR>

<TR>

<TD>3</TD><TD>Biogon 75mm f4.5</TD>

<TD>f11</TD><TD>43.6</TD><TD>43.6</TD><TD>38.8</TD>

</TR>

<TR>

<TD>4</TD><TD>Biogon 75mm f4.5</TD>

<TD>f22</TD><TD>48.8</TD><TD>38.8</TD><TD>34.5</TD>

</TR>

<TR>

<TD>5</TD><TD>Biogon 75mm f4.5</TD>

<TD>f32</TD><TD>38.8</TD><TD>34.5</TD><TD>27.5</TD>

</TR>

<TR>

<TD>1</TD><TD>Planar 135mm f3.5</TD>

<TD>f11</TD><TD>50.8</TD><TD>45.3</TD><TD>40.3</TD>

</TR>

<TR>

<TD>2</TD><TD>Planar 135mm f3.5</TD>

<TD>f22</TD><TD>40.3</TD><TD>45.3</TD><TD>28.5</TD>

</TR>

<TR>

<TD>3</TD><TD>Planar 135mm f3.5</TD>

<TD>f32</TD><TD>32.0</TD><TD>28.5</TD><TD>25.4</TD>

</TR>

<TR>

<TD>4</TD><TD>Fujinon W 180mm f5.6</TD>

<TD>f5.6</TD><TD>53.4</TD><TD>47.6</TD><TD>30.0</TD>

</TR>

<TR>

<TD>5</TD><TD>Fujinon W 180mm f5.6</TD>

<TD>f22</TD><TD>47.6</TD><TD>47.6</TD><TD>42.5</TD>

</TR>

<TR>

<TD>6</TD><TD>Fujinon W 180mm f5.6</TD>

<TD>f32</TD><TD>42.5</TD><TD>42.5</TD><TD>37.8</TD>

</TR>

<TR>

<TD>1</TD><TD>Sironar 180mm f5.6</TD>

<TD>f5.6</TD><TD>53.3</TD><TD>47.6</TD><TD>47.6</TD>

</TR>

<TR>

<TD>2</TD><TD>Sironar 180mm f5.6</TD>

<TD>f22</TD><TD>42.4</TD><TD>42.4</TD><TD>42.4</TD>

</TR>

<TR>

<TD>3</TD><TD>Tele-Arton 250mm f5.6</TD>

<TD>f5.6</TD><TD>54.3</TD><TD>48.1</TD><TD>43.0</TD>

</TR>

<TR>

<TD>4</TD><TD>Tele-Arton 250mm f5.6</TD>

<TD>f22</TD><TD>43.0</TD><TD>48.1</TD><TD>43.0</TD>

</TR>

</TABLE>

</p>

[wieslaw1]

To me more important than fractional numbers of lines/mm resolved is barrel/pincushion distortion of the above lenses. Any idea?

[matthew_brain]

Wieslaw,

 

Both the schneider website and paul butzi's rodenstock information contain the theoretical barrel and pin-cushion distortion measurements for their lenses. Test charts are available to do this I beleive. I just accept the lenses distortion as a characteristic of the lens. To your first answer, I only became interested in doing the tests out of curiosity and some availability of several lenses. Knowing the optimum aperture is useful but these are usually published anyway, there are many factors that influence sharpness in a scene and I usually err toward composition even if it means f32 or more.

 

Michael,

 

Thanks for tabulating the results, they are much clearer.

 

Joe,

 

I read somewhere that the maximum lp for f32 is 40 something as this is diffraction limited but that the smaller aperture compensates for the difficiencies of the lenses at the edge. Rodenstock suggest two optimum apertures for their lenses eg the sironar is f16 when no movements are used and f22 when movements are used.

[j._mose]

Matthew,

 

I always find these comparisons interesting! The only one that surprises me is the Biogon test. I have a 75mm Grandagon N f/4.5 and a 75mm Biogon. Although I have not utilized resolution charts for comparison, I have performed shots of same subjects using each lens. I did the shots in total darkness, open shutter, electronic flash, tripod, indoors on concrete floor, ground glass focusing, etc. All for purposes of trying to eliminate ANY vibration. I found that the Biogon was much better from f/4.5 through f/8 (no surprise) and was equal with the Grandagon at f/16. I guess the great feature about a Biogon is being able to use it at any aperture and yield great results! Close up performance is also great on the Biogon! It reminds me of a Leica lens -- I NEVER feel shy about using my 24mm Elmarit ASPH or 35mm Summilux ASPH wide open.

 

Of course my above comparisons do not yield resolution figures...perhaps my Grandigon and Biogon do not exceed your resolution figures on the Biogon. All I can do is compare it with results from several other LF lenses, which always places the Biogon more on top.

 

Thanks for taking the time to do this.

 

Best regards,

 

JP Mose

[matthew_brain]

I could not tell the images apart for sharpness with a light table and a loupe until I used a microscope at a pathology lab. They are all pretty sharp and it is comforting to know that f32 provides a reasonably small drop in sharpness for the increase in DOF available. I certainly could not tell a shot at f32 from one at f16 with a 10x loupe.

[mclaine]

G'day J. Mose,

 

My Biogon is the one Matt tested. It is a 1976 version. It had been locked away in a box since the '70's, (along with a Tech V, plus the 135mm Planar, 180mm Sironar and 250mm Tele-Arton that appear in Matt's list), until just a few weeks ago, when I bought it. All of the gear had a thin film of crud on it from 20+ years locked away unused. I only gave the front and back a very light clean, but it had virtually never been used before it was forgotten and locked away in the late 70's, so I'd be hopeful that after I send it away for a professional CLA, it should sing as sweetly as yours.

 

Also bear in mind that Matt makes no claim that the figures are absolute, just of interest for comparison of a bunck of different lenses (basically all of Matt's and mine). They are just one TMX frame each, of inkjet facsimiles of the USAF Chart on office copy paper, sticky-taped to Matt's back wall, focused on Matt's GG, developed according to his usual method.

 

Just for fun.

 

Cheers,

[mclaine]

Err..that's a bunch of lenses. Not a bunck. In case anyone is reaching for their dictionary.

 

John Mc.