Docter Optic Tessar, Apo-Germinar MTF curves

[Dan Fromm]

Ian, the 1971 Yakovlev catalog, pp. 368 ff, reports that at f/10 the RF-5 resolves 35 lp/mm at the center, 15 lp/mm in the corners. Stopped down a little it is usable on 2x3. Claimed coverage is 43 degrees, a little over 350 mm at infinity. I have the impression, not checked, that it is somewhat dimmer than the nominal maximum aperture would suggest. Mount threads are M72x1, and quite a distance from the end of the lens; front-mounting it will require a deep cup-shaped adapter.

I have an RF-5, have shot it. I'd rather use a good grade of dialyte type process lens or an Apo-Saphir than my RF-5, which I bought before I found an affordable 480/9 Apo-Nikkor. If I hadn't found that Apo-Nikkor I'd have used the RF-5 and would not have felt sorry for myself for having to use it.

To educate yourself about Soviet lenses, visit http://www.lallement.com/pictures/files.htm and download the 1963 GOI and 1970 and ’71 Yakovlev catalogs and visit http://www.photohistory.ru/index.php, which covers fewer lenses but includes newer ones than the GOI and Yakovlev catalogs.

Arne, aren't you the person who convinced me to look harder at my RF-5 to see the very weak reflection from the inner groups' glass-cement-glass interfaces?

[Ian Greenhalgh]

Cheers Dan. I had read that there were earlier and later versions of the LOMO process lenses with the earlier ones being dialytes and the later ones being 6/4 double gauss.

My interest is in using it on 5x7, I think it's about the longest that I can fit on my Lancaster Instantograph comfortably, although I need to measure just how much bellows extension it actually has.

I might just wait until an 18 inch Wray Lustrar or Ross Xpres process lens comes along, I have 10/300 Process Lustrar and Xpres lenses and both are excellent, I have much more faith in the consistency of QC with Wray and Ross than the Soviets.

[Arne Croell]

Arne, aren't you the person who convinced me to look harder at my RF-5 to see the very weak reflection from the inner groups' glass-cement-glass interfaces?

I am embarassed to say that I only dimly remember something to that effect now that you mention it. I guess by not learning from history I had to repeat it...

[Emmanuel BIGLER]

Dan said:
aren't you the person who convinced me to look harder at

In our series: Secrets of the Cold War Unveiled, here are the persons who easily convinced Dan in 1963 to look harder at [TOP SECRET]

Many thanks to Arne and Dan for pointing us to so many fascinating documents !
Too bad that I can't read Russian ..

[ridax]

Maybe you also know the answer to a recent question on another lens we discussed here, its about the unknown logo of a company, see: http://www.largeformatphotography.in...ufacturer-logo

Arne, I've seen the thread but sorry that was the first time I saw a logo like that. In fact, I always was interested in practical usage of lenses (and I tested a great deal of those myself. Generally, I try to avoid posting anything from books whenever possible - just my own first-hand opinions) but not in the historical nuances so I'm afraid I can't help too much on the points like that logo. I am pretty sure you know the Russian photo gear history way better then myself. Still I'd be glad to help if I can; please feel free to PM me as I may be away from the forum for months sometimes.

the 1971 Yakovlev catalog, pp. 368 ff, reports that at f/10 the RF-5 resolves 35 lp/mm at the center, 15 lp/mm in the corners.

Dan, I think it was you that pointed once the Russian resolving power figures were always for the lenses shot wide-open. That is very true. And that is just what matters for aerial lenses (the main concern of the Soviet optical industry) but isn't much help in evaluating any other types of glass.

And there is more. As decades passed, the industry standards changed. There were periods they used white test targets with black lines (= a lot of stray light lowing the image contrast which in turn lowered the resolving power figures), and periods when black test targets with white lines (resulting in very little stray light in the system) were used. The films used in the tests changed, too. (And they usually were pretty low-resolution films, BTW - hence the low photographic resolving power figures published for lenses.) To make the things even worse, the popular literature almost never mentioned the film types used. And often there were mixed figures from different tests (made with different films and different test targets, decades apart) in the same book or magazine article (and later, on websites such as the KMZ's). One can see photographic resolving power figures ranging from, say 16 to 48 line-pairs/mm for exactly the same lens shot at exactly the same f-stop (wide open). I wouldn't value that numbers too much. :)

Also, the quality control is almost never a problem with Russian optics (as opposed to cameras and some other gear - including the lens irises, focusing mounts, etc.). For their cheap high-volume production lines, they just were smart enough to chose the lens designs that had large tolerances from the start, so visible sample to sample variations in sharpness are in fact extremely rare (that does not apply to coating, stray light and image contrast, though).


As for the RF-3,4,5 lenses, they were reported to be somewhat better for process work (but I didn't care to verify that) then the Industar-11M line but for general photography, those RF's are equal to the I-11s in sharpness. The RF's are also bigger, have less coverage due to vignetting with their longer barrels (both the 9/300 Industar-11M and the 9/300 Apo-Tessar cover 8x10" at infinity; the RF-3 does not - if I remember correctly, as I've got rid of all my RF's), and are worse in color rendition (greenish-yellowish) due to their inferior coating (2 layers deposited from an ester solution by coagulation).

The RF-240 and RF-300 are plasmats. Their quality is OK but nothing special. Plasmats have anastigmatic menisci next to the iris, and those make great reading glasses. This is what the RF-300 (as well as the Apo-Graphigon, etc.) is the best for IMO. One can buy some 5 to 8 RF's for the price optical shops charge for just one uncoated German-made consumer Punktalglas (anastigmatic meniscus) here in Russia!

For taking pictures, I personally prefer the 9/300 and 9/450 Induatar-11M's - the earlier ones with better (the pale looking vacuum-deposited) coating. They are dirt cheap and sharp enough, and their barrels are pretty easy to unscrew just a bit to change the distance between the 1st and the 2nd glass elements - the old trick to change the out of focus rendition (and add softness if you wish). The resulting background blur is still not on par with a Dagor's though, and the resulting foreground blur is not as good as a Convertible Symmar's at f/5.6 (but the Symmar isn't a good foreground-maker stopped down, and the Industar is). The later version 4/4 ("decemented" - but still tessars) f/9 GDR Apo-Tessars are slightly sharper then the I-11's but the price difference is just enormous compared to the sharpness difference, and the lens elements in the later Apo-Tessars can't be moved - as well as the Apo-Germinars' and Apo-Ronars' elements. The Apo-Germinars (and Apo-Ronars, and etc.) are still more sharp but the celor-type ones have far less coverage. Tessar type process lenses are sharp up to the edges if well stopped down; celor type ones are not.

The longer 6-elements Apo-Germinars at smaller f-stops are pretty sharp up to the edges of the illuminated field. My 12/1000 (CZJ GDR, in barrel) actually covers 1.25 meter diagonal at f/32 at infinity. That's WAY more then the published figures, and sharpness is great not only for contact prints at all.

The 600mm and longer Industar-11 lenses are rare as they were replaced with the 10/600 and 10/750 celor-type "O-2" lenses at least as early as the 1960's. The O-2's do not contain any heavy glass types so they are not as sharp as the Apo-Ronars but also not as yellowish (especially if you get an amber-coated one). That's a GREAT advantage for me as I really love blue-sensitive film. (That also applies to the Industar-11's which are among the most transparent process lenses in the Blue and UV regions of the spectrum.) And the lens element separation in the O-2's is changeable as easily as in the I-11's, with the same effect.

The 600mm O-2 is very common and extremely cheap; the 750mm one is harder to find and is several times more expensive. The later 10/800 "O-6" (same type, same glass as the O-2) is still harder to obtain; my 1979 one has coating even better then any of my (selected) O-2's from the 1960's. It is also lighter in weight: the O-2's are in Aluminium barrels, but the glass elements themselves are in brass; the O-6's barrel is all Aluminium. The O-6 lens elements are easy to move, too. In fact, that's the latest barrel still employing the 'each glass threads in' principle that I know of (though the pre-WWII German lenses were usually in that type of barrels). And though my 600mm is not going to get much usage while I do have my 9/505 Protar IIIa (still better background, still more coverage - and less weight, too), I would never trade my 800mm O-6 for any other process lens (but then go offer me the 820mm IIIa instead...).

Some weight figures for the longer ones - all lenses in barrels, no flanges, no caps; read from a primitive kitchen scale, in kg; and mounting threads:

Apo-Tessar_____9/600:__1____kg;__M_90x1_(later 4/4 CZJ GDR version)
O-2___________10/600:__1.5__kg;__M_90x1
Industar-11M___9/600:__2____kg;__M_90x1_(1956 all-brass barrel)
Apo-Germinar___9/600:__3____kg;__M_90x1_(CZJ GDR)
Apo-Tessar_____9/750:__1.8__kg__________(later 4/4 CZJ GDR version)
O-2___________10/750:__2.75_kg;__M110x1
О-6___________10/800:__2.2__kg;__M110x1
Apo-Germinar_12/1000:__3.5__kg;__M110x1_+_M115x1_(CZJ GDR)
Apo-Tessar___11/1200:__6____kg__________(later 4/4 CZJ GDR version)
Apo-Ronar-CL_16/1200:__6.5__kg;__M120x1_(6/6, #10,xxx,xxx)

[Dan Fromm]

Ridax wrote:

Also, the quality control is almost never a problem with Russian optics

Interesting, if true. Here are two observations that disagree with you:

First, some years ago a number of 200/6.3 Orions (Ia, I think, could be mistaken) came to market. Buyers complained loudly about defective front elements.

Second, this http://web2.ges.gla.ac.uk/~gpetrie/P...y_fulltext.pdf reports severe centering problems with Russar mapping lenses.

About the value of resolution tests taken wide open, well, in general the lens will never do worse than that. Exceptions include high performance macro lenses, where image quality degrades visibly on stopping down. Few of us work in that neighborhood.

Further on this point, my Uran-27 is ok but no better at the apertures I normally use than a much lighter lens of the same focal length, tessar or plasmat, in shutter. So the Uran-27 almost always stays home, as does my 4"/2.0 TTH. The GOI catalog reports the Uran-27's resolution (center/corner) wide open as ~ 48/~8 lp/mm wide open. Not that wonderful, but the worst it will do.

Cheers,

Dan

[Ian Greenhalgh]

I've owned over 50 Russian lenses, never had a bad one yet. My friend Attila has tried over 200 Russian lenses, he reported only two duds, and both had probably been taken apart by a previous owner and reassembled wrongly.

The claims of highly variable quality are greatly exaggerated.

[Armin Seeholzer]

Its good to see that Zeiss in east germany also only cooking with water, it I look on the MTF graphs!

Cheers Armin

[Arne Croell]

Its good to see that Zeiss in east germany also only cooking with water, it I look on the MTF graphs!

Cheers Armin

The laws of physics apply everywhere...
That being said, one has to keep in mind that MTF curves are usually calculated and depend on the input parameters, e.g. the spectral distribution of the light etc. It is therefore difficult to exactly compare curves from different sources. When I look at the Docter curves, I would say the following: The basic shape of the Tessar curves is pretty characteristic of the type with the intermediate dip, the Schneider curves for the Xenars looked similar; also, one has to keep in mind that these are at f/8, not f/22! The Apo-Germinar curves are pretty similar to the Apo-Ronar curves, but one has to look carefully: for instance, for the same lens type the MTF curves usually go down with increasing focal length, this is true for the 240-360mm Apo-Germinar, too. The following one, 450mm, is however higher than the 360mm curve. When looking at the Apo-Germinar W, the comparison would be G-Claron or Apo-Gerogon lenses, which have similar specs (60°-70°, f/8-f/9), but a different construction. Here the Apo-G. W is clearly superior, especially away from the center.

[ridax]

Ridax wrote: "Also, the quality control is almost never a problem with Russian optics."

Interesting, if true. Here are two observations that disagree with you: First, some years ago a number of 200/6.3 Orions (Ia, I think, could be mistaken) came to market. Buyers complained loudly about defective front elements. Second, this http://web2.ges.gla.ac.uk/~gpetrie/P...y_fulltext.pdf reports severe centering problems with Russar mapping lenses.

About the value of resolution tests taken wide open, well, in general the lens will never do worse than that. <...> my Uran-27 is ok but no better at the apertures I normally use than a much lighter lens of the same focal length, tessar or plasmat, in shutter. So the Uran-27 almost always stays home, as does my 4"/2.0 TTH. The GOI catalog reports the Uran-27's resolution (center/corner) wide open as ~ 48/~8 lp/mm wide open.

Dan, my remark was about cheap and simple mass-produced lenses. Mapping lenses aren't pieces of high-volume production. And they aren't cheap either. That is especially true for the Russar-29 which was the very first lens ever made according to the Roosinov's principle of aberrational vignetting (his own term). And btw, the Russar-29 was designed in 1945. So the well-proven method to choose a high-tolerance lens design instead of bothering to carry out the extreme QC in actual production, was unavailable for that lens. There was nothing to choose from. The design was the only one available those days.

Besides, I've got the impression the paper at the link you posted states the lenses they tested were not manufactured out of the specs but rather were intended to be like that from the start, and other 'external to the lens' means were to be used to correct for the distortion. So that's actually not a QC issue. (Sorry I hadn't enough time to read the .pdf thoroughly enough and can be mistaken on this.)

As for the Topogon-type Orions, those were used not only for aerial photography but also for low-grade (wideangles are not the way to go when one wants the best quality reproduction) process and copy work and even for inexpensive consumer projectors (like those used in schools before the computer era), too. Those pieces of glass were certainly much much inferior to exactly the same Orions made for mapping.

And more on the 'resolution' (another slang word... and with quite a bit of that nasty digital smell to it ;). For a lens, the correct term actually is photographic resolving power). Another too often forgotten problem with the 'resolution' (as well as the MTF) is that it's relevant to a flat subject only. Any bit of field curvature ruins that 'resolution' digits and MTF graphs. That's just the right way to judge a process lens, an enlarging lens and an aerial lens. But that's often close to nonsense for evaluating a general-purpose lens that deals with the 3-D world. And the Urans are among many lenses that are actually sharper in the field then the tests suggest (but certainly stopped down, they aren't as sharp as less-fast lenses of the same quality. Everything else being the same, the smaller the max. aperture, the better the sharpness. That rule is older then photography. And yes it's still valid).

On the other hand, there are also a lot of lenses that look much better in 'resolution' tests then they actually are. One reason for this is the field curvature; the second is, some lenses were (and still are, worldwide) actually made to look nice in tests rather then to be good for taking real-world pictures. Tests are great for marketing. And who cares if the consumer is unhappy with his/her new lens - since the purchase is made already? And though the Soviet economy didn't care for marketing, the lens designers and the industry bosses did care for high-figure reports and praises and rewards coming with those. And as there were well established testing standards within the industry, those guys often made their lenses to be excellent in the center and in one exact point of the field - the one to be used for resolving power testing, wide open, and didn't care much about the rest of the field as well as the rest f-stops. There is a ton of USSR era publications stating that according to the very accurate tests, this or that Soviet lens was far better then its CZJ analog. But the real world Russian photographers were willing to pay 5 to 10 times more for the CZJ glass nevertheless.

The Russian lenses are generally not great. But they are usually decent and quite cheap and really stable in quality, and they are indeed very usable. But the Russian resolving power figures are not. And all the other resolving power figures aren't either IMHO.