Confusion over 90mm Leica lenses and APO?

[Freakscene]

APO aint apochromatic, though.

Here: http://leica-users.org/v13/msg13490.html Erwin Puts points out:
"Now there is no industry norm that first describes which glasstypes you have to use to make a lens an "apo" and secondly describes which numerical deviations are required for such a designation."

If you use the absolute definition: i.e. an optical system corrected so that
it gives three images of identical size for three different spectral lines
or regions, you'll find that there are no APO lenses in current production, including those made by Leica. There are, however, several lenses that
are lens that has been corrected to a greater degree than in most other
lenses for the three primary spectral colors, including the R&M 90 APO
ASPHs. The differences in image size and therefore the 'degree' to which
the Leica APOs are apochromatic is an interesting study in itself. One thing is for
certain, they are better colour corrected (closer to the absolute APO) than
the majority of other current lenses.

Anyone with sufficient masochistic tendencies can calculate the (apo)chromatic error.

The other little-discussed influence is that the chromatic error decreases with stopping down, at least until diffraction makes it worse. At f8 or thereabouts, it is minimal, in all optical systems.

The true APO lenses for my microscopes cost about twice as much for a lens as an M9, but the lenses are about the size of your thumb and don't have an aperture or focusing mechanism. It is very difficult to achieve with camera lenses and not really necessary. 'Good' correction, such as is offered by the Leica APOs (excellent, really) will very, very rarely show any chromatic abberration.

Superachromatic lenses are corrected for four spectral lines but are mostly theoretical unless you have access to the kind of lens systems used in space, high end military . http://en.wikipedia.org/wiki/Superachromat There are Zeiss Superachromats for the Hasselblad: http://lenses.zeiss.com/photo/en_DE...ad/500_series/telesuperachromatt56350cfe.html but I don't know how fully corrected they are across all four spectral line. Very well, I guess, but I don't know if they really do all focus to a single point.

Marty

 

[Freakscene]

75 Summicron (aspheric, because they all are).

There is also the Leitz Midland 75/2.4 APO, which is not asph, but it is incredibly rare.

Marty

 

[ka7197]

Most apochromatic (Apo) lenses bring three colours to identical focus ...

No, most don't. In fact, most lenses designated 'apo' really are achromats only, not apochromats in the word's narrower sense, but with a lower-than-usual secondary spectrum (i. e. very low residual chromatic aberration). True apochromatic lenses for photographic purposes exist but are very rare.

... hence the paint-filled, engraved red, green and blue lines on the mount of both original and modern Apo-Lanthars ...

That's just a marketing ploy, supposed to make you believe these were true apochromats. Obviously it worked ...

... but a few apparently bring four colours to a common focus ...

Those are superachromats, and they are even rarer than true apochromats. They might be common (kind of) for industrial, scientific, or military purposes but for photography I am aware of only two superachromats, the Zeiss Super-Achromat 250 mm and 350 mm CFE telephoto lenses for Hasselblad medium-format cameras.

It is perfectly possible to build apo lenses with only common (spherical) curves.

As a matter of fact, the curvature of the elements has nothing to do with chromatic aberrations or the correction thereof, so aspheric elements won't help against chromatic aberrations. Clever combination of different kinds of glass (different ratios of refraction vs. dispersion) does. Aspheric elements help against spheric aberrations only. As I said: apo and asph are two entirely different things.

The other little-discussed influence is that the chromatic error decreases with stopping down, at least until diffraction makes it worse. At f/8 or thereabouts, it is minimal, in all optical systems.

This is only half-way true. Longitudinal chromatic aberrations will get reduced by stopping down but lateral chromatic aberrations won't.

'Good' correction, such as is offered by the Leica 'apochromats' (excellent, really) will very, very rarely show any chromatic abberration.

Sigh ... if only that was true. Unfortunately, it isn't. When looking closely then you'll see both longitudinal and lateral chromatic aberrations in pictures taken with current Leica Apo-Summicron-M Asph or Apo-Telyt-M lenses just as with any non-apo lens. Sure, their chromatic aberrations are pretty low, definitely lower than in non-apo lenses, but not lower by one or more orders of magnitude ... it's maybe half or one-third as much. That is a significant improvement for sure but still far from perfect.

 

[Freakscene]

This is only half-way true. Longitudinal chromatic aberrations will get reduced by stopping down but lateral chromatic aberrations won't.

Of course, but this is why manufacturers choose to control different aberrations differently. You can detect some CA in the leica APO lenses wide open, but while at f8 I can measure in on my optical bench or detect it using the colour picker tool in PS at 100% pixel view, I can't see it in prints.

Sigh ... if only that was true. Unfortunately, it isn't. When looking closely then you'll see both longitudinal and lateral chromatic aberrations in pictures taken with current Leica Apo-Summicron-M Asph or Apo-Telyt-M lenses just as with any non-apo lens. Sure, their chromatic aberrations are pretty low, definitely lower than in non-apo lenses, but not lower by one or more orders of magnitude ... it's maybe half or one-third as much. That is a significant improvement for sure but still far from perfect.

They're good enough for me, although I must admit my own interest in CA only goes as far as wanting sharper edges, because I mostly use B&W for my own photography and when I work professionally it isn't using Leica camera lenses.

If you can show us an image from a leica APO that shows clear chromatic aberrations that can't be attributed to sensor bloom, I'd be interested to see them.

Marty

 

[Sonnar Brian]

An Apochromat is corrected for three wavelengths, usually within the three primary colors, three zero crossings where the chosen wavelengths are in agreement. That is the accepted optical definition. An Achromat is corrected for two wavelengths to have zero crossings. The deviation from the crossings varies greatly among lenses. A Super-Achromat is corrected for four crossings. I suspect most people use "color" and "wavelength" interchangeably. The optics are corrected for specific wavelengths to have zero-crossings, not ranges of colors to all agree perfectly.

The IR index of the type 1 Rigid Summicron, an Achromat, is within the F2 DOF marks. Many lenses, also Achromats, have the IR index at between F4 and F8. The deviation from the crossings is much higher than the Summicron.

Some manufacturers use the term loosely, a highly corrected Achromat with low deviation from the crossing being labeled "APO". It is not, the difference may not be noticeable, but if you run the calculations- it was designed as an Achromat and the rest is marketing.

 

[Freakscene]

I suspect most people use "color" and "wavelength" interchangeably. The optics are corrected for specific wavelengths to have zero-crossings, not ranges of colors to all agree perfectly.

The normal technical terminology is "spectral lines" and there is no standardised approach there for lenses either - thre three or four spectral lines with zero crossings of an apo or superchromat may not be the same spectral lines between manufacturers.

Where the IR index lies, is, I think, an interesting point - it is possible to have a very lens that is very well corrected for visible light that is not well corrected outside that range, and vice versa - the new Summicrons are better corrected for the visible range, but shift noticeably more in the IR range.

Certainly interesting if you're shooting IR.

Marty

 

[Sonnar Brian]

The optical design software that my engineers used (ASAP 11.0) asked for input in wavelength. The patent for the Pentax Ultra-Achromat uses wavelength in microns.

I'm used to seeing spectral lines with names for the transition that produces them. Perhaps different software packages for certain applications allow for specific emission lines to be input for zero-crossings, that would make sense in many applications.

 

[Freakscene]

The optical design software that my engineers used (ASAP 11.0) asked for input in wavelength. The patent for the Pentax Ultra-Achromat uses wavelength in microns.

I'm used to seeing spectral lines with names for the transition that produces them. Perhaps different software packages for certain applications allow for specific emission lines to be input for zero-crossings, that would make sense in many applications.

The software that we use (supplied by Zeiss) is set up so that we input those spectral lines - it seems that it can be heavily customised, but we only tend to use it for a single purpose. I haven't used or seen ASAP for a long time, but I remember now that it worked that way - and I can see a lot of situations where that could be more convenient too.

Marty

 

[1joel1]

I have an older 75mm Summilux that I used with the R-D1. However, I just sold that camera and am only a few days away from my "new" M8 arrival. The 75 was perfect for many uses on the Epson and I would imagine that it will still be the best lens on the M8. I use it as a normal lens most of the time as I "see" things more telephoto than wide.

Joel

 

[Tom Niblick]

Thanks for the info guys. I will probably wait until I upgrade to a. M9 or M10 a few years from now before going for a 90mm. I think 75mm is as long as I'll want to go on then M8 due to the crop factor.

I loved the 90 Elmarit M on my M8. The EFV is 120 which is perfect for tight head shots and a good general purpose telephoto. But I have to admit, it's pretty fine on the M9.