Makro-Symmar HM 120mm

[karlmera]

Karlmera, does the distance between lens groups increase or decrease for digital sensors? If it increases, we could do that modification ourselves by slightly unscrewing the front/rear components of the Makro-Symmar.

Increase. The front group 1-2mm.

[Lou Jost]

Thanks karlmera. Was this modification done to account for the thickness of the sensor filter pack? If so, I wonder what thickness of filter pack they are designed for.

[karlmera]

Was this modification done to account for the thickness of the sensor filter pack? I

No, its a only a correction of the beam path.

[RobertOToole]

Would be interesting to see how actual lens measured data compares to OEM's published data for various OEMs and lenses. Maybe someone here is/was a recent lens engineer designer that can provide some "inside" information on what actually goes on with the published lens data vs. measured results.

Best,

Hi Mike,

One of the manufacturers, I believe it was Rodenstock, replied to someone online that they expect a lens to perform within 10% of the published MTFs, which are an ideal example.

But I think production sample MTFs are going to be far off from the real world MTF since you have to deal with the camera's sensor glass filter/cover stack. Also I have read that MTF results can easily vary 10% between test runs. The Raw convertor used can make a huge difference in MTF results, the results can vary 10-50% with different RAW convertors so this would cause even more variation in the results.

This would be a good question for Miljenko.

[RobertOToole]

Thanks karlmera. Was this modification done to account for the thickness of the sensor filter pack? If so, I wonder what thickness of filter pack they are designed for.

There are some old Lensrentals.com blog posts on this subject of sensor cover glass. One mentions that Nikon bodies having different glass thickness so I this must not make much difference to Nikon but Sony on the other hand supplied LR a custom filter stack from a A7RII to compensate for the lack of cover glass on their test bench to improve their MTF performance. This means Sony optimizes their lens designs for certain camera bodies.

According to SK Optics in the US, the APO-Digitar N and M were designed for digital backs, so that means PhaseOne.

Robert

[karlmera]

I have the information directly from Schneider that it is a correction of the beam path for a smaller field of view.

[RobertOToole]

I have the information directly from Schneider that it is a correction of the beam path for a smaller field of view.

Yes, true, but you don't need any classified top secret info, it's in the APO-Digitar brochure:

[enricosavazzi]

Thanks karlmera. Was this modification done to account for the thickness of the sensor filter pack? If so, I wonder what thickness of filter pack they are designed for.

The thickness of the filter pack is mainly relevant with very fast lenses (f/1.4 or faster). It is not a major concern with an f/5.6 lens, even one designed for high resolution.

It is more likely that the different spacing between groups is an optimization for a given magnification range. This allows the lens maker to use a single basic lens design, change a single parameter and use the results for a series of lenses that differ from each other in optimal magnification (and possibly just a little in focal length).

The "famous" Jenoptik 60 mm Apo f/4 was reportedly designed for an average thickness of the filter pack of digital cameras at the time of design, noting that the actual thickness may differ by up to a couple of mm in different cameras (with smaller sensors using thicker filter stacks, in part because a thicker stack puts the front surface of the stack farther from the sensor and slightly lessens the negative effects of minute dust particles on this surface). However, CoastalOpt (the original designer/maker) also used to say that the lens would perform optimally within the whole range of filter pack thickness used at the time by large camera manufacturers. This lens, incidentally, uses a fixed rear subassembly (2 elements) and a moving front subassembly (8 elements) to correct aberrations at different magnifications, using the same principle of varying the distance between two subassemblies to optimize at different magnifications.

Some macro lenses use three (or possibly more) moving subassemblies. The Olympus 60 mm f/2.8 macro uses three internal moving subassemblies (3 + 3 + 1 elements, I don't know whether all three subassemblies move on different cams) for focusing and compensating, and two fixed ones (4 + 2 elements) at front and rear of the barrel. The Olympus 30 mm f/3.5 macro uses two moving (3 + 1 elements) and two fixed (2 + 1 elements) subassemblies.

[karlmera]

Very good information. But also interesting: the "old" macro-symmar HM 120 for a bigger film format is also very good, hardly visible the difference to the digitar on the D810.

[RobertOToole]

Thanks for the input Enrico.

The "famous" Jenoptik 60 mm Apo f/4 was reportedly designed for an average thickness of the filter pack of digital cameras at the time of design, noting that the actual thickness may differ by up to a couple of mm in different cameras (with smaller sensors using thicker filter stacks,.....

Now that you brought up the 60/4 lens (the designer of this lens used to comment here on the forum from time to time BTW) has helped LensRentals with sensor stack info.

Check out the full post here on the LensRentals blog:

https://www.lensrentals.com/blog/2014/0 ... it-matter/

[RobertOToole]

Another excellent post with more sensor stack info:

https://www.lensrentals.com/blog/2014/0 ... ed-lenses/

And another:

https://www.lensrentals.com/blog/2014/0 ... it-matter/

[rjlittlefield]

Robert, thanks for the references.

To make quick sense of the overall shape of the graphs...

When the ray cone is perpendicular to the added glass, having the wrong thickness introduces spherical aberration but nothing else. Spherical aberration diminishes as f-number raised to the 4th power, so stopping down is a powerful way of getting rid of it. This is why in the first graph, HERE, the effect of a 4mm stack is still huge at f/1.4 but essentially gone at f/2.8. That 2X increase in f-number turns into a 16X decrease in SA.

But when the ray cone is not perpendicular to the added glass, the added glass adds other aberrations that diminish more slowly. This is why in the second graph, HERE, contrast drops off from center to edge, falling more with steeper angle due to both off-axis distance and exit pupil distance.

The same effects are seen when you're shooting into an aquarium. That's why it's recommended to use a long lens, perpendicular to the glass. See HERE for some discussion and illustration.

--Rik

[elimoss]

So which lens do you think would perform better in a lens stack?

Makro-symmar 1x 120, APO-Digitar Macro 120 1x or APO-Digitar normal 120?

Thats the question

Not sure, but Aperture Mount NK-0 is the device you want in order to use a normal Digitar. This is Schneider's Copal 0 compatible aperture. I don't think they make a Componon style or Makro Iris style barrel version

[RobertOToole]

So which lens do you think would perform better in a lens stack?

Makro-symmar 1x 120, APO-Digitar Macro 120 1x or APO-Digitar normal 120?

Thats the question

Not sure, but Aperture Mount NK-0 is the device you want in order to use a normal Digitar. This is Schneider's Copal 0 compatible aperture. I don't think they make a Componon style or Makro Iris style barrel version

Thanks, yes, that does look like the correct mount.

Best,

Robert

[nords]

I have the same Makro Symmar 120mm f/5.6. At f/5.6, the diaphragm blades are not completely closed (probably more correctly say: not fully "open"). Does this have to be so or the mechanism has shifted?