RF lens distortion v DSLR

Steve_F said:
Since Ive been using prime Leica I've really noticed any minor distortions in my Nikkors. It really stands out now.
Click to expand...

Same here - my M-Hexanons are excellent too.

BTW, I use the distortion/aberration/perspective correction software PTLens to correct all critical pictures I shoot using my Nikon. This software has a better distortion correction algorithm than Capture NX2!

What's more, this software has a very powerful perspective correction module that beats practically all older Photoshop versions, so it is even useful if you're using primes that are free of distortion.
 
Wow

Wow

Guys, guys, guys, this is fantastic information. I sincerely appreciate all of your efforts to explain this to me. I really love the look of the images coming from the 25 Biogon. The fact that the edges don't render people's heads or bodies oblong is really nice. Once again the forum has come through for me!

Have a great weekend.

Chris
 
I've seen plenty of posted examples of RF lenses, and seen plenty of distortion. Both the Nokton 35 1.4 and Color skopar 35 2.5 both have it, and so do some Leica lenses. Conversely, there are plenty of SLR lenses that have very little distortion. I also believe, as others have pointed out, that it is a choice made by lens manufacturers. Low distortion is expensive, especially with fast glass.
 
I have a question regarding software.

If software can correct barrel or pincushion distortion, does it add distortion to the rest of the image? I've never tried this, and that's why I ask.
 
What rest of the image? The software simply does it for the whole image. It's just most apparent toward the edges.
 
By the way, I used to think that there was a fundamental difference between retrofocus and non-retrofocus regarding distortion, but apparently the elimination of barrel distortion is just harder for a retrofocus design.

That there is a design decision involved can be seen when comparing published distortion specs for Leica vs. Zeiss M-Mount lenses. The Leica ones always tend to have more distortion (with some exceptions), so clearly their designers have optimized something else at the cost of distortion...
 
Eliminating barrel distortion is also more difficult for faster lenses. Example: the Canon FD 50 1.4 is sharper than the 1.8, but exhibits higher barrel distortion. The Leica Elmar-M 50 2.8 is renowned for its lack of distortion, but that's because it's a slow lens.
 
Firstly, a zoom lens has all sorts of optical compromises that don't really make it directly comparable to a prime lens. If comparing an SLR and RF lens of the same spec however, there are differences.

kossi008 said:
By the way, I used to think that there was a fundamental difference between retrofocus and non-retrofocus regarding distortion, but apparently the elimination of barrel distortion is just harder for a retrofocus design.
Click to expand...

It is, but note that the Zeiss 25 mm is a semi-retrofocus design, that is some of the front elements are negative. Look at the difference between the 25 mm and the 21 C lens.

kossi008 said:
That there is a design decision involved can be seen when comparing published distortion specs for Leica vs. Zeiss M-Mount lenses. The Leica ones always tend to have more distortion (with some exceptions), so clearly their designers have optimized something else at the cost of distortion...
Click to expand...

Flatness of field, primarily, but this influences edge sharpness if you photograph flat things.

There is a technically competent, though somewhat unclearly written explanation of the performance issues faced by retrofocus lenses here: http://www.camerarepair.com/Retrofocus-Design-Problems-A-Synopsis-T37.html

It's harder to correct the negative front elements than it is to maker a symmetrical design. The more negative, the harder it is. In this case, hard means "difficult to design, manufacture and sell profitably at a given price point". That's why the Zeiss has less distortion than an SLR lens. If you gave Zeiss $100 million to design and build you ONE lens they could make a 25 1.4 or 1.2 with less distortion, for an SLR or RF camera.

Marty
 
Last edited:
All I can say is that my Zeiss ZE 35mm distagon for canon DSLRs surely has a fair bit more distortion than the ZM biogon of the same focal length.
 
A easy way to think of it is that broadly, you need more glass for speed (including freedom from vignetting); more glass for reduced aberrations; and more glass for telephoto or reverse-telephoto (Retrofocus) construction. Try to get all of it perfect, and you end up with a big, expensive lens, and ever-diminishing returns. Most people run out of money or patience before the theoretical or even practical limits are reached.

Remember, too, that the only purpose of the extra glass(es) in a telephoto lens is to decrease the overall length of the lenses, and the main purpose of the extra glass(es) in a reverse-telephoto design is to leave room for the flipping mirror (though it can help control vignetting as well, which is one reason why even RF lenses often have some degree of reverse-telephto construction).

In other words, all this extra glass does nothing for distortion or resolution, and indeed, often necessitates still more glass(es) to clean up problems introduced by the negative elements behind the image forming group (telephoto) or in front of the image forming group (reverse telephoto or Retrofocus).

Cheers,

R.
 
Roger Hicks said:
Remember, too, that the only purpose of the extra glass(es) in a telephoto lens is to decrease the overall length of the lenses, and the main purpose of the extra glass(es) in a reverse-telephoto design is to leave room for the flipping mirror (though it can help control vignetting as well, which is one reason why even RF lenses often have some degree of reverse-telephto construction).
Click to expand...

These days, in addition to less optical and mechanical vignetting, you have the issue of telecentricity when using a digital sensor. Semi-retrofocus lenses will be standard for rangefinder wide angles until someone makes a sensor that has a wider cone of acceptance for the light striking it. Of course working with the lenses is cheaper and easier for now.

Marty
 
Freakscene said:
These days, in addition to less optical and mechanical vignetting, you have the issue of telecentricity when using a digital sensor. Semi-retrofocus lenses will be standard for rangefinder wide angles until someone makes a sensor that has a wider cone of acceptance for the light striking it. Of course working with the lenses is cheaper and easier for now.

Marty
Click to expand...

Dear Marty,

Indeed, which is why I said 'one reason why even RF lenses often have some degree of reverse-telephoto construction'. The trouble is that telecentrics are harder to correct for chromatic aberration, so that's yet another trade-off.

Cheers,

R.
 
Roger Hicks said:
Indeed, which is why I said 'one reason why even RF lenses often have some degree of reverse-telephoto construction'. The trouble is that telecentrics are harder to correct for chromatic aberration, so that's yet another trade-off.
Click to expand...

I didn't read it that way at the time, yes, I see what you mean. Adding aspherical element(s) to correct the spatial optical aberrations makes correcting that additional chromatic aberrations even harder too. It's all interlinked.

Marty
 
You must log in or register to reply here.
Back
Top Bottom