Aspherical lens elements — always good?

[folderholder]

Since I got my 7D in December I've been it's been fun to try out an assortment of lenses on it — some brand new, others as much as 30-40 years old. One common feature of all the recent lenses is their excellent resolution. Even a cheap lens like the Canon EF-S18-55mm IS is commendably sharp over most of its focal range. The workmanship, resolution and bokeh on my older Canon 20-35mm L is impressive. The Canon 28-135mm IS is a fine optic, too. Yet, after a good deal of comparative testing I've settled on an outdoors kit of old Takumar and Yashica ML lenses. Why is this?

First off, since most of what I use the camera for is video the highest levels of resolution are probably wasted on me (HD video is roughly 2 MP/frame so a lot of information is ultimately not being used) — and the older prime lenses hold their own in sharpness anyway. Second, all these older lenses — even a couple of real cheapo offbrands — that I tried out have a significantly greater sense of depth and three dimensionality than the newer models. Now, I'm not claiming this is a universal condition, but it was true for the specific lenses I tried.

This got me to wondering about a cause. One possibility is that all the new lenses I tried use aspherical lens elements to achieve their sharpness. The older lenses do not. Could it be that the aspherical lens elements come at the price of some 3-dimensionality even as they add sharpness and contrast to the image? Is this idea totally out of left field?

[indianadinos]

Hi,

I don't know how much an aspherical element can affect the 3D effect ...
I would be more prone to say that old lenses have a more pronounced 3D effect because they are faster (f/1.2 ... f/2.8 ) than most of the modern lenses (f/4 ... f/5.6), thus offering a reduced DOF to play with ...

What are the f/stops of the lenses you are using ?

[folderholder]

The 20-35mm L lens is a constant f/2.8 so I felt comfortable comparing it with 28mm f/2.8 and 35mm f/2.8 Yashica lenses.

[dnhkng]

I find that aspherical elements don't improve sharpness, but correct for spherical and other aberrations from high contrast scenes.

I had the canon 55mm F1.2, and the 55mm F1.2 aspherical lenses, and here's what a crop looks like:



The non-spherical lens on the left is definately sharper, but has lots of blooming from the bright white areas.

Personally, I don't see the "3d effect". All I see is a shallower depth of field, which you get from all fast lenses. Maybe it has something to do with a smoother bokeh, or maybe its the feelings lightness you get from an empty wallet after buying zeiss gear...

[folderholder]

I may be on the wrong track in attributing greater resolution to aspherical lens elements. Yet, I'm still trying to figure out why these old lenses produce such deeply spatial images compared to the more recent optical designs I looked at.

[Orio]

[indianadinos]

Hi,

[folderholder]

The Distagon has no aspherical elements, does it? Is it a coincidence that Distagons provide such an excellent sense of dimensionality? This is the quality that is lacking in the various current lenses I tried.

[Bruce]

what of the lenses "minimum focus distance" and relationship to 3D dimensionality? Orio's "Hollwood Distagon" has that floating "close focus"
element.

And how apropos Orio uses his Hollywood for the cinematographer

[folderholder]

None of the "older" lenses I tried had floating elements, yet they all produced a good sense of space. The Distagon may be at the the top of the heap but this is a quality that used to be, I think, more common. Yes, I agree that having a strong foreground element does much to give a composition a strong sense of depth. Some lenses accentuate this quality, others do not despite having high resolving power.

[poilu]

[folderholder]

Just curious: does anyone here see a different style of dimensionality in the photo that Orio posted that was taken with the 28mm Distagon than with this series of 35mm Distagon lens images? One lens uses aspherical glass while the other doesn't. I would love to see some side-by-side comparisons between these two wonderful lenses. Of course they are two different focal lengths and that may be the overriding factor in this comparison. Very impressive images Poilu, thanks.

[dnhkng]

Well, I had a look at quite a few "3d effect" pictures now, and I would have to say I think its derived from just two factors. The first of course the shallow DOF, which I mentioned earlier, and the second is the short lens length.

The larger angle-of-view with these wide angle lenses compresses distant objects, making them seem further away.

Designing and constructing fast wide angle lenses is very difficult, as the light pathways have to be 'bent' much more than a fast telephoto lenses, giving rise to many aberration problems. The tricks used to correct there problems are expensive, like using low-dispersion glass and aspherical lenses. So only a few companies ended up building such lenses that are good quality. Zeiss being of course one of them. But I don't think there is any other magic pixie dust needed to give the "3D effect".

[no-X]

I think there are more factors... microcontrast (coating) and some aspects of optical design. I remember an interview with Sigma employee, who mentioned, that for wide-angle lenses it is not only important to calculate the optical formula for sharpness, but it's favourable not to bend the rays too much.

Zeiss Distagons were one of the first retrofocus lenses, which utilized this approach - that could be the key reason of their 3-dimensionality.

[Esox lucius]

FWIK, aspherical elements correct aberrations but do not increase sharpness, in fact decrease detail: adding more glass between film/sensor and subject is bound to remove some detail.

Lots of old glass with great detail but plenty of aberrations. In the days dominated by black and white film no-one pretty much cared about chromatic aberration.

[no-X]

That is quite logical statement, but there are too many exceptions...

All old Meyer B/W era alu lenses have less CA, than the newer serie for colour photography:

Primagon 35/4.5, Helioplan 40/4.5, Trioplan 50/2.9, Primoplan 58/1.9, Primoplan 75/1.9 + 80/1.9, Trioplan 100/2.8... All of them are exceptional in terms of CA. On the other hand, the 29/2.8, 50/1.8 and 135/2.8 are average in terms of CA.

[spleenone]

[Orio]

[dnhkng]

If you have shots from two lenses of the same length and used at the same aperture comparing the "3D effect" side by side, I would really like to see them.

[Orio]

[my_photography]

[folderholder]

First: Best wishes to Orio's mother.

Here are two images that I think exhibit some of this difference. Both are shot at f/8. One is taken with a 28mm Yashica ML and the other is from a 20-35mm f/2.8 L Canon zoom lens set at 28mm.
[img][url=http://parkfilms.com/IMG_1032_1600.jpg[/img][/url]
[img][url=http://parkfilms.com/IMG_1039_1600.jpg[/img][/url]
I won't say which is which yet. On my own screen a couple of people who have seen these two samples said the difference (besides color cast) is visible. Does anyone else think this is true, too?

OK, the uploaded images aren't showing up. What did I do wrong? Thanks for any assistance.

[visualopsins]

From my undertanding aspheres do the job of multiple elements; fewer elements = greater efficiency, more light gets transferred.

Please refer to pp 14-15 of "Optics Made Easy", "Aspheres" and "Aspherized Achromats" http://www.edmundoptics.com/techSupport/resource_center/downloads/optics-made-easy-guide.pdf

3 varieties of aspheres and their benefits.

Nothing is said about "3D" capability.

[Orio]

[Bruce]