Swirly Bokeh: Modifying Lenses Non-Destructively to Get it

[goamules]

I'm not sure who this is I found a few weeks ago, but his Vivitar is swirling quite a bit:
http://www.flickr.com/photos/ghosstrider/3581026115/

He mentions "optical vignetting" as the cause, with his definition; "trying to make a lens cover a bigger piece of film (or digital sensor) than it was meant to."

That doesn't make much sense to me, many lenses will cover more than their recommended size, but we don't do it because of the loss of sharpness. Anyway, it's another example of a design doing the spins.

[Dan Fromm]

We dealing with people who don't have the concept "off-axis aberrations." They think that coma is comma misspelled ... and they work, possibly unwittingly, to use the Internet to spread errors.

[Archphoto]

Dan, it belongs to photography in general: experimenting with optics that are not ment for their new use......
And sometimes something usefull and interesting comes by.
LF uses at time projection lenses, fitted to shutters that were never intended for it, and we love the results.
Being curious is a part of photography, when you loose that you are just making a living or even not that.
The great thing with LF (and digital) is that you can experiment all you want.
I did some testing with the 58XL, the 72XL and the 115 Grandagon on 8x10 just to see what I could with them and found out that that the 72XL and 115 are usable for panorama's with the propper cropping in absence of having a 6x17 camera.

Peter

[Dan Fromm]

Peter, I have no problems with experiments in general or with the experiment that launched this thread, although I think what the OP sees is more in his mind's eye than in the image he posted. I'm glad that you experiment, am surprised that it took experimentation to discover that cropping to a smaller format will give the same results as shooting on the smaller format. But then, I shoot a 38/4.5 Biogon on 2x3, which it absolutely positively doesn't illuminate, let alone cover.

I have major problems with the OP's explanations of what causes the effect he wants. Also with the explanation that goamules reported and carefully didn't endorse.

Cheers,

Dan

[Mark Sawyer]

I have major problems with the OP's explanations of what causes the effect he wants. Also with the explanation that goamules reported and carefully didn't endorse.

Cheers,

Dan

Cheers, Dan! You have a good bit of knowledge in optics, and as this has been a subject of debate and speculation for some time, could you offer any insights into what specific factor or combination of factors causes the infamous "swirlies" to occur?

[Dan Fromm]

Mark, I can't offer much beyond "lens abuse," using a lens with bad (but which ones?) off-axis aberrations on a format larger than the format it covers.

That said, I have some test shots that show what I diagnose as astigmatism (manifested as different resolution with horizontal than with vertical lines) and bad distortion (e.g., vertical lines strongly slanted) at the edges of the field. Toss in a lot of coma, which gets worse the farther we go off-axis, and the combination might do it.

There's an example of, IIRC, trees floating around that's said to show the swirlies very strongly. Every time I look at it I feel queasy. And every time I look at it the nasty little voice in the back of my mind murmurs "long exposure. wind. the leaves moved."

Cheers,

Dan

[Mark Sawyer]

Darn, no clear answer there, Dan! So the theorizing goes on...

I've thought of astigmatism occurring in a circular rather than linear pattern, but astigmatism seems to happen on a fairly small scale, too small to be the cause of the big swirlies we've seen.

Mechanical vignetting (the "cat's-eye" shaped off-axis aperture) seems the most likely culprit, but the Petzval swirls so dramatically compared to the minimal-to-nonexistant swirls from Rapid Rectilinears and Aplanats which would have near identical off-axis apertures.

And they really don't look like any coma I've ever seen...

No answers here, either...

[dh003i]

Yea, everyone jumps all over me for "wishful thinking", but no-one seems to have a clear authoritative explanation and reference.

I think that the article I linked to is at least partially explanatory.

I also posted negative controls with my 58/1.2 and 50/1.4 without the rear mask. I think they exhibit some of the same effect without it too. I think it looks like the milder swirl from the Petzvals in the center of the shot.

If I cropped out those petzvals to the middle 1/3rd to 2/3rds of the shot and said there was "mild swirl", I think people was say I was hallucinating too.

[dh003i]

A bit like a reversed Imagon ????

Peter

That's another good idea. The other link I had in the 1st post was to "heart-shaped" bokeh patterns; this was caused by simply making a heart-shaped cutout on a piece of paper, and then making that a circle a little larger than the lens, and twist-tying it to the end of the lens.

I suppose a disc with a central hole and many smaller surrounding holes, like the imagon could also be made. Although I seem to remember reading someplace that the position of the diaphram matters, and that for most lenses, optimal position is where it is actually placed ("inside" the lens), and it will be sub-optimal in front of the lens.

Another interesting idea would be center spot filters, with a transparent center, and opaque edges; graduated in between. This would create silkier bokeh in highlight points, like with the Sony 135mm f/2.8 [T4.5] STF.

[Mark Sawyer]

That's another good idea. The other link I had in the 1st post was to "heart-shaped" bokeh patterns; this was caused by simply making a heart-shaped cutout on a piece of paper, and then making that a circle a little larger than the lens, and twist-tying it to the end of the lens.

The heart-shaped patterns are lens flare reflections of the aperture shape, and occur equally throughout the frame. The "swirlies" occur at the outer edges of the image circle. Two different phenomena, thugh the underlying causes may be related (or not...)

I suppose a disc with a central hole and many smaller surrounding holes, like the imagon could also be made. Although I seem to remember reading someplace that the position of the diaphram matters, and that for most lenses, optimal position is where it is actually placed ("inside" the lens), and it will be sub-optimal in front of the lens.

With a single lens grouping, the aperture's optimal position is in front of the lens. That's where it is on Imagons, Kodak Portrait lenses, landscape lenses, etc. A misplaced aperture will adversely affect field curvature and coma.

Just more unhelpful information!