Iscorama 36 on & off comparison

Here is a comparison that some of you may find interesting. This is a comparison of what a bunch of tins sitting on my kitchen table look like when shot with a Nikon 50mm 1.8 versus a Nikon 50mm 1.8 with an Iscorama 35 anamorphic adapter screwed on to it.

I carried this out because I wanted to see what exactly the Iscorama was doing to the image on the mundane levels of softening, light loss, contrast, and barrel distortion - as opposed to the obvious and sensationalised transformations that are well documented: anamorphic stretch, flaring, oval bokeh, and light rendering.

All shots were at f1.8, shutter 1/50, ISO 640, and show the lenses focused at 2ft, 3ft, 4ft and 6ft. In order to achieve this focal distance with the Isco (which focuses from around 6ft to infinity) I've used a +2 and +1 diopter (Kenko brand) and a +0.5 diopter (Tokina brand). The final shot is the Isco with no diopter.

I brought all clips into a 1920x1080 project but - importantly for pixel peepers - conformed the Isco clips to the correct aspect ratio, thus cropping off the left and right portions of the image and lowering resolution.

The original full res stills are here: http://rapidshare.com/files/325080708/I ... _1080p.zip

Here are the downrezzed comparisons:










Caveats:
- My reflection is in the frontmost tin and shifts between shots
- The TV was on so maybe there are some illumination differences
- There is a little bit of voodoo involved in focusing the Isco. You're supposed to set the lens to infinity and focus the Isco. But with my fotodiox adapter I believe my Nikon focuses beyond infinity, and so there is a grey area where both lenses can be focused to a degree to give varying results.
- The Isco's alignment is slightly off

My conclusions:
- Sharpness is good as far as I'm concerned, even with diopters
- Light loss is minimal
- There is some pronounced barrel distortion with the Isco
- It's very difficult to get correct alignment, what looks right on the LCD can be still significantly off (and the image becomes skewed)

But what is a revelation for me, is that the squeeze factor is not 1.5x......... Looking at the comparisons in Premiere, I tried a pixel aspect ratio of 1.422 (PAL Widescreen) and it was much more accurate, at least in the center of the frame. That is what the above images have been conformed to. I'll have to do a little more testing with some square objects or circles.

This yields an aspect ratio of 2.53:1 as opposed to the 2.66:1 I had previously been using. In a way this is good news as it is less of a crop to go to 2.39:1 or 2.35:1. In fact if you conform the footage to a pixel aspect ratio of 1.35:1, this will yield a 2.4:1 image which is stretching the image vertically a bit but no-one would ever notice, probably. This corresponds to a very convenient 1920x800 picture size for HD delivery.

Thanks for doing these tests - very useful!

OK here are some more. This time on a Tair 11a 135mm 2.8 (nice lens) at ISO 1250:

(Full res are here: http://rapidshare.com/files/325125309/I ... onTair.zip)








The Isco gives a slightly wider field of view and a noticeable light loss. Most significantly there is greater depth of field with the Isco on. No harm too, at this focal length on a full frame body.

I always thought anamorphic adapters would give really different results (quiet distorted), but these images look quiet similar

The anamorphic images are cropped from 23:9 to 16:9.

Large parts of the left and right sides are missing

The anamorphic images are cropped from 23:9 to 16:9.

Large parts of the left and right sides are missing

ohh ok, I see. what's the point in doing that though if the result is nearly the same? or did you just want to show how similar it is with the adapter + cropped vs. original without the adapter?

Yes, basically I wanted to see what the difference was once the bells and whistles of aspect ratio and flares had been left out.

Even with this prosaic test material I do see features of the Isco that are desirable. I prefer the how rendering of the bokeh is altered by the Isco in the Tair 11a samples.

Thanks for doing these tests, very interesting. I wasn't expecting to get a wider depth of field with the isco! You can see the contrast loss, but not a huge deal.

I got my russian dipters yesterday. They are marked up as 1m and .80m. Not sure how that compares to +.5 or +1 etc. They were made for FOTON anamorphics.

Jim,

Great test & great lens. Thanks for this it really cleared some things up. I really do love this lens.
Here is a recent test I shot on my 7D.

http://vimeo.com/8420199

Cheers,
Hector
OSCARwao

Thanks for doing these tests, very interesting. I wasn't expecting to get a wider depth of field with the isco! You can see the contrast loss, but not a huge deal.

I got my russian dipters yesterday. They are marked up as 1m and .80m. Not sure how that compares to +.5 or +1 etc. They were made for FOTON anamorphics.

The 1m would probably be a +1 and the 0.8m would probably be a +1.5