Group: Moderator
Post: 2,040 (563 liked)
Join date: April 2014
Status: Long, long time Contax and Yashica user; glad to be here and hope to contribute.
|
|
on Aug 11, 2019 10:28:05 GMT
Posted: Aug 11, 2019 10:28:05 GMT
A couple of months ago I took this piccie of these two lenses, made in 1959, which boast the same optical formula as far as I know; they are pictured with their apertures set to f5.6.
9 vs 6 blades
The one on the left in the photo has 9 blades whereas the other has the more common 6. One would expect the bokeh of the 9-bladed 55mm to be softer than its twin but what I had not considered until taking this photo was the amount of light that would be let through; it seems that using the same external meter to determine the correct shutter speed at f5.6 would result in a slight over-exposure in one or under-exposure in the other. I hope to check this out shortly but it does look like these two lenses, using the same optics and set to the same focus and aperture settings, will produce images where it will not only be the oof elements that are slightly different. |
|
Group: Member
Post: 132 (14 liked)
Join date: September 2017
Status:
|
|
on Aug 11, 2019 17:54:51 GMT
Last Edit: Aug 11, 2019 17:56:25 GMT by matthiasausk
You "only" have to compute the area of a Reuleaux-polygon with 5 and 9 edges. (to be honest: I did not find the formula in a quick search)
The difference should be no more than about 1/10 step.
On the other hand, the engineers should have taken this into account, so that the diameter of the Reuleaux-polygons should be slightly different at same aperture.
|
|
Group: Administrator
Post: 1,371 (303 liked)
Join date: February 2017
Status: Failed treatment for L.B.A. and G.A.S,
|
|
on Aug 11, 2019 18:55:44 GMT
Last Edit: Aug 11, 2019 18:58:01 GMT by lumiworx
I would probably make a lousy optical engineer, but the part of my brain that works pretty well with geometry leads me to think that these 2 shapes would likely be close in terms of their opening's area in mm 2 on a 2D plane, as matthiasausk suggests - BUT - they might not be, depending on the actual measurements. If you go to the extreme in the opposite direction - from a 6 blade iris down to 3 - you'd end up with a Carl Zeiss 35mm f/1.4 HFT from Rollei. It was made in both 6 and 3 blade versions. As for the 3-blade Bokeh, here's a YouTube clip to illustrate. A series of area calculators can help with measurements... www.omnicalculator.com/math/area |
|
|
Deleted
Group: Member
Post: 0 ( liked)
Join date: January 1970
Status:
|
|
on Aug 12, 2019 14:37:33 GMT
Posted: Aug 12, 2019 14:37:33 GMT
I think all lenses I've ever worked on have some way of calibrating the opening of the aperture and I guess these would be the same. So the opening size would be adjusted at manufacture to give the correct exposure in each case. But the difference, if there is any, is going to be too small to be of concern and would probably fall within the tolerance allowed anyway. Manufacturing tolerances of camera gear is surprisingly large sometimes. Though we should remember that most of it is produced as consumer items, not scientific equipment, so large tolerances maybe shouldn't be surprising.
|
|
Group: Moderator
Post: 2,040 (563 liked)
Join date: April 2014
Status: Long, long time Contax and Yashica user; glad to be here and hope to contribute.
|
|
on Aug 12, 2019 15:31:52 GMT
Posted: Aug 12, 2019 15:31:52 GMT
Well, I decided to use a Minolta Flashmeter IV with its Booster II to measure illumination at the film plane on the Pentamatic.
After doing the maths and rechecking several times, I was surprised to learn that the difference between the two lenses set at f5.6 was almost exactly half a stop. It should be noted though that the film-plane attachment on the Booster II is in effect a spot-meter so I can't attest to evenness of illumination right across the plane.
At some stage, I'll pop a film in the camera and alternate the lenses at different f-stops with each shot to see just how noticeable this is; I doubt it would create much of an issue with colour negatives but based on previous experience, it might be a little more evident on transparencies.
One lives and learns... |
|
|
Deleted
Group: Member
Post: 0 ( liked)
Join date: January 1970
Status:
|
|
on Aug 12, 2019 16:01:17 GMT
Posted: Aug 12, 2019 16:01:17 GMT
I'm guessing the nine bladed one had the more exposure. Just by looking at your picture the size of the aperture on the nine blade one looks bigger. But it doesn't surprise me, especially with an older lens. Again, they are probably within the tolerance allowed at the time. Many spec's of the time, and later, allowed errors that would give half a stop or more exposure error. But in this case, it could be one of the lenses has been dismantled at some point and the aperture adjustment disturbed. You might be interested in something I wrote about metering accuracy. It's specifically about the 139 but still shows the sort of tolerances that were used. It's at www.contax139.co.uk/exposure-accuracyYou might also be interested in a blog post of mine that shows the Zeiss 50/1.7 and 18/4 lenses under expose by half a stop. This is for totally different reasons but it again shows what manufacturers considered acceptable. It's at contax139.blogspot.com/2019/01/some-zeisscontax-lenses-under-expose-by.html |
|
Group: Member
Post: 132 (14 liked)
Join date: September 2017
Status:
|
|
on Aug 12, 2019 19:16:34 GMT
Last Edit: Aug 12, 2019 19:22:20 GMT by matthiasausk
I'm guessing the nine bladed one had the more exposure. Assuming identical diameter, from a mathematical point of view this should be correct, because the area of the Reuleaux-Poligons is increasing with the number of edges.The area of the circle with the same diameter is the upper bound. biggles3 : The ratio of the area of circle : reuleaux-triangle of identical diameter is about 1.113, so the ratio of 9-angle to 6-angle must be less than that, because 9- and 6-angle are between circle and triangle. This is far less than an f-stop. There should be another reason for the result. |
|
|
Deleted
Group: Member
Post: 0 ( liked)
Join date: January 1970
Status:
|
|
on Aug 12, 2019 19:30:52 GMT
Posted: Aug 12, 2019 19:30:52 GMT
I'm guessing the nine bladed one had the more exposure. Assuming identical diameter, from a mathematical point of view this should be correct, because the area of the Reuleaux-Poligons is increasing with the number of edges.The area of the circle with the same diameter is the upper bound. I'm just judging them visually. It looks like one or the other is out of adjustment. |
|
Group: Moderator
Post: 2,040 (563 liked)
Join date: April 2014
Status: Long, long time Contax and Yashica user; glad to be here and hope to contribute.
|
|
on Aug 12, 2019 19:38:11 GMT
Posted: Aug 12, 2019 19:38:11 GMT
I'm guessing the nine bladed one had the more exposure. Just by looking at your picture the size of the aperture on the nine blade one looks bigger. But it doesn't surprise me, especially with an older lens. Again, they are probably within the tolerance allowed at the time. Many spec's of the time, and later, allowed errors that would give half a stop or more exposure error. But in this case, it could be one of the lenses has been dismantled at some point and the aperture adjustment disturbed. You might be interested in something I wrote about metering accuracy. It's specifically about the 139 but still shows the sort of tolerances that were used. It's at www.contax139.co.uk/exposure-accuracyYou might also be interested in a blog post of mine that shows the Zeiss 50/1.7 and 18/4 lenses under expose by half a stop. This is for totally different reasons but it again shows what manufacturers considered acceptable. It's at contax139.blogspot.com/2019/01/some-zeisscontax-lenses-under-expose-by.htmlAha! I should have known which Peter you are...
A great website for any Contax user, not just for 139s; I found that exposition on the Zeiss under-exposure most informative.
|
|
Group: Moderator
Post: 2,040 (563 liked)
Join date: April 2014
Status: Long, long time Contax and Yashica user; glad to be here and hope to contribute.
|
|
on Aug 12, 2019 19:44:22 GMT
Posted: Aug 12, 2019 19:44:22 GMT
Hi guys,
It would be great to carry out a more comprehensive test using a larger cohort of lenses but sadly I only have one of each, along with their respective Pentamatics. Finding another 9-bladed Yashinon may not be easy though there are quite a few of the standard version available. If I can find some at a sensible price here in the UK, I'll try to grab a couple and extend the comparison...
|
|
