Light Loss with an adaptor ring

[Patrick Neary]

Hello all-

 

there has been an ongoing discussion here about this PL adaptor ring:

 

http://www.xl1solutions.com/ADAPTERS.htm

 

the big mystery seems to be why (as they claim) there is a 1 to 2 stop loss of light-when obviously this is just a bare naked, hollow, metal ring.

[Patrick Neary]

also I should note for anyone who didn't catch it in the other thread, I did send a note to the company asking specifically what causes the light loss, the reply was:

 

"Hi open up the film stop...1 or 2 stops. Thanks"

[Jason McKelvey]

What about this... I'll put it in an annalogy:

If you put a light meter in a dark room 3" away from a wall and cut a hole in the wall 2" in diameter, take a reading of the light coming from the room next door, then cut a hole 3" in diameter, would the light meter read higher foot candles? If the ccd is only seeing a fraction of the 35mm image, the ccd isn't seeing as much as a film frame which sees the whole 35mm lens image. Some of the light spills out past the CCDs in all directions. There's your lost 1 or 2 stops. Or, I'm an idiot.

 

j

[Patrick Neary]

Nope. (thanks for the try though!)

 

Light transmission from a lens is not affected in any way by the size of the surface where it's image is landing.

 

I think this whole thing might be just an issue of poor ad copy, rather than any interesting video/optical phenomenon.

Edited March 5, 2005 by PatrickNeary

[Laurent Andrieux]

Patrick is right. If you consider the illumination is the same in every point of the image that is produced by the lens, it is the same whatever the size of the used image by the sensor.

 

This is why you can put a 35 mm lens on a 16mm camera (assuming you have the proper mount) the f stop of the lens will be as good for 16 mm.

 

Your analogy is not much usable, because it all depends on the fact of source being punctual or wide... and here the source is a lens that is closer to a puntual source.

 

Also it depends on the relative distance of the hole to the source and to the wall light is projected on. You considered 3" in the camera, it's like the 0", since the considered point is in the same plane as the "hole".

 

If you have a punctual source and say, flags at some distance of it, the illumination of projected light will remain the same even if you limit the projected surface with the flags. When you have a wide source, and if the flag is close to the projecteur, the flag will cut the source's surface and the lumination will decrease.

 

The size of the frame in a camera is like a flag that would be aside of the considered illuminated point, it would have no consequence on it.

[Laurent Andrieux]

Sorry I made a mistake - too late to edit the post - The lens shouldn't be considered as a puctual light. The rest is correct.

[Michael Nash]

I think this whole thing might be just an issue of poor ad copy, rather than any interesting video/optical phenomenon.

<{POST_SNAPBACK}>

 

No, it really is an optical phenomenon that you have to deal with in photography/cinematography. I'm sorry I can't explain how the optics work, but any time you move the lens physically farther away from the imager (chip or film) you lose light. Bellows lenses and macro lenses often include scales marked on them for the exposure compensation per distance.

 

"Range extenders" that you mount between a lens and the camera (which this essentially is) follow the formula: Aperture is reduced by the square of the magnification. For example a 2x extender reduces the light by 2 stops (2 squared equals 4; four times reduction of light equals two stops).

 

Again, sorry I can't explain why this is, but rest assured it IS an actual optical phenomenon.

[Laurent Andrieux]

Yes Michael you are right, it is a Lambert low.

 

The illumination is by definition, the luminous flux that lights a surface devided by the surface it lights.

 

That gives actually, an other explanation why the image obtained by a lens has the same illumination, whatever the frame size is : the surface has decreased, but the flux on the surface has decreased also (a part of it is outside the frame, as Jason said).

 

You can consider also that the illumination is the source intensity divided by the square of the distance beetween source and lighten object (ie here, the sensitive surface) since these parameters don't change, the illumination doesn't change.

 

The fstop reading on the lens is ment for an image that is done at the given distance of its typical flange focus distance.

 

When you add an extender, you increase the distance beetween the lens and the gate or pick-up device.

 

This time, the flux decreases because the distance changes. As you said, the light loss is proportional to the square of the distance.

 

one can also consider that since the surface didn't change but since the distance changed, this time, the flux has changed (light is less concentrated as we go far from the source) and therefore, the illumination has changed.

 

This hypothesis for the phenomenon Patrick and I are discussing about is exactly the one I put in the original thread Landon created (it's there : http://www.cinematography.com/forum2004/in...showtopic=5298)

 

This explanation couldn't satisfy us because we are not shure at all the focal flange of the system is not the typical one of the PL mount.

 

Still searching...

Edited March 5, 2005 by laurent.a

[Patrick Neary]

No, it really is an optical phenomenon that you have to deal with in photography/cinematography. I'm sorry I can't explain how the optics work, but any time you move the lens physically farther away from the imager (chip or film) you lose light. Bellows lenses and macro lenses often include scales marked on them for the exposure compensation per distance.

 

"Range extenders" that you mount between a lens and the camera (which this essentially is) follow the formula: Aperture is reduced by the square of the magnification. For example a 2x extender reduces the light by 2 stops (2 squared equals 4; four times reduction of light equals two stops).

 

Again, sorry I can't explain why this is, but rest assured it IS an actual optical phenomenon.

<{POST_SNAPBACK}>

 

Yes, this is absolutely true when you are talking about physically pulling the lens farther away from it's intended target (the film plane or CCD imager), but the by-product of that is that you are turning the lens into an extreme macro, incapable of focusing more than a few inches from the front of the lens. And it doesn't take much of an extension to do this. If you have an SLR at home, loosen the lens mount, set the lens at infinity and then slowly pull the lens away from the camera, you'll see that in only a few millimeters, you can't focus any farther out than a few inches. And you do lose light as soon as you begin pulling the lens also, a standard close-focus chart (as well as hands-on practice) shows this.

 

The goofy thing about this particular adaptor is that it is not an extension tube, apparently it's intended for normal-distance filming. Their frame-grab examples clearly show "normal" shots and not macros of insects.

 

And unlike a tele-extender, which is made up of a small grouping of optics, this thing has no optics.

 

I honestly don't know why this thing has me so worked up. I just want some video/optical engineer to chime in and say "oh yes, it's the Narshompsky Theorem" and be done with it...

 

The other solution is for Laurent to go buy one and try it out and report back! :)

Edited March 6, 2005 by PatrickNeary

[Laurent Andrieux]

I think better Landon should do that ! He's the one who was asking at the first place ! :D

 

I really don't need this kind of stuff, really. But if you read my last post in the original thread, I was suggesting you Patrick to write again to the Savant Cosinus who made you this wonderfull reply :lol:

[Laurent Andrieux]

Funny, the thing was sold on ebay for 375 $ and they sell it 499 $ !

 

I actually sent a e-mail to the person who bought it, he or she might tell us more if is coming around :)

[Patrick Neary]

http://www.cinematography.com/forum2004/in...?showtopic=3095

 

scroll down to the bottom of this (linked) thread...a few more comments about the same company's adaptors, or more to the point, their weird claims about their adaptors...

 

why does this feel like one of those "the hunt for bigfoot" shows?

[Jason McKelvey]

I would like to get some advice from you guys on doing some controlled tests. The guys at XL1solutions have graciously agreed to loan me a beta product... a PL to B4 mount with a back focus adjustment. By the way, they are really nice guys. Anyway, I will be going to a local rental house who has agreed to let me shoot some tests with Arri PL lenses free of charge. I'll be using an SDX900 at 50mbs mode 16x9, Progressive. I want to set up some test shots that include a subject of some kind with forground and background elements. What would be the best way to compare the PL lenses to my Canon 21x7.8 HD lens. Should I use primes lenses or a Arri compact zoom? My thinking was to set up a shot with a prime lens, lock off the tripod, shoot some tests, leave the camera in the same position and using the zoom on the HD lens, frame up the same shot and repeat. I'll post frame grabs of my test for you guys to look at.

 

Thanks,

Edited March 6, 2005 by JasonMcKelvey

[Patrick Neary]

good, an end to this ridiculous thread.

 

You might want to also just shoot a resolution chart, and compare both lenses at different stops, especially wide open, and at different distances. Maybe take your PL lens and see if the measured distances match up to the actual lens markings. I'd also drag it out to the parking lot and see if you can get an infinity focus with the adaptor.

 

Do post the results!

[Laurent Andrieux]

Yes, shooting a resolution chart is a good idea, but please mind that the ratio is always the same, ie have frame marks on your board/chart and precisly respect them.

 

Also it would be a good idea to compare the f stop values, with a waveform monitor. Can you do that ?

 

When you set the back focus, check that the focus mark is still usable and of course, shoot an outside infinite shot, wide open please.

[Jason McKelvey]

Yes, shooting a resolution chart is a good idea, but please mind that the ratio is always the same, ie have frame marks on your board/chart and precisly respect them.

 

Also it would be a good idea to compare the f stop values, with a waveform monitor. Can you do that ?

 

When you set the back focus, check that the focus mark is still usable and of course, shoot an outside infinite shot, wide open please.

<{POST_SNAPBACK}>

 

If I use a 50mm prime, should I set the HD lens zoom the the middle somewhere and move the camera in and out to match frame, or what?

 

I have a Hamlet portable waveform that I can use to set iris on charts.

 

Forgive my lack of film knowledge... the distance markers on the prime lens: is it the distance from the back of the lens or film plane? I'll get the wide open shot by using the shutter speed to control exposure. I'll get the infinity shot too.

[Patrick Neary]

I'd match the field of view of your zoom to the prime lens you are testing it against, so don't move the camera.

 

In filmdom you measure from the focal plane mark on the camera, where the film travels through the gate.

[Laurent Andrieux]

The thing is, comparing with an HD zoom won't tell us if the system with a prime lens is of great value on an xl2 compared with its default zoom (what they put forward). Anyway, let's try to have something of interest comparing it to the HD zoom... What comes to my mind as a proposition :

 

1) We have to make sure of the focus calibration of the system.

 

- Go outside, focus on the infinite, set the back focus, then inside, check a closer distances by eye on a chart that is placed at an engraved distance of the lens with a tape measure. Let's say 50 X the focal length (2.5 m for a 50 mm). Is focus the sharpest when set to that distance ?

 

- If there is defenetly a difference beetwen engraved distance and eye focus, do the contrary : set focus ring at the distance, set back focus, go outside, can you have the infinite sharp by eye ?

 

All of this being done wide open (I like ND filters better than shutter, myself...)

 

2) It will be of interest to see if the field of view is altered by the system. So we should compare a prime lens to an engraved focal length on the zoom, at the same distance lets say 50 X the focal length.

 

May be it would be worth doing it at all the engraved focal lengths on your zoom.

 

3) comparing definition with the HD zoom : if field of view matches at the same focal length and focus is good, the above test will do. If not, as Patrick says, it's very important that the field of view is the same, then we have to cheat something. I don't know if setting the zoom's focal length to match the same FOV is better than moving the camera a bit (and refocus), may be there won't be much of difference but if you zoom a bit, you get rid of the focus changed parameter so maybe it's better... Write down the changed parameter anyway.

 

4) f-stop test : within the light that gives you the wide open value of the HD zoom, have a chart exposed corectly (white = 100 IRE), put the prime with system. What IRE do you get for the white then ? How many stops compensate ?

 

5) As was said here somewhere people have a doubt concerning the red color of the material. Have a look on the color rendering. Maybe it would be interesting to see what it looks like with volonteer flare, and in high key situations, outside, wide open for instance.

 

What do you guys think ?

[Laurent Andrieux]

The guy who bought it on ebay sent me a reply. He said he didn't get it yet but would given me more information when he does. He didn't say if he comes here, I sent him links to both topics, we'll see. Don't know what camera he's got, may be interesting.

[Martin Hesse]

The Light loss is most likely due to the prism block in the XL-1. Obviously ther is no Prism block in a film camera.

 

As far as lens comparisons go... You may want to consider that the prism block has an effect on the image and video lenses (for 3CCD cams) are designed to compinsate for chromatic aberation caused by the prism block.

 

 

I found this link at:

 

http://www.dvinfo.net/conf/showthread.php?s=&threadid=40603

 

"Look here at a side by side test done with a Fujinon broadcast (video) lens and a Zeiss lens (not a "cheapandnasty").

This is with a hacked Sony FX1 HDV camera, so the CA is more prominent because the higher resolution.

With a SD camera setup, the CA could be less visible:"

 

http://www.eidomedia.com/hdve/ziess_fuji.htm

 

 

 

 

 

 

Report this post to a moderator | IP: Logged

[Laurent Andrieux]

Sorry, but as someone said before, any loss by the camera (prism block or whatecer)) is "included" in the camera sensitivity. Actually, the manufacter of the device doens't say the light loss is due to the video camera and neither says it depends on the camera, though it then sould be the case.

 

The t stop on a lens is a photometric value and there is no reason why an adpatator that is only mecanichal, ie with no dioptric system should occure any light loss, unless the focal flange is changed...

 

The aberations caused by the prisms in a video camera are supposed to be compensated by the white and black shading, not the lens.

 

In the pictures that can be seen on the link you are giving, many factors such as black balance and both white and black shadings as well as the chromatic gamma (gamma is different for each R,G and B channel) can be involved. The test would be interesting if comparing the two lenses on the same camera. I don't mean the same model I mean the same one.

 

Anyway, I don't understand if the zeiss lens was used with the pl mount we are talking about. Is it the case ? Otherways, what is the CA you are talking about (excuse my possible misunderstanding due to the fact english is not my natural language).