EOS lens

Discussion in 'Casual Photo Conversations' started by russell_fill, Feb 3, 2007.

  1. I am using Canon EOS 1dsM2 and other Canon EOS cameras to photograph time lapes sequences and
    work them up in to movies. Im trying to control the f/stop of the lens in smaller movements than 1/3 of
    a stop. Any ideas would be welcome. Im tring to find out how the eletronic apetrure is controled or how
    to contol it manually. What I need is to be able to control it in as small amounts as possible. Maybe
    1/64th of a stop at a time or smaller still.

    How do the apetures work?

    And is there any one out there who knows if it is possible to run a external control through the contacts of
    the lens to control the apeture in smaller amounts than 1/3rd of a stop.

    Thanks for any info
    Russ Fill
  2. There is no simple solution that I know of, and no commercial solution.

    Canon have not released any details of their electronic lens interface, but it's obviously been reverse engineerd by the 3rd party lens makers.

    I'd guess that the aperture could be controlled in finer than 1/3 stop steps, but you'd have to build your own rig to do it and I'm not sure you could actually do it with the lens attached to a camera.

    Why do you need 1/64 stops? There has to be a better way to do fades than slowly changing the f-stop. Post production maybe?
  3. It has been speculated that the EF lenses are cabable of at least 1/6 stop resolution, because many EOS cameras can work in either 1/3 or 1/2 stop increments.

    If you post process RAW images with DPP, you can "adjust" the exposure in 1/6 stop increments, although you're obviously not changing the actual lens aperture.

    I wonder if the lens is capable of precision down to 1/64 stop anyway, even in nominal 1/6 stop steps?
  4. You are being a little optimistic:


    Also, bear in mind that quantum shot noise is proportional to the square root of the signal. Therefore, if a pixel has a full well capacity of say 80,000 electrons, at full well the noise is about 0.35% of the signal. Three stops down (10,000 electrons), the noise is at 1% - or 1/64th stop.

    Theoretically, with a film camera by using half stop exposure compensation intervals and combining with the 1/3rd stop interval between ISO setting, you can control exposure to 1/6th stop (i.e. maximum error of 1/12th stop) in Tv mode - but you can't do this with a DSLR because the effective ISO of the sensor is changed when you change the ISO (whereas for film it merely affects the metered exposure).
  5. bear in mind that quantum shot noise is proportional to the square root of the signal.
    Sez who?
    Kidding! Although I did get a mild headache trying to comprehend your answer, that was an interesting link. ;-)
    So what this amounts to is obviously Canon is "ripping us off" by denying us smaller aperture increments! Why, I oughta...
  6. I'm kinda simple minded, but couldn't you just adjust this is Photoshop? Am I missing something?
  7. Thanks all for all the info.

    I have gone around and around with adapter rings and second party lens etc etc etc...
    What I am up to now is to think of a way to open the lens and put some sort of a control
    arm/ring through a port so it could be controlled from the out side manually. And
    hopefully with the resolution I need for my work. That is why I have purchused a old
    28-70 3.5-4.5 lens to tear apart to see what might be able to be done. This is why Im
    asking how the iris part of the lens works. Magnets activated by a current or small motors
    controlled by a driver or computer on the camera or what?

    Does anybody know of a lens builder/repair or customizer.

    Im also in the process of using a photocromatic transition lens to see if it will handle the
    change from day to nite, maybe 3 stops...maybe. Here is the reason for the 1/64th or even
    higher resolution. When I say shoot a sunset from full sun to stars coming out I need
    about 6 to 8 stops to keep correct exposure and this would be approx. 300 to 1000
    frames/photos. In say 30 to 60 images that represents about 2 seconds of a clip in TV
    time. If you say make a 1/3rd stop change each frame if you had that many stops, you
    will notice a gradual change. If you have say 1/3rd of a stop every 4 or say 8 frames you
    will have large changes of exposure in the clip as the frames are all run continuously at 24
    to 30 frames per second. With extreme over exposing and having the light go to extreme
    under exposure I can correct for about 2 to 3 stops on each end of that clip using post
    adjustments, photoshop/After Effects. After that the noise starts to get un useable for the
    start of the clip and the last part of the clip. My hope is to be able to connect a stepper
    motor with a gear drive and a light sensitive sensor to control the amount of movement
    with the iris over the change in light from sunset to dark or when a large cloud goes by or
    for any light change..

    Still hoping for some way to get about 5 stops more from my clips.
  8. I think Popular Photography has a holder for EOS lenses that allows external control of the iris. They use it in lens testing. However I don't know what sort of resolution it has. I'd assume it's not 1/64 stop since there wouldn't be any point in that. I don't know if they'd tell you how they did it or who made it for them. It might have been Canon I guess.
  9. It's a long shot but I was thinking you might contact some of those guys (Russian?) who "hacked" the early Digital Rebels to make "cheap 10D's". I have no idea who they are or if I'd let them play with my $7,000 1Ds MarkII camera - but they obviously have a bit of experience in playing with Canon EOS digital cameras in ways other than Canon intended. Good luck!
  10. Can you just use old manual focus lens with an proper EOS adapter? Use those that have no aperture indentation (e.g: Older Preset lens) or remove the aperture indentation (simple enough).
  11. I don't know if you can find one that is suitable for you, but there are continuously variable neutral density filters. I don't remember where I've seen them, but it might be worth a web search. Depending on exactly what you're doing, a variable ND filter might not only solve the exposure problem, but also have the benefit of retaining the same depth-of-field throughout. Good luck.
  12. Thinking about it, what you might be able to use is a filter graduated in small steps. Here is one source: http://www.thorlabs.com/NewGroupPage9.cfm?ObjectGroup_ID=1624&PN=NDL-10S-2&rpLink=2

Share This Page