1. Reminder: Please use our affiliate links for holiday shopping!

Olympus Sensor Shift To Compensate For Earth Rotation

Discussion in 'Astrophotography' started by AG_Alex2097, Dec 18, 2015.

  1. AG_Alex2097

    AG_Alex2097 Mu-43 Regular

    157
    Dec 18, 2015
    Alex S.
    I was wondering about this, why is this not a thing? Would it be too inaccurate? Drain battery life too fast perhaps? Too much location/ orientation dependant?

    Ps first post, hi all! ^^
     
  2. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    It's too location dependent, star tracking mounts require a lot of calibration and a fixed position once calibrated to work, in a mobile camera it's not really feasible.

    The sky rotates around the poles also, so the camera would have to know where the pole was in the frame (or out of frame) and rotate the sensor around this point. Any distortion in the lens would make this inaccurate also (same for wrong focal length etc).
     
  3. AG_Alex2097

    AG_Alex2097 Mu-43 Regular

    157
    Dec 18, 2015
    Alex S.
    Could a GPS accessory not resolve this though? Heck you don't really need an accessory, it could work via a smartphone app which provides the data to the camera. (Most have a compass/positioning built-in) It seems weird that with such a great IBIS system they couldn't work something out? :/
    I guess the lens thingy does come into play though, but isn't Hi-Res mode facing the same problem then?

    Thx for your reply btw ^^
     
    Last edited: Dec 18, 2015
  4. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    The problem is knowing exactly where the lens is pointed and where that is relative to the poles. High res is moving the sensor half a pixel, it's actually less of an issue than occurs during normal stabilization.
     
  5. Petrochemist

    Petrochemist Mu-43 Top Veteran

    651
    Mar 21, 2013
    N Essex, UK
    Mike
    Pentax do actually do this with their sensor shift on K5 & later models if you add the GPS accessory.
    The only reason I can think of for ever getting the GPS, but I'd have to upgrade my DSLR body first!
     
    • Agree Agree x 1
  6. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    Perfect example, it works pretty well at shorter exposure times (5 minutes is the limit) however every time you move the camera position you have to recalibrate again just like a star tracker. I'm unaware of how it copes with distortion however Pentax are pretty good at making lenses so apart from fisheyes I doubt it would be a serious problem.
     
  7. AG_Alex2097

    AG_Alex2097 Mu-43 Regular

    157
    Dec 18, 2015
    Alex S.
    A gyro could do that i guess, i'm no expert but isn't there a gyro in the camera anyway? as it's able to detect movement/shaking for the IBIS in the first place?

    Aah i see thanks for the input, so it IS possible, here's to hoping Olympus will maybe introduce something in the future ^^

    5 minutes would be more than plenty for my needs (i prefer the landscape + sky kind of astrophotography, although it has its charms i'm not a big fan of deepsky), it also kind of completely removes the whole disadvantage of ยต4/3 for astrophotography (Shorter shutter times/noise) vs other sensor sizes

    The only problem left then, as you said, is lens aberrations, but if Pentax can do it, i'm sure Olympus can too! Maybe one day :)
     
  8. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    The hard part isn't the camera movement (there will be none at all, this will only work on a tripod) which is what a gyro detects, the hard part is detecting the camera orientation relative to the stars (well, in this case the GPS satellites).

    The main reason the Pentax requires you to recalibrate every time you move the camera is not the camera movement (as I understand it), it's the GPS module alignment to the camera itself. If the module rotates 1 degree on the hotshoe it will think the satellites (and thus the stars) are in a different position to the camera than they actually are. Recalibrating every time makes sure the module and camera are aligned.
     
    • Agree Agree x 1
  9. AG_Alex2097

    AG_Alex2097 Mu-43 Regular

    157
    Dec 18, 2015
    Alex S.
    But it would be nice to have it all in one (relatively) small package :)

    I see, thanks for the heads up and i did realise you don't astrophotography with your hands :p I was referring to the E-M5 II having a "levels" display which shows pitch and yaw on screen (so you can make sure your shot is horizontal/aligned), sorry if it was not clear ;)
     
  10. oldracer

    oldracer Mu-43 All-Pro

    Oct 1, 2010
    USA
    Yes. What about this:

    A small clock drive and wedge with internal GPS, attitude sensor, and compass. Use the camera LCD to cue the user/manage the alignment of the wedge with the earth's axis of rotation.

    Strong odds are that the attitude sensors and the (magnetic) compass would not give high accuracy such as is needed for hours-long exposures, but it would seem that "close enough" might be adequate for much longer exposures than would be feasible without any compensation. It might also be feasible to adjust calibration through some kind of successive-approximation scheme again using the LCD.

    </bubble>
     
  11. GFFPhoto

    GFFPhoto Mu-43 All-Pro

    Feb 24, 2013
    Isn't that how we came up with pneumatic tires?
     
    • Like Like x 1