Olympus-OM
[Top] [All Lists]

Re: [OM] OT Lei*a designed for digital

Subject: Re: [OM] OT Lei*a designed for digital
From: Kennedy McEwen <rkm@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 11 Jul 2003 00:43:18 +0100
Cc: Pschings@xxxxxxx
In message <3682EC47.16B8415C.0080BFFB@xxxxxxx>, Pschings@xxxxxxx writes
In a message dated 7/10/2003 1:35:20 PM Eastern Standard Time, rkm@xxxxxxxxxxxxxxxxxxxx writes:

<snip>

The only problem I couldn't overcome was getting the CCD to the film
plane itself - and I certainly wouldn't recommend taking the CCD out of
the package

What about a fiber optic light pipe, like NPC uses for their Polaroid backs?

You really need to have the fibre bundle intimately mating with the CCD for that to work, which is quite a demanding optomechanical task. On a standard CCD, like those you can buy off the shelf, the 2-3mm gap between the window face and the CCD focal plane would cause too much loss of resolution to be worth having a high pixel count in the first place. Even if the light from the fibre ends could be constrained to 10deg cones, which is a tough target to meet, each single point in the image would subtend around 400um at the focal plane or, in the case of the Philips chip, about 36pixels in diameter. Since it is more likely to be a 60-90deg cone, the problem is fairly obvious. You need to buy the CCD with the fibre bundle fitted instead of the window, and these are not readily available.

I've read some initial reports from users of the full frame Canon that they *are* seeing chromatic aberration problems at the corners of the frame with super wides.

I am sure it can happen, and I did expect to see some - as soon as there is a finite distance between the colour filter plane and the pixel plane it becomes inevitable. However, despite my own expectations, I found no evidence of it with the devices I used and what is pretty close to the widest angle lens in the OM range. A bit like WMD in Iraq really. :-)

Interestingly, since the Foveon chip relies on the penetration depth of the light to discriminate colour, rather than a filter, it would not exhibit chromatic problems at all, irrespective of distance of the optical vertex from the focal plane.
--
Kennedy


< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >


<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz