----- Original Message -----
From: Javier Perez <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, December 16, 2002 5:10 AM
Subject: EOS AF or MF confirm accuracy Vs manual focus accuracy


> Hi Everyone
> for awhile I've been noticing that when I set my AF
> lenses to MF on any of my OESes there is a small range
> in terms of barrel turning where the confirm light
> stays on. Yet the split image can narrow it down far
> more. It's only 2 degrees or so with MF confirm but
> far less with the split image. The question is, is MF
> more accurate than AF or MF confirm?
> Has anyone else observed this?
> See Ya
> Javier

Hi,
I thought it was well documented (from the original Minolta 7000 onwards) that
all common AF systems had accuracy limited to somewhat lax depth-of-field
figures, but that in many situations this was of little consequence as the
subject's depth ensured that good focus was found somewhere on the subject.
After some years of AF usage, I recall being surprised recently to find that
when field testing a s/h Nikon FG with a 50 f2 Nikkor, using the split-image I
could reliably distinguish between "real" infinity and the corner of a building
350m away. (I was so surprised that I checked the 350m range figure on a
detailed map.)  Haven't been able to replicate this with my EOS3, and so far
haven't tried it with my EFM and its split-image.
I guess it's one reason why FTM is so valued on Canon's USM lenses thus allowing
habitual full aperture workers to get the best out of the system.
On a similar tack, it's well known that lenses occasionally shift effective
focus when stopped down so one way in which AF systems could be improved would
be to store the focus shift corrections in the lens' chip and transmit this to
the camera as a fine correction when stopped down.  Something for Contax (Off
Topic) perhaps?!

Malcolm
Milton Keynes, UK

*
****
*******
***********************************************************
*  For list instructions, including unsubscribe, see:
*    http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************

Reply via email to