[MF3D.FORUM:1185] Re: How much shift can be necessary toset the window?

[Paul Talbot <ptww@xxxxxxxxx>]

"Michael K. Davis" wrote:
> 
> Having read your comments about three times and played with the numbers a
> bit, I feel dashed at the prospect of settling on 50x50.  Letting go of any
> hope of conformity with other peoples' viewers, I suppose I could do 50x63
> with a mirrored viewer.

With a mirrored viewer you could use virtually any format.
Why stop at 50x63?

Consider the advantages over 50x50:

> 50x63 would make an 80mm lens more like a 43mm on 35mm format, vs. 49mm.

I think we could argue whether that is an advantage.  IMHO,
many photos could be improved by concentrating on the most
interesting part of the scene, instead of just trying to go
for the wider view.

> 50x63 would offer a 13.75% larger diagonal vs. 50x50
> 50x63 would increase area by 26.0% vs. 50x50 (3150 sq. mm vs. 2500 sq. mm)

Use of diagonal and area probably overstate the significance of the
difference.  Richard Rylander came up with an "immersion factor"
formula the last time this subject was discussed.  I suggest studying
the many messages in the late December-early January time frame.
This can easily be done using the date-order index in the archives
at Marco's site http://www.pauck.de/archive/mailinglist/mf3d/mf3d.html
by date:
http://www.pauck.de/archive/mailinglist/mf3d/mhonarc/idx_date1.html

Also, John Bercovitz kindly created a spreadsheet that utilizes
Richard's formula, and has it available for download at:
http://www.angelfire.com/ca2/tech3d/techdocs.html
(see the line near the bottom "What solid angle does your
format subtend at your eye? Rylander calculation. (Excel file)"
which points to:
http://www.angelfire.com/ca2/tech3d//FormatImmersion.xls
Note: this is one person's attempt at quantifying "immersion factor."
I don't even understand it's working very well, and I'm not trying
to present it as "gospel truth."  Something to consider, however.

> 50x63 yields the dominant, if not esthetically preferable, 4:5 aspect ratio

I'm not even going to touch that one.  ;-)  I'm just glad you
aren't trying to sell the 2:3 aspect ratio of full frame 35mm.
I don't care for that format at all.  I like the 4:5 format,
but I don't know whether it's my favorite, and I certainly
won't try to make the decision for anyone else!

> 50x63 would allow a mix of portrait vs. landscape orientations

As does 40x50.  Maybe that's one reason we also have 40x50 and
50x40 mounts in the 80x132 size.  :-)

> 50x63 would simply make the best possible use of a 6x7 frame

I'd be tempted to utilize more of the vertical dimension than just
50mm.  If you have a means of assuring good vertical alignment in
the cameras, there is no reason to crop off a full 5mm from the
height, if image maximization is the goal.  So you could use a
format like 53x57 in lieu of 50x63.  It would achieve some of the
same goals as your proposed format, come close on others, allow
more window setting flexibility, and attain a slightly higher
immersion factor rating under Richard's formula (as presented in
John B's spreadsheet; there may be some kinks to work out yet,
however).  Perhaps most importantly it would have a better chance
of being compatible with other folks' viewers.

Of course you could also consider 53x63, but the gains would
be small relative to what you'd be giving up, IMO.

I think that there are some differences in viewing 2D images
versus 3D images that are not being accounted for in your
analysis.  More on that another time...

Paul Talbot