[photo-3d] Re: stereo prints

["Thom Gillam" <ThomGillam@xxxxxxxxxxxxxx>]

--- In photo-3d@xxxxxxxxxxx, "Todd Leghorn" <musclepuppy@xxxx> wrote:

> It is correct that a 5-perf negative in an 24x36mm negative frame 
> will result in unpredictable or variable results due to the
> adjacent negative image areas on a given strip: The photo sensors 
> read red, green, blue, and density. The *slightest* change in 
> positioning the negative will impact color balance and print 
> density.

Precisely!

> The other big factor is the space between the negatives: the 
> unexposed clear area. I guarantee that the printer's photo sensors 
> do not know how to automatically ignore that impact. It requires a 
> manual override on the part of the printer operator, otherwise the 
> print will be apt to print too light.
 
Of course, because the 36mm wide mask will include that space. The 
only solution is to mask off to 5 perf.

> It takes a override on the part of the printer operator who is 
> competent in what the industry calls "subject classification". A
> skilled printer technician knows that 90% of the exposures on an
> average roll of film do *not* print their best if the printing 
> machine is left at "Normal".

Which implies that 90% of ALL automatic output is inferior! The best 
photofinisher I ever had the pleasure to deal with (The Photo Center 
in Brick, NJ - *unsolicited advertisement*) had a technician who 
inspected EVERY negative, all the time (not just stereo).  Very few 
redos.  When it came time to do stereos, it was relatively simple and 
not much more effort.

> When I worked in a lab, I made a simple mask with black opaque 
> plastic tape that framed in the left and right sides of the 24x36mm 
> film aperture so that I could print Stereo Realist format.

Very similar to the solution arrived at by the Photo Center before 
they got a new machine :^(.  The technician rigged a broken 8 perf 
mask with cardboard, effectively creating a 5 perf mask.

> It also required the operator (me) to write down and track the 
> subject classification value overrides for each left-right pair so 
> that they have a chance of matching. It's conceptually easy but it 
> is much more cumbersome and time-consuming than printing a regular 
> roll of film where you do not have to match each print's density
>and color balance.

Though I never asked, I think he had a similar system.  The big 
problem is that most if not all of these machines do not allow 
settings to be locked in.

> Also remember please that this is production equipment: time, 
> resources, and efficiency impact money! A dollar per print that one 
> chap gets is a bargain when you look at how it impacts efficiency.

As I said in an earlier post, for any service to continue it must be 
profitable.  And at typical one-hour prices, even for reprints, that 
just ain't gonna cut it!

Todd, thank you for your insight into this problem, and thank you
Dave Williams for your earlier illuminating post.

The bottom line is that, in the real world, one-hour labs cannot do 
realist prints.

Period.