[MF3D.FORUM:500] Re: Exact fl match for paired lenses.

["Bill Glickman" <bglick@xxxxxxxx>]

David

      Once again, you offer some fabulous suggestions.  You must have read
my mind, because I just completed what you had suggested.  I shot a grid
with both lenses and then shot some stereo pairs to see if they look right.
I am hoping I do not have the problem of different size apt. between the two
lenses....but now that you suggested it, I will always get to the same f
stop from the same direction.  One less variable to worry about.   I also
agree, I doubt its worth altering these lenses.

      Had I been a bit more knowledgeable, I maybe should have sent Mamiya
in Japan my 80 mm lens and had them match it at their factory.  With an
optical bench this is not difficult, but this random testing is a very
unscientific way to determine such.   It may still be a good idea to do
this, assuming my grid test fails.  I am hoping I can avoid this.

      I will report back when I get the film back.  Thank you David...

Bill G


>
> I would agree with most of what Greg has said about matching lenses.
Having
> matched 6 pairs of medium and large format cameras and lenses with mixed
> success, I have sympathy for your plight. Especially considering the cost
of
> those lenses. I thought about buying a pair of those cameras but
eventually
> balked at the cost and ended up buying 2 large format lenses which the
> Schneider people were kind enough to match for me at the factory. (They
> measured them down to a tenth of a mm (123.6mm to be exact), which works
out
> to .1% difference at most.
>
> As for the differences that the eye will tolerate, the researcher Bela
> Julesz who wrote the seminal book, "Foundations of Cyclopean Vision,"
found
> that the brain could fuse images (random dot in this case) up to something
> like a 20% difference. So obviously one would have no trouble comfortably
> fusing any difference in a modern lens with reasonable standards (as I
have
> no doubt that the Mamiya 7 has). But, as Greg said, the problem lies in
the
> differences at the edges of the mounts. I certainly won't tolerate much
> myself, even though I mostly make prints and could actually vary the
> enlarger position to make the images exactly the same size. Having done
this
> a few times, though, I have had enough of it and now insist on matched
> lenses. It took me 5 Yashicamats to get close enough, although I had
better
> luck with the others.
>
> Where I differ from Greg is that I would not even think of putting one of
> the lenses out of focus to get the same image size. One only uses medium
> format if one wants absolutely sharp images, and while depth of focus
> (similar to, but not exactly the same as, depth of field) will cover it to
> some degree, I can't imagine that you would find that satisfactory after
> spending some $2000 dollars for the pair of lenses. And while one might
> reasonably shim an old Ricoh or  Yashicamat lens, I can't imagine trying
> that with a Mamiya 7 lens.
>
> As for testing, I did very similar to what you are planning on, that is, I
> shot a newspaper (to check the sharpness) with a MacBeath color checker (a
> grid) on it. I then put one on top of the other and it was a simple matter
> to see whether they were the same size. I can't imagine that you will see
> any change in size with different f stops, but I have found with several
> pairs that there is a difference in aperture size between cameras, and
also
> a difference in the same camera depending on whether you open up to get
the
> f stop, or close down to get to it. So I have a note on the back of the
> cameras telling me how much more to open or close a particular aperture,
and
> if they are the same, I have gotten into the habit of either opening or
> closing both of them, as opposed to opening one and closing the other to
get
> to the same aperture.
>
> As for interpreting the results of the test, you might, on the same rolls,
> shoot some actual stereo shots, mount them, and see if they are acceptable
> to your eye. If they are not, I would cross that bridge when I came to it
> and hope for the best in the meantime. By the way, I suspect that if you
> simply got one additional lens it would very likely be acceptably close to
> one of the ones you have. We are, after all, talking about some of the
> sharpest and best lenses ever made. Then you could sell the odd one for  a
> slight loss (or put it one eBay and make a profit :-))
>
> David Lee
>
> ----- Original Message -----
> From: Bill Glickman <bglick@xxxxxxxx>
> To: Medium Format 3D Photography <MF3D.Forum@xxxxxxxxxxxxx>
> Sent: Thursday, May 18, 2000 5:14 PM
> Subject: [MF3D.FORUM:496] Re: Exact fl match for paired lenses.
>
>
> > Greg
> >
> >    These lenses are for the Mamiya 7 rangefinder camera.  They have leaf
> > shutters.  I will visit your site.
> >
> > A suggestion that seemed sensible to me was to shoot a grid with each
lens
> > and then lay the chromes on top of each other over a light box to tell
if
> > the fl are identical.  I will try this at each f stop to be sure there
is
> no
> > changes.  Of course if they are not, I don't know how to interpret the
> > shortcomings to know if it is acceptable.   What do you think about
this?
> >
> > Thanks you.
> >
> > Bill G
> >
> >
> >
> > > >Greg
> > > >
> > > >Thank you for the input.  How did you discover the % difference in
fl's
> > of
> > > >the 5 lenses?   Did you shoot test targets without moving the camera
> > body?
> > > >What tolerance would you say would be acceptable after you advise me
of
> > how
> > > >best to test the lenses?
> > >
> > >   Read my web page on how I compared the FLs (basically
> > > built a jig with shutter to hold the lenses and then
> > > projected a target onto a large sheet of paper).
> > >
> > >   I suspect everyones tolerance for missmatch is
> > > different. For me, how the scene is placed behind
> > > (or cutting) the window is very important to me
> > > (ie I look at the edges of the photo) so unequal
> > > cropping due to size missmatch is a big deal for
> > > me. People who just enjoy the 3d scene without
> > > looking too hard at the edges could tolerate and
> > > enjoy much more missmatch than me.
> > >
> > >
> > > >      If the fl's are off by more than the tolerable %, can I send
them
> > off
> > > >to a lens shop and have them make the same adjustment you did?  If
so,
> > does
> > > >anyone know of a lens shop that could accomplish such a task?  I can
> not
> > get
> > > >5 Mamiya lenses to choose from, I would be forced to buy them all
then
> > sell
> > > >4 of them at a big loss!
> > >
> > >   Shimming one lens out to improve the size match
> > > has the sideeffect of moving the focus point.
> > > So the two lenses are no longer focused at the
> > > same distance. If you like to shoot wide open
> > > or have to shim a lot then this may not work
> > > for you.
> > >
> > >   In my case the 4 thou shim I moved makes the
> > > focus points mismatch by 3 (I think) inches at
> > > 2m distance. It was barely noticable in my test
> > > shots at f3.5 (newspaper laying on floor shot at
> > > 45 degrees) with a 10x loupe. At f8, the widest
> > > I'm likely to shoot for real photos I couldn't
> > > see the difference. And an MF viewer is only about
> > > 3x, not 10x.
> > >
> > >   But if you have significant size missmatch the
> > > focus shift from shimming may be greater than the
> > > DOF you have at semi-close distances and wider
> > > apertures.
> > >
> > >   I could crunch some numbers for you if you get
> > > a measurement for the missmatch (shoot some test
> > > rolls). And tell me what apertures you think you'll
> > > use.
> > >
> > >   I don't know how the Mamiya's are constructed but
> > > a camera tech could likely tweak it for you.
> > >
> > >   Are these Mamiya TLR's you have or SLRs?
> > >
> > > Greg
> > >
> >
>