Header banner

<< Previous Thread DoF calculations for the TL-120 Next Thread >>

Subject: DoF calculations for the TL-120
Date: 2006-11-09 03:13:24
From: Michael K. Davis
Dale,

At 06:29 PM 11/6/2006, you wrote:

Has anyone computed a hyperfocal table specifically for this
camera?

The following formula gives you the enlargement factor of a viewer:

Viewer Angular Magnification = (254mm / Viewer FL) + (254 / (Image
Distance * 25.4))

If you intend to shoot for a stereo viewer that has 75mm lenses focused at
Infinity, your viewer angular magnification (enlargement factor) will be 3.39x.

The following formula gives you the proper CoC diameter to use for DoF
calculations:

Max. Permissible CoC Diameter = 1 / Desired Resolution / Enlargement Factor

I don't know how aggressive you want to be in selecting your desired
resolution, but I would recommend using a CoC diameter that would support
at least 4 lp/mm as seen in the viewer. Here are the CoC diameters you
should use in DoF calculations to support various resolutions in a 75mm
viewer focused at Infinity:

2 lp/mm - 0.148 mm
3 lp/mm - 0.098 mm
4 lp/mm - 0.074 mm
5 lp/mm - 0.059 mm
6 lp/mm - 0.049 mm
7 lp/mm - 0.042 mm
8 lp/mm - 0.037 mm

The choice is yours, but if you're concerned about focusing errors, film
flatness, views being shown in viewers that can focus more closely than
Infinity (viewers offering higher magnification) or any additional slop
that's beyond your control, give yourself a safety margin by selecting a
CoC that will support a higher resolution than you might choose otherwise.

Note: There are many factors that influence the total system
resolution. This discussion is limited in scope to the impact of Circles
of Confusion. (I had to throw that in because someone might think I'm
claiming you can achieve any resolution you desire just by making the CoC
small enough....)

OK, having decided which CoC diameter to use, you can plug it into any
available DoF calculator that supports user-specification of the maximum
permissible CoC diameter (in addition to the camera focal length). A DoF
calculator that does not allow you to specify the CoC diameter is next to
worthless, in my opinion.

I recommend you try the freeware available at http://www.dofmaster.com/ to
produce a custom spinning disk calculator. It's superior to working with
tables, in my opinion. (I have a lot of opinions, huh?)

Here are some numbers to get you rolling prior to making your own DoF chart:

Assuming a desired resolution of 5 lp/mm as seen in a 75mm viewer that's
focused at Infinity, using a Maximum Permissible CoC Diameter of 0.059 mm
and a camera focal length of 80mm, here are the Near and Hyperfocal
distances (in feet) at various f/stops, for scenes that include Infinity
subjects:

Stop Near Focus Far

f/2.8 62.8 125.7 Infinity
f/4 44.4 88.8 Infinity
f/5.6 31.4 62.8 Infinity
f/8 22.2 44.4 Infinity
f/11 15.7 31.4 Infinity
f/16 11.1 22.2 Infinity
f/22 7.9 15.8 Infinity

Mike Davis
Subject: Re: DoF calculations for the TL-120
Date: 2006-11-09 13:36:47
From: John Hart
--- In MF3D-group@yahoogroups.com, "Michael K. Davis"
wrote:
>> If you intend to shoot for a stereo viewer that has 75mm lenses
focused atInfinity.... but I would recommend using a CoC diameter
that would support at least 4 lp/mm as seen in the viewer. Here
are the CoC diameters you should use in DoF calculations to support
various resolutions in a 75mm
> viewer focused at Infinity:
>
> 2 lp/mm - 0.148 mm
> 3 lp/mm - 0.098 mm
> 4 lp/mm - 0.074 mm
> 5 lp/mm - 0.059 mm
> 6 lp/mm - 0.049 mm
> 7 lp/mm - 0.042 mm
> 8 lp/mm - 0.037 mm
>
> The choice is yours

Hi Mike,

Are you suggesting that in practice a good MF viewer system will
typically max out at 8 lp/mm? Or, a 50mm wide image can be observed
at about 500 lp across (sorta like XGA). That seems woefully low?

Of course, if you plug in a CoC for 32 lp/mm the DOF even at f22
gets darn small. Hmmmmm.

John
Subject: Re: DoF calculations for the TL-120
Date: 2006-11-09 15:22:27
From: Michael K. Davis
Hi John!

I heard great things about the work you showed at the 2006 NSA Convention.
I wish I had been there because at the 2005 NSA Convention, I genuinely
found your very colorful macrostreography (of crystals) to be the "best of
show" in my opinion. Considering the low resolution and luminance of the
digital projector, it was a triumph of spectacular artistic content over
inadequate display technology. I was impressed. I'd love to see those
images displayed with the level of detail they deserve.


On Thu, 9 Nov 2006, John Hart wrote:

> --- In MF3D-group@yahoogroups.com, "Michael K. Davis"
> wrote:
> >> If you intend to shoot for a stereo viewer that has 75mm lenses
> focused atInfinity.... but I would recommend using a CoC diameter
> that would support at least 4 lp/mm as seen in the viewer. Here
> are the CoC diameters you should use in DoF calculations to support
> various resolutions in a 75mm
> > viewer focused at Infinity:
> >
> > 2 lp/mm - 0.148 mm
> > 3 lp/mm - 0.098 mm
> > 4 lp/mm - 0.074 mm
> > 5 lp/mm - 0.059 mm
> > 6 lp/mm - 0.049 mm
> > 7 lp/mm - 0.042 mm
> > 8 lp/mm - 0.037 mm
> >
> > The choice is yours
>
> Hi Mike,
>
> Are you suggesting that in practice a good MF viewer system will
> typically max out at 8 lp/mm?

The formula I provided for calculating the maximum permissible CoC
diameter for use in DoF calculations included only two variables - Desired
Resolution and Enlargement Factor. I suspect your question stems from my
failure to emphasize that the "Desired Resolution" is the resolution you
hope to achieve as seen in the viewer after magnification - not the
resolution you hope to achieve on-film before the enlargement factor is
applied.

For example, if you were to specify a maximum permissible CoC diameter of
0.037mm, a resolution of 8 lp/mm will be supported (per my table above),
as seen in the viewer after magnification. For a viewer that imposes
3.39x magnification, 8 lp/mm would equate to an on-film resolution of 3.39
* 8 = 27.12 lp/mm. I do not recommend the selection of a CoC diameter
that would require an on-film system resolution greater than 45 lp/mm with
the best color reversal films and lenses, or 60 lp/mm with the best black
and white films and lenses. 45 lp/mm equates to a CoC diameter of 0.0222
mm. 60 lp/mm equates to a CoC diameter of 0.0167mm. (Just take the
reciprocal to convert between lp/mm and diameter.)

I calculate my CoC's with a desired resolution of 7.62 lp/mm (in the final
print or in the stereo viewer, after enlargement). For my SaturnSlide
viewer, which has 78mm lenses and variable focus producing magnifications
as high as 4.21x (when focused at an image distance of 10 inches (instead
of Infinity), the CoC I use for DoF calculations works out to a very small
0.03115mm - close to the 0.03mm CoC commonly used for 35mm format cameras.
0.03115mm supports my desired resolution of 7.62 lp/mm, as seen in the
viewer after magnification.

Why did I stop at 7.62 lp/mm instead of 8.0 lp/mm? Had I attempted to
make my CoC's any smaller than 0.03115mm, I would not be able to use f/22
(the smallest aperture on my camera lenses) without forcing the diameter
of diffraction's Airy disks to exceed the diameter of my CoC's. Stopping
down makes your CoC's get smaller, but it simultaneously causes Airy disks
to get larger. Stopping down for "more DoF" is counterproductive when
doing so would force Airy disks to grow larger than the CoC's produced at
the f-stop you're considering.

Another formula: The f-number (N) at which diffraction's Airy disks will
exceed the size of your chosen maximum permissible CoC diameter can be
calculated as follows:

N = Max. CoC Diameter / 0.001353831438675

23 = 0.03115 / 0.001353831438675

Someone might ask, why f/23, instead of f/22, when I said my cameras
stop down no further than f/22? For the purpose of such calculations, I
always use the actual ratio of focal length to aperture had for a lens
marked f/22: That's 22.6 (not 22.0) which rounds to 23. :-)

Thus, for a Maximum Permissible CoC Diameter of 0.03115mm, I can not stop
down below f/22 (as marked) without forcing Airy disk diameters to
compromise a desired resolution (after 4.21x magnification) of 7.625
lp/mm. For the magnification imposed by my viewer, there's no point in
seeking a resolution greater than 7.625 lp/mm, as seen in the viewer.
The diffraction experienced at f/22 would not support a higher resolution.

Mike Davis
Subject: Re: DoF calculations for the TL-120
Date: 2006-11-09 16:56:03
From: John Hart
--- In MF3D-group@yahoogroups.com, "Michael K. Davis"
wrote:
> I'd love to see those images displayed with the level of detail
they deserve.

Thanks, and glad you liked them so much! For 2006NSA I did a little
test you might have been interested in. I had slides.com make best-
possible-quality MF film-recorder slides of some of the crystals
(which start off as huge 60+ MB digital files) and viewed them in a
quality MF viewer. We compared that with 8.5x11's printed on an R800
and viewed with the MirScope Viewer (see www.mirscope.com ).
Everyone who saw this demo said the print viewing was better. I
don't think such is in fact better than hand viewing original MF's,
but it clearly is superior to outputing digital to MF slides for
handviewing.

I hope to bring a couple of the 13x19 print viewers to NSA2007.
Prints in that are something else. Maybe I could interest you (or
others in this group) in generating some material for it?

And thanks for the clarification on your calculations. It's all
clear now.

John
Subject: TL120 stereo mounter
Date: 2006-11-09 21:33:55
From: Dale Yingst
I'm trying to make a viable slide mounter out of this. Pretty easily, I
modified the base to accommodate a gage from the SAM mounter. It works
pretty well and looking at the included slides from 3DWorld, I can see
that not much attention was given to the stereo window, just the
vertical alignment. The SAM gage appears to have an at-window vertical
line spacing of about 62mm. Would that be appropriate spacing for the
MF slides? If not, I could scribe my own lines. My problem is that I
need to raise the lenses about 3.5mm to get them back into focus. I
tried scoring the base of the lenses to free it but fear they may have
been ultrasonically welded down to the mounter base, so that a more
destructive means may be needed to separate them from the base and add a
spacer. Would I need a tiny hobby hacksaw to do this? I think if I can
do this, I would actually prefer this to the SAM mounter that I modified
for MF mounting.

DAle Yingst
>
>
> _,___
Subject: Mirscope
Date: 2006-11-10 16:05:27
From: Michael K. Davis
Attachments :
    John,

    The 13x19 exhibition version is a beast, but I would love to see it.  Where is the NSA convention in 2007?  (I could probably find out by searching for it...)

    Thanks,

    Mike Davis

    At 04:51 PM 11/9/2006, you wrote:

    --- In MF3D-group@yahoogroups.com, "Michael K. Davis"
    wrote:
    > I'd love to see those images displayed with the level of detail
    they deserve.

    Thanks, and glad you liked them so much! For 2006NSA I did a little
    test you might have been interested in. I had slides.com make best-
    possible-quality MF film-recorder slides of some of the crystals
    (which start off as huge 60+ MB digital files) and viewed them in a
    quality MF viewer. We compared that with 8.5x11's printed on an R800
    and viewed with the MirScope Viewer (see www.mirscope.com ).
    Everyone who saw this demo said the print viewing was better. I
    don't think such is in fact better than hand viewing original MF's,
    but it clearly is superior to outputing digital to MF slides for
    handviewing.

    I hope to bring a couple of the 13x19 print viewers to NSA2007.
    Prints in that are something else. Maybe I could interest you (or
    others in this group) in generating some material for it?

    And thanks for the clarification on your calculations. It's all
    clear now.

    John

    Subject: Re: Mirscope
    Date: 2006-11-10 17:18:22
    From: David W. Kesner
    --- In MF3D-group@yahoogroups.com, "Michael K. Davis"
    wrote:
    >
    > Where is the NSA convention in 2007?

    That would be Boise, Idaho, July 10-17,2007. Don't forget that it is
    a joint event between the 33rd NSA Convention and the 16th ISU
    Congress. This should be the largest gathering of stereo enthusiasts
    the world has ever seen.

    For more info check out the official website:

    http://2007.nsa3d.org

    There are plans for not only a large format stereo exhibit that John
    Hart is heading up but also a medium format exhibit that I hope Sam
    Smith will be heading up.

    Thanks,

    David W. Kesner
    Chair, 2007 NSA/ISU Convention/Congress
    http://2007.nsa3d.org
    chair@dddphotography.com
    Subject: NSA 2007 - in Boise
    Date: 2006-11-10 19:18:43
    From: Michael K. Davis

    Thanks David!

    At 05:04 PM 11/10/2006, you wrote:

    --- In MF3D-group@yahoogroups.com, "Michael K. Davis"
    wrote:
    >
    > Where is the NSA convention in 2007?

    That would be Boise, Idaho, July 10-17,2007.

    http://2007.nsa3d.org

    Subject: Re: Mirscope
    Date: 2006-11-15 18:57:07
    From: John Hart
    --- In MF3D-group@yahoogroups.com, "Michael K. Davis"
    wrote:
    >
    > John,
    >
    > The 13x19 exhibition version is a beast, but I would love to see it.

    Yeah, installation-art. Be great if you can get to NSA2007.

    Curiously, although I haven't actively hawked these things yet, I've
    had a lot of inquiries from people who are interested in the big dude.

    John
    www.mirscope.com