idea for multi-shot film digitizing...?

[jtk]

https://www.amazon.com/SWEBO-TC18-p...BRY9V2F9DWY&psc=1&refRID=22ET5DG5KBRY9V2F9DWY

Would be better if motorized X/Y and that was digitally controlled for perfect replication...

I did play with an X/Y axis microscope platform years ago to make multi-image slides...more like this:

 

[glbeas]

Perfect for focus stacking. Got a cheaper version on Ebay a while back, worked just fine using it manually.

 

[removed account4]

this also looks like it would be fantastic for doing STEREO photography...

 

[NortheastPhotographic]

I also have looked for 2 way rails for this exact purpose but haven't made any moves just yet. My advice is still to get a camera with multishot built right in, which means either the most recent A7R or my personal choice, the S1R. If you were to combine the two methods so that you capture the film with pixel shift and stitch, well then people would be throwing out their Tangos pretty soon. Personally though I shoot a lot of 6x6, 4x5, and 8x10. A single pixel shifted S1R capture gets me a 360ppi print at the maximum size my 24" Epson can handle. 2 captures get me at or near 44" printers, which I don't have. Plus I'm getting true-RGB capture with extremely low noise even when doing massive shadow recovery. It's pretty great!

 

[jtk]

I'm mainly thinking about multi-shot "scanning" from the very solid Durst stand that I use for single-shot.

For example with one click make 16 computerized-position digital shots from one original 6x7 frame. I think a much finer version of that already exists for digitally-controlled microscopy.

My Samsung can supposedly do focus tracking using its existing autofocus without mechanical stacking devices if i want to struggle with a camera software tweak (not sufficient interest, too old etc).

 

[jtk]

https://www.amazon.com/SWEBO-TC18-p...BRY9V2F9DWY&psc=1&refRID=22ET5DG5KBRY9V2F9DWY

Would be better if motorized X/Y and that was digitally controlled for perfect replication...

I did play with an X/Y axis microscope platform years ago to make multi-image slides...more like this:

I'd move the Omega and Durst film holders over the 5000k LED light panel the way I do manually now.

 

[focus_on_infinity]

Looks neat, but considering that a 4x5 is 15 times the size of a full sized sensor, and 34 times the size of an APS-C, and that's a lot of repositioning.

Also, this is a better link to the product.

 

[jtk]

Looks neat, but considering that a 4x5 is 15 times the size of a full sized sensor, and 34 times the size of an APS-C, and that's a lot of repositioning.

Also, this is a better link to the product.

I don't think your math works, but that's irrelevant here.... Yes, "a lot of repositioning" is exactly the reason for digital motor control for multiple positions. That has been central to robotic manufacture for decades, and computers don't care about "a lot." That's half of what they're about. Computerized motor control would be infinitely more accurate than what I showed in the links. At least as accurate as micrometer measurement, for example (thickness of a hair, etc).

 

[jtk]

In the 70s, when we needed to put 9 different images in one 35mm slide, we did it with a Kodalith mask, 9 Ektachrome slides, and bits of tape. Fussy work, that's why we charged $35 per slide, vs $0.35 for standard slide dupes (plus $10 per glass slide mount . In addition to the Kodalith, we processed all the Ektachrome in house our E4 home brew (typically 60 rolls per day, all pro).

 

[jtk]

Here's the local genius, one-man-band, triple Oscar winner (for color, positioning, software etc), photographer and monster inkjet printer and very cool guy who doesn't waste much Nikon sensor real estate making his 16-68 frame 8' long (and soon a lot longer) panos... doesn't use other vendors (other than Nikon), used his own software for a while but probably uses PS these days, since it comes with the $10/mo package

His taste in color etc is his taste, but the prints are miracles (Canon inkjet). And he sells lots of them. The smallest I've seen is about a yard long.

http://sumnerdene.com/artist-details.php?artistID=90

http://www.unit16.net/Adobe Web Gallery/content/index.html

 

[GLS]

I'd move the Omega and Durst film holders over the 5000k LED light panel the way I do manually now.

Maybe I'm being dense, but I don't see how this would work to get the camera's lens axis along the "Z" orientation relative to that contraptions "X/Y" adjustments. You would need a camera L-bracket which oriented along the lens axis (i.e. with one foot of the "L" covering the back of the camera) rather than perpendicularly.

 

[focus_on_infinity]

I don't think your math works, but that's irrelevant here....

It's correct. A 4x5 is 20 sq inches. A 35mm/FX is 1.35 sq inches. An APS-C is .6 sq inches.

 

[jtk]

It's correct. A 4x5 is 20 sq inches. A 35mm/FX is 1.35 sq inches. An APS-C is .6 sq inches.

OK...but not relevant. Most LF lenses and many MF lenses are unsharp compared to most 35mm primes, mot to mention dslr primes.

 

[gorbas]

Here's the local genius, one-man-band, triple Oscar winner (for color, positioning, software etc), photographer and monster inkjet printer and very cool guy who doesn't waste much Nikon sensor real estate making his 16-68 frame 8' long (and soon a lot longer) panos... doesn't use other vendors (other than Nikon), used his own software for a while but probably uses PS these days, since it comes with the $10/mo package

His taste in color etc is his taste, but the prints are miracles (Canon inkjet). And he sells lots of them. The smallest I've seen is about a yard long.

http://sumnerdene.com/artist-details.php?artistID=90

http://www.unit16.net/Adobe Web Gallery/content/index.html

Thank you for the link JTK!
Does he use motorized rig to move the camera? With clouds and fast moving sun there's not much time to be wasted fiddling with gears.

 

[jtk]

Thank you for the link JTK!
Does he use motorized rig to move the camera? With clouds and fast moving sun there's not much time to be wasted fiddling with gears.

Good question. No. He's not a clouds type of photog. He's devoted to the final heavily manipulated decorative kind of guy. So he sells a lot of huge prints.

 

[jtk]

Maybe I'm being dense, but I don't see how this would work to get the camera's lens axis along the "Z" orientation relative to that contraptions "X/Y" adjustments. You would need a camera L-bracket which oriented along the lens axis (i.e. with one foot of the "L" covering the back of the camera) rather than perpendicularly.

I don't understand. There is no such thing as a Z axis...unless you're adding focus tracking.

 

[GLS]

I don't understand. There is no such thing as a Z axis...unless you're adding focus tracking.

If the aim is to use the two knobs on this gizmo to control the movement of the digital camera across the plane of the film being digitised (i.e. in the X and Y axes of the film plane), then by necessity the optical path of the digital camera lens would have to be perfectly perpendicular to both, in other words pointing along the Z axis. I can't see how you could attach a camera to the Arca style clamp pictured on the device in the OP and achieve this orientation of the camera, unless using a custom-made L-bracket.

 

[gorbas]

Good question. No. He's not a clouds type of photog. He's devoted to the final heavily manipulated decorative kind of guy.

But clouds and sun are moving during his 16-68 shots
Yes, I noticed "heavy manipulation) kind of HDRs

 

[jtk]

If the aim is to use the two knobs on this gizmo to control the movement of the digital camera across the plane of the film being digitised (i.e. in the X and Y axes of the film plane), then by necessity the optical path of the digital camera lens would have to be perfectly perpendicular to both, in other words pointing along the Z axis. I can't see how you could attach a camera to the Arca style clamp pictured on the device in the OP and achieve this orientation of the camera, unless using a custom-made L-bracket.

I think a near-ultimate answer would entail a very long macro lens, such as 200mm.

 

[ssssjjjj]

I tried to shoot 4x5 films in 1:1 by the same idea.

Photo merged by photoshop

https://www.flickr.com/photos/186734762@N07/49475618263/in/dateposted-public/

 

[jtk]

If the aim is to use the two knobs on this gizmo to control the movement of the digital camera across the plane of the film being digitised (i.e. in the X and Y axes of the film plane), then by necessity the optical path of the digital camera lens would have to be perfectly perpendicular to both, in other words pointing along the Z axis. I can't see how you could attach a camera to the Arca style clamp pictured on the device in the OP and achieve this orientation of the camera, unless using a custom-made L-bracket.

Sorry for delay response. Answer: camera remains fixed...XY Axis only moves the slide mount or film holder.

 

[David A. Goldfarb]

I’ve seen quite a few discussions about building precision alignment easels for DSLR scanning, but haven’t found them necessary myself. Photomerge in Photoshop is designed to align misaligned frames digitally. Get it in the ballpark, and Photomerge can do the precision alignment. Maybe if you were planning to DSLR scan hundreds of transparencies at high resolution and wanted to minimize digital processing time and reduce necessary overlap it could help, but I only do multi-shot digitization occasionally.

 

[jtk]

I’ve seen quite a few discussions about building precision alignment easels for DSLR scanning, but haven’t found them necessary myself. Photomerge in Photoshop is designed to align misaligned frames digitally. Get it in the ballpark, and Photomerge can do the precision alignment. Maybe if you were planning to DSLR scan hundreds of transparencies at high resolution and wanted to minimize digital processing time and reduce necessary overlap it could help, but I only do multi-shot digitization occasionally.

That's reasonable...probably best for most panos etc. However if a person wants to scan multiple sections of subject slide/slides, put them all in one file for purposes other than pano, precision becomes important.

 

[David A. Goldfarb]

That's reasonable...probably best for most panos etc. However if a person wants to scan multiple sections of subject slide/slides, put them all in one file for purposes other than pano, precision becomes important.

I’m usually doing this to get a hi-res scan of a neg or transparency, so I might use a 35mm macrophoto lens on a DSLR to take 9 shots of a 35mm slide to get a stitched file of around 5000 ppi. This is easier for Photomerge than stitching something like an in-camera landscape pano made by rotating the camera on a conventional tripod head, since the slide is flat and doesn’t require any perspective correction.

 

[ssssjjjj]

Share my current experience of multi-shot film.

1. Determine the focal plane
I think this is the most important step. Once you have set up the focal plane, you don't need to make focus stacking.

2. Move the film instead of camera

3. A copy stand that can be fine-tuned (Z-axis)

4. A lens with a flat image field and good in distortion control

5. Even light source
Since the 4x5 film is much larger than 135 or 120, if your light source is uneven, it will affect the result clearly.
This will be obvious in the bright part(sky) of film.

Now I only need to take 6 photos to merge a 300MP 4x5 film, no need focus stacking.