I want to use an R-D1 as a poor-man's digital back for my view camera...

[Audii-Dudii]

as well as occasional hand-held use with a couple of CV lenses, but before I buy one, I'm doing some research to make sure the R-D1 will actually work when used in this manner.

My plan is to compose 6x4.5 images on my Toyo 23G's ground glass, then swap it for the R-D1 body and use the Toyo's rear shift and rise movements to capture a series of eight images that cover the 6x4.5 composition. Once I return home, I can then reassemble the images in Photoshop to create the final 25+MP image. In addition to my large-format lenses, I also plan to use a couple of Bronica GS-1 lenses, which have a sufficiently large image circle (6x7 and then a bit) and enough resolution to work well for this application, and are available inexpensively enough these days to fit my budget.

I originally planned to use a Leica M8 and purchased a used one, but it arrived DOA and the seller has refunded my money accordingly. However, I'm having second thoughts about spending $3700+ on an M8 when an R-D1 will cost half as much (if that!) and by all accounts, should work more than half as well an M8.

So, what I need to confirm is that 1) the R-D1 shutter can indeed be fired without a lens attached; 2) the shutter speeds are both accurate and repeatable enough so that eight images shot over approx. one minute won't vary in exposure by more than 1/10th of a stop or so; and 3) there's nothing else unique about the R-D1 that will prevent it from being used in the manner I've described above.

Thanks in advance in for any and all input! 🙂

 

[Uwe_Nds]

I am just thinking: the sensor will be some 28 mm behind the ground glass. How did you consider this?

Best regards,
Uwe

 

[wontonny]

I am just thinking: the sensor will be some 28 mm behind the ground glass. How did you consider this?

Best regards,
Uwe

Yeah, good luck.

 

[jmkelly]

I don't see why the 28mm can't be factored in with a spacer for the gg. Yes, the shutter will fire without a lens attached. I can't speak to the 1/10th stop reproducibility.

 

[Audii-Dudii]

I am just thinking: the sensor will be some 28 mm behind the ground glass. How did you consider this?

I'm not sure exactly what you mean, but it's not too difficult to space the Toyo's ground glass backward by however much is necessary to position it in the same plane as the R-D1's sensor. It will require a bit of trial-and-error to fine-tune, of course, but it's not an impossible task by any means. Also, the fact that the Toyo's rear standard will accept a flat aluminum panel in place of its OEM ground-glass assembly, which itself spaces the ground glass back approx. 10mm from the mounting surface, means that I'll only need to space it back an additional ~18mm rather than the full ~28mm distance. Also, remember that the lens register for the GS-1 lenses is more than 80mm, so there will be plenty of room between the camera body and the rear element of the lens.

FYI, my last attempt at this project was to use a DMC-L1 body and compose my images on the LCD using its Live View facility. Although this approach worked, I was trying to use Leica R SLR lenses and the difference in lens register between the body (~39.7mm) and the lenses (47.5mm) simply didn't leave enough room for the rear elements of the wide-angle lenses I needed to use to compensate for the 4/3 format's 2x focal length multiplier. However, because with the R-D1, I will be shooting several images and stitching them together, instead of shooting a single image as I was doing with the DMC-L1, the focal length multiplier is now a non-issue and I won't need to use wide-angle lenses that are half as long in order to achieve wide-angle compositions.

Oh, and if it turns out to be necessary, I also have 5x7 and 8x10 Toyo rear standards, which will allow me to position the body ahead of the standard by the appropriate distance rather than spacing the ground glass further back. However, this will be much more complicated than mounting the body to a flat panel via an LTM->M bayonet adapter and it will also result in my having to carry a much larger camera around, so it isn't my first choice.

 

[Audii-Dudii]

I don't see why the 28mm can't be factored in with a spacer for the gg. Yes, the shutter will fire without a lens attached. I can't speak to the 1/10th stop reproducibility.

Thanks for the info. I assumed as much, but as I've never been able to put my hands on an R-D1, let alone see one in person, I wanted to confirm this before I laid out the cash.

 

[iamzip]

hm, interesting idea and it might work in theory, but since you're planning on taking several snapshots and reassembling thwm in photoshop, what would the advantage be over simply using the Epson with its intended lens to take several pictures to reassemble in photoshop? That seems like it would be much simpler and yield results at least as good if not better than what you are proposing.

 

[squirrel$$$bandit]

hm, interesting idea and it might work in theory, but since you're planning on taking several snapshots and reassembling thwm in photoshop, what would the advantage be over simply using the Epson with its intended lens to take several pictures to reassemble in photoshop? That seems like it would be much simpler and yield results at least as good if not better than what you are proposing.

That's what I was thinking--it seems like a solution in search of a problem.

That said, I like weird projects like this...

 

[Audii-Dudii]

hm, interesting idea and it might work in theory, but since you're planning on taking several snapshots and reassembling thwm in photoshop, what would the advantage be over simply using the Epson with its intended lens to take several pictures to reassemble in photoshop? That seems like it would be much simpler and yield results at least as good if not better than what you are proposing.

When you re-aim the camera and take additional shots (or move the front standards on a view camera), as you propose, you are also changing the relationship of the lens to the subject, which creates parallax effects that the stitching software must compensate for in order to stitch the images together seamlessly. This means the stitching software must manipulate the image pixels around each seam and in my experience, whenever this happens, fine details are lost.

However, when you use the rear movements on a view camera to reposition the camera to capture different parts of the same image circle being projected by the lens, the relationship of the lens to the subject doesn't change. With each successive image, you're simply capturing different "slices," if you will, of the image circle. As a result, you don't need stitching software to manipulate the image pixels in order to blend the separate images together into a seamless whole, you can simply layer them one on top of the other in Photoshop, line them up using the difference mode to an accuracy of within a few pixels, and flatten them. No image pixels need to be manipulated this way and as a result, the maximum amount of fine detail is preserved at every point of the image.

While it may or may not matter whether the pixels have been manipulated for most images, for the sort of stuff I like to shoot, fine details are critical, not just important. Mind you, I'm coming to digital capture not from the perspective of 35mm film and 11x14 prints, but from 4x5 and 8x10 film and 24x30 and 40x50 prints. The original reason for this exercise was to find a way for me to make quicker, less costly "sketches" while I was shooting with my 8x10 -- at $12+ per image for film and processing, being creative becomes very costly, very quickly! -- but as digital capture technology has improved in quality and come down in price over the past few years, I'm now looking at it as a way to replace shooting film altogether.

This will be my sixth (and final!) attempt to make a poor-man's digital view camera and if it doesn't work, then I'll give up and wait for today's 20+MP single-shot digital backs to depreciate into the mid-five figure range. We'll see...

 

[tomkai]

I have used my Canon DSLRs on a Sinar to do the same thing. It works very well with studio macro shots where the distance between the rear lens and the camera-body's front is long enough.
The SLR has the advantage that I can focus through the camera's viewfinder and I don't have to change between the ground glass and the "back" and re-focus. On the other side I have to move the back around to compose, and I can never see the whole picture. But then, while stitching the files together in PS, it is fun and pleasure to see how the picture grows and grows. And the prints from these files are incredible.

I can't see a technical reason why it should not work with the RD-1. But one thing you should keep in mind when you shoot outdoors: a possible change of light and a movement of your subject (i.e. clouds) between the single shots. Even if your fast, there's some time between the first and the last exposure. That's the reason why I use mine only for studio shots.

Tom

 

[uhligfd]

Duct tape will hold everything together, I think.

 

[Audii-Dudii]

The SLR has the advantage that I can focus through the camera's viewfinder and I don't have to change between the ground glass and the "back" and re-focus.

If I get everything just right, then no re-focusing will be necessary when I swap the ground glass for the camera body.

On the other side I have to move the back around to compose, and I can never see the whole picture. But then, while stitching the files together in PS, it is fun and pleasure to see how the picture grows and grows. And the prints from these files are incredible.

By composing in a 6x4.5 frame imposed upon the ground glass, I will be able to see the whole composition before I capture slices of it with the camera body. I tend to compose my images fairly precisely, right out to the edges, which is another reason the usual panorama-type approach where the final image can't be seen until all of the pieces have been stitched together won't work for me.

But one thing you should keep in mind when you shoot outdoors: a possible change of light and a movement of your subject (i.e. clouds) between the single shots. Even if your fast, there's some time between the first and the last exposure. That's the reason why I use mine only for studio shots.

I agree that this is potentially a problem, as will be variations in exposure due to shutter-speed tolerances. Fortunately, very little of my images include any sky or clouds and my test runs thus far have confirmed that I should be able to get the necessary eight images in approx. one minute, which should be quick enough most of the time. (By comparison, when I was using the Betterlight scanning back, the single exposure sometimes took as long as three minutes, which was too long for my purposes.)