One screenshot of SW part for something larger than large

my gradual development

Hope that it is understandable

Luisifer wrote:
my gradual development

Hope that it is understandable

Guess I'm being slow here.

Stool manager?

It is software part of device automatically digitizing large format sheet film via DSLR with macro lens (MP-E) piecewise (on screenshot in matrix of rectangles one rectangle = one shot / digitized part)

and one draft of "stool bones"

- free space from bottom is for strobe light
- top construction will hold the camera with macro lens
- "gold board" will hold the sheet film movements (stepper motors, linear guideways, gear racks, etc.)

I'm also curious.

And with some light modification (and glass holder) there will be possible to digitize smaller formats too.

To the draft i added the most expensive parts. It is necessary to save a bundle to them but hope some months later they will be sent to me from the store

Soooo ... this project is intended to be a replacement for a "drum scanner" for medium and large format film?

Interesting, and hope it works out.

At the beginning there were idea to digitize (only) sheet film with available equipment (ff dslr + flash + macro lens + computer) with relatively minimal additional budget.

When i tested it "on my knee" i found out that i can not move the sheet by hand and it is necessary to automatize it.

And with automation it is possible to do it in "extreme mode" with making of thousands shoot parts from one sheet with the 5:1 magnification.

The unnecessarily large resolution from this digitization will be used for "transfer" to high bit depth via downsampling.

test "on my knee":

Some news:
now i know the maximal size of possible digitized sheet film. This size is 12x20" given by length of linear guideways which i just ordered.

one grouping for an example

hands test...

I found that there is necessary to exchange one trolley from ZA version to ZF version without passive resist

and get couple of better cables...

two parallel rails are already fastened to the base plate:

Some week later

Two more photos on the end of the week

And one more screenshot of updated webapp.

changes for some parameters to enumerated (like magnification), changes of way back from time parameter to speed parameter, etc.

On this screenshot is setting for 12x20" sheet film digitized on 5:1 magnification with number of frames and expected duration of digitizing.

First test via mobile:

http://youtu.be/dgYHI-oEI34

Most people scan in a simple raster, but that can cause issues....

When I do automated scanning (e.g., using a camera on a computer-controlled telescope mount), I always use a space-filling curve with higher spatio-temporal locality rather than a raster. Think Hilbert/Peano or Z curves.  Just a suggestion....  I think Gigapan still uses a simple raster, so if you don't wnat to deal with the fancy scan orders, you're still in good company.  

I expect to use the raster (x1, y1; ...; xn, y1; x1, y2; ...; xn, y2; ..... ; xn, yn). But there is no problem to add some other orders if these orders will be better.

Which issues do you mean?

I expect to use autopano giga and files will be with applicated flat frame.

Luisifer wrote:

I expect to use the raster (x1, y1; ...; xn, y1; x1, y2; ...; xn, y2; ..... ; xn, yn). But there is no problem to add some other orders if these orders will be better.

Which issues do you mean?

I expect to use autopano giga and files will be with applicated flat frame.

It's more critical with a potentially changing scene, but even with a fixed scene and lighting you can accumulate small positioning errors and lighting can slowly change (due to bulb life, heating, etc.). Using the kinds of orders I suggested minimizes the impact of such variability.

Vacuum tube has about 100k flashes life period.

I expect flash light (but studio flash is still only on budget queue) with power controlled over application where will be one flat frame for each row => new flat frame after about 100 frames max. (after 50 frames on average)

And flash unit about 500Ws with enough distance from hole with part of digitized film => minimal (or without) temperature impact (power of the flash set close to minimal and active cooled flash unit).