dorsey film conversion questions and estimate

[Pat Warren]

Thanks Rory, Mark, and Tom

That gives a real good idea of what we've got.

If I were to do the digital work, I'd think I'll go with this ballpark
guess: 5x7 - around 50kb ea., 20,000x.05= 1GB, 2 CDs

> So although the handwritten cards are hard
> to read, it probably doesn't matter too much as
> they are only duplicates of words on the type-written
> cards anyway.  If we suppose each word is represented
> by both a handwritten card and a type-written card,
> we should estimate about 10,000 words in the
> collection. (Rory)

Well, I'd be scanning everything in there. You don't want to redo work like
this, so I prefer to get it all in one go and preserve as much integrity of
the collection as possible. So the number of cards is what I'm looking at
now, redundant or not. But once it's digitized, then it's all about the
content and making it useful, while still having the reproduction  (of the
reproduction) of the collection to refer to, as it is physically
structured.

> Each frame includes typically 4, but often fewer, index card images.
(Rory)

> The lens on my recycled microfilm reader is not marked for magnification.
> 1-1/2 frames appears on each screen. However, I move the carriage from
side
> to side to focus directly on one card within a four card frame for best
> clarity. (Mark)

Well, if there are generally multiple cards per frame, then this will go
way faster. Microfilm resolution gives about 200dpi output at max. When
digitizing you don't zoom in on each card on the microfilm reader, you just
take the whole frame as it fills the image capture area. Zooming in doesn't
actually give you better quality, it just gives the same effect as scanning
the whole frame at once and zooming in on the digitized images on the
computer.

> The images are black with white lettering (t.s. and m.s.) (Mark)

> If it's of any help, I ordered my 8 Dorsey reels with "reverse imaging",
that is black letters on a white background. They were, in my opinion, far
easier to read that > way. Perhaps the reverse imaging would enhance the
scanning process? Don't know. If so, I have them available. (Tom)

Well, the images would be archived as positive to have the smallest file
size, but if it's easier to read it as negative it could be converted to
negative jpg's for display. It doesn't make any difference whether the film
is positive or negative for scanning quality or image size.

> For total frames, I think we can figure about
> 5,500.

I have now learned that my math skills when sleep deprived are rather
fanciful. At least I used a calcumalator in figuring out the MB.

Should have been 20000 images / 100 images per session = 200 sessions / 4
sessions per week = 50 weeks. Though now I'm more sleep deprived than
earlier, so maybe I should redo this later. It's a good thing the slips
weren't filmed individually! Well, I'll try again with the 5500 number:

5500 images / 100 images per session = 55 sessions / 4 session per week =
14 weeks / 4 weeks per month = under 4 mo.

So I'd say for the Dhegiha film with around 20,000 images in 5,500 frames
it would take four months to scan and process and we'd have the whole thing
on 2 CDs.

Pat