Olympus-OM
[Top] [All Lists]

Re: [OM] Scanning Question

Subject: Re: [OM] Scanning Question
From: Moose <olymoose@xxxxxxxxx>
Date: Tue, 14 Jan 2014 16:20:42 -0800
On 1/14/2014 2:16 PM, Jon Mitchell wrote:
> Hi Tina,
>
> Thanks for the quick reply.  That was my initial thinking too, but over
> 200MB per scan seems high.

It's all very straightforward. 24x36 mm @ 5400 dpi = 39 Mpixels. Times 3, for 
RGB, is 117 Mdata points. at 8 bits, 
that's 117 MB. Double, two bytes for each value, goes to 234 MB. Add an IR 
channel, and you get 312 MB.

Based on my file sizes, I'm guessing the IR channel is always 8 bit; that's all 
that needed. That would give a total RAW 
size of 273 MB. Actual sizes are generally slightly less for negs, and even 
smaller for mounted slides, with the image 
area lost to the mount and crop settings for mounts with rounded corners.

> Can I ask, what sort of file sizes you get for your scans ?

I just posted them. They are in line with the above math, for 4000 dpi.

> I would welcome any else's comments on my scanner, but I believe it scans at
> 5400 dpi rather than interpolating.

True

> In VueScan I ...
>
> I understand the logic of greater resolution, and greater bits per pixel,
> meaning bigger file sizes.  One bit that I can't fathom is how greater bits
> per pixel makes it considerably sharper.  That one has me scratching my
> head.

One may theorize, but pragmatic experience trumps all theory. It may, for 
example only apply to that Minolta, perhaps 
with that particular software. I've not noticed that with my Canon, but then 
I'm not sure I've ever scanned at 8 bit. 
:-)   But it's there, so one uses it.

I do know that with really good original color negs, multiple scanning passes 
can add additional fine detail with my 
scanner, but the practical difference is small.

The other thing, since you mention older, family heirloom slides, is that a lot 
of those older slides will have less 
detail. I would still scan them at the full 5400 dpi, and process them 
otherwise the same.

Then, say looking at a fair to large number of same era, lens quality, film, 
etc. find a combo of downsampling, NR and 
resharpening that retains all the actual detail on the film, while 
significantly reducing file size. You could then 
apply that formula as a batch process.

As Ken and Tina say, there's no point in throwing away potentially useful data 
in choosing dpi. OTOH, once one knows how 
much actual detail data is there, it makes sense to throw away meaningless 
extra data.

I determined that some late '30s KR slides from my parents' archive have no 
more than 12 MP of data, or a file size of 
about 36 MB.

How Big Moose

-- 
What if the Hokey Pokey *IS* what it's all about?
-- 
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz