Scanning 35mm Slides

On Sat, 26 Jun 2004 16:49:27 -0400, "Michael A. Covington"
wrote:


"Roger Halstead" wrote in message
.. .

My Coolscan III captures everything that's on the slide. In fact I
usually
use it at half resolution (1350 dpi) to hide grain. Remember that
digital

2700 dpi should not be capable of capturing any where near all the
information on the slide unless it's very high speed film.

2700 dpi is more than 100 lines per mm, more than 50 line pairs per mm.

100 lines is only about 33 line pair. There has to be a space between
the lines detectable space between the lines.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

In article , Roger Halstead
writes
On Sat, 26 Jun 2004 16:49:27 -0400, "Michael A. Covington"
wrote:


"Roger Halstead" wrote in message
. ..

My Coolscan III captures everything that's on the slide. In fact I
usually
use it at half resolution (1350 dpi) to hide grain. Remember that
digital

2700 dpi should not be capable of capturing any where near all the
information on the slide unless it's very high speed film.

2700 dpi is more than 100 lines per mm, more than 50 line pairs per mm.

100 lines is only about 33 line pair. There has to be a space between
the lines detectable space between the lines.

No, 100lpmm is 50lp/mm *not* 33lp/mm. That is an old chestnut which is
completely misleading and usually only raised by people who actually do
know better as a means of testing whether the person has learned the
meaning of the two parameters correctly. It is a bit like the old joke
of giving away $10 out of $20 by counting down and then adding the
remainder to "prove" that there were $21 to begin with.

The lines and line pairs refer to measurements over extended distance.
Your estimate assumes only 3 lines and is still inaccurate since the 3rd
line is actually half of the next line pair. For example, if there were
5 lines in the test target, then your method would conclude 2 line pairs
and hence a line pair would be 40% of the number of lines, not 33.33%, 7
would be 3 line pairs, concluding a line pair as 42.86%, 9 would result
in 44.44% and so on. Thus the conversion figure produced by your method
is inconsistent, it is a function of the size of the test target and
hence number of lines present, but converges on 50% as the number of
lines increases. However, for measuring resolution, such a conversion
is useless because the same limiting resolution results in a different
measured line pair per mm depending on how many lines were present in
the first place.

Counting the 3rd, 5th, 7th and subsequent line in any sequence as part
of the *next* line pair means that 3 lines corresponds to 1.5 line
pairs, 4 to 2pairs, 5 to 2.5pairs, 6 to 3pairs and so on - a fixed
conversion ratio of 50% which is independent of the distance over which
the test pattern extends. Thus any resolution measured is also
independent of the number of lines or the size of the test target being
used, which is essential for a meaningful measurement.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's ****ed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)

On Sat, 26 Jun 2004 16:49:27 -0400, "Michael A. Covington"
wrote:


"Roger Halstead" wrote in message
.. .

snip
How things look at 2700 or 1350 depends more on the physical
characteristics of the scanner, but you should not have to scan at
half resolution to hide grain unless it's coming from the scanner and
not the slide.

It's real film grain (Ektachrome 200, typically).


Real film grain is not uniform and will have patterns. I have a good
many examples from ASA 400 I've scanned and the irregularities are
quite prominent at high magnigication. What I see from the 4000 dpi
scans of ASA 100 and Even Kodachrome 25 is a very uniform grain.
There are no patterns discernable. (Kodachrome 25 with grain?)

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

Again, I agree with Phillip that scanning at a higher resolution than
needed and down sampling gives better results than scanning at a
lesser resolution.

In theory, yes. But it depends on the MTF of the entire system. I get
satisfactory results for most purposes -- not super-large exhibition
prints -- by scanning at half resolution on my Coolscan III.


Clear skies,

Michael A. Covington
Author, Astrophotography for the Amateur
www.covingtoninnovations.com/astromenu.html

On Sun, 27 Jun 2004 08:58:00 +0100, Kennedy McEwen
wrote:

In article , Roger Halstead
writes
On Sat, 26 Jun 2004 16:49:27 -0400, "Michael A. Covington"
wrote:


"Roger Halstead" wrote in message
...

My Coolscan III captures everything that's on the slide. In fact I
usually
use it at half resolution (1350 dpi) to hide grain. Remember that
digital

2700 dpi should not be capable of capturing any where near all the
information on the slide unless it's very high speed film.

2700 dpi is more than 100 lines per mm, more than 50 line pairs per mm.

100 lines is only about 33 line pair. There has to be a space between
the lines detectable space between the lines.


First I want to thank you for the information. Back in college
photography classes they just glossed over the line pair definition
and gave it to us pretty much the way I quoted. Most likely as that
wasn't part of the course goals it hadn't been ... researched, or part
of the prepared materials.

No, 100lpmm is 50lp/mm *not* 33lp/mm. That is an old chestnut which is
completely misleading and usually only raised by people who actually do
know better as a means of testing whether the person has learned the
meaning of the two parameters correctly. It is a bit like the old joke
of giving away $10 out of $20 by counting down and then adding the
remainder to "prove" that there were $21 to begin with.

snip good information

The other is the grain issue.
I use a Nikon LS5000 ED.
It produces a very fine grain like texture with extreme enlargements.
But, this time I decided to go farther.

Using Fuji 5-100 ( ASA 100 negatives) I picked an image with a lot of
sky as I find grain clumping is easier to find in a lightly tinted
area of a constant color. Then with the grid turned on I started
stepping up the view.

With 47 grids corresponding to the roughly one inch dimension of the
negative that makes each grid approximately 0.54 mm across.

At the point where each grid is 1/2 inch wide on the screen the image
takes on a grainy appearance, but it is very uniform with no patterns.

This time I took it one step farther which put the grid at 13/16
(0.81) inch wide. At this magnification the grain patterns and
clumping do become evident "if you are looking for them". As there
are 25.4 mm in an inch and I have one mm spread across 0.8 inches
(rounding off as I can't measure that close any way) It takes
roughly 20X enlargement before the grain becomes evident. Near as I
can tell the grain is one tenth to one twentieth the distance across
the grid. The Fuji and Kodak ASA 400 grain is much larger and shows
plainly with much less magnification.

When the grain, or the appearance of a very uniform grain turned up I
had quit looking farther, but it only took about 50% more
magnification to bring out the real grain.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

In article , Roger Halstead
writes

First I want to thank you for the information. Back in college
photography classes they just glossed over the line pair definition
and gave it to us pretty much the way I quoted.

I am shocked to hear that - but on perusal, perhaps I shouldn't be. I
well recall many lecturers using terms like "obviously" or "it is
trivial to show that" when they didn't really understand the issues
involved themselves. Often, you will find that explaining something to
someone else helps to make you understand it better yourself, so I hate
using such glossy terms.

The other is the grain issue.
I use a Nikon LS5000 ED.
It produces a very fine grain like texture with extreme enlargements.
But, this time I decided to go farther.

Using Fuji 5-100 ( ASA 100 negatives) I picked an image with a lot of
sky as I find grain clumping is easier to find in a lightly tinted
area of a constant color. Then with the grid turned on I started
stepping up the view.

With 47 grids corresponding to the roughly one inch dimension of the
negative that makes each grid approximately 0.54 mm across.

At the point where each grid is 1/2 inch wide on the screen the image
takes on a grainy appearance, but it is very uniform with no patterns.

This time I took it one step farther which put the grid at 13/16
(0.81) inch wide. At this magnification the grain patterns and
clumping do become evident "if you are looking for them". As there
are 25.4 mm in an inch and I have one mm spread across 0.8 inches
(rounding off as I can't measure that close any way) It takes
roughly 20X enlargement before the grain becomes evident. Near as I
can tell the grain is one tenth to one twentieth the distance across
the grid. The Fuji and Kodak ASA 400 grain is much larger and shows
plainly with much less magnification.

When the grain, or the appearance of a very uniform grain turned up I
had quit looking farther, but it only took about 50% more
magnification to bring out the real grain.

With Photoshop? I am pretty sure that this creates the zoomed views
from precomputed cached scaled copies, which have been filtered
reasonably correctly for the appropriate scale. So you may well find
that the single step in magnification changes the cached image the
display is produced from.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's ****ed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)

I dont often reply to posts here but I happen to be dealing with slides
right now so maybe something I have to say will help you.

Getting slides developed can be tricky with a mass market store. Pro labs
are much better and if you shoot or plan to shoot a lot you can cut costs by
only having them develop. Then you can mount them yourself which will insure
quality. I would recomend Dale Labs in Hollywood Florida or other similar
full feature pro labs.

I use the Minolta Scan Dual III and it is great. I scan my slides myself
although I do get them printed at a pro lab anyway. I do my own scanning so
I can be precise with settings and import directly into photoshop.

If I can offer any more help drop me an email.

Rick
"MATT WILLIAMS" wrote in message
...
In the past Costco has been able to scan my slides and make prints and
enlargements for a very reasonable cost about three dollars for a 12X18.
They have a new processing machine which has done a great job on my
Digital
Enlargements and my own medium format scans that I do on my old Epson
2450.
The Epson 2450 is just not cutting it on the 35mm slides However, my
slides
come out with bad color balance and fuzzy with the new machine from
Costco.
They say that they have tried to get it fixed, but because there are so
few
people with slides now that it is not a priority. I know that there are
other photo places that I can use, but they are much more expensive. I was
wondering if the Minolta IV scanner at 3200dpi would give good enough
results to have a slide enlarged to 12x18. I am shooting Kodak VS100 slide
film. Thanks for the info. Matt

On Thu, 1 Jul 2004 14:54:17 -0400, "Rick Davis"
wrote:

I dont often reply to posts here but I happen to be dealing with slides
right now so maybe something I have to say will help you.

Getting slides developed can be tricky with a mass market store. Pro labs
are much better and if you shoot or plan to shoot a lot you can cut costs by
only having them develop. Then you can mount them yourself which will insure
quality. I would recomend Dale Labs in Hollywood Florida or other similar
full feature pro labs.

I went to purchasing film in 100 foot rolls and processing them
myself.
I scan them as film strips (rolls are a pain to store) and mount only
the ones I want to mount which are few.

Using the three step process it takes about 10 minutes from start to
hanging out to dry.

I use both NikonScan and ViewScan with a Nikon LS5000 ED scanner.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

I use the Minolta Scan Dual III and it is great. I scan my slides myself
although I do get them printed at a pro lab anyway. I do my own scanning so
I can be precise with settings and import directly into photoshop.

If I can offer any more help drop me an email.

Rick
"MATT WILLIAMS" wrote in message
. ..
In the past Costco has been able to scan my slides and make prints and
enlargements for a very reasonable cost about three dollars for a 12X18.
They have a new processing machine which has done a great job on my
Digital
Enlargements and my own medium format scans that I do on my old Epson
2450.
The Epson 2450 is just not cutting it on the 35mm slides However, my
slides
come out with bad color balance and fuzzy with the new machine from
Costco.
They say that they have tried to get it fixed, but because there are so
few
people with slides now that it is not a priority. I know that there are
other photo places that I can use, but they are much more expensive. I was
wondering if the Minolta IV scanner at 3200dpi would give good enough
results to have a slide enlarged to 12x18. I am shooting Kodak VS100 slide
film. Thanks for the info. Matt