"Shough, Dean" wrote:
> The optics in a scanner cause an image to be formed on the scanner's CCD.
> This image will have various aberrations and diffraction artifacts that
> cause the image to not be a faithful reproduction of the original piece of
> film.
I'd call that things like "distortion",
Mike wrote:
>Exactly why one might use an anti-aliasing filter ahead of the CCD which
>I and others have mentioned (and which one person called "cheating" which
>it is, in the sense of the winner of a race "cheated" by running faster).
When I was reading something someone else wrote on this topic
I agree. I do find the information fascinating. How do I apply it?
Dave Dillon
-Original Message-
From: Bond, Alistair [mailto:[EMAIL PROTECTED]]
Sent: 06 December 2000 11:02
To: '[EMAIL PROTECTED]'
Subject: RE: filmscanners: RE: Film Scanners and what they see.
>>D
Dean Shough wrote:
> Mathematically, for a linear-shift invariant system,
> the image seen by the CCD is:
> image ** pixel ** sampler
> where '**' is a two dimensional convolution, 'image'
> is the image at the CCD, 'pixel' is the response of a
> single pixel due to its size and shape, and
> Alaising and filtering are two completely different animals. (Sounds of
> someone ducking underneath his desk)
Regrettably not, as you seem to have missed that the CCD itself produces aliased pixel
values as a result of this filtering. You are looking at things only in the frequency
domain
> My god!
>
>
> *** REPLY SEPARATOR ***
>
> On 05/12/00 at 12:54 Shough, Dean wrote:
>
> >Alaising .
> .
> >(sin(x)/x)^2 with the first zero occurring at the pixel separation.
Quite!. Let's all just try and get out more :-)
Regards
Tony Sleep
http://www.halftone.c
> Exactly why one might use an anti-aliasing filter ahead of the CCD which
> I and others have mentioned (and which one person called "cheating" which
> it is, in the sense of the winner of a race "cheated" by running faster).
It doesn't buy you any more information in your scan, all it does is a
> > Alaising and filtering are two completely different animals. (Sounds of
> > someone ducking underneath his desk)
>
> Regrettably not, as you seem to have missed that the CCD itself produces
> aliased pixel
> values as a result of this filtering. You are looking at things only in
> the fre
> > My god!
> >
> >
> > *** REPLY SEPARATOR ***
> >
> > On 05/12/00 at 12:54 Shough, Dean wrote:
> >
> > >Alaising .
> > .
> > >(sin(x)/x)^2 with the first zero occurring at the pixel separation.
>
> Quite!. Let's all just try and get out more :-)
>
Sorry. I will (
Tony Sleep wrote:
>
> > Exactly why one might use an anti-aliasing filter ahead of the CCD which
> > I and others have mentioned (and which one person called "cheating" which
> > it is, in the sense of the winner of a race "cheated" by running faster).
>
> It doesn't buy you any more information
Rob Geraghty wrote:
>
> Mike wrote:
> >Exactly why one might use an anti-aliasing filter ahead of the CCD which
> >I and others have mentioned (and which one person called "cheating" which
> >it is, in the sense of the winner of a race "cheated" by running faster).
>
> When I was reading somethi
CD player do NOT oversample. The input data is at 44.1kHz. What they do
is interpolate the data at typically 8x the input frequency. That is NOT
oversampling, it is interpolating. What that does is minimize the
requirements on the analog output filter design.
--
> I couldn't
> hel
> The oversampling business in CD players is mostly a method
> to save as much as maybe a dime in their production costs to reduce the >
> cost of the analog output reconstruction filter.
Not quite. There is no oversampling in a CD player, it is interpolation.
And it's not primarily money, i
It's basic sampling theory that I was taught in the early
70's and still holds today.
As an aside, this reminds me of the caption of a cartoon:
"It was a long time ago, but at the time it seemed like the
present."
Bill Ross
Austin wrote:
> CD player do NOT oversample. The input data is at 44.1kHz.
> What they do is interpolate the data at typically 8x the
> input frequency. That is NOT oversampling, it is
> interpolating. What that does is minimize the
> requirements on the analog output filter design.
I know it
>Austin wrote:
>> CD player do NOT oversample. The input data is at 44.1kHz.
>> What they do is interpolate the data at typically 8x the
>> input frequency. That is NOT oversampling, it is
>> interpolating. What that does is minimize the
>> requirements on the analog output filter design.
>I k
Austin wrote:
>>My point was - I wonder whether digital interpolation would be useful
in
>>a scanner design to smooth the output?
> Every scanner I know does offer interpolation as how they do their
> higher than native output... Why would you want to smooth the
> output? If you wanted to do tha
>Putting a smoothing function into the scanner's own interface would be
> much simpler from the user perspective.
Simpler, yes, but how many people would actually use it?
> Why would you want to do it... haven't we been talking about increased
> apparent grain for a couple of years now?
How doe
Austin Franklin wrote:
>
> > The oversampling business in CD players is mostly a method
> > to save as much as maybe a dime in their production costs to reduce the >
> > cost of the analog output reconstruction filter.
>
> Not quite. There is no oversampling in a CD player, it is interpolation
> My point was - I wonder whether digital interpolation would be useful in
> a scanner design to smooth the output?
That's what anti-aliasing filters in image editing software do, interpolate pixels to
achieve smoothing.
Regards
Tony Sleep
http://www.halftone.co.uk - Online portfolio & exhibi
> Further, at least at first,the
> "oversampling" CD players were low end units
That's not quite true, they were mid range units, and it was because the
initial interpolation filters were quite bad, and were only 2x to 4x, and
certainly did not meet the audio quality that was achievable withou
Rob Geraghty wrote:
> When I was reading something someone else wrote on this topic I couldn't
> help wondering about the kind of oversampling used in CD players to filter
> the output. I wonder if similar technology could be used to smooth the
> output from a scanner - maybe some scanners alrea
Austin wrote:
>Rob Wrote:
>>Putting a smoothing function into the scanner's own interface would be
>> much simpler from the user perspective.
>Simpler, yes, but how many people would actually use it?
I would for one! Last I checked, I wasn't the only one appalled by the
coarse "grain" appearing
Tony wrote:
>Rob Wrote:
>> My point was - I wonder whether digital interpolation would be
>> useful in a scanner design to smooth the output?
>That's what anti-aliasing filters in image editing software do, interpolate
>pixels to achieve smoothing.
Sure, but as I pointed out elsewhere:
1) There's
on 12/7/00 3:51 PM, Rob Geraghty at [EMAIL PROTECTED] wrote:
> Do the filter in Photoshop work with 16bit data yet?
The important ones (Gaussian Blur, Add noise and most importantly Unsharp
Mask) sure do in Photoshop 6!
Andrew Rodney
Pete wrote:
> Oversampling in audio is a hardware implemented function, which is
> really only of any worth when it's applied at the recording stage.
> The primary use is to make the 'brick wall' low-pass filter more
> effective (and cheaper to make) prior to the A to D conversion stage.
Er. Sor
Rob Geraghty wrote:
>
> Mike wrote:
> > Well, not really. The oversampling business in CD players is mostly
> > a method to save as much as maybe a dime in their production costs to
> > reduce the cost of the analog output reconstruction filter.
>
> Regardless of why they use a particular techn
Austin Franklin wrote:
>
> > Further, at least at first,the
> > "oversampling" CD players were low end units
>
> That's not quite true, they were mid range units, and it was because the
> initial interpolation filters were quite bad, and were only 2x to 4x, and
> certainly did not meet the audi
Rob Geraghty wrote:
> >A researcher at Microsoft has shown, crudely, how oversampling can be
> > used to reduce aliasing in digital HDTV without increasing the
> > bandwidth to accomodate high sampling rates
>
> I'd be interested to know how this was done.
Much the same as I described for scan
Pete wrote:
>Much the same as I described for scanners. More bandwidth and samples
> at the camera end, then downsampled before recording/transmission.
Oh. I didn't realise you were talking about a system that required a change
in the signal from the camera onward.
>> Unfortunately, the hardwar
- Original Message -
From: "photoscientia" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, December 10, 2000 1:40 PM
Subject: Re: filmscanners: RE: Film Scanners and what they see.
>
> > > It wouldn't take much to combine an area of 4 s
toscientia" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Sunday, December 10, 2000 1:40 PM
> Subject: Re: filmscanners: RE: Film Scanners and what they see.
>
> >
> > > > It wouldn't take much to combine an area of 4 samples into one,
&
Michael Moore <[EMAIL PROTECTED]> wrote:
> [...] we professional photographers and imaging types need to remind these
> technodictators that they still haven't solved the archival problem as of
yet, at
> least from what I have been following on the CD thread...
*shrug* 50+ years from a CDR sounds
.
- Original Message -
From: "Rob Geraghty" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, December 11, 2000 11:42 AM
Subject: Re: filmscanners: RE: Film Scanners and what they see.
> Michael Moore <[EMAIL PROTECTED]> wrote:
> > [...] we profess
After what was I reading on the CD storage thread, particularly about the work
the Canadians are doing on determining the long term stability of vaious brands
of CD's, plus the other factors of recorder type, speed, labels, etc., as well
as the simple fact that they ain't no CDR that's been around
> Epson used to sell film scanners...maybe its time they got back
> into the market.
Epson withdrew because they were disappointed by sales volume with the FS200. At that
point they decided that dual-purpose flatbeds were the way to go.
Regards
Tony Sleep
http://www.halftone.co.uk - Online p
Chris McBrien wrote:
>
> Surely longevity is the key word here and not pure capacity.
>
> As the amount of stored data increases we do not want to have to spend
> a bulk of our time copying all the library CDs onto the latest media,
> we want to be creating and storing the latest
> information/m
> FYI, if you check out the archiving standards set by HABS,
> which is the national group that sets up standards for US archives, Black and
> White archivally processed silver based images are the rule...
All is not entirely rosy even there! See the several papers by Douglas Nishimura and
other
Chris wrote:
>Surely longevity is the key word here and not pure capacity.
>As the amount of stored data increases we do not want to have to spend
>a bulk of our time copying all the library CDs onto the latest media,
>we want to be creating and storing the latest information/music/photos.
Er yes
Mike wrote:
> as the simple fact that they ain't no CDR that's been around
> as long as I have and Kodachrome has, all that makes me doubt
> that CDR's are my archiving technology of choice.
If you're looking for a solution which is "write once and keep forever",
CDRs probably aren't the solution
Hi Rob.
Rob Geraghty wrote:
> Oh. I didn't realise you were talking about a system that required a change
> in the signal from the camera onward.
C'mon Rob, you're windin' me up entcha?
I'm sure you know that I meant the HDTV camera, and not our still cameras.
> Are the CCD elements small eno
The reason I made my comments was that I ( and I imagine a lot of other
photographers/historians, etc.) have to make decisions today that will affect
people in the distant future who may want or need to access the images we are
creating today. The nitrate film used for years by the motion picture
Tony...
Thanks for the link... I have known for some time that certain toners would extend the
archival life, and that you have to be careful about glass framed prints that the glass
doesn't touch the surface and they have a little breathing space... Who know? Maybe all
that's changed by now... An
Pete wrote:
>Rob Geraghty wrote:
>> Oh. I didn't realise you were talking about a system
>> that required a change in the signal from the camera onward.
> C'mon Rob, you're windin' me up entcha? I'm sure you
> know that I meant the HDTV camera, and not our still cameras.
I knew you were talking
Tony wrote:
> Epson withdrew because they were disappointed by sales
> volume with the FS200.
Given that it was really only a 1200dpi scanner, the price was hard to justify.
If they'd made something which gave "real" 2400dpi results, it would have
been much more worthwhile.
> At that point they
Laurie wrote:
> I am sorry but I fail to see how higher speed and higher
> density archival storage lessens the amount of time it
> takes to copy data from one media to another, which I
> take it was part of Chris's point.
A CDR which I write at 2X can be read at 40X. If in five years time I hav
I am sorry but I fail to see how higher speed and higher
density archival storage lessens the amount of time it takes
to copy data from one media to another,
Once the price of copying time is paid, dividends are paid
every time the image is accessed thereafter on the mo
o: <[EMAIL PROTECTED]>
Sent: Tuesday, December 12, 2000 5:07 AM
Subject: RE: filmscanners: RE: Film Scanners and what they see.
> Laurie wrote:
> > I am sorry but I fail to see how higher speed and higher
> > density archival storage lessens the amount of time it
> > takes
One caveat, although you did mention "archivally processed". Wilhelm,
the same guy now most noted for his work on inkjet inks and papers, has
for years been researching the longevity of photographic materials. In
spite of earlier reports to the contrary, it is now known that a large
number o
PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Monday, December 11, 2000 11:42 AM
> Subject: Re: filmscanners: RE: Film Scanners and what they see.
>
>
>
>> Michael Moore <[EMAIL PROTECTED]> wrote:
>>
>>> [...] we professional photographers and im
Chris McBrien <[EMAIL PROTECTED]> wrote:
> also the more images that we can store on a piece of media,
> the more 'we' are liable to loose should the media fail. I can just
> get one image from my FujiFilm MX2900 Zoom onto a 1.44MB floppy. It
> could be argued that I may just loose one ima
ECTED]>
Sent: Tuesday, December 12, 2000 2:12 PM
Subject: Re: filmscanners: RE: Film Scanners and what they see.
> Chris McBrien <[EMAIL PROTECTED]> wrote:
> > also the more images that we can store on a piece of
media,
> > the more 'we' are liable to loose
There are some companies that are specializing in saving old hardware and
software, like 5.5" floppies and readers, and using them to access the
stored data. The companies are major accounting firms (e.g. Price
Waterhouse) and the area is loosly called "computer forensics." The need is
mostly dr
- Original Message -
> From: "Rob Geraghty" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, December 12, 2000 2:12 PM
> Subject: Re: filmscanners: RE: Film Scanners and what they see.
>
>
> > Chris McBrien <[EMAIL PROTECTED]> w
now or in the future will probably not be able to afford such
services.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Gordon Tassi
Sent: Tuesday, December 12, 2000 10:57 AM
To: [EMAIL PROTECTED]
Subject: Re: filmscanners: RE: Film Scanners and what they
Art wrote:
> Only fiber based B&W silver prints (properly processed)
> are considered stable.
Ah, crud. (sorry, that's an expression of frustration - I'm not saying
you're wrong :)
This means I have to buy a print washer and find an FB paper I can actually
work with. :(
Rob
Rob Geraghty [EMAI
ortunity to introduce some lightness into
the conversation. :-)
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Rob Geraghty
Sent: Monday, December 11, 2000 11:08 PM
To: [EMAIL PROTECTED]
Subject: RE: filmscanners: RE: Film Scanners and what they see.
Laurie wro
Chris McBrien wrote:
> also the more images that we can store on a piece of media,
> the more 'we' are liable to loose should the media fail. I can just
> get one image from my FujiFilm MX2900 Zoom onto a 1.44MB floppy. It
> could be argued that I may just loose one image if a disc goes ba
nsity archival storage and may only be a phantom.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Bill Ross
Sent: Tuesday, December 12, 2000 2:36 AM
To: [EMAIL PROTECTED]
Subject: RE: filmscanners: RE: Film Scanners and what they see.
I am sorry
er 12, 2000 9:35 PM
To: [EMAIL PROTECTED]
Subject: Re: filmscanners: RE: Film Scanners and what they see.
Art wrote:
> Only fiber based B&W silver prints (properly processed)
> are considered stable.
Ah, crud. (sorry, that's an expression of frustration - I'm not saying
you&#x
Laurie Solomon wrote:
> from that media; but the copying of the 20 CDR onto the new media in five
> years from the older media even at 40X still will take a fair amount of time
> even though it will be less time per CDR than it took to originally record
> the data on a 2X CDR. However if one has
may only be a phantom.
>-Original Message-
>From: [EMAIL PROTECTED]
>[mailto:[EMAIL PROTECTED]]On Behalf Of Bill Ross
>Sent: Tuesday, December 12, 2000 2:36 AM
>To: [EMAIL PROTECTED]
>Subject: RE: filmscanners: RE: Film Scanners and what they see.
>
>
> I am sorry
Bad luck for most of the people outside this list! I could not be more
enthusiastic about a single feature of a scanner. Maybe mine works better
than some... I don't know, but the time it has saved me (even on clean negs
let alone fungus-ridden old slides) would dwarf the 50% longer scan time
I think the issue here is not that people were unaware that RC papers
were less archival than fiber based. I think this was pretty much a
given, although I do recall reading at one point about the advantages of
RC papers because they did not (in theory) absorb the thiosulfate
radicals into th
Laurie Solomon <[EMAIL PROTECTED]> wrote:
> the data on a 2X CDR. However if one has to continually recopy the same
> data over and over every 5 years with the development of new media and
> technologies (even granted an increase in speed and density) there is no
> lessening of time being spent a
Laurie Solomon <[EMAIL PROTECTED]> wrote:
> I am surprised you did not know that [RC prints] were not as archival
> as fiber based prints and that you are surprised by this. However, I
> bet you knew it all along and are just pulling out legs. :-)
Yeah, I knew that only the archival reliability
Digital storage is both a blessing and a curse. The ability to make
"virtually" identical copies means the ability to make nearly perfect
copies without generational loss, which is a great advantage in video,
for instance, where editing and copying require several generations to
be produced u
In a message dated 12/13/2000 6:41:32 AM EST, [EMAIL PROTECTED] writes:
> Bad luck for most of the people outside this list! I could not be more
> enthusiastic about a single feature of a scanner.
I have to agree - the infrared dust removal is a really nice feature
in a film scanner. It only
At 03:18 PM 12-12-00 -, you wrote:
>Rob,
>I've just had a failure, yesterday, with a Kodak Gold Ultima
>CDR. I put about 50 images on it, sealed it, so that I can add more
>later, came to re-open it and it will not allow me any more access.
>I'm just using a Yamaha 4416 (I think) with
the printing process uses little to no silver in the paper emulsions.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Rob Geraghty
Sent: Wednesday, December 13, 2000 7:27 AM
To: [EMAIL PROTECTED]
Subject: Re: filmscanners: RE: Film Scanners and what they
EMAIL PROTECTED]
Subject: Re: filmscanners: RE: Film Scanners and what they see.
I think the issue here is not that people were unaware that RC papers
were less archival than fiber based. I think this was pretty much a
given, although I do recall reading at one point about the advantages of
RC p
> The ScanWit 2740 is a special case, since it does an entire separate
> scan pass just to get the infrared channel.
Interesting. Do they just flip an IR filter in front of the CCD and use the red
channel
output then?
Regards
Tony Sleep
http://www.halftone.co.uk - Online portfolio & exhibit;
In a message dated 12/13/2000 3:18:02 PM EST, [EMAIL PROTECTED] writes:
> > The ScanWit 2740 is a special case, since it does an entire separate
> > scan pass just to get the infrared channel.
>
> Interesting. Do they just flip an IR filter in front of the CCD and use
the
> red channel outp
h does impact on such things as the
meanings of break even points and the time costs of the archiving task.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Mike Kersenbrock
Sent: Wednesday, December 13, 2000 1:58 AM
To: [EMAIL PROTECTED]
Subject: Re: fi
Scanners and what they see.
Just a quick point here with respect to the obsolescence of media. It is
true, but some backward compatibility is usually carried forward for at
least a generation. i.e., most of us can still read a 3 1/2" diskette, and
DVD will be able to read a CD, for marketing re
Arthur Entlich wrote:
> reading formats become obsolete, the other problem is that some of the
> methods of storage can be disrupted with a simple loss of a few bits of
> data, making the whole thing unreadable. A scratch in a photographic
> print rarely makes the image undiscernible, a scratch o
Ed Hamrick wrote:
> The ScanWit 2740 is a special case, since it does an entire
separate
> scan pass just to get the infrared channel.
The Minolta Elite seems to apply ICE in different ways. If you have
done a preview of a slide with ICE on, the final scan is done in a
single pass. If you h
In a message dated 12/14/2000 5:27:42 AM EST, [EMAIL PROTECTED] writes:
> The Minolta Elite seems to apply ICE in different ways. If you have
> done a preview of a slide with ICE on, the final scan is done in a
> single pass. If you have done a preview with ICE off but then switch
> it ba
On Wed, 13 Dec 2000 16:37:53 EST ([EMAIL PROTECTED]) wrote:
> I was told that the main light goes out during the infrared pass.
> I suspect it turns on an infrared LED during the infrared pass.
Ah, same consequence, different means. Cunning, and should be copied widely where
registration issu
Laurie Solomon wrote:
> I am not sure that I understand what you are saying here or how it relates
> to my points. We are talking about the historical archiving of data across
> changing technological developments in digital media on which the data is
I mean that I agree that having to change f
Hi Tony.
Tony Sleep wrote:
> > The ScanWit 2740 is a special case, since it does an entire separate
> > scan pass just to get the infrared channel.
>
> Interesting. Do they just flip an IR filter in front of the CCD and use the red
>channel
> output then?
Nope.
They've fixed up an IR LED array
Hi Ed et Al.
[EMAIL PROTECTED] wrote:
> I assume Acer fixed the multi-pass registration problem of the ScanWit 2720,
> since the
> dust removal won't work right if the infrared and color passes aren't
> perfectly aligned.
The mechanism of the 2740 appears identical to the 2720.
I think the pro
Mike wrote:
> Well, not really. The oversampling business in CD players is mostly
> a method to save as much as maybe a dime in their production costs to
> reduce the cost of the analog output reconstruction filter.
Regardless of why they use a particular technique, the fact is that a filter
*is
TECTED]
Subject: Re: filmscanners: RE: Film Scanners and what they see.
Chris wrote:
>Surely longevity is the key word here and not pure capacity.
>As the amount of stored data increases we do not want to have to spend
>a bulk of our time copying all the library CDs onto the latest media,
> Er - I thought we were talking about *decreasing* the
> apparent grain by filtering?
I must have misread the post i replied to...I thought I read
INcrease, which made no(t much) sense to me...
-
Sent using MailStart.com ( http://MailStart.Com/welcome.html )
The FREE way to access your m
Austin wrote:
>> Er - I thought we were talking about *decreasing* the
>> apparent grain by filtering?
>I must have misread the post i replied to...I thought I read
>INcrease, which made no(t much) sense to me...
I probably referred to the fact that aliasing in a scanner *increases* the
apparent
86 matches
Mail list logo
