Why would one buy a used/refurb Nikon Super Coolscan 4000 vs new Coolscan V?

S

Steve Jobs

First of all, pardon my ignorance on this, but much of the film
scanner techno-speak is over my head. In the approximately $500-$600
price range, I see one can buy one of the two in the subject line and
I'm trying to understand why one might choose an older model with a
slower interface versus purchasing a brand new one. I'm looking to
scan a bunch of old 35mm film strips.

Any guidance would be appreciated.

TIA,
Steve
 
K

Kennedy McEwen

Steve Jobs said:
First of all, pardon my ignorance on this, but much of the film
scanner techno-speak is over my head. In the approximately $500-$600
price range, I see one can buy one of the two in the subject line and
I'm trying to understand why one might choose an older model with a
slower interface versus purchasing a brand new one. I'm looking to
scan a bunch of old 35mm film strips.

Any guidance would be appreciated.
First things first - the older model *doesn't* have a slower interface!
The LS-4000 (Coolscan 4000) is a Firewire *only* interface model. The
LS-50 (Coolscan V) has a USB-2 interface which is almost, but not quite,
as fast as Firewire, but intrinsically compatible with USB1.0, a *much*
slower interface. So you can, and will if you have any USB1 devices
connected on the same port, find that the LS-50 actually has a slower
interface than the LS-4000.

As far as the scanners themselves are concerned, the main advantages of
the LS-4000 over the newer LS-50 come from the fact that the LS-4000 was
the top of the range model in its time and consequently was compatible
with Nikon's top of the range features, such as single pass
multiscanning (improving the signal to noise ratio by simulating a much
less noisy CCD) and support for bulk scanning adapters such as the SA-30
and SF-200.

Multiscanning can be simulated in the LS-50 by a multipass technique,
but this is significantly inferior to the "real thing" - the improved
noise is vastly inferior due to technical reasons and the image blurs
due to mis-registration of the multiple passes.

The main advantage of the LS-50 over the LS-4000 is that the newer
scanner supports DEE, which claims to improve the density range of the
image. In addition, the USB2 interface is much more common on modern
PCs than Firewire, although the cost of a Firewire card (if not included
with the used LS-4000)is trivial.

In addition, buying new means you know what the scanner has been
through, whilst buying used is always a bit of a gamble - will it need
servicing or cleaning as soon as you get it? Buying a refurb does, of
course, level this out quite a bit, since it will probably come with
some warranty and certainly should not require cleaning or servicing.

So there are advantages to both scanners but, assuming the costs were
within $100 either way, for my money I would buy a refurbished LS-4000
over the newer LS-50 any day.
 
S

Steve Jobs

Thx for the thorough reply. Not sure what I was thinking when I said
"slower" interface since it's pretty much a wash. I was more
interested in whether the quality of the newer LS-50 scanner was equal
to that of the older LS-4000, and I take it from your post that it
isn't.

I don't want to start a holy war in this forum, but might I ask your
opinion on the LS-4000 vs. Minolta DiMAGE Scan Elite 5400?

Thx again,
Steve
 
K

Kennedy McEwen

Steve Jobs said:
Thx for the thorough reply. Not sure what I was thinking when I said
"slower" interface since it's pretty much a wash. I was more
interested in whether the quality of the newer LS-50 scanner was equal
to that of the older LS-4000, and I take it from your post that it
isn't.

I don't want to start a holy war in this forum, but might I ask your
opinion on the LS-4000 vs. Minolta DiMAGE Scan Elite 5400?
If the Minolta had been around when I bought the Nikon, that is probably
what I would have gone for - higher resolution and more bits, with all
of the same features, such as ICE etc. and a few additional ones like
the Grain Dissolver, or Scanhancer.

However, much of the benefit of those additional 2-bits on the ADC of
the Minolta is lost when compared against the LED illumination of the
Nikon - this produces very pure colour separations with minimal matrix
manipulation. So the SNR of the ADC is what you get on output of the
Nikon, rather than an arithmetically reduced equivalent as a result of
colour correction on traditionally filtered CCDs used on other scanners.

Similarly, the 5400ppi resolution of the Minolta is pretty much lost
unless you traditionally shoot slower films on a tripod. It is real,
and it can be recorded on film, but it is a marginal step up from
4000ppi, given the MTF of the film and camera lens, and hence only makes
a difference on the very best of images.

Similarly, the Grain Dissolver, while creating a cleaner image certainly
is not without its downsides - extending scan times significantly.

Nevertheless, whilst all of these advantages of the Minolta over the
Nikon are marginal in real terms, when combined they would probably have
been enough to tilt my choice in favour of the Minolta - because
sometimes they would have been useful. The biggest con against the
Minolta that I am aware of is the inability to engage ICE without the
Grain Dissolver in the native software, making ICE scans unnecessarily
long. It would be a close call though - and as the market is now, with
the Minolta pitched against the Nikon LS-5000, I would be back with
Nikon again: 2 real extra bits on the data compared to what I have now
would swing it.
 
C

closecall

Kennedy said:
If the Minolta had been around when I bought the Nikon, that is probably
what I would have gone for - higher resolution and more bits, with all
of the same features, such as ICE etc. and a few additional ones like
the Grain Dissolver, or Scanhancer.

However, much of the benefit of those additional 2-bits on the ADC of
the Minolta is lost when compared against the LED illumination of the
Nikon - this produces very pure colour separations with minimal matrix
manipulation. So the SNR of the ADC is what you get on output of the
Nikon, rather than an arithmetically reduced equivalent as a result of
colour correction on traditionally filtered CCDs used on other scanners.

Don't quite understand what you meant by the last sentence above. Both
Nikon's LED and Minolta's 5400 light source can be adjusted in hardware
to change the exposure.
Similarly, the 5400ppi resolution of the Minolta is pretty much lost
unless you traditionally shoot slower films on a tripod. It is real,
and it can be recorded on film, but it is a marginal step up from
4000ppi, given the MTF of the film and camera lens, and hence only makes
a difference on the very best of images.

With 5400ppi, a full frame scan can be cropped somewhat and can produce
a 13"x19" print at 300dpi printer resolution without resampling. Can't
do that with 4000ppi.
Similarly, the Grain Dissolver, while creating a cleaner image certainly
is not without its downsides - extending scan times significantly.

Nevertheless, whilst all of these advantages of the Minolta over the
Nikon are marginal in real terms, when combined they would probably have
been enough to tilt my choice in favour of the Minolta - because
sometimes they would have been useful. The biggest con against the
Minolta that I am aware of is the inability to engage ICE without the
Grain Dissolver in the native software, making ICE scans unnecessarily
long. It would be a close call though - and as the market is now, with
the Minolta pitched against the Nikon LS-5000, I would be back with
Nikon again: 2 real extra bits on the data compared to what I have now
would swing it.

Minolta 5400 offers auto focus as well as manual focus controls at any
spot of an image. An user can produce two scans based on the two
different controls and pick the better one. Not sure if the Nikons have
anything equivalent. Nikons are known to have a shallow dof which can
result in uneven sharpness across a scan. Some attribute the shallow dof
to the LED light source.

Last but not least, the 5400 currently costs roughly half as much as the
Nikon LS-5000.
 
K

Kennedy McEwen

Don't quite understand what you meant by the last sentence above. Both
Nikon's LED and Minolta's 5400 light source can be adjusted in hardware
to change the exposure.
What I mean, and what you seem to be unaware of, is that the data from a
filtered CCD has poor colour purity. In particular, the blue channel
often has a significant response in the red, while the green shows
response in both blue and red regions. This impurity can be, and
requires to be, corrected by matrix manipulation of the actual data
produced by the CCD channels. For example, subtracting a fraction of
the red channel response from the blue will produce a more accurate blue
output, and so on. Indeed, this is exactly what a scanner profile does.
However, subtracting one signal from another, where both contain noise,
adds the noise of both signals. Thus the colour correction of the
filtered CCD reduces the effective number of bits in the data. With
the LED illumination system, the colours are implicitly pure due to the
near monochromatic nature of the light source, and hence the only colour
impurity is what already exists in the dyestuffs of the film emulsion -
which is exactly what you are trying to reproduce.
With 5400ppi, a full frame scan can be cropped somewhat and can produce
a 13"x19" print at 300dpi printer resolution without resampling. Can't
do that with 4000ppi.
Of course you can - you seem to be under the illusion that resampling is
somehow inferior to scanning a soft image at a higher resolution!

If the image contains no more than 1500cy/in of information either
because of the use of high speed film, less than perfect optics or
simply camera shake then you will get no more information in a 5400ppi
scan than you will in a 4000ppi scan. Furthermore, at f/16 the
diffraction limit of a *perfect* lens is only about 20% at 4000ppi - and
zero at the red end of the visible spectrum - so if you shoot anything
with a practical lens close to f/16 or smaller you have already run out
of resolution - ignoring completely the resolution limits of the film or
the scanner optics!
Minolta 5400 offers auto focus as well as manual focus controls at any
spot of an image.

And your point is what, exactly? Manual focus at any point on the image
has been in Nikon scanners before Minolta even appeared in the market.
An user can produce two scans based on the two
different controls and pick the better one. Not sure if the Nikons have
anything equivalent. Nikons are known to have a shallow dof which can
result in uneven sharpness across a scan. Some attribute the shallow dof
to the LED light source.
Some do, others just use the manual film holder if the film is curved -
of course you wouldn't know about that since a manual film holder is all
the Minolta has...
Last but not least, the 5400 currently costs roughly half as much as the
Nikon LS-5000.

Indeed, and I took that into consideration in my previous comparison.
Based on the LS-5000 and Minolta 5400, I would choose the LS-5000 every
time. If the Minolta had been around at the time when I bought my
LS-4000 (which was priced at the same level as its successor) I would
probably have bought the Minolta, even though, in practice, its benefits
are quite marginal.
 
R

rgou4576

Kennedy McEwen said:
If the Minolta had been around when I bought the Nikon, that is probably
what I would have gone for - higher resolution and more bits, with all
of the same features, such as ICE etc. and a few additional ones like
the Grain Dissolver, or Scanhancer.

However, much of the benefit of those additional 2-bits on the ADC of
the Minolta is lost when compared against the LED illumination of the
Nikon - this produces very pure colour separations with minimal matrix
manipulation. So the SNR of the ADC is what you get on output of the
Nikon, rather than an arithmetically reduced equivalent as a result of
colour correction on traditionally filtered CCDs used on other scanners.

Similarly, the 5400ppi resolution of the Minolta is pretty much lost
unless you traditionally shoot slower films on a tripod. It is real,
and it can be recorded on film, but it is a marginal step up from
4000ppi, given the MTF of the film and camera lens, and hence only makes
a difference on the very best of images.

Similarly, the Grain Dissolver, while creating a cleaner image certainly
is not without its downsides - extending scan times significantly.

Nevertheless, whilst all of these advantages of the Minolta over the
Nikon are marginal in real terms, when combined they would probably have
been enough to tilt my choice in favour of the Minolta - because
sometimes they would have been useful. The biggest con against the
Minolta that I am aware of is the inability to engage ICE without the
Grain Dissolver in the native software, making ICE scans unnecessarily
long. It would be a close call though - and as the market is now, with
the Minolta pitched against the Nikon LS-5000, I would be back with
Nikon again: 2 real extra bits on the data compared to what I have now
would swing it.

Kennedy
which scanner would you favour for B/W negs? - I'm intending scans of
several thousand B/W negs, and I don't have much colour material to
scan. I've heard suggestions that the Minolta is better for B/W
because of the more diffused lighting, the higher resolution to reduce
grain aliasing further, and the grain dissolver option.

Your thoughts?

Thanks
Ross Gould
 
K

Kennedy McEwen

rgou4576 said:
Kennedy
which scanner would you favour for B/W negs? - I'm intending scans of
several thousand B/W negs, and I don't have much colour material to
scan. I've heard suggestions that the Minolta is better for B/W
because of the more diffused lighting, the higher resolution to reduce
grain aliasing further, and the grain dissolver option.

Your thoughts?
With traditional B/W (ie. not chromagenic emulsions that are C-41
compatible) ICE doesn't work, so that removes one of the main advantages
of the Nikon and Minolta scanners and I would certainly lok at scanners
which have a more diffuse light source for such material.

I'm not sure that the Minolta is intrinsically less collimated than the
Nikon in this respect, but it does have the grain dissolver, which
effectively converts it to a very diffuse source. If the GD can be
engaged without ICE (it appears that ICE cannot be engaged without the
GD in the native software, I am not sure about the other way round
though) then that would probably be the better choice for your
application.
 
W

Wilfred van der Vegte

Kennedy said:
I'm not sure that the Minolta is intrinsically less collimated than the
Nikon in this respect, but it does have the grain dissolver, which
effectively converts it to a very diffuse source. If the GD can be
engaged without ICE (it appears that ICE cannot be engaged without the
GD in the native software, I am not sure about the other way round
though) then that would probably be the better choice for your application.

Yes, the grain dissolver can be activated without ICE, even in the
Minolta software.
 
P

Paul

Wilfred van der Vegte said:
Kennedy said:
Yes, the grain dissolver can be activated without ICE, even in the
Minolta software.

Does anyone know where you can get refurbsihed nikon scanners in the
UK? Particularly the LS-4000 and the LS-8000?

Many thanks

Paul
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top