Nikon super coolscan 5000 ED B&W film compatabilites

  • Thread starter Thread starter JimKramer
  • Start date Start date
rafe b said:
There's nothing wrong with NikonScan -- once it's
properly installed. I've been using it for years
and it's stable as Gibraltar.
Unfortunately, while this may be true for the single frame LS-8000 you
are using Rafe, it isn't true when the film strip loader is used in the
35mm scanners.

Nikon themselves have confirmed the problem I have mentioned is
reproducible, and that it can cause a system crash. This was in
response to the bug report I sent them within days of v4.02 being
released.

Apart from that though, I don't have any problems with NikonScan, so I
really wish they would fix this film polarity issue.
 
JimKramer said:
I've already had some additional issues with it running ICE. It won't
finish the scan, runs up to almost 100% and then stalls.
ICE will take a very long time to complete *if* you accidentally enable
it with silver loaded monochrome film, because it is interpreting
millions of pixels as defective and interpolating each one. That is the
only time I have ever seen the symptom you describe above.
I think a 2.8GHZ PIV with a GB of Ram is sufficient overkill for a
program that was copyrighted in 2003.
What OS are you using?

I have a similar configuration (3G-P4 & 1G RAM) running on W98SE without
any problems other than the film polarity one I already mentioned. So
your system certainly should be capable.
 
I recently got a 5000 ED and have been very pleased with the results
for most things....

I'm a bit behind on the thread and have tried to read all the posts,
but probably missed some.
But it will not scan any of my Fuji Acros 100 films. All sorts of
errors: program crashes, "no film detected", "can't autofocus"

One shortcoming of Nikon Scan is it has a strong tendency to crash
when it gets confused which makes it a royal pain with a the auto
slide feeder (when you have a batch of slides that are warped).

There are a number of things to try most of which have already been
brought up. Scan something a similar density and turn the auto focus
off. The problem is it still may get confused as to where one
negative ends and the next begins with any contrasty lines.
It has the same problem with color slides containing high contrast
lines. I had a bunch of slides from the Detroit Auto Show from a few
years back that about drove me nuts. I re scanned them using VueScan
and they came out fine.
Any thoughts?

I'd get hold of several of the other software packages and try them to
see if you have any different results.
Has anyone else run across films that simply are incompatible with this
scanner?
Not the scanner. It works with every thing I've fed it, but it seems
like each software package has it's own set of "glitches", Nikon
Scan's" being it's propensity for crashing, or freezing up with
contrasty lines in a photo. High contrast images in B & W would be
included in this..

Oft times it becomes necessary to close the software, turn off the
scanner, turn the scanner back on and then restart Nikon Scan.

OTOH I've gone through hundreds of rolls of color negatives with the
only problem being very dark areas, but VueScan seems to have more of
a problem with those than Nikon Scan, or at least the versions I've
used. Most of the time either gives excellent results with one
filling in where the other has a problem.

I said "hundreds of rolls"; I'm approaching about 30,000 images
scanned. I've lost track of how many of which, but the mix is
probably close to equal between slides and negatives. Only a few have
been B & W negatives but having ICE on or off made no difference in
either the results (It doesn't work on normal B & W film) or the
crashing.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com
 
There's nothing wrong with NikonScan -- once it's

Ah, but there is and they admit it.
properly installed. I've been using it for years
and it's stable as Gibraltar.

There are issues with Nikon's uninstaller, but then,
I've had similarly infuriating issues with other
uninstallers, eg., from Epson, HP, Canon, et. al.

NikonScan wants a powerful CPU and as much memory
as you can throw at it.

It doesn't need any where near as much as Photoshop.

It works as well on my old 2 Gig Athlon with 512 megs as it does on
this 64 bit, 3.4 Gig Athlon with one Meg on board cache.
Now for multiple image scans
The "key" to uninstalling NikonScan is to drill
down to:

[root]\Documents and Settings\All Users\Application Data\Nikon

Remove the files from that folder, or the folder itself.

This is quite common with much of today's software. Uninstalling any
Symantec software leaves tracks (files) all over the place and you
have to edit the registry as well.

Then, if using any form of XP don't forget the ability to "roll back"
as nearly all of those deleted files are out there in the hidden files
you can not edit.
Also, Nikon supplies (on the CD that came with your
scanner) a utility for cleaning out the registry
after a driver uninstall. At least they did with
my 8000.

And here's a kicker, that even I had trouble believing.

About a year or so ago, NikonScan started acting flaky.
But this was after installing a new motherboard. After
trying everything under the sun -- including a new
system power supply -- the problem was eventually

Typically if the PS is dumping the system it'll take out something
really important after four or five failures. With the switching PS
it just depends on where it is in the cycle when it fails. I think
the smallest PS here is now 480 or 530 watts. Now I have two UPSs that
are telling me it's time to replace the batteries. I think I'll just
go purchase some industrial Gel Cells and mount them outside the
cases.
traced to an improperly installed heat sink on the CPU.

I wouldn't blame anyone for disbelieving this story,
but it's the truth.

Computers were my profession and where I have my degree. I built all
four I have running now and I'd believe most any tale about them.
Improperly installed heat sinks are a big (and common) problem with
today's high powered systems that have lots of muscle. The thing is,
the more power you run the more cooling air it takes and the more air
you move the more likely the system is to load up with dust and go
down from an over temp which leads to the necessity for regular
maintenance shut downs and cleanings.

My most recent, well, the previous most recent problem was a computer
that would run a while and then crash. The problem? The case fan
(not the one in the power supply) was drawing excessive current. I
had the system spread all over the place and it was working. I
finally got it all back together except for that fan. When I hooked up
the fan everything quit. Unfortunately that was the straw that broke
the camel's back. It took out the PS which took out about everything
except the processor and hard drives as I recall. My memory isn't as
good as it used to be on remembering those kind of details.

My current problem isn't actually the computers, but the Internet.
About the start of Christmas vacation the number of attempted
intrusions went from no more than a few hundred total over the past
couple of years to so many they sometimes slow my connection down.
This one has been hit with well over one and a half million attempts
since then and one of the others has been hit nearly one million
times. So far they still show as being clean, but I run a lot of
stuff to *try* to keep them that way.Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com
 
In message said:
I've already had some additional issues with it running ICE. It won't
finish the scan, runs up to almost 100% and then stalls.

I think a 2.8GHZ PIV with a GB of Ram is sufficient overkill for a
program that was copyrighted in 2003.

The memory required is down to the data being generated and processed,
so if you are scanning massive images then you need masses of RAM.
 
I noticed after the first round of uninstall/reinstall that it was
"kind enough" to preserve all of my user setting from the last set of
scans

Yes, it does leave "droppings" behind but they are really harmless.
It's disconcerting to see the last preview show up after an uninstall
but that's just cosmetic.
I hope Nikon wasn't hoping to win an award with this software.

Actually, NikonScan is very solid and well written.

There's no excuse for not fixing the bug Kennedy wrote about but I
guess now that the scanning "fad" has subsided (Minolta just got out
of the scanner business completely) Nikon probably feels it's not
worth the trouble.

Don.
 
Actually, NikonScan is very solid and well written.

Accept when you trigger a bug that makes it crash consistently. Or that
makes it misbehave consistently.

NikonScan does have bugs, and if you trigger them, it may be tricky to
get around them.
 
Accept when you trigger a bug that makes it crash consistently. Or that
makes it misbehave consistently.

NikonScan does have bugs, and if you trigger them, it may be tricky to
get around them.


Guess I must be lucky then.

These are the bugs I'm aware of:

1. The banding issue on the original LS-8000,
which is fixed by selecting SuperFine Scan.

2. Occasionally poor recognition of frame
boundaries on 645 film strips, which is
dealt with by adusting the film-strip
offset value.

3. Histogram "bunching" on C41 (negative)
scans, which can be dealt with by scanning
negatives as positives.


I'm told that NikonScan 4, on the LS-9000,
addresses problems #1 and #3.


rafe b
www.terrapinphoto.com
 
Accept when you trigger a bug that makes it crash consistently. Or that
makes it misbehave consistently.

NikonScan does have bugs, and if you trigger them, it may be tricky to
get around them.

That's just demonstrably not true.

There has only been *one* reported bug so far which causes NikonScan
4.2 to actually crash. Kennedy has quite comprehensively explained
both the bug and the workaround. Nothing tricky about that. As
workarounds go it's quite pedestrian, but annoying nevertheless.

!=> Now then, since you made the claim can you substantiate it with
some facts? What other bugs are there which cause NikonScan to crash?

In the interest of full disclosure I don't even use NikonScan so I'm
not "defending" it but just stating facts. However, I have tested it
extensively even going as far as disassembling it. I tried fooling it
by manipulating the preview file and the program displayed the image
with new dimensions without a hitch showing it's written properly.

I also modified the TWAIN binaries to extend the AG and it worked like
a dream (I can go up to +16 rather than the default +2). Again, the
graphic front end didn't blink an eyelid and extended all the graphics
elements accordingly showing its flexibility and proper coding.

NikonScan can handle multiple objects simultaneously, etc. etc. It's a
very solid and well written program. It handled everything I threw at
it no matter how "unfair" e.g. modifying the binaries!!!

Now, NikonScan may be stubborn and cranky e.g. it resists when you try
to turn AutoExposure off, but that's not a failure of the program but
intentional design because that information is stored in the scanner
itself. I suspect, but don't know this for a fact, it's all a part of
Nikon's "copy protection" so people don't turn on multiscanning in
devices like the LS-50 by making simple software modification (the
very reason why I disassembled it in the first place).

Don.
 
Guess I must be lucky then.

And you're not alone! Indeed, you're in good company! ;o)
These are the bugs I'm aware of:

1. The banding issue on the original LS-8000,
which is fixed by selecting SuperFine Scan.

I don't have an LS-8000 but I understand (from reading here) that's
not a NikonScan bug but a hardware problem of the LS-8000. However,
not having the scanner I did not pay that much attention so I yield to
Kennedy, because he knows much more about it than I do...
3. Histogram "bunching" on C41 (negative)
scans, which can be dealt with by scanning
negatives as positives.

Or by simply turning AutoExposure off!!!

The problem is apparently due to how NikonScan reverses the image and
sets the black and white points. It's really more a domain of image
editing/processing rather than the actually scanning. But if
AutoExposure is turned off then there is no clipping.

Don.
 
That's just demonstrably not true.

The most interesting bug I had was one where the SA-21 did not position
correctly. Previews were no problem, scans were off. In the end I had
to use the film strip holder.
There has only been *one* reported bug so far which causes NikonScan
4.2 to actually crash.

I never reported any bugs to Nikon. So that doesn't say much. Anyhow, I
was talking about bugs in general. Not about the subset of bugs that just
happen to crash the application.

I think that are other bugs that make it impossible to use NikonScan but
that don't actually crash it.
 
Don said:
I suspect, but don't know this for a fact, it's all a part of
Nikon's "copy protection" so people don't turn on multiscanning in
devices like the LS-50 by making simple software modification

Yeah, but they didn't reckon on the "moron with a multimeter" working
out how to turn an SA-21 into an SA-30. ;-)
 
Did anyone actually get that to work?
Assuming you mean turning an SA-21 into an SA-30 (it helps if you quote
the statement you are questioning) then yes - I made it work, even
though I have both. The mod is trivial.
 
Have you tried running something other than Nikon Scan for a
comparison?
The memory required is down to the data being generated and processed,
so if you are scanning massive images then you need masses of RAM.

Yup! and even with 4000 dpi and 16 bit color depth you only have about
128 Megs per image. With a 5 image strip that is 640 Megs. Of course
if the image is being directly imported into Photoshop you also have
Photoshop and it's work space.

I was regularly running that combination on a system with one gig of
RAM. With no problems, that is until I started running a bunch of
other *stuff* such as word, excel, and OE at the same time. Even then
the memory management would "usually" just swap out the app not being
used, BUT at times I'd end up with it page file swapping the images
out while working on them. If you didn't know what had happened you'd
think it had stalled as it would take forever, plus a little to get un
constipated. Some times you could watch it draw the image line by
line with it taking 2 or 3 seconds per line.

Prior to that I ran the same system with 512 Meg of RAM, but without
going directly into Photoshop. It worked well, but going to one Gig of
RAM made a marked difference in processing time for those long
negative strips.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com
 
The most interesting bug I had was one where the SA-21 did not position
correctly. Previews were no problem, scans were off. In the end I had
to use the film strip holder.

There are bound to be positional differences between the preview and
the full scan simply because of scale. A preview is a low resolution
scan and is never going to be as accurate as the full resolution scan.
Usually, if the preview is off by 1 pixel the final scan will be off
by about 8 pixels (depending on preview resolution, of course).

Considering SA-21 has a drive to draw in the film, all that just adds
to the inaccuracy. That's why there is "Strip Film Offset" to adjust
the thumbnails. I've never used the SA-21 (except once when I got the
scanner) but the implementation of the above offset adjustment just
reinforces how well written NikonScan is as they accounted for that!

So, positioning has nothing to do with NikonScan. It's just mechanics.

BTW, I suffer from this all the time because I only use the holder.
Since the holder has some "play" when inserted in the slide adapter I
have to tap it left and right until it's lined up. But, even when I've
aligned the preview "perfectly" (and my software previews at higher
resolution than NikonScan!) the full scan is usually still off by a
bit which is only to be expected with 4000 dpi precision.

That's why implemented another workflow but that's a different story.
I never reported any bugs to Nikon.

I wasn't talking about Nikon (their so-called "support" is
pathetic!!!) but about what people have reported here.
So that doesn't say much. Anyhow, I
was talking about bugs in general. Not about the subset of bugs that just
happen to crash the application.

Actually, you did (you said "crash consistently") :

--- start ---
Accept when you trigger a bug that makes it crash consistently. Or that
makes it misbehave consistently.
--- end ---
I think that are other bugs that make it impossible to use NikonScan but
that don't actually crash it.

Again, I'm not aware of any bugs per se, but I do find quite a few
major annoyances, the most important (for me) being NikonScan's
reluctance to turn off AutoExposure.

Don.
 
Yeah, but they didn't reckon on the "moron with a multimeter" working
out how to turn an SA-21 into an SA-30. ;-)

ROTFL! Indeed!

I wish I had more time because I would really, really, really... like
to crack the firmware and implement my "twin scan" on board.

Don.
 
There are bound to be positional differences between the preview and
the full scan simply because of scale. A preview is a low resolution
scan and is never going to be as accurate as the full resolution scan.
Usually, if the preview is off by 1 pixel the final scan will be off
by about 8 pixels (depending on preview resolution, of course).

Yeah. But by the time, the scan is half a frame off compared to the preview,
something weird must be going on. Ejecting and re-inserting the strip
did not help.
Actually, you did (you said "crash consistently") :

You are right. I think that did happen also at some time.

I am not saying that NikonScan is particularly unreliable. I have scanned
lots of films without triggering any strange bugs. But it does happen
every once in a while that NikonScan (or the scanner) gets into a strange
state.
 
my software previews at higher
resolution than NikonScan!

I find that hard to believe, Don!

Whilst your software may preview at higher resolution than the *initial
default* preview resolution of Nikonscan, the preview size of Nikonscan
can be set to any size up to almost the entire desktop area. On my
system, a Nikonscan preview has been set to 800x550 pixels for about
three years, and can readily be expanded to 1350x880 pixels without
changing desktop size and a full 100% scan by changing the desktop area
if I want, although that defeats the objective of a "preview".
 
Yeah. But by the time, the scan is half a frame off compared to the preview,
something weird must be going on. Ejecting and re-inserting the strip
did not help.

Oh, I see. It's that extreme! Have you asked here if anyone else has
experienced it? (In order to rule out a hardware problem or something
unique to your installation.)
You are right. I think that did happen also at some time.

I am not saying that NikonScan is particularly unreliable. I have scanned
lots of films without triggering any strange bugs. But it does happen
every once in a while that NikonScan (or the scanner) gets into a strange
state.

I have used NikonScan quite a bit and (as I mentioned) actually took
it apart. It really is a very well written program. Any problems
people experience are usually due to "user error" or misunderstanding
e.g. clipping of negatives. But as far as programs go, NikonScan is
quite solid.

Don.
 
Back
Top