Nikon LS40 Inconsistency of Frame offset
Hiroaki YOSHIFUJI
But I have trouble of Frame offset. So I cannnot leave PC while scanning.
It is inconsistent, and sometimes scan starts before the edge of film,
the minimum frame offset -37.941 is not enough, I can not
scan full frame.
I thought film should slip, so
I have cleaned all rollars and 3 photo sensers of SA-21.
But nothing improbed.
Vuescan 7.5.17 Linux and Windows have same trouble.
Nikon Scan 3.1 doesn't have this trouble.
So I think my LS40 doesn't need repair.
Hiroaki YOSHIFUJI
Jeff Randall
> I think batch scan of vuescan is nice feature.
> But I have trouble of Frame offset. So I cannnot leave PC while scanning.
>
> It is inconsistent, and sometimes scan starts before the edge of film,
> the minimum frame offset -37.941 is not enough, I can not
> scan full frame.
Hiroaki:
This is a known problem with VueScan and Nikon scanners. I think it
still is on Ed's "to do" list.
I've found the following to be a workaround (found in the User's
Manual). 1) trim off any excess blank leade. 2) preview frame 2 and
adjust the frame offset. 3) go back and scan frame 1 and other frames
if you are batching.
Jeff Randall
Jeff
filmstrip holder, instead. Tell VScan that it's "slide". The FH-3
often holds the film a bit flatter as well. Everything works
perfectly - an amazing thing.
My scanning is for fun, so I'm not worried about productivity in
volume.
Hiroaki YOSHIFUJI
jran...@ch2m.com (Jeff Randall) wrote:
>
> This is a known problem with VueScan and Nikon scanners. I think it
> still is on Ed's "to do" list.
Nikon service.
>
> I've found the following to be a workaround (found in the User's
> Manual). 1) trim off any excess blank leade. 2) preview frame 2 and
> adjust the frame offset. 3) go back and scan frame 1 and other frames
> if you are batching.
Thanks I will try.
Maybe off topic:
Today I newly installed vuescan on another Linux PC. I just tested
once, it's frame was perfect without any setting.
Hiroaki YOSHIFUJI
jeff...@mac.com (Jeff) wrote:
>
> Same problem occurs w/my LS4000. I still use VScan, but with the FH-3
> filmstrip holder, instead. Tell VScan that it's "slide". The FH-3
To get raw file?
I really don't understand.
> often holds the film a bit flatter as well. Everything works
> perfectly - an amazing thing.
>
> My scanning is for fun, so I'm not worried about productivity in
> volume.
My scanning is for fun, too. I want to use batch scan as to make
contact prints.
I havn't used FH-3 yet, I will use it to scan precisely for large
print as you suggested, thanks.
I used Nikon Coolscan II last year, borrowed from my friend, and which
filmstrip hoder masked a little bit, I could not scan full frame.
FH-3 doesn't have this problem?
ThomasH
This is an long-term problem with NikonScan. Vuescan can not determine
the 1st image only! Once you set the offset for the first frame,
Vuescans autocrop does a magnificent job on the entire roll of film.
Some more details and examples about NikonScan and its handling of
thumbnails are on http://www.pbase.com/phototalk_thh/sa30ns312
We have been reporting this problem to Nikon Tech Support even since
LS-4000ED is on the market, I even got once a "level two" ticket for
this particular problem. They are just so clumsy with their software
and seemingly unable to improve.
The scanner itself + the SA-30 are one of my best pieces of equipment
and it is a pleasure to play with it.
Thomas.
Jeff Randall
> This is an long-term problem with NikonScan. Vuescan can not determine
> the 1st image only! Once you set the offset for the first frame,
> Vuescans autocrop does a magnificent job on the entire roll of film.
>
> Some more details and examples about NikonScan and its handling of
> thumbnails are on http://www.pbase.com/phototalk_thh/sa30ns312
>
> We have been reporting this problem to Nikon Tech Support even since
> LS-4000ED is on the market, I even got once a "level two" ticket for
> this particular problem. They are just so clumsy with their software
> and seemingly unable to improve.
Thomas:
I understood Hiroaki to be asking about VueScan's sometime inability
to accurately find and crop the first frame of a film strip using the
SA-21 adaptor. NikonScan works completely differently than
VueScan--it scans the entire strip and then breaks it down into
thumbnails while VueScan finds the first sprocket hole and "counts"
holes to find and preview/scan a specific frame. Why would anyone
report a VueScan frame offset problem to NikonScan? If NikonScan also
has a frame offset problem then OK, but the VueScan and NikonScan
problems would not be related. Sorry if I misunderstood.
Jeff Randall
ThomasH
>
> ThomasH <Tho...@nospam.net> wrote
>
> Thomas:
>
> I understood Hiroaki to be asking about VueScan's sometime inability
> to accurately find and crop the first frame of a film strip using the
> SA-21 adaptor. NikonScan works completely differently than
> VueScan--it scans the entire strip and then breaks it down into
> thumbnails while VueScan finds the first sprocket hole and "counts"
> holes to find and preview/scan a specific frame. Why would anyone
Jeff,
the both SA-21 and SA-30 are virtually identical, except that
the SA-30 has a small brush on the entry. They both use rubber
rolls to advance the film and they do not 'see' the sprocket
holes (unfortunately...)
My point is that Vuescan never "finds" the first frame. The offset
must be adjusted, unless it happen to be is close to null. But
once you adjust it on one frame, Vuescan auto crop will perfectly
follow further frames along the entire film strip or even the
entire roll of film. It is cool, it works.
This necessity to adjust the first frame offset is a minor
inconvenience. In contrast to this, NikonScan has a much more
nasty problem because it keeps spreading wrong offsets among all
thumbnails.
I assume that for the use with the SA-21 the film strips will be
cut to the image edge, and thus the begin of the first image has
an offset close to zero. In addition, both adapters grab the film
somewhat randomly. A 0.5-1 mm difference in offset between feeds
of the same film is not so uncommon. It appears to me that
Hiroaki's case with the correct offset the other day is not
related to his software update.
> report a VueScan frame offset problem to NikonScan? If NikonScan also
> has a frame offset problem then OK, but the VueScan and NikonScan
> problems would not be related. Sorry if I misunderstood.
>
> Jeff Randall
Entirely my fault, I hope that this explains the offset issue.
By the way, I wish that the SA-nn film adapters would use sprocket
holes, just like the cameras do! I am puzzled about the reason that
they use the rubber rolls, the more that they depress the entire
surface of the film and not only the film edges.
Thomas.
Hiroaki YOSHIFUJI
jran...@ch2m.com (Jeff Randall) wrote:
>
> Thomas:
>
> I understood Hiroaki to be asking about VueScan's sometime inability
> to accurately find and crop the first frame of a film strip using the
> SA-21 adaptor. NikonScan works completely differently than
> VueScan--it scans the entire strip and then breaks it down into
> thumbnails while VueScan finds the first sprocket hole and "counts"
> holes to find and preview/scan a specific frame. Why would anyone
> report a VueScan frame offset problem to NikonScan? If NikonScan also
> has a frame offset problem then OK, but the VueScan and NikonScan
> problems would not be related. Sorry if I misunderstood.
>
> Jeff Randall
I think
Vuescan and Nikonscan problem might related.
I mean it could be a problem of farmware or hardware of Nikon scanner.
I found Thomas's trouble in my LS40 FH3 Vuescan linux, I tried to scan
index file from 6 frames strip(batch scan). First photo was repeated
and 6th was not scanned.
It is as same as Thomas claimed at
http://www.pbase.com/phototalk_thh/sa30ns312.
Adn also I encoutered starange behavior of Nikon scan last night.
And frame inconstency happenen even Nikon scan.
I have not scanned so many times with vuescan and Nikon scan yet and
error happens inconsistently. So I don't know definitely repeatable
situation. (BECAUSE IT IS SO INCONSISTENT.) I WILL TRY TO FIND IT OUT
AND WANT TO LIVE WITH IT.
I don't know the system of scanner and how scanner count and know the
frame.
Onece LS40 acts starange, there is no way to recover until
power off. I believe use Nikon scan is to cure. Is it right?
Where frame offset value is stored, only in vuescan inifile and Nikon scan
setting file or LS40's RAM of flash RAM?
Paul Turtle
Hi,
I've got the Nikon LS-4000 and I've reported the frame inconsistancy
before also. I'll try the workaround shown above.
I've found that it takes a bit a work to get the first frame aligned
(seems to be almost spot on and can just move crop box with mouse or
it's about 3mm out and I have to adjust the frame offset and do
another preview to be able to see the whole frame then adjust crop
box). But I've repeatedly found that after correctly aligning the
first frame then the offset to the next frame is not correct. Offsets
between frames 2, 3 and 4 are exact but Vuescan doesn't seem to be
able to align accuratly between the first two frames. Many times I've
had to align and scan frame 1, then align frame 2 and scan frames 2-4
in one batch. Any solution to this would speed my scanning
considerably. Maybe Ed can implement the auto detection of the first
frame (by scanning while moving the whole film strip?) like the Nikon
software.
People say that the CoolScan doesn't count or measure the sprocket
holes in the film, but I still think it has something to do with it.
But I've also had inconsistant results (different first frame offset
by 3mm or more) by inserting the same strip again.
Regards
Paul Turtle
Hiroaki YOSHIFUJI
Hi Paul
I am still doing many try and error. The best way I found at this moment is
to follow Jeff's suggesion.
I really don't know why it works, but it works fine.
1) trim off any excess blank leade. 2) preview frame 2 and
adjust the frame offset. 3) go back and scan frame 1 and other frames
if you are batching.
According to Japanese manual of LS-4000 and LS-40, NikonScan software>
scanner setting > frame offset,
"In case of using SA-21 and SA-30, NikonScan automatically detect
frame and display separetely, but if border of image is not clear,
frame might be wrong, in this case frame offset should be used to
correct frame.
So I think Nikon Scan detect frame by CCD sensor. and forwarding
distance might counted by pulse motor.
There is 3 photo sensors in LS-40, might have somoting to do with count
and measure sproket holes. Because my manual says film which has
extra holes between sproket holes can not be used by SA-21, and in such a
case and extra holes are only one side, extra holes side should not be
inserted photo sensor side.
MY EXPECTAION
If vuescan scan whole strip film from edge to edge as one raw file,
then I can crop as I like, there is no worry of frame inconstency ;-)
best regards
Hiroaki