Gee,
I was thinking about this scanner for a future purchase. Thanks for the
warning. That does beg the question of batch scanning. My Minolta Scan
Dual 2 does up to 6 frames, but you have to put the strip into a carrier
that isn't tight enough to hold the film still so the film will creep
sometimes causing less than full frame scan. It also has to do an index and
then the real scan. That means you have to nurse it along. No walking
away. So, it is better than 1 at a time, but not by much. I'm not that
happy with it. Anybody out there happy with a sub $1000.00 scanner?
Bruce Dayton
Sacramento, CA
----- Original Message -----
From: "Gerald F. Cermak" <[EMAIL PROTECTED]>
To: "Pentax Discussion List" <[EMAIL PROTECTED]>
Sent: Thursday, May 24, 2001 10:21 PM
Subject: Kodak RFS 3600 Scanner comments (cont.)
> I've owned this scanner since earlier this year, having no scanner and no
> patience to wait for the new high priced Nikons. For those interested,
here
> are some further comments about it.
>
> If you are thinking of scanning mounted slides, it works well out of the
> box, and the rest of this is immaterial to you. However, the film
transport
> and focusing mechanisms have design and manufacturing problems. The
latest
> software has numerous annoyances. Together these will make life difficult
> for film scanning folks, to say the least.
>
> 99% of my images are on negatives and unmounted slides in strips. So this
> capability is very important aspect of a film scanner for me. In fact,
the
> RFS 3600's ability to scan a whole 36 film strip appealed to me from a
time
> saving perspective.
>
> The product turns out to be a real dichotomy, in that it performs best on
> mounted slides, yet offers no slide feeder capabilities at all. With
that,
> it's primary appeal would be to people scanning film strips, especially
with
> the desire to do a whole roll at once (a unique feature in the market
> currently). Face it, scanning film, especially at higher resolutions,
such
> as the 3600 dpi the RFS3600 offers take a lot of time. The more automated
> the process, the less invasive it is to your life. God knows we all spend
> too much in front of Photoshop as it is, why add more the basically
> mechanical process of converting analog images into digital form with a
> fixed recipe.
>
> My plan was to get all film back from the lab uncut, feed it into the RFS
> 3600, do some initial adjustments on the first frame, and then tell it to
> scan every frame to disk with the same settings, returning 1 or 2 hours
> later to do any editing. That was the plan, the reality follows:
>
> In the RFS3600, the focusing is performed using a plastic frame holder
that
> is pushed down on each corner in up to 20 steps (film strips) or 40 steps
> (mounted slides). The differences in focusing steps between mounted film
> and unmounted film will be obvious in a moment.
>
> Focus setting "1" (of 1 to 20) is perfectly aligned vertically with the
film
> transport guides and transport rollers on either side of the focusing
frame.
> The transport's internal film guides run outside the image area, along the
> sprocket borders, with the pinch roller running on the film sprockets
> themselves. All well and good, so far. However, there has been velvet
> installed on the top and bottom portion of the transport over the film's
> image area. The felt is so thick that is rubs on both the emulsion and
> non-emulsion side of the film both on the intake and outtake sides of the
> transport. The inbound and outbound pinch rollers are connected in tandem
> rotation by a toothed-belt driven by one stepper motor. This means the
> right and left sides of the film always move in the same direction.
>
> When the RFS 3600 focuses, it moves just the center frame downward using
the
> central frame holder. The central frame holder rides on pins, while the
> whole thing suspended with some rather fine springs. My unit only had 2
> pins at diagonal corners, yet there were mounts and holes for 4 pins, one
at
> each corner. As the film is pushed downward to focus it, tension is
created
> by the fact that the pinch rollers on the left and right side of the unit
> each have a fairly good grasp on the film. As the focus approaches the 20
> mark (around 15 or greater), the film strip is actually under so much
> tension that the frame image will buckle in the middle, completely
changing
> the image's distance from the lens.
>
> Now imagine this, when scanning the first or last frames of the roll, the
> tension only comes from one side. In this case the films goes through
> different gyrations during focusing, partially due to the pair of missing
> vertical guide pins, and some missing screws holding the center frame
holder
> together. For the first frame, with the tension on the left only, the
right
> side of the frame would be pushed down further than the left side, causing
> nothing but a narrow portion in the middle to be focused. The more the
> focus value approached its limit of 20, the more tilted the image became.
>
> [ this is longer than I had planned, so I'll cut to the chase ]
>
> After repeated scans of the same image, the heat from the lamp under the
> film causes it buckle slightly, changing focus over time.
>
> When scanning long strips, 36 images is the software's limit (37 doesn't
> work), and often on Win98 (occasionally on Win2k), the scanner software
> would crash after frame 16 or 17.
>
> The velvet used collects dust quickly, and will add dust and scratches to
> clean film strips. Removing the felt causes the unit to malfunction and
not
> recognize film, as the light sensors think the plastic is film, and tries
> forever to eject a non-existent film strip upon first power up.
>
> Forgoing the warranty (instead thinking class action lawsuit), I
endeavored
> to see if an otherwise fabulous lens and scanner element could be made to
> work better.
>
> I found 2 M3x25 roller pins, ground them down slightly the same height as
> the other two, and put them in the unit. The bound the center image
frame,
> so I driller the holes larger where needed so the binding was gone. I
added
> the 2 missing tiny screws holding the back edge of the center image frame
so
> it wouldn't split under film tension. I removed all the velvet, and above
> each of the 3 sensors I drilled 1/4" holes so I could small squares of
> velvet above them without touching the film.
>
> I also played with the image sizing and focusing (complicated to explain
> here, but requires a caliper to get the focus properly set). I moved the
> proper focus range as close to the "1" setting as possible. "4" turned
out
> to be achievable. This reduced the tension on the film considerably, and
> finally consistent focusing was achieved, though the first, middle, and
last
> frames all had different ideal focus points, but consistent from film type
> to film type, which is better.
>
> To anyone who has gotten this far, you probably realize that the RFS 3600
is
> a disaster, and should be pulled from the market and considerably
> reengineered before allowed to be sold again.
>
> I should add that I've had 2 different 1 hour phone calls with Kodak about
> the film strip focusing problem, and each time that claim no one else had
> problems with focusing using the newest software (which I have), and they
> quickly dismissed my suggestions that it has a design or manufacturing
> problem.
>
> Cheers,
> Gerald
> -Your neighborhood tech reporter on the beat. :)
-
This message is from the Pentax-Discuss Mail List. To unsubscribe,
go to http://www.pdml.net and follow the directions. Don't forget to
visit the Pentax Users' Gallery at http://pug.komkon.org .
