Post your drum scans (aka the first official Drum Scanners thread)

tsiklonaut said:
I'd actually recommend installing OS9 in the same machine and the correct OS9 SCSI drivers for your Adaptec card.
Click to expand...

Scanner is visible under OS 9 ! Now don't know what to do - my drum has no bar code. I did preview scan but lamps does not work.
 

Attachments

  • Untitled-1 copy.jpg
    Untitled-1 copy.jpg
    32.8 KB · Views: 0
jzagaja said:
Scanner is visible under OS 9 ! Now don't know what to do - my drum has no bar code. I did preview scan but lamps does not work.
Click to expand...

Good to know, told ya OS9 would help ;)

If no lamp with transmission try previewing in reflection mode as well to see if the other lamp works.

If not then check if the wires in lamps are not broken (take out the lamps and see them close), if they're good then check the wiring going into the lamps and fuse for the lamps.

Good luck,
Margus
 
Can preview be stopped? How? Two lamps looks ok but doesn't work. Will check power supply board. Is bar code on drum important? What about focus calibration?
 
Margus,

Thank you for "filling" in the missing information, as I do not have a true drum scanner.

For all 3 of my scanners, I am either using OSX in its various flavor due to software legacy that doesn't go beyond certain point.

At least for the latest software that was written for OS10.4.12 Tiger and Leopard for my Scitex & Imacon. Eventually most will become obsolete as scanners are no longer made.



tsiklonaut said:
If it does rotate during boot it's a major sign it still at least partly lives. I'd try Adaptec under OS9 to be sure.

Yes, "ScanMate 11000" should show in System Profiler under SCSI/ Parallel SCSI / Power Domain XXX (XXX - your SCSI card model).

Since OSX can have its quirks with SCSI hardware I'd actually recommend installing OS9 in the same machine and the correct OS9 SCSI drivers for your Adaptec card. I've found OS9 can solve many questions about things running or not since it's a "simpler" OS so to speak. Under OS9 Adaptec drivers create "SCSI Probe" program - with this you can see the card you have connected and if you have the scanner connected onto this particular card. I.e. if it shows the correct SCSI card but no scanner then start messing with the SCSI cables, ID and terminators. If no card then it's the card issue. Etc etc, much better for fault-finding analysis than OSX.

You can keep OSX as well - just use Startup Disk function to boot into one or another.

Hope this helps,
Margus
Click to expand...
 
jzagaja said:
Can preview be stopped? How? I probably lost one of the dichroic filter (green) during opening - must double check. Two lamps looks ok but doesn't work. Will check power supply board.
Click to expand...

You can cancel but it's pointless, since it takes ages. Better use i.e. 1-2MB preview size, so it's quicker, but still couple of minutes. Remember everything is like in slow motion in drum scanning (except the fast rotating drum) so be patient and don't damage anything - you just create additional problems with rushing things :D

Ouch with dichrotic filters - they are precision filters so you need to put it back exacly the same how it was inside.

Here's a pic of my SM11K sensor unit and the dichrotic filters, one I took out for inspection of dust or wear. I used the touch marks on the filter surface (that touches to the soft surface on one side) to put them back inside with the correct "side" up. Do not mess up anything there or you're in trouble with messed up RGB channels or twisted overtoned colours.

attachment.php



attachment.php


Hope this helps,
Margus
 

Attachments

  • SDI1114.jpg
    SDI1114.jpg
    40 KB · Views: 0
  • SDI1117.jpg
    SDI1117.jpg
    34.5 KB · Views: 0
Evanjoe610 said:
For all 3 of my scanners, I am either using OSX in its various flavor due to software legacy that doesn't go beyond certain point.

At least for the latest software that was written for OS10.4.12 Tiger and Leopard for my Scitex & Imacon. Eventually most will become obsolete as scanners are no longer made.
Click to expand...

It's the same with ColorQuartet 5.3.1 running ScanMates - OS10.4.12 is the latest.

But for me there's no worries about the OP system or scanning software, the best scans always come in linear mode so you can even scan in DOS OPsystem if you have to as long as you have decent optical signal going into AD converters, focus and 16-bit scanning ability :D Basically there's no need to run along with the latest OS or scanning software. A pro 16/32-bit photo editor such as Photoshop does the same tricks (sharpening, levels, curves, highlight/shadow etc etc) A LOT better than most (with just handful of high-end exceptions) of the modern scanning softwares offer and with PMT drum scanning you do not need multi-scan like CCD scanners need to get good scans since you already get very "organic"-looking and excelleint shadow detail scans, IMO.

So being "obsolete" in scanning software or op-system wise is not really an issue - they're good for decades to come - the scanning files are still the same while post processing editing softwares progress rapidly. Only storing and exchanging big files with your main (modern) workstation computer can be a bit hassle with older systems connected to your scanner.

IMHO of course.

Margus
 
jzagaja said:
Can I touch them?
Click to expand...


NOOO! :p

Only with clean brand new cotton "photolab" gloves that leave no traces of fat/dirt.

Or if you're used to handling small bits - holding them from the extreme corners (wash your hands throughout before :D), never touch the middle with your own fingers! The middle is where the optical signal has to selectively reflect and work through the surfaces.

With antistatic brush or air carefully blow clear any dust particles they may have on before putting back. They must be explicitly clean.

Their order and positions you must figure out yourself! I may be a physicist, but I don't have a sixth sense to know what you're up there :D

Cheers,
Margus
 
Ten years ago I was interested in colorimetry, writing Photoshop plugins for inverting negatives etc. Do you remember that old discussion about image manipulation in linear vs. logarithmic domain? ;)

Lamps not working - contact were dirty :) Have you tried install LEDs?

Two questions - do you have below shown halos around edges and how do you set focus on non barcoded drums?

I also have problem with preview - reflective original placed in the middle of a drum but preview shows one half on left and another half on right side.
 

Attachments

  • Untitled-1 copy.jpg
    Untitled-1 copy.jpg
    47.5 KB · Views: 0
jzagaja said:
Ten years ago I was interested in colorimetry, writing Photoshop plugins for inverting negatives etc. Do you remember that old discussion about image manipulation in linear vs. logarithmic domain? ;)

Lamps not working - contact were dirty :) Have you tried install LEDs?

Two questions - do you have below shown halos around edges and how do you set focus on non barcoded drums?

I also have problem with preview - reflective original placed in the middle of a drum but preview shows one half on left and another half on right side.
Click to expand...

As long as it doesn't clip any useful info who cares about the editing domain, be it linear, logarithmic or even log-linear modelled domain ;)

LEDs - unless you can find diodes that match the ScanMate's specified halogen lamp's spectrum then let me know, since all the optical signal path and color space is optimized around this particular spectrum. I don't think you can find such LEDs.

CQ previews tend to create halos with SM11k (while no halos with SM3000 previews). I guess it's some quick-scan cap with some channels or part of the info skipped. Real SM11K scans have no halos.

If CQ can't read the barcode it should save it automatically under default drum and use focus data for this, so it's no real problem. You can also save that non-barcoded drum under any number 1-10 of drums in CQ. See CQ user manual for this info.

Media "placement" on drum can have 120-degrees of difference depending how you mount the drum on the motor.

The correct placement is when the recessed dot matches the dot on the motor wheel. But you can always rotate by 120 degrees if you somehow misplaced the media.

Always do it in drum-lock mode. With the lid open use the left-side "ScanFlow" button on scanner so it recesses the moter into "lock" position and rotate the motor wheel into lock-mode with your hands, then remove /or/ mount the drum and see the markings between them.
 
jzagaja said:
First scan - Superia 200:

https://docs.google.com/file/d/0ByP8kkW_h00uN1pQRWRTTHQwWWs/edit?usp=sharing
http://www.youtube.com/watch?v=WZiBTWFDB0A

Will check focus later today.
Click to expand...

Yep, somethings not right there. Definitely not fair to get your drumscanner's azz handled by a mere V700 :D

You can see my ScanMate 3000 vs V700 comparisons on the first page of this thread. I haven't even compared ScanMate 11000 with Epsons since I already know 11000 is a lot better than the 3000 I used before so there's no competition.

So did you use CQ for negative inversion or you scanned them as linear positives and inverted them later in PP? CQ is not the best solution for negatives - I always scan C41 as positives.

I can already see you have the "zaggy" problem - i.e. the contrasty black brake cables on the bike have noticable zag lines. This is caused either a loose drum mechanism, unbalanced drum (place media on both sides of the drum if you have to), shaky table where the scanner sits or not tightly mounted negative (too tight creates Newton rings, so it's practice to get it right alignment and tightness - wet-mount always works the best avoiding those two problems). THis also can also be a SCSI transmission issue or dying-DC motor, but highly unlikely.

For the best results clean all optics - sensor head glass (very carefully!), light tube glass, opical fibre end from the lamps side.

Calibrate white (from completely clear part of the drum) before doing any color work.

Check your drum's alignment - did you get the focus calibration target with the scanner? If yes first place it on the left side of the drum. Do autofocus, then scan center part of it. Then place the same target on the right side of the drum, and scan the same part (NB! w/o focusing again). Then compare the two scans, if one is more blurry than the other you need to align the drum, using the same drum "two extreme sides" method step-by-step adjusting, autofocusing again and comparing the two scans till you have perfectly even focus on both sides of the drum.

Since it has very shallow DoF - overall focus is critical, autofocus that uses grain is very good on ScanMates but some very detail critical jobs sometimes require some fine adjustments using focus-elevation function.

And now you owe me a beer :D

Cheers,
Margus
 
Ok so come one day to Szczecin for a beer - this year we had Tall Ships' Races 2013 :)

I've scanned in 16 bit TIFF and inverted in PS. 16 bit linear TIFF was written as 8 bit - don't know why :) Negative has gamma 0.6 if I remember corectly so I inverted that but log curve is probably best (from linear RAW):

1) Linear RAW
2) White point on non exposed part
3) Log curve
4) Invert
5) White point (levels) manually
6) ICC profile for particular CCT (correlated color temperature) made on reflective target

Example:
https://docs.google.com/file/d/0ByP8kkW_h00udFc1ZDNVdVhhZFU/edit?usp=sharing

I have no focus target so what can I do?
 
jzagaja said:
Ok so come one day to Szczecin for a beer - this year we had Tall Ships' Races 2013 :)

I've scanned in 16 bit TIFF and inverted in PS. 16 bit linear TIFF was written as 8 bit - don't know why :) Negative has gamma 0.6 if I remember corectly so I inverted that but log curve is probably best (from linear RAW):

1) Linear RAW
2) White point on non exposed part
3) Log curve
4) Invert
5) White point (levels) manually
6) ICC profile for particular CCT (correlated color temperature) made on reflective target

Example:
https://docs.google.com/file/d/0ByP8kkW_h00udFc1ZDNVdVhhZFU/edit?usp=sharing

I have no focus target so what can I do?
Click to expand...

For me it's 16bit TIF, white point from clear area of the drum - not from the clear area of the film. Non exposed (backing) film area you correct later in PP to remove coherent color overtone of the film backing (see "smart inverting" text below). Drum scanners have their pure-white level optimized to pure optical signal through the system - meaning completely clear and clean area of the drum where's no film placed. White you don't need to calibrate often, just when you switch between wet- (then through the wet-clear area) or dry mount (just clear area of the drum). When you work with just one mounting method you don't need to white-calibrate often.

Curves in CQ are a bit pointless since it scans linear with hardware and modifies curves through software before saving file. PP does the same curve modification with better quality. Ditto black/white point temperatures, levels, sharpening etc etc.

Inverting is the biggest trick in getting fine C41 results from linear (positive) 16bit TIFs. My warmest recommendation is NOT to use direct-invert, do it with curves and smartly since you will not lose so much info in multiple-process steps you'd need to to with direct-invert to have it balanced later. Smart inverting is something you must discover yourself, do your homework and see tutorials in the net etc - smart inverting is where the high-quality C41 world really will "reveal itself" so to speak IMO. ;)

Critical focus is something that needs to be checked regularly, since different films act slightly differently although focus place is the same in the camera the emulsion's "depth" (i.e. E6 vs C41 vs B&W emulsions) can play it's part on finding the sharpest detail focus.

Focus target - maybe contact the guys at ABC-scan, they can send you a target? Not sure how much it'll cost. This you can probably replace with a similar VERY sharp target. I.e. a standard resolution target with vertical and horizontal lines can work.

Can't see the pic - needs permission/login.

Margus
 
jzagaja said:
Agree with all of this :) When I first did negative scan in CQ image wasn't inverted and when I saved TIFF it was inverted and washed out. Don't know how to save pure linear raw file without power function applied.

I unlocked the file - this is a scan from Epson:

https://docs.google.com/file/d/0ByP8kkW_h00udFc1ZDNVdVhhZFU/edit?usp=sharing
Click to expand...

Simple, just turn off all "open-auto-close" functions in any of the adjustment windows in CQ (if you've made them or set-up them before). If you haven't saved any of the presets then just select the crop you want to scan, double click on this crop opens the scan dialog - select 16bit TIF, turn off sharpening, put required dpi values and color space you require and scan.

Now you have a 100% linear scan - exacly what the scanner "sees" optically and outputs it to 16bit TIFF w/o any software manipulations.

The rest do in PP. As said C41 is very sensitive to inversion, and smart inverting is the key for good results. Google and see some tutorials posten in RF forums as well about inverting a color negative from flat-linear scanning files. It's all about how to use curves.

With B&Ws you can do direct invert function (and desaturate if you like to). And use curves to get the tonality right for your taste.
 
I feel comfortable with negs. The only drawback is that you work with large linear 16-bit files (that’s why movies are stored 10-bit log), transfer from Mac to PC for better comfort.

How do you mount drum when zero line is missing?
How do you wet mount without station? – I have cosmetic naphtha that seems ok.

All my current scans are dry mounted.
 
You must log in or register to reply here.
Back
Top Bottom