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

Thank you Jack!

I use CQ 5.3.1 on Mac OS X 10.2.8 and it runs fine (within the known scan size limitations), but I'll definitely have a look at ColorTrio on Windows. :)
 
Difference between old and new AD converters (underexposed Velvia). Old SCSI boards had mods probably for stripes in shadows issue. We could try this mod on newer boards.

11212293704_ea61f27c53_b.jpg
 
Ed Bray said:
Some lovely imagery there Margus, really love the feel of the 1st image.
Click to expand...

Thanks! Good to know you're getting there with yours :cool:


jzagaja said:
Difference between old and new AD converters (underexposed Velvia). Old SCSI boards had mods probably for stripes in shadows issue. We could try this mod on newer boards.
Click to expand...

Seeing the pics of both boards the newer board seems quite different though, from ADs to some other parts and routings in the circuitry. Probably need some very good electronic person to compare two of them and "crack" the secret of the old board mod hoping it's applyable to the new one.


Fernando2 said:
I have a < 200 board, probably (never had banding).
Click to expand...

Or maybe you already got your scanner with the modification done. Check the year and serial written on the back of the scanner. All the newer "stock" control boards have this issue and can be modified by ABC-scan, costs around €400 - I know since I had to do this mod on mine.
 
All the newer "stock" control boards have this issue and can be modified by ABC-scan, costs around €400 - I know since I had to do this mod on mine.
Click to expand...

Problem is ABC don't want do this - they want complete scanner and €800. I don't know how to speak with them.
 
Color Quartet 5 parallel port dongle through Unitek USB adapter does not work :( Not seen through Sentinel Advanced Medic.
 
Very chuffed today, managed to get my first two 11,000 dpi scans from 35mm negatives. They were done in RGB Lab, 2750%, 200 lines and quality of 2.0 bits, no sharpening and an aperture of 2. They gave file sizes of 550MB.

Film was Tri-X @ 400 in Leica M4 with Voigtlander 21mm f4 Color Slopar processed in Divided Pyrocat HD.

I did convert one to mono and set it at 8 bit which gave an 82MB file, no other PP was carried out.


11000dpi 35mm Tri-X by Ed Bray, on Flickr
 
jzagaja said:
Color Quartet 5 parallel port dongle through Unitek USB adapter does not work :( Not seen through Sentinel Advanced Medic.
Click to expand...

No need for a dongle in OS X if you're resourceful. Unless the dongle does part of the image processing... does it do any?
 
I've tried Mac and no thanks. Using 10 years old machine is a pain. Scanner only worked with MacOS9. This is stupid that Sentinel does not allow USB-parallel adapters :(

Anyone tried hack CQ for Windows?
 
jzagaja said:
Color Quartet 5 parallel port dongle through Unitek USB adapter does not work :( Not seen through Sentinel Advanced Medic.
Click to expand...

Don't know anything about Windows CQ. But keep searching, maybe you'll find a solution.


Ed Bray said:
Very chuffed today, managed to get my first two 11,000 dpi scans from 35mm negatives. They were done in RGB Lab, 2750%, 200 lines and quality of 2.0 bits, no sharpening and an aperture of 2. They gave file sizes of 550MB.

Film was Tri-X @ 400 in Leica M4 with Voigtlander 21mm f4 Color Slopar processed in Divided Pyrocat HD.

I did convert one to mono and set it at 8 bit which gave an 82MB file, no other PP was carried out.
Click to expand...


I reckon it's a 8 bit Lab scan then to get more than 10 000 px a side? You've lost some of tonal information on inversion in this case.

Overall a very decent detail pulled out there Ed. I never considered the Tri-X a high-resolving film, but it does the job on wide-range of subjects. It's a grainy film and most of the subjects is shot through haze in distance, hence contrast problems, but the grille in the foreground below and the scratches+dust in the frame give you a glimpse of the scanner's true resolution. Scratches going through the emulsion diagonally give you around 1-1.5 pixel edge-sharpness, while the grille for example is close to the 1-2-pixel range edge sharpness, hence the resolving power is close to the actual 11000 ppi. And this isn't interpolated source like dslr-scanning CCD/CMOS etc, where sensor interpolates out perfect-straight lines or perfect 1-pixel sharpness 'cutoffs' over long distances - a drum scanner is "blind" to interpolation since it scans one-pixel-at-a-time, it doesn't know how to cheat you with digital interpoation or enhanced edge-sharpness. Multiple sources have measured ScanMate 11K to be around 6000-9000 ppi depending on the direction (you still have to scan at full 11K to get this max resolution btw). Some mechanically tight and well calibrated units with excellent overall optics condition could probably score closer to the 11K in all directions, but I guess most "run-in" units would fall into that 'practical resolution' range of 6-9K range.

Margus
 
Thanks for the comments Margus, the scratches are on the drum, that is what happened when the drum came off the revolving mount. It is okay for 2 strips of 35mm (I mounted 4 because its a PITA wet mounting with 35mm) or 1 of 120, for any large format scans I will use the other 'new' drum.

The tonal range is where I am having difficulty at the minute, I am having to open the 'film type' box with custom and sort of guess using the graphs until it looks about correct. I am always losing a bit of shadow and highlight detail and even though there is visible definition in the negative I just can't seem to pull it out at the moment (that's the next improvements in my scanning).

It's amazing how they can build such bendy turrets and stop them falling over ;-)

Lovely colours in that 2nd image too.
 
Ed,

I can give you a workflow for negatives (RAW 16 bit scans with ICC profile embedded). PM me if you like. I also have Negative Inverter plugin somewhere (at least source code for Filter Meister).
 
You must log in or register to reply here.
Back
Top Bottom