Linotype-Hell Tango blog

[georgl]

I've started a blog to collect and share my experiences with the Tango and other scanners:
http://drumscan.blogspot.com

Currently it discusses the Tango and a brief comparison to the Howtek D4000, Scitex / Creo Eversmart Pro II, Minolta Dimage 5400 and the Imacon Flextight 646.

It's my first blog, so please bear with me, still have to correct many things, I still hope it's helpful to some and I will add further posts in the future!

[buggz]

Nice, thanks for this.
If only I could afford one, then shipping would as be as much as the machine, sigh, one day, one day...

[cosmicexplosion]

thats a great thing you are doing, well done.

the dimage looks the best, and at lower res than tango , why is that?

[georgl]

the dimage looks the best

Sadly, these scans were made years apart and not all of them were saved without processing. So the only chance for me was to sharpen and process all of them - which doesn't work perfectly and the second crop already shows much softer rendering on the Dimage. The second image is done under almost perfect circumstances and I try to do all future comparisons in a similar way but I didn't want to keep the broader first results from you, either.

[Joerg Krusche]

Hi,

which software on the Tango .. Linocolor or Newcolor Pro ? .. did you ever run a USAF test chart on the Tango ? .. and which max resolution did you obtain ?

best,

joerg

[Lenny Eiger]

The Eversmart II and Tango are at the top of their line. Why use a D4000, which is three or four generations back. Why not compare to an Aztek Premier?


Lenny

[cosmicexplosion]

Are you sure it's an 5400 I can only find info on machines that do up to medium format?

[georgl]

I run the Tango with Linocolor 6.0 on a Mac G4 and more recently with a modern PC on Newcolor (which runs under Win XP). Linocolor is only 8bit while Newcolor adds some sharpening which makes comparisons difficult and doesn't scan beyond 1GB (strangely, the old G4 with Linocolor can scan a 8x10"-slide @2400ppi!). I will add an article over the software later.

The Minolta Dimage 5400 was a popular 35mm scanner with 5400ppi max. resolution - This (http://www.filmscanner.info/MinoltaDimageScanElite5400II.html) was the last model.

The Aztek Premier is nearly unobtainable here in Germany, I have a Howtek D4000 myself and will discuss differences in future articles but I also had the chance to test this one slide on a D4000 in mint condition with an experienced operator. In theory, the Premier should offer higher resolution with 8000ppi and a 3µm aperture but if that's actually the case is hard to tell. In my experience, the Tango gains most information out of the originals despite it's 10µm aperture (vs. 6µm on the D4000). It also runs much smoother due to it's higher build-quality, I guess this also affects actual resolution. I don't have the USAF-target but when somebody is willing to send it to me (also with test slides of your own) I would be happy to test it. I will add another article regarding the aperture and resolution as well.

[cosmicexplosion]

not sure how you relate the dimage to large format?

[georgl]

Not at all :-) The blog is meant to be about the Tango which I also bought because I'm getting started in large format photography - but the Dimage was tested in different times...

[Lenny Eiger]

The Aztek Premier is nearly unobtainable here in Germany, I have a Howtek D4000 myself and will discuss differences in future articles but I also had the chance to test this one slide on a D4000 in mint condition with an experienced operator. In theory, the Premier should offer higher resolution with 8000ppi and a 3µm aperture but if that's actually the case is hard to tell. In my experience, the Tango gains most information out of the originals despite it's 10µm aperture (vs. 6µm on the D4000). It also runs much smoother due to it's higher build-quality, I guess this also affects actual resolution. I don't have the USAF-target but when somebody is willing to send it to me (also with test slides of your own) I would be happy to test it. I will add another article regarding the aperture and resolution as well.

The Premier is nearly unobtainable anywhere. Rarely does a used one come up for sale, as they are usually kept by their owners. However, that said, I am thoroughly sick and tired of Luminous Landscape's comparing the top-of-the-line digital camera to a med-level or consumer level scanner, often with a poor operator. So this has become one of my pet-peeves.

As Tim Parkin will attest (I have just done over a hundred scans for him for free), I am happy to scan things for testing purposes for people that are writing articles in the media. We need to have fair comparisons to there if we want to see film continue - at least for a bit. If you want some assistance, let me know.

Lenny

[georgl]

I have to make a few reference slides/negatives, I ruined previous attempts. I would scan these for myself and when they meet your expectations, I could also send them to you.

I'll have to wait till my dealer has the new Apo-Summicron 50mm for a few shots - seems like the ideal lens for this purpose :-)

[georgl]

I finally found some time to dance some tango ;-) I had to use the Mac-version of Newcolor (the 16bit-capable successor to Linocolor) on my old G4 (with Mac OS 10.2.8) because the Windows-version has serious bugs (sharpening cannot be turned off, no scans beyond 1GB) - I will discuss my experiences in my blog in a few days including some samples.

But here it is, my first 8x10"-slide and also my first "real" scan:

http://www10.pic-upload.de/03.07.12/6nanfarquog.jpg (http://www.pic-upload.de/view-14960945/8x10bmp.jpg.html)

And some 100% crops:
http://www7.pic-upload.de/03.07.12/jduz6uc2929y.jpg (http://www.pic-upload.de/view-14960947/crops.jpg.html)

It's the "christmas-tree" at Potsdamer Platz, Berlin. 8x10" Provia 100F, Sinar adhesive holder, Schneider 150XL, F22, ~10sec.

The scan was made @ 1920ppi, max. resolution at this size without stitching, resulting in 1.7GB (no fun on a 14 year old G4!) at over 300 Megapixels. I added some brightness to the overall quite dark slide and sharpened it (150/0.3) - nothing else. Colors are 99% right, highlight & shadow-rendition similar to the slide on the light-table! Clearly superior to my previous scan on the Eversmart (stray-light is an issue with these extreme slides).

[georgl]

Can you guys see the images? It works on my computers but maybe I have to use a different upload-service?

[pinup tragic]

I get pic - upload.de
site don't allowed ------ a 404 image i think - even stranger English :)

[MisterPrinter]

Can you guys see the images? It works on my computers but maybe I have to use a different upload-service?

I can see them fine.

[georgl]

I also updated my blog (http://drumscan.blogspot.de/)with this scan and a brief comparison to the Eversmart Pro II.

[EOTS]

Hi Georg!

Thanks for sharing your findings (here and also in your excellent blog)!


The scan was made @ 1920ppi, max. resolution at this size without stitching, resulting in 1.7GB (no fun on a 14 year old G4!) at over 300 Megapixels.

Regarding the max. resolution of 1920dpi limitation:
Is this caused by a limit within the Newscan software or by some kind of hardware limit?
As you tried Silverfast (following your blog), do they have the same limitation?

Best regards and thanks,
Martin

[SergeyT]

Tango has a hardware limitation of 16000 pixels per line in drum rotation dimension.
So with an 8x10in film one should expect up to 16000/8 = 2000 dpi resolution.
For a 4x5 it will be double of that (16000/4=4000 dpi) and so forth while it reaches its 11000 dpi with a 35mm film.

SergeyT.

[EOTS]

Thanks SergeyT, that explains it!

[Leszek Vogt]

Has anyone seen any tests done on the new 35/120 Plustek ? I can appreciate the tests and your blog, though I would rather have someone scan my 4x5 work vs investing in software, hardware, high end computer/monitor, and all the time and frustrations to get every morsel-atom out of the scan. Sure, I enjoy quality, but frankly I'd rather use that time to be out there and enjoy the nature and photography. By the way, much like Nikon scanners, I just heard of two Dimage type bit the dust - they are no longer supported and obtaining parts is increasingly difficult.

Les

[Karl Hudson]

Great Blog Georgl! I'm in Germany too...in Kiel (birthplace of the Tango)...would enjoy the chance to meet sometime.
Karl

[Nathan Potter]

Georgl, I think one can just roughly scale a magnified section of your photo above from the 8 X 10 original.
Your small sections are about 16X of originals as shown. If I measure the width of about the finest line resolved on the magnified sections I get about 0.5 mm (500µm). That would translate to about a 15 µm line on the 8 X 10 original, albeit of quite low contrast I suspect. A 15 µm line is 75 lines/mm or 37 lp/mm really quite exellent for an 8 X 10 image. This would equate to about 1660 spi. as reproduced. This seems to be a high quality 8 X 10 original.

Nate Potter, Austin TX.

[Nathan Potter]

Moderators: you may want to delete my post immediately above; it is in error.

OOps, I should respond to my thread above. What was I thinking last nite?

If I adjust the image size on my screen to about 8 X 10 then the enlarged section is about 7 mm wide on the 8 X 10. Going to the enlarged section on screen then, that represents the 7 mm wide section on the original 8 X 10.
The finest line I can see in the enlarged section is perhaps 0.5 mm (500 µm) wide and appears to be about 110 mm wide on my screen. Scaling the 0.5 mm as a fraction of 110 mm we have 0.5/110 = .0045. Thus the width of the finest line on the original is .0045 X 7mm or .0045 X 7000 µm = 30 µm. 30 µm is nominally about 33 lines/mm. or 16 lp/mm. This is a more likely practical limit for 8 X 10.

The caveat here is that we are looking at a scanned image so is there a resolution limit imposed by the scanner or by the camera and lens? If it were imposed by the scanner then the implication is that the scanning was done at about 840 spi. Georgl reports that it was done at much higher spi (about 1800) so it seems likely that the resolution limit is with the camera and lens arrangement, which contrary to my previous post puts the rez on film not so high and may not be evaluating the optimum performance of the scanner.

Nate Potter, Austin TX.