Kodak TMAX Scan Resolution

[bonk]

If I have some sharp and well developed TMX/TMY negatives what would approximately be the minimal resolution I would need to scan them in to not lose anything? Approximately at what resolution the filmgrain is larger than the pixels for TMX and for TMY? Did anyone experiment with this?

It be interested in your numbers for Kodak VC and NC too.

 

[wildbill]

sounds like a question for hybridphoto.com

 

[HerrBremerhaven]

It's not like using an enlarger and a grain focuser. TMax has a ton of resolving power, probably way beyond your lenses, unless you have something truly rare. As with most films, use your non-camera gear at the maximum settings to retain the most information. Quite likely the films will be better than the capabilities of your other gear, unless you have an ICG drum.

Ciao!

Gordon Moat Photography

 

[keithwms]

If I have some sharp and well developed TMX/TMY negatives what would approximately be the minimal resolution I would need to scan them in to not lose anything?

You'll always lose something, but the importance of what you lose will decline rapidly at about ~4000 dpi, 16-bit, with a drum scanner. This translates to ~21 mp for a 35mm frame. If not using a drum, you'll likely need to oversample substantially e.g. to 9600.

Why not print them traditionally!

 

[Bruce Watson]

Approximately at what resolution the filmgrain is larger than the pixels for TMX and for TMY?

Grain clump size (dye cloud size in color films) is a variable. The range is huge -- several orders of magnitude or more for normal exposures and development.

Scanners, OTOH, scan with a fixed size -- sensor size and spacing for CCD scanners, or aperture size and step size for drum scanners.

I'm just sayin' that it's impossible to match scanner resolution to grain clump size. Can't be done.

But, it's not important to do so. Image information is not at all the same thing as film grain. It takes a lot of grain clumps to define even the smallest bit of image information. This is why scanning is possible - you can scan to recover the vast majority of image information. Not all, but close. What you can't do is scan to recover even a decent portion of the film grain (besides, film grain itself is fractal in shape and 3D, where pixels are square, of uniform color, and 2D by definition).

What resolution to scan at is of course a religious argument. Depends on what you want, and what you believe.

But few will argue the fact that printing conventionally will use more of the image information (for a given level of enlargement) and capture more of the look and feel of film grain, as compared to scanning. Some people find this a very important distinction. Like the majority of APUGers I think.

 

[FilmLives!]

Cannot speak directly for Tmax, but I can tell you that ISO 400 35mm film (shot in a Rollei 35S) scanned in a true, old-school film scanner (in this case an underrated-in-its-time Hewlett-Packard S-20 scanner that is now what, at least 10-years old?) scanned at its max optical resolution of 2400dpi is enough for a high-quality 8x10 print (with no cropping). This is in comparison with two 8x10's hanging on the same wall that came from a Canon 20D digital SLR from images shot in RAW.

I also have a scan of a Panatomic-X (ISO 32) B&W negative scanned more recently at a claimed 7000dpi with a Plustek film scanner that produced a stunning 11x14 print that is on my wall and a 16x20 from the same negative that is just as nice too (that's the one that would be hanging if I could find a reasonably priced 20x24 frame & mat for it). The fine grain of the Panatomic-X film was rendered very nicely by the Plustek, I imagine Tmax would produce a scan as nice or nicer. You really begin to see any flaws in your technique at these sizes.

I recently scanned two other images of my Dad's that he shot in the early 70's on the old H&W Film (basically, the original Technical Pan film) on a trip we took to Washington DC. That image showed me how much detail the Plustek was capturing since it took me literally 8 hours each to touch out not only all the dust specs, but also specs of grit and tiny little scratches in the film from what I remembered as pretty careful home processing. It also showed the old Yashinon lens on the rangefinder camera Dad used that was so spectacularly sharp back then was in fact not that great. He's getting a set of these in 11x14 to replace the 11x14 optical prints I'd made for him back then but were stolen off his office wall at work just before he retired 15 years ago.

Bottom line, in a true film scanner (not a flatbed scanner), in my opinion 2400dpi is your minimum for an 8x10, and 5400-7000dpi is likely enough to max out what a home film scanner can capture.


jZ

 

[nickandre]

You always lose "something" in scanning. Any time you transfer a medium save vector graphics or uncompressed forms of digital duplication you are "losing something."

 

[JBrunner]

You've got some good answers here. Further discussion should be taken to www.hybridphoto.com, the APUG sister site dedicated to film/digital hybrid processes.