Film and Digital Prints

[pico]

Advertisement (gone after registration)

One thing that should be mentioned is that other than contact prints, film is always projected and fidelity of the original film is degraded by the projection lens. The degradation is usually insignificant with a good lens, and of course degree of enlargement and viewing distance are significant. I can make some determinations of final print quality by looking at the negative under magnification.

 

Now I look at an image from a converted DNG from the M9 on an Apple display of 101ppi. I'm not even seeing all that is there, am I? Is it true, then, that prints done, say at 240 to 360ppi will have more resolution? More gradient fidelity than what I can see on the monitor?

 

It seems that way to me. Prints have no back light to add to their impression, but they still look better at, say, 16"x20" than monitor versions.

 

All this regardless of the fact that the human eye resolves at what, 6lp/mm? (Which is strictly speaking, not true. It depends upon the texture (shape), color and contrast of the objects.)

[adan]

First, let's eliminate the potential "film vs. digital" aspect of the thread - no point in getting people's knickers in knots. What applies for a digital image also applies for a 4000ppi digitized 35mm film image, in terms of how a screen view compares to a digital print.

 

(and this thread probably belongs in the "digital darkroom" forum - not that I'm complaining!)

 

"Now I look at an image from a converted DNG from the M9 on an Apple display of 101ppi. I'm not even seeing all that is there, am I?"

 

Yes and no.

 

Yes - if you are viewing at a "100% pixel" view. In fact you can probably see detail at 100% on-screen that won't ever be visible in a print unless it is a really big print (print resolution no higher than 240 ppi - and better yet, 180-200 ppi).

 

The reasons being: ink scatter, dithering, and diffusion of the ink on the paper. Take a 10x loupe to an inkjet print to look at the ink dots, and you will see they are not perfect dots. There is a fair amount of smudging and overspray around each dot.

 

The nominal resolution of an Epson at the 1440 ink dots per inch setting is 240 ppi - but those factors I just mentioned mean the pixels will still be blurred slightly at that rez.

 

Always remember the old saw about "300 ppi" for prints was developed about 1990 for high-end glossy magazine printing at 150 ink dots per inch with rotated half-tone screens on a printing press. It really is not an appropriate standard for sprayed ink from an inkjet printer

 

To summarize (M9 or 4000 ppi film scan, print dimensions in inches):

 

print at 300 ppi = 11 x 17 print, with a lot of finest detail "lost" in the print

print at 240 ppi = 14 x 20 print, with a bit of detail in the image "lost" in the print

print at 180 ppi = 20 x 30 print, with probably most/all of the original image detail visible

print at 150 ppi = 23 x 35 print, with every detail reproduced

print at 101 ppi = 34 x 52 print, exactly replicating a "100% view" on-screen with an Apple 101 pixel-pitch monitor (actually my 27" Apple iMac display is pitched at 110 ppi - screen res keeps creeping up)

 

Note that the last two resolutions not only guarantee that every detail is visible, but also that every artifact (moire and other aliasing) is visible - thus 180-200 ppi is the more "ideal" range for "highest visible detail without artifacts").

 

Those resolutions also assume that each image has an "appropriate" amount of output sharpening. the bigger the pixels, the lower the sharpening needs to be.

_____

 

No, if you are viewing at "print size" view. As you surmise, you are then viewing a picture (M9 or 19 Mpixel scan from 35mm) 3456 pixels tall, "dumbed down" to 1616 pixels tall (16" x 101 screen pixels per inch). So you really are "not even seeing all there is" in that case.

[pico]

print at 240 ppi = 14 x 20 print, with a bit of detail in the image "lost" in the print

print at 180 ppi = 20 x 30 print, with probably most/all of the original image detail visible

 

Thanks very much, Andy. I think the above will be my targets for a preview portfolio. I'll give it a try tomorrow, after I get the profile for my printer-guy's printer. B&W is coming out too dark on it now.

 

And I had forgotten about the very fine reproduction techniques of the better publishers. It really is rocket science to me.

[johnloumiles]

Pico and Adan,

 

I'm writing to you in here because my question is somewhere in the vicinity of what you are talking about and I trust your answers.

 

I'm working with a M6 recently and after having the first few rolls developed and scanned I'm not seeing the magic. Can you let me know your ideal scan settings?

 

Thanks

[adan]

That's a really broad question.

 

Without knowing specific scanner and film type (brand, B&W/color neg/color slide, ISO), the only "general-purpose" settings I can recommend are: use the maximum resolution setting and the highest bit-depth, and don't apply any unsharp masking when scanning (Photoshop gives better sharpening control).

 

Realistically, my "settings" were mostly made on the fly, per image, just as I would try different things in the darkroom with a given neg. I never batch-scanned.

 

I also learned to shoot and develop B&W negs differently for scanning than I did for chemical printing (a bit less exposure and development).

[johnloumiles]

I had a whole explanation with the last post but then I deleted most of it because I didn't know if it was just my eyes and not the scans.

 

Basically I'm using a M6 with great Leica Glass and the images are looking rough. Not in a film grain sense, there just ins't a smooth sharpness to them if that makes sense. I'm hoping its the scans from the lab but I rescanned with my Microtek Scanmaker i900 at 400 dpi and I'm not seeing that much better results.

 

Its been about 5 years since my film days but I didn't scan back then, prints or contact sheet only. I'm hoping I'm doing something wrong with either shooting or scanning.

 

Here is a photo I'm talking about below. M6 with Elmarit 28m/Kodak Ektar 100. I just feel like if I shot that with my M8 the result would have had a smoother draw to it ( I'm sorry its hard to quantify in words what my eye is seeing).

 

There is the possibility lack of sleep is clouding my vision so if thats the case I apologize ahead of time.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[ron110n]

Advertisement (gone after registration)

I rescanned with my Microtek Scanmaker i900 at 400 dpi and I'm not seeing that much better results.

 

 

John,

 

1. We usually scan *at least* at 1200 dpi on a dedicated negative scanner. You need to collect as much information on the negative as much as you can.

 

2,. I like to stay away from Digital Ice because it compromises the quality, so I pass the negative to a Static Vac then blow it with a Dry Compressed Air before I scan, then Photoshop can clean what's few that is left.

 

Hope this helps.

 

-Ron

[johnloumiles]

Very informative thank you Ron. I'll give it a go and see what happens!

[plasticman]

Hi John

 

I'm by no means an expert, as I'm a very green film newbie, but I hope you don't mind me contributing some ideas?

 

- I'd start by trying another film: I personally use Portra, both 400 and 160, and while I haven't tried Ektar (which is supposedly even more smooth-grained), the slightly more muted color of Portra may contribute to the 'smoother' look you seem to be striving for.

 

- I scan at 4000dpi. If something like this is possible with your scanner, then try it.

 

- invest in a room air cleaner if dust on your negs is a problem. I bought a beautiful unit from a Japanese designer called plusminuszero, but I'd guess that any model would do the job. I rarely had dust on my negatives before, but now I most often have none, whatsoever.

 

- If you're looking to emulate the smoothness of an image from a digital sensor, then you're going to need to switch up to MF, I'm afraid.

 

Now, I'm currently struggling to analyze why I prefer the look of images from film over the smoother, more detailed images I get from digital, but I think that the subtle texture in the image and precisely the absence of an artificial smoothness is one of the important contributory factors.

By this I mean that if I simply look at skin or a wall or even a glass filled with water, the acuity of my vision discerns a texture rather than a simple graduated flatness, and the fine grain in a film image reflects this 'reality' more accurately than the smooth interpolations of a Bayer grid. (I'm sure there will be plenty of people that disagree - this is the beginning of my struggle to articulate why film better reflects the world I see).

 

Best of luck!

[adan]

mani - a good point. One reason why Tri-X pictures (printed well) often "looked" as sharp or sharper than Pan F or Panatomic X, even though the latter usually had demonstrably more detail. If the Tri-X grain was crisp to the eye, the picture looked "sharp."

[johnloumiles]

Hi Plastic

 

I've run the Porta gamut in the past and 800vc was my goto for a long time but I tried Ektar this time because it's supposed to be even smoother. I don't think so at this point. I don't care for the color temperature either.

 

The only other rolls I've ran through the M6 is PAN 50 and Fuji Pro 800. Contrast on 50 and color on 800 were fine but still looked rough.

 

When I get home I will be scanning highest possible res. The scanner is supposed to be great, it has a separate drawer for negatives so it doesn't scan through glass.

 

I've shot film for years, mostly Mamiya's , so I know what I looking at is not what I should be seeing. The M6 should be rendering even better then a RB or RZ if only because of the lenses. That's why I hope it's the scanning because at this is point my Yashica T3 is the reigning king in my film collection!

 

On my end I don't love any photo I've shot so far so I have some work to do.

[pes084k1]

One thing that should be mentioned is that other than contact prints, film is always projected and fidelity of the original film is degraded by the projection lens. The degradation is usually insignificant with a good lens, and of course degree of enlargement and viewing distance are significant. I can make some determinations of final print quality by looking at the negative under magnification.

 

Now I look at an image from a converted DNG from the M9 on an Apple display of 101ppi. I'm not even seeing all that is there, am I? Is it true, then, that prints done, say at 240 to 360ppi will have more resolution? More gradient fidelity than what I can see on the monitor?

 

It seems that way to me. Prints have no back light to add to their impression, but they still look better at, say, 16"x20" than monitor versions.

 

All this regardless of the fact that the human eye resolves at what, 6lp/mm? (Which is strictly speaking, not true. It depends upon the texture (shape), color and contrast of the objects.)

 

Projected slides with Colorplans are about equivalent to a 12-14 Mp resolution for each color (50+ lp/mm on film). At 1 m distance from the screen the eye resolution equals 2 lp/mm (100 ppi) for moderate contrast lines, i.e., about the typical monitor resolution. But the largest PC monitors have only 2 Mp/colour and a much smaller gamut. The 4K screen will have similar resolution than 35 mm slides slides, but still less gamut and less brilliance. Projector still wins with good media.

[pes084k1]

mani - a good point. One reason why Tri-X pictures (printed well) often "looked" as sharp or sharper than Pan F or Panatomic X, even though the latter usually had demonstrably more detail. If the Tri-X grain was crisp to the eye, the picture looked "sharp."

 

It comes from acutance of high energy developers. But a TMax 100 or a Pan F+ used with high acutance developers like Gradual ST-20 or TMax will have far more detail, crispness and less grain than Tri-X: the difference with D76 is really impressive on enlargements.