Noritsu scanner - showing rebate question

koraks said:

Their not offering tiff would be a dealbreaker for me; 8 bit color space is a problem. The sRGB vs Adobe RGB isnt much of an issue; there's very, very little real world gain in that and arguably, working with color negative film, there's no benefit at all to adobergb. What's more relevant is that they probably send you inverted, color-balanced digital files, which means that most of the flexibility is already gone from whatever they send you.

Click to expand...

Interesting that you believe no big difference between sRGB and Adobe RGB 1998 for color neg scans. I have always used RGB 1998 for everything (scans, Photoshop, etc) to make sure I have as much color info as possible. I'll have to do a comparison. Maybe I'll see you're right.

I think most labs use sRGB because their printers use sRGB. If scan is RGB 1998, they have to then convert to sRGB for printing.

Also, so many people use their scans for internet where, as I understand it, sRGB is more universal, and you are less likely to get surprising colors.

calebarchie said:

Loading your film in camera slightly askew can cause this, also it's just a drop down menu.

Click to expand...

What is just a drop down menu? Do you work with Noritsu scanners?

@calico

1. Yes HS-1800
2. In EZcontroller once the 120-AFC is setup selecting wide framing is just an option in a drop-down menu
3. The carrier is auto-framing, if the neg is underexposed it struggles to detect the edges - this can be turned off and manually framed by operator and works better uncut
3½. In this old software, the above process is called "manual cut" which may be where some confusion is coming from
4. Unlikely but they may be using the rare MFC which affords more control as expense of time and for cut strips only
5. My original point, have a look at your negs if it was misloaded (eg exposure is too close to one edge) not much can be done in that case

@brbo Like he said, if the equipment is poorly maintained all of the above is a bit hit and miss.

calebarchie said:

@calico

1. Yes HS-1800
2. In EZcontroller once the 120-AFC is setup selecting wide framing is just an option in a drop-down menu

Click to expand...

Interesting. So they don't have to install different film holder or anything. Maybe they had told me the extra charge to include rebate was for "extra setup," and I conflated that with what another lab told me about having to use different film holder which will show rebate. That other lab mentioned how expensive those film holders were (and hard to find?), which is reason most labs don't have them. Maybe they were referring to Frontier, not Noritsu.

Incidentally, doesn't seem like changing a selection in pull-down menu should cost $7 per roll.

calebarchie said:

3. The carrier is auto-framing, if the neg is underexposed it struggles to detect the edges - this can be turned off and manually framed by operator and works better uncut

Click to expand...

The images for the scans I received with the wide rebate on only two sides were quite dark, which seemed weird to me, because I shoot Portra 160 at 100, and I use a hand held meter and am usually good at getting exposures right. It's possible I somehow screwed up and underexposed, but I wondered if the darkness was due to how it was scanned.

They still have the film, so I can't check it yet.

calebarchie said:

3½. In this old software, the above process is called "manual cut" which may be where some confusion is coming from

Click to expand...

Re cutting the physical film or not: I always ask for uncut, because my scanner can only fit three 6x6 frames, and sometimes labs will cut in fours, despite instructions.

So their first try at scanning was with uncut, and then they requested permission to cut for second try.

They did not say anything about "manual cut" (software) -- so no confusion there.

calebarchie said:

4. Unlikely but they may be using the rare MFC which affords more control as expense of time and for cut strips only

Click to expand...

Interesting.

calebarchie said:

5. My original point, have a look at your negs if it was misloaded (eg exposure is too close to one edge) not much can be done in that case

Click to expand...

I will definitely check the negs when I get them back -- see if rebate wider on left or right of image. I don't think there is a lot of give on the Hassy film insert which would allow the film to be too much to one side. But maybe enough to cause a problem in the Noritsu?

I am out of town right now, and don't have a Hassy with me. But I will check when I get home.

In the scans, one of the thick rebate edges is always on the left, so maybe position of film in film insert is an issue. The other wide rebate in the scans is usually on bottom of image.

calebarchie said:

@brbo Like he said, if the equipment is poorly maintained all of the above is a bit hit and miss.

Click to expand...

Ah, that makes sense. I would expect this lab to maintain their equipment well, but you never know.

Thanks for your help.

calico said:

I think most labs use sRGB because their printers use sRGB.

Click to expand...

Most labs use srgb because the rest of the world mostly use srgb, including virtually everything shown on the web. So if you give customers srgb, you can pretty safely assume that the customers will see the colors as they were intended by the lab, which avoids complaints. That's probably the main reason for relying on that particular encoding.

If you look at the color space of srgb vs. AdobeRGB you'll notice that the difference is mostly on the green part of the spectrum, with magenta, yellow, blue etc. all having more or less the same color space. You'd have to plot the gamut that the scanner can output based on the scan of a negative or slide on top of those two gamuts to see where you might lose data. Things get really complex here because you're basically trying to figure out how real-world hues encode into the gamut of a slide or color neg, which is then picked up by RGB sensor sites with a certain peak sensitivity and dynamic range into a digital file, and then in the case of CN film this is post-processed to produce a positive image, again with a certain gamut. The question of color profiles now also becomes one of what happens under the hood; if you receive an AdobeRGB vs. an srgb output from the lab, this only says something about the encoding of the final output file. But what were the different gamuts of the process steps the image information/data went through? My feeling is that throughout this chain, there are considerable losses of gamut, starting with a very dramatic loss during the capture stage, followed by further compression steps (the scanning step itself being also quite dramatic). Ultimately, it is questionable whether outputting to AdobeRGB consistitutes anything resembling a preservation of color data that would otherwise be smashed flat into sRGB, or whether this constriction has already happened to excess and outputting to AdobeRGB only re-codes the data into a theoretically lager color space without actually using the part of this space that's not covered in sRGB. I don't know for sure, but I expect that the (theoretical) loss is so slight that it'll never really show up in practice and if it does, the only way to see it is to do a side-by-side comparison of a very particular test setup. I.e. it's not a practical issue for the kind of thing you're trying to do.

Of course, if you then go to print, the gamut shrivels back into a fraction of either sRGB or AdobeRGB since reflective media generally don't even come close in offering the kind of chroma that a transmissive (e.g. color slide) or emissive (monitor, which really is a bit like a low-res slide) device allows. Now, there's a reasonable case to be made for marginal gains when working in a larger color space even when outputting to printed media - see e.g. the argument put forth in this video. Although much of that seems to be down not so much to gamut restrictions in digital space, but the question of how color data are translated between different stages of a digital imaging chain.

This one is really a question of theoretical vs. real-world gains. I have doubts about the latter.
But hey, nothing wrong with working in the widest gamut possible, of course. As long as intepretation and translation keep into account the color spaces used.

koraks said:

This one is really a question of theoretical vs. real-world gains. I have doubts about the latter.
But hey, nothing wrong with working in the widest gamut possible, of course. As long as intepretation and translation keep into account the color spaces used.

Click to expand...

Thanks for all of your thoughts re sRGB and Adobe RGB 1998.

Yeah....."real world." Have to factor that in.

But I figure I may as well work in the widest gamut possible. Why not.

In the past when I compared scan from lab and my scan of same image (at same resolution) on my Eizo, my scan looked so much better. I don't know if it's my 16 bit vs their 8 bit, my tiff vs their jpeg, or my RGB 1998 vs their sRGB, or quality of scanner, or how I balance the colors, or what. Probably combination.

I'm out of town right now, but when I get back I might do a test just for sRGB vs RGB 1998. I had always just assumed RGB 1998 was better. But then, as you say, there is the "real world," separate from my fancy Eizo monitor : )

If you want to work in a large gamut and basically be done with it all, just use ProPhotoRGB. I get the argument of basically trying to remove this factor from the equation; I don't see anything wrong with it either. I just wanted to point out that for me, personally, a lab returning in sRGB would not be a dealbreaker - but a lab returning only 8 bit would be a problem for me.

Then again, my own preference is to scan as a positive and then proceed manually from there using curve adjustments. This means that as long as the color space the positive scan comes in doesn't lop off anything that the scanner puts out, it's OK by me.

koraks said:

If you want to work in a large gamut and basically be done with it all, just use ProPhotoRGB. I get the argument of basically trying to remove this factor from the equation; I don't see anything wrong with it either. I just wanted to point out that for me, personally, a lab returning in sRGB would not be a dealbreaker - but a lab returning only 8 bit would be a problem for me.

Then again, my own preference is to scan as a positive and then proceed manually from there using curve adjustments. This means that as long as the color space the positive scan comes in doesn't lop off anything that the scanner puts out, it's OK by me.

Click to expand...

Thanks for your further thoughts : )