An alternative to Negative Lab Pro and Lr has to exist (C-41 reversal and orange mask removal)?!

JWMster · Oct 9, 2019

Just kidding. Didn't mean to offend, but clearly I have. My apologies and good wishes to you and your ventures.

FWIW, I wonder whether there is a relationship beyond the name of this software https://www.supershareware.com/info/simple-image-tools.html and the software you're writing / using? The link looks to be something from long ago (2013) and far away and would appear to have a different screen name for the author. But if I've followed the description there, the author is at least involved in a similar endeavor... though probably not at your level or with your focus in proprietary usage. For my part, I am at least as sleery of these things for the viruses they may have accumulated as long forgotten shareware files on unknown servers. You seem rightly to be in terms of seeking to carve out your niche and no more, and stick there. That's fine. Why not? Then again for my part I do use shareware like Quadtone Print in addition to ImagePrint - though the latter is commercial software and remains a tad pricey enough.

I'm sure your software is wonderful. Doesn't help me though. I develop and scan my own stuff. But as much as I may be late to the party, to my ear, the OP's inquiry wasn't answered per se - except in the negative... unless he writes his own.

Helge · Oct 10, 2019

JWMster said:
Funny comment a while back: Why is it Vuescan and Silverfast never tied into DSLR scanned images? Clearly a lack of demand?

Likely more that the original developers left the project, and now they are just coasting along on fumes.
There has been precious few improvements in the last few years.
A major overhaul would be too expensive and beyond the capabilities of the current people probably.

Adrian Bacon · Oct 10, 2019

JWMster said:
Just kidding. Didn't mean to offend, but clearly I have. My apologies and good wishes to you and your ventures.

FWIW, I wonder whether there is a relationship beyond the name of this software https://www.supershareware.com/info/simple-image-tools.html and the software you're writing / using? The link looks to be something from long ago (2013) and far away and would appear to have a different screen name for the author. But if I've followed the description there, the author is at least involved in a similar endeavor... though probably not at your level or with your focus in proprietary usage. For my part, I am at least as sleery of these things for the viruses they may have accumulated as long forgotten shareware files on unknown servers. You seem rightly to be in terms of seeking to carve out your niche and no more, and stick there. That's fine. Why not? Then again for my part I do use shareware like Quadtone Print in addition to ImagePrint - though the latter is commercial software and remains a tad pricey enough.

I'm sure your software is wonderful. Doesn't help me though. I develop and scan my own stuff. But as much as I may be late to the party, to my ear, the OP's inquiry wasn't answered per se - except in the negative... unless he writes his own.

No offense taken, I just don’t want there to be any confusion.

None of my stuff is related to that shareware. I didn’t even know that it existed. Before I started using the name simple image tools a few years back, I did a cursory google check to see if it was being used for anything reasonably popular, and that tool didn’t pop up. This was back in 2016-early 2017 time frame. My software is a lot more like a raw file processor that has been optimized to deal with color negative scans than a general purpose image manipulation utility, though in recent versions I’ve added the ability to process raw files generated by Vuescan and turn them into very high resolution floating point CFA DNG files. Scanning 8x10 color negatives at 2400 dpi RGB and getting 460MP DNG files that look like they where shot with a super high resolution digital camera is pretty awesome.

Helge said:
Likely more that the original developers left the project, and now they are just coasting along on fumes.
There has been precious few improvements in the last few years.
A major overhaul would be too expensive and beyond the capabilities of the current people probably.

That’s likely. People don’t realize this, but it’s actually incredibly expensive to support lots of hardware. In my case, that also means making emulsion profiles for all that hardware since I can’t really leave it to the end user to figure out on their own.

e2b · Oct 23, 2019

I came across this thread by accident and have found it both fascinating and informative, including the video (https://ethw.org/Wesley_T._Hanson). So, thanks to all contributors.

Getting right back to original question about an alternative to Negative Lab Pro (NLP) and Lightroom, I'd like to put in a word of support for Colorperfect (www.colorperfect.com) as a (non-free) alternative, which can also use PhotoLine in place of Photoshop. I was happy to read in posts #34 and #64 that I'm not the only one who thinks it does a "pretty good" job.

As an example, about ten years ago I scanned a couple of hundred rolls of 35mm color negative (dating from the 1990s) using a Nikon 5000ED with Vuescan. The Vuescan inversions were terrible, but I archived the linear RAW scans and about 18 months ago I decided to re-process these scans using Colorperfect, which I've used from time to time over the past ten years. I ran Colorperfect in batch mode within Photoshop, setting the parameters for each roll on an "average" image. This worked well and I'd say less than 0.5% of images needed individualized inversion. I aimed at a neutral sort of inversion and completed final individual adjustments on selected images in Lightroom. A couple of hundred rolls is not a high-volume production situation, but it's not negligible and Colorperfect made the task manageable under the circumstances.

When I subsequently became aware of NLP I tried it on a few of the same archived scans. While I was happy enough with the results, I didn't think they weren't sufficiently better to make it worthwhile re-doing what I had already done with Colorperfect. I remain undecided as to whether it would be worth making the change to NLP if had many more rolls of color negatives to scan (which I don't). My impression is that it's easier than Colorperfect in dealing with a small number of images (and it doesn't have such a steep learning curve), but I didn't get a strong feeling that it would be much quicker or easier in dealing with the numbers I'd been working with.

However, considering what I've spent on cameras, lenses, film, scanners, computers, monitors, Vuescan Pro, Colorperfect and ongoing subscriptions to Photoshop and Lightroom and so on (not to mention printing), and having some idea of the work involved in creating a product for a rather niche market, the additional cost of buying NLP seems fairly modest.

Les Sarile · Oct 29, 2019

e2b said:
As an example, about ten years ago I scanned a couple of hundred rolls of 35mm color negative (dating from the 1990s) using a Nikon 5000ED with Vuescan. The Vuescan inversions were terrible, . . .

I bought the Coolscan 5000 when it was first released and tried a few scans using Nikonscan and Vuescan. I thought that perhaps the built-in film profiles and modifiers in Vuescan would be an advantage over Nikonscan. Below, I used the built-in profile for Kodak 160VC in Vuescan as well as using the various modifiers compared to one neutral scan using Nikonscan.

Todate I have almost 40K scans using Nikonscan on various films and still maintain a Windows Vista machine to run it.

e2b · Oct 30, 2019

Interesting. Your experiences with inverting negatives in Vuescan seem quite similar to mine. Actually, the original reason I changed from Nikonscan to Vuescan wasn't do to with negatives, but rather because Vuescan performs much better IR cleaning of Kodachrome transparencies (and also gives a consistent, if idiosyncratic, interface for all my scanners). Then having achieved success with ColorPerfect both for inverting negatives and color balancing slides, I never changed back to Nikonscan, and I never bothered to see if Vuescan inversions had been improved.

Manufacturers' lack of support for old scanners, as in I "still maintain a Windows Vista machine to run" Nikonscan, certainly left a market gap that Vuescan fills.

Les Sarile · Oct 30, 2019

e2b said:
Actually, the original reason I changed from Nikonscan to Vuescan wasn't do to with negatives, but rather because Vuescan performs much better IR cleaning of Kodachrome transparencies (and also gives a consistent, if idiosyncratic, interface for all my scanners).

Nikonscan vs Vuescan ICE

Full size -> http://www.fototime.com/B52423F2F398DA3/orig.jpg

Coolscan 9000 + Nikonscan ICE is great

Full size -> http://www.fototime.com/36BED059AD8E686/orig.jpg

On Kodachrome too

Full size -> http://www.fototime.com/E3B7FD459E9EA9F/orig.jpg

e2b · Nov 4, 2019

Well, that looks pretty good! Is there usually such a big difference between the Nikon 5000 and 9000 scans (without ICE)?

Les Sarile · Nov 4, 2019

e2b said:
Well, that looks pretty good! Is there usually such a big difference between the Nikon 5000 and 9000 scans (without ICE)?

What big difference do you mean specifically?

e2b · Nov 4, 2019

On the set of Kodak 160VC scans under the heading "Coolscan 9000 + Nikonscan ICE is great": comparing left-hand-top, and and left-hand-bottom scans, scratches less visible with Coolscan 900 No ICE.

Les Sarile · Nov 5, 2019

e2b said:
On the set of Kodak 160VC scans under the heading "Coolscan 9000 + Nikonscan ICE is great": comparing left-hand-top, and and left-hand-bottom scans, scratches less visible with Coolscan 900 No ICE.

Compared to the Coolscan 9000, the 5000 can exagerate the appearance of dust, scratches and grain. When scanning fine films (Ektars, Velvia and such), the difference in results is not apparent.

Alain Deloc · Dec 26, 2019

Hi,

It seems people is looking for a NLP alternative. I wrote a Photoshop app for myself to invert digital camera scans. where I should post about that?
Thanks,

Helge · Dec 26, 2019

Here would seem the ideal place...

jtk · Jan 10, 2020

Alain Deloc said:
Hi,

It seems people is looking for a NLP alternative. I wrote a Photoshop app for myself to invert digital camera scans. where I should post about that?
Thanks,

Photoshop natively inverts B&W negs...and NIK provides infinite renditions as well as dodge/burn...if you specify it will give specific film renditions...back to PS you can then tone however you want. Ive never seen much need for color neg film tho I have some old c41 negs so will download free NLP to evaluate.

Film Rescue · Jan 15, 2020

I've spent probably 100 hours trying to make this work well over the last couple of decades. While i could on occasion get an okay inversion on some images nothing was reliable or consistent and I would classify nothing I did as comparable to our Creo, Fuji sp-3000 or even our Epson v-750. That was until Negative lab pro. It's the only software that comes close and in my opinion it's not worth the time or effort to use any other method...and no, I have no association with the people who make this software. Still though, DSLRs aren't really an optimal tool for other than with B&W negatives. They are fast though.

runswithsizzers · Jan 16, 2020

ludwighagelstein said:
[...] That is coming from someone who gladly paid 80€ for Colorperfect and 100€ for NLP. These programs are worth their money. [...]

ludwighagelstein, since you have both, I would be curious to hear your opinion about the relative merits and deficiencies of each.

To the forum, at large:
When considering what software to use to convert negatives to positives, there are two goals - the acceptability of the results - and what changes to my workflow are needed to get there.

I own and use the Negative Lab Pro software, and it's OK in the results department. So far, I have processed only a few rolls of negative film which I copied with my digital camera. Color inversion results are not bad, but I'm finding the NLP controls are not sensitive enough when making fine adjustments to colors in the mid-tones, highlights and shadows. That is, if the slider starts at "0" (zero), and I increase it to "1" - that is too much correction. I need some way to make smaller adjustments. I have contacted the developer about this (Nate Johnson), and I have high hopes that future versions will correct this.

NLP is relatively new, and each update has added significant features and improvements. I was pretty excited about the metadata editor, which allows the user to add film camera data to your digitzed image file. This is a benefit if you want to seach your digitized image catalog to find images taken with a certain film camera or lens or shutter speed. However, I was less excited after actually using the metadata editor. It is rather tedious to imput the data (I am a terrible at the keyboard) - and then one must take a few extra steps to create a TIFF/JPEG version that includes the film camera data before exporting to a service like Flickr or SmugMug. I have a Lightroom plugin that can convert my edited RAW files to JPEG and export to SmugMug in one step, without needing to keep a JPEG version in Lightroom.

Which brings up the workflow issue. As someone who uses my Lightroom library to organize, catalog, keyword, and edit all my images, I see two very different possibilites for dealing with "scanned" negatives:
1. Import the raw dSLR image files into Lightroom, and make the inversions and corrections from within LR, or
2. Invert the RAW dSLR images with Photoshop, make some/most corrections in Photoshop, then import a TIFF version into my Lightroom catalog for keywords, captions and minor edits.

Obviously, NLP is designed for workflow #1, and ColorPerfect would probably be better for workflow #2 (I've not used ColorPerfect, so I don't really know anthing about it).

What I am trying to avoid is a workflow which results in several different versions of the same image in my Lightroom catalog. I definitely do not want to try to keep track of a RAW:NLP version and a TIFF or JPEG version in Lightroom. Unfortunately, NLP almost forces the user to do just that. As long as I can get to an acceptable image from the NLP interface, no problem. But as soon as I want to make a minor tweek to the NLP-converted file using the vastly more versatile and competent Lightroom controls, I enter a pushmi-pullyu, upside-down and opposite bizarro world. The only work-around is to create a TIFF version (or JPEG), and edit that, normally, with the Lightroom controls.

So workflow number 2 is starting to look more attractive to me. But before I make the effort to learn how to use ColorPerfect, can anyone tell me if there is anything significant - pro or con - I need to know about it first? Any fatal flaw or magic bullet?

GLS · Jan 16, 2020

runswithsizzers said:
So workflow number 2 is starting to look more attractive to me. But before I make the effort to learn how to use ColorPerfect, can anyone tell me if there is anything significant - pro or con - I need to know about it first? Any fatal flaw or magic bullet?

I have used ColorPerfect for quite some time now. No fatal flaws to speak of, although the interface and workflow is frankly a bit poor and unintuitive, but once learned it is not hard to use.

I only use ColorPerfect to do the inversion and minimal adjustments; namely to make sure I am not clipping either highlights or shadows (the default preview settings always clip a bit in order to give a punchier result), and possibly a slight colour tweak if I detect any remaining trace cast (honestly though this is rare, and trivial to remove if present). If this process is done properly you will get a 16-bit TIFF which probably looks a little dull and lacking in contrast, but contains all the available data. I tend to then perform final tweaks in the Camera Raw plugin before final sharpening and dust removal.

I have briefly used a trial of NLP, and was never happy with the results.

runswithsizzers · Jan 16, 2020

GLS said:
I have used ColorPerfect for quite some time now. No fatal flaws to speak of, although the interface and workflow is frankly a bit poor and unintuitive, but once learned it is not hard to use.

I only use ColorPerfect to do the inversion and minimal adjustments; namely to make sure I am not clipping either highlights or shadows (the default preview settings always clip a bit in order to give a punchier result), and possibly a slight colour tweak if I detect any remaining trace cast (honestly though this is rare, and trivial to remove if present). If this process is done properly you will get a 16-bit TIFF which probably looks a little dull and lacking in contrast, but contains all the available data. I tend to then perform final tweaks in the Camera Raw plugin before final sharpening and dust removal.

I have briefly used a trial of NLP, and was never happy with the results.

Thanks for your reply. I am in the middle of installing a trial version of ColorPerfect - which is not as simple as I might have hoped. About how big are your 16-bit TIFF files? Do you store your images as 16-bit TIFFs, or do you downsize them to 8-bit after all adjustments are done? Unless the files are huge, I would prefer to store the images as 16-bit TIFFs, but as I recall, some adjustments in Photoshop work poorly or not at all on 16-bit files (?)

GLS · Jan 16, 2020

runswithsizzers said:
About how big are your 16-bit TIFF files? Do you store your images as 16-bit TIFFs, or do you downsize them to 8-bit after all adjustments are done? Unless the files are huge, I would prefer to store the images as 16-bit TIFFs, but as I recall, some adjustments in Photoshop work poorly or not at all on 16-bit files (?)

I save the final edits as 16-bit PSD files. My digitised frames are typically about 22-25 megapixels in resolution, and PSD filesize varies primarily depending on the number of layers the PSD contains, but will typically be between 500-700 MB each. The source RAW files are about 40-45 MB each.

I'm not aware of any processes in Photoshop that don't work with 16-bit files. There is certainly no way I would ever archive final versions of my photos in 8-bit.

Oldwino · Jan 20, 2020

If there was a stand-alone version of NLP, I’d buy it in a heartbeat. I don’t use Photoshop or Lightroom.

Helge · Jan 20, 2020

Oldwino said:
If there was a stand-alone version of NLP, I’d buy it in a heartbeat. I don’t use Photoshop or Lightroom.

Absolutely! I wouldn't think twice about paying double up for a standalone version. Or a version that hooks onto a free image editor.

Helge · Jan 20, 2020

Film Rescue said:
I've spent probably 100 hours trying to make this work well over the last couple of decades. While i could on occasion get an okay inversion on some images nothing was reliable or consistent and I would classify nothing I did as comparable to our Creo, Fuji sp-3000 or even our Epson v-750. That was until Negative lab pro. It's the only software that comes close and in my opinion it's not worth the time or effort to use any other method...and no, I have no association with the people who make this software. Still though, DSLRs aren't really an optimal tool for other than with B&W negatives. They are fast though.

There should be nothing about DSLRs that would make them, less well suited for colour "scanning" than line CCD based scanners. Apart from the bayerfilter, which is realtively easily circumvented or compensated for.
Of course a monochrome sensor with variable backlight would be optimal for everything, but then we are on to something else entirely.
Did you have any particular problem in mind?

dkonigs · Jan 20, 2020

Helge said:
Absolutely! I wouldn't think twice about paying double up for a standalone version. Or a version that hooks onto a free image editor.

Me too. And I actually do use Lightroom. Its just that Lightroom is the wrong part of my workflow to want to do the scan-processing in. NLP is also a bit more annoying to use with a scanner that outputs TIFF (or TIFF-like) data, since it needs an extra processing step to get good results.

At the moment, I'm using ColorPerfect in Photoshop to do my initial processing and then just bring the output into my Lightroom catalog. (The original uninverted flat scans get stashed off to the side, and might even get deleted once I need the space back.)

Adrian Bacon · Jan 20, 2020

Film Rescue said:
DSLRs aren't really an optimal tool for other than with B&W negatives. They are fast though.

That depends on how much resolution the DSLR has and the quality of the lens being used, and the light being used. It's a bit more finicky to get right for sure, but once you're there, the results tend to be more than good enough for the vast majority of uses.

The thing to keep in mind is even though other scanning technologies exist, DSLRs are the only technology that is currently being actively developed and improved upon over time. Drum scanners? Going away. Lab scanners? When's the last time Fuji or Noritsu released new scanners? Going away. Flatbed scanners? Not really going away, but definitely plateaued. DSLRs/Mirrorless? New models with more dynamic range, more resolution, less noise, better optics, better processing software every year, and continuing that way for the foreseeable future. Do they have issues? Yes. So does every other scanning technique.

Adrian Bacon · Jan 21, 2020

runswithsizzers said:
ludwighagelstein, since you have both, I would be curious to hear your opinion about the relative merits and deficiencies of each.

To the forum, at large:
When considering what software to use to convert negatives to positives, there are two goals - the acceptability of the results - and what changes to my workflow are needed to get there.

I own and use the Negative Lab Pro software, and it's OK in the results department. So far, I have processed only a few rolls of negative film which I copied with my digital camera. Color inversion results are not bad, but I'm finding the NLP controls are not sensitive enough when making fine adjustments to colors in the mid-tones, highlights and shadows. That is, if the slider starts at "0" (zero), and I increase it to "1" - that is too much correction. I need some way to make smaller adjustments. I have contacted the developer about this (Nate Johnson), and I have high hopes that future versions will correct this.

NLP is relatively new, and each update has added significant features and improvements. I was pretty excited about the metadata editor, which allows the user to add film camera data to your digitzed image file. This is a benefit if you want to seach your digitized image catalog to find images taken with a certain film camera or lens or shutter speed. However, I was less excited after actually using the metadata editor. It is rather tedious to imput the data (I am a terrible at the keyboard) - and then one must take a few extra steps to create a TIFF/JPEG version that includes the film camera data before exporting to a service like Flickr or SmugMug. I have a Lightroom plugin that can convert my edited RAW files to JPEG and export to SmugMug in one step, without needing to keep a JPEG version in Lightroom.

Which brings up the workflow issue. As someone who uses my Lightroom library to organize, catalog, keyword, and edit all my images, I see two very different possibilites for dealing with "scanned" negatives:
1. Import the raw dSLR image files into Lightroom, and make the inversions and corrections from within LR, or
2. Invert the RAW dSLR images with Photoshop, make some/most corrections in Photoshop, then import a TIFF version into my Lightroom catalog for keywords, captions and minor edits.

Obviously, NLP is designed for workflow #1, and ColorPerfect would probably be better for workflow #2 (I've not used ColorPerfect, so I don't really know anthing about it).

What I am trying to avoid is a workflow which results in several different versions of the same image in my Lightroom catalog. I definitely do not want to try to keep track of a RAW:NLP version and a TIFF or JPEG version in Lightroom. Unfortunately, NLP almost forces the user to do just that. As long as I can get to an acceptable image from the NLP interface, no problem. But as soon as I want to make a minor tweek to the NLP-converted file using the vastly more versatile and competent Lightroom controls, I enter a pushmi-pullyu, upside-down and opposite bizarro world. The only work-around is to create a TIFF version (or JPEG), and edit that, normally, with the Lightroom controls.

So workflow number 2 is starting to look more attractive to me. But before I make the effort to learn how to use ColorPerfect, can anyone tell me if there is anything significant - pro or con - I need to know about it first? Any fatal flaw or magic bullet?

I do option number 3, which isn't really something you can do unless you write some code, but I take the raw scanned image and turn it into a floating point positive image that is in the native Adobe LR DNG format. From there, you can simply import it into LR and treat it just like any other raw file from a digital camera. Bit of a best of both worlds, if you will. No need to keep multiple versions around if you don't want to. All of your edits and everything goes into the DNG and stays with it.

An alternative to Negative Lab Pro and Lr has to exist (C-41 reversal and orange mask removal)?!

Privacy & Transparency

Privacy & Transparency