What software do you use after scanning negatives with a camera to turn them into positives?

[PhilBurton]

Maybe it got lost in the comments, but Negmaster and Negative Lab Pro both deserve a closer look for inverting color. I have provided both the endorsement and full-sized sample scans you can evaluate.

Old Gregg,

What do you see as the tradeoffs between these two products?

 

[wiltw]

Maybe it got lost in the comments, but Negmaster and Negative Lab Pro both deserve a closer look for inverting color. I have provided both the endorsement and full-sized sample scans you can evaluate.

OK went back and read every one of your posts. I saw a comment that programs like Negmaster need a lot of tweaking (not good) and 'One final word: avoid Negadoctor Darktable plugin. '

I don't interpret either to be an endorsement. Am I missing something, point me to a post number because I'm sure not seeing one.

 

[Nelari]

I
Have you (or anyone else) used VueScan to open a camera-copy RAW file of a color negative, and then converted that to a positive in Vuescan?

I have. Some negatives are inverted pretty well, but not all are. The RAW files are from a Pentax camera (type .PEF).

I
Would anyone care to speculate what would cause VueScan to render a RAF like this?

Perhaps you could ask Vuescan, the company?

 

[wiltw]

Fair enough. My point was that all color inversion tools require tweaking, including Silverfast, supposedly the top choice among film scanner owners. Which is a way is a good thing, because a negative is just a starting point for creative interpretation. If your preference is for a reasonable one-click inversion, you can probably ignore my input because I have never seen that work to my liking, neither with scanners nor cameras.

OK, now understand where you are coming from.

The scanner image taken with MP Navigator EX that I posted (image 3 in Post 54) did not involve much effort to tweak at all, and it came out looking similar to the print made from the same negative. So it is possible to get highly acceptable results without great time and effort tweaking a converted image.

I am trying to find something comparable in result to the MP Navigator scanned image, when using a dSLR captuure of the negative, but thus far have found no suggestions with endorsement about ease of process.

Getting close, as seen in the photo 3 result upon initial conversion, much better than photo 2 of Post 54, is my goal...otherwise I might as well simply take the time to use the scanner, scanning at 4800 dpi (resulting in 30MPixel image) takes 8 minutes to do a one-pass scan, or 40 sec. per image. Shooting with dSLR and then converting and tweaking would be hard pressed to equal the less-than-a-minute result achieved per image with a scanner.

But I have the out of owning a scanner, unlike many who find themselves with old negatives and no scanner, and not a big enough budget to send all of them out to a commercial digitizing service.

 

[grat]

I'd like to comment about Darktable + negadoctor-- it is a somewhat peculiar interface. It started as a Lightroom clone, but then went it's own direction. Negadoctor provides the ability to fine-tune your results pretty well-- pick your film base color, your highlight and shadow colors, and well, there's your inversion. Tweak as needed. Then, because of the way Darktable works, you can copy that recipe and apply it to the rest of the images from that roll.

It may not be intuitive, but it's pretty simple once you understand the process. There are a couple good videos on using it out there.

My problem with DSLR digitizing, and I'm not too proud to admit it, is that the scanned negatives suffer from too much outside light and wrong color balance. I'm still fine-tuning my process, but as I've said elsewhere the Epson produces good results with less effort, and I suffer from a high moment of inertia.

 

[wiltw]

@wiltw I must say your sample definitely looks pretty good for a 100% hands-off inversion. The closest equivalent in camera-scanning world IMO would be NLP, which can operate in fully-automatic mode, here's a sample that I spend zero time on: invert+apply. Kodak ProImage 100.

That looks pretty darn good! Maybe in addition to giving RawTherapee a try, NLP gets consideration, too....
nope, just re-read the comments about NLP in post 21!

 

[wiltw]

My problem with DSLR digitizing, and I'm not too proud to admit it, is that the scanned negatives suffer from too much outside light and wrong color balance. I'm still fine-tuning my process, but as I've said elsewhere the Epson produces good results with less effort, and I suffer from a high moment of inertia.

That pretty well summarizes my feeling, too...if it ain't broke, change only if it saves you time or level of effort to get 'the same' end result. And your comment about color balance mirrors my own concern about 'dSLR converion' of negatives.

 

[Huss]

Still not as good as it should be-- which is kind of my point. DSLR digitizing can indeed produce superior results-- but a flatbed will produce very usable results for much less effort.

Super low effort for me with digicam scanning.
For 35mm film - 60mm AF macro lens on my Nikon. Screw on the ES2 film copier on the end of that. Point camera at light source (a $20 light pad). AF focuses on the grain instantly, take snap of image. Repeat until all 36 are done. (Takes maybe 5 minutes to capture 36 8200x5300 images).
Then download onto computer - 2 mins. Open in LR and following NLP instructions, place the white balance dropper on an unexposed bit of the film to get the WB for that particular film stock. That takes a few seconds and if you save it as a WB profile, you only need to do that once per film stock. Convert in NLP. Done.

How long does it take for you to make 1 scan at 8200 pixels long using a flatbed?

For other formats I use the above, but replace the ES-2 film holder with a copy stand (not inverted tripod - the copy stand is much easier/quicker and more accurate to use - $150 on ebay new), and whatever film holder you like. I use Lomo Digitiliza ones.

35mm film example (Nikon F6, 85 1.8G, Kodak ProImage 100) scanned with digicam @8200x5300



100% crop from above:

 

[Huss]

Xpan scan using my digicam:



More 35mm:



Rolleiflex scan:




Lomo LC-A 120 scan:



35mm:



All converted in seconds with negativelabpro.

 

[wiltw]

Huss,

7 minutes total for 36 exposures...pretty good turnaround, less than 15 sec per image.
But your shot of Golden Gate Bridge reveals color issues..GG bridge too pink..photo .too cyan overall. not enough orange in the bridge color

 

[MattKing]

Back to OP, I have tried built in capability within Paintshop Pro to convert negative to positive. The result came out inferior

That "negative to positive" facility in Paintshop Pro is not designed to convert an orange masked film negative to a positive. It is designed an unmasked digital image.
You can use Paintshop Pro in a manner similar to Photoshop to remove the effects of the mask - essentially creating a layer based on a sampling of the unexposed rebate and then merging the image and mask layer using a division process. I've only tried it a couple of times but found the process unweildy.

 

[wiltw]

That "negative to positive" facility in Paintshop Pro is not designed to convert an orange masked film negative to a positive. It is designed an unmasked digital image.
You can use Paintshop Pro in a manner similar to Photoshop to remove the effects of the mask - essentially creating a layer based on a sampling of the unexposed rebate and then merging the image and mask layer using a division process. I've only tried it a couple of times but found the process unweildy.

Thx for explaining the rationale for the Paintshop Pro conversion...so you effectively undo the mask first, then apply the negative-to-positive conversion?! Interesting to hear your assessment as 'unweidly'.
If reversal of negatives can be done pretty well and easily by scanners, I wonder why no application software has been deleloped to undo the mask and convert dSLR-imaged negatives in a process as easy as the scanners seem to do it?!

 

[wiltw]

I do not understand where this method of removing orange mask came from, I suspect it originated from the older days of hybrid workflow. It is much easier (and mathematically equivalent) to simply set the wide balance on a RAW file using the "pipette" on a film rebate. There's one caveat though: some RAW converter + light source combinations will not have the adjustment range to cancel a heavier mask of some films (Fuji 400H Pro). For this, I have installed a CC40C filter into my light source.

Hmmm...I have a reason to pull out my color printing filter set which I used well before getting a dichoic head enlarger!

 

[grat]

If reversal of negatives can be done pretty well and easily by scanners, I wonder why no application software has been deleloped to undo the mask and convert dSLR-imaged negatives in a process as easy as the scanners seem to do it?!

There are like 4 or 5 packages that have been discussed that do this process-- with varying levels of difficulty. Silverfast has done the fine-tuning to produce curves for each film-- so as long as Silverfast has a profile for your film, you're good to go (although in my experience, it gets the colors "close"-- the colors in my converted cat image aren't quite true to life, but neither are the converted images I've produced from the DSLR).

If Silverfast doesn't have a profile for your film stock, you're outta luck.

 

[grat]

I do not understand where this method of removing orange mask came from, I suspect it originated from the older days of hybrid workflow. It is much easier (and mathematically equivalent) to simply set the wide balance on a RAW file using the "pipette" on a film rebate. There's one caveat though: some RAW converter + light source combinations will not have the adjustment range to cancel a heavier mask of some films (Fuji 400H Pro). For this, I have installed a CC40C filter into my light source.

The orange mask is designed to interact with the three color dyes (and tungsten light) in such a way as to produce a reasonable facsimile of the originally captured colors. When printing, this is fine.

When imaging digitally, it's not enough to deal with the mask-- you have to compensate for the interaction between the mask and the dyes as well, and that's where the complexity comes in.

 

[wiltw]

You are confusing me again.

As I stated earlier, what someone posted in post 21 made me think NLP is not the solution for me:

"That is exactly how Negative Lab Pro complicates my workflow in Lightroom.
I digitize film with my digital camera, and import the camera's RAW files into Lightroom. If I don't duplicate the NLP-converted file as a TIFF/JPEG, then normal Lightroom tools work backwards, so further editing with Lightroom tools is difficult or impossible. I use Lightroom's local adjustment tools a lot - adjustments which are impossible using the NLP module. So, using NLP really forces me to duplicate the RAW files as a TIFF/JPEG."
​

I want to bring in files to LR in order to be able to make further adjustments when necessary.

 

[wiltw]

@wiltw Yes, but that's because they also wanted to make additional custom tweaks on top of NLP, but also (for some strange reason) want to keep the original file. It's like you scanning into a TIFF, keeping it, and making and keeping another TIFF with additional edits. That's a really strange corner case.

Will just have to try out RawTherapee, and then try out the trial copy of NLP to decide for myself if a reasonable solution exists that rivals scanning on my Canon 8800F with NP Navigator software for time, effort, and quality of result. Equal is not good enough, if you have to pay money for the NLP. and you already own the scanner-based solution.

 

[grat]

Sorry, this is incorrect. This interaction (when the mask cancels out dye imperfections) happens when light travels through film during digitization, exactly the same way it happens in an enlarger. After that, it's just a trivial WB correction, a digital equivalent of color filters in an enlarger. Even when inverting manually, it's not rocket science: set the WB, invert, adjust R/G/B layers to a common white point, clip on both ends, then adjust gamma on R+B channels a bit, and here's the end result.

So you maintain there's no difference in sensitivity between the curves for Red, Green and Blue with RA-4 photosensitive paper and a CMOS sensor?

That's news. I guess Silverfast wasted all their effort creating custom curves for each channel for each film type.

You're describing the process I follow for manual inversion when I'm trying to do it within Affinity-- sometimes it works well, sometimes it works badly. Sometimes, everything comes out green.

 

[Huss]

Huss,

7 minutes total for 36 exposures...pretty good turnaround, less than 15 sec per image.
But your shot of Golden Gate Bridge reveals color issues..GG bridge too pink..photo .too cyan overall. not enough orange in the bridge color

The GG bridge photo is the one taken with the Lomo.
The Lomo's lens introduces crazy colour saturations. It is not the camera to use for 'accurate' representations. It is the camera to use to get lovely crazy Lomo represetnations.
That is why I also included one taken by my Rolleiflex.

 

[Huss]

I feel awkward. Is this not polite to point out that colors/contrast/grain/saturation on these are way off? I don't even know what went wrong here, could be botched development maybe?

Don't feel awkward. I create the images the way I want them to look.


The one area where I don't add my vibe to is grain. Do you have an issue with the grain in the 100% crop? If so, show me what the grain should look like from a 100% crop of an 8500x5300 scan of a 35mm Kodak ProImage capture.

 

[wiltw]

So you maintain there's no difference in sensitivity between the curves for Red, Green and Blue with RA-4 photosensitive paper and a CMOS sensor?
That's news. I guess Silverfast wasted all their effort creating custom curves for each channel for each film type.

It used to matter if I printed on Fuji vs. on Kodak paper, same film two looks. So a single profile to miimic a single film ignores the fact that what is recoreded by film is NOT necessarily whoat someone sees on final print! Dyes and sinsitivity proviles exist in the film AND also in the paper it is printed on!

 

[runswithsizzers]

As I stated earlier, what someone posted in post 21 made me think NLP is not the solution for me:

"That is exactly how Negative Lab Pro complicates my workflow in Lightroom.
I digitize film with my digital camera, and import the camera's RAW files into Lightroom. If I don't duplicate the NLP-converted file as a TIFF/JPEG, then normal Lightroom tools work backwards, so further editing with Lightroom tools is difficult or impossible. I use Lightroom's local adjustment tools a lot - adjustments which are impossible using the NLP module. So, using NLP really forces me to duplicate the RAW files as a TIFF/JPEG."​

I want to bring in files to LR in order to be able to make further adjustments when necessary.

Post 21 is mine. Let me try to clarify:
If you don't mind Exporting the NLP edits to your camera RAW file as a TIFF - and then re-importing the TIFF back into Lightroom - then most of my objections go away. (The NLP software makes the process of exporting and re-importing simpler than it sounds.) Whether you keep the camera RAW file, or trash it, is up to you. If you keep it, then there is a cost in storage. If you trash the RAW file, then you won't have the option re-convert it again with NLP using some other combination of settings.

@wiltw Yes, but that's because they also wanted to make additional custom tweaks on top of NLP, but also (for some strange reason) want to keep the original file. It's like you scanning into a TIFF, keeping it, and making and keeping another TIFF with additional edits. That's a really strange corner case.

As mentioned above, the main reason to keep the RAW file is to preserve the option to use NLP to convert it again later. There are many different NLP conversion settings and each combination produces very different results. After you get better at choosing the settings that produce results you like best, you might want to go back to the RAW file for a do-over. Throwing away the RAW file is more like throwing away the negative after you've made a print.

 

[wiltw]

Post 21 is mine. Let me try to clarify:
If you don't mind Exporting the NLP edits to your camera RAW file as a TIFF - and then re-importing the TIFF back into Lightroom - then most of my objections go away. (The NLP software makes the process of exporting and re-importing simpler than it sounds.) Whether you keep the camera RAW file, or trash it, is up to you. If you keep it, then there is a cost in storage. If you trash the RAW file, then you won't have the option re-convert it again with NLP using some other combination of settings.
As mentioned above, the main reason to keep the RAW file is to preserve the option to use NLP to convert it again later. There are many different NLP conversion settings and each combination produces very different results. After you get better at choosing the settings that produce results you like best, you might want to go back to the RAW file for a do-over. Throwing away the RAW file is more like throwing away the negative after you've made a print.

Thx for more detail. That raises questions about a scenario like this.....

1. If I start with a film negative, when A) printing it I can make Print 1, then B) alther the WB and the exposure and make Print B
   Print A might be brighter/darker than Print B. Print B might be better color balanced than Print A.
2. If I take that same negative and digitize it via my dSLR, if I use NLP to interpret my RAW file,
   can I make one color balance and exposure and then
   easily make a different color balance and lighter/darker exposure than the first?

 

[runswithsizzers]

Just throw away the inverted TIFF and keep the original RAW file. I get a bunch of TIFFs in Lightroom all the time, because this dumb thing can't really do much anyway, it's just an NLP launcher. Every time I edit a file in Affinity Photo I end up with yet another temporary TIFF.

What inverted TIFF?

Maybe my understanding of how Negative Lab Pro and Lightroom work is all wrong?

As I understand it, all edits to RAW files in Lightroom are kept as a set of instructions that are not actually applied untill/unless the file is Exported, right? So yes, if you "Edit with..." Affinity Photo, or Photoshop, or any other bit editor, that involves generating a TIFF. I assume the NLP conversion is like any other edit in Lightroom - that is, a set of instructions and not an actual duplicate file - unless I tell NLP that I want it to create the TIFF, right?

But the vast majority of the editing I do can be done entirely in Lightroom, so I almost never* have a TIFF in Lightroom that duplicates a RAW file. And that's the way I like it. A RAF file from 16 MP Fuji camera is about 33MB. But if I tell NLP to make a TIFF copy, that TIFF copy is 90MB. So if I keep the RAW, what was 33MB is now 123MB.

* Exceptions: If making camera copies of b&w or slides I usually start in Photoshop and get the file pretty close to how I want it before taking it into Lightroom as a TIFF for cataloging, captions, keywords, etc. Because all the heavy lifting was already done in Photoshop, I don't feel the need to keep the RAW file for those in Lightroom, because only minor future edits are likely. I will do the same with camera-copies of my color negatives if I can find some way to do the color inversion that I can live with and that can be done outside of Lightroom. Still trying out some of the non-Lightroom options mentioned here.

 

[Huss]

@Huss Phew, thanks for not holding it against me, I always enjoy your posts. Speaking of grain, I suspect the difference is in our approach to RAW processing. Here's a full-sized scan of ProImage. The grain looks softer for some reason, and also I downsample my scans to 5000px wide, that could also explain some.

Hey Gregg, sorry if I came over as defensive, not the intention! My image does look sharper and grainier; it most probably has to do with the processing. Using NLP I allow for the default sharpening setting, which is Leave as Set. Other options (and honestly I have never looked at this before!) are Off, Lab and Scanner.

Now I'm curious, and will process one image using the four different settings and will compare.