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

Adrian Bacon

Subscriber
Joined
Oct 18, 2016
Messages
2,086
Location
Petaluma, CA.
Format
Multi Format
So, since this thread has been started, I've actually gotten a fair amount of contact privately asking to come up with a way for people to be able to use my tooling without having to send me their film. I'm technically not averse to this, however, this represents a *really significant* level of effort/learning/know-how on my part to get it to where it is today, and I've been paying for that effort through processing/scanning film for other people.

I'm not that wild about going with a paid software model as all it takes is one guy to put a download link up somewhere and I can kiss ever getting compensated for my time and effort goodbye. I'm not saying anybody in this group would do that, but I've been in the technical software game for a really long time and have seen human nature at work. It's all too easy to lose your shorts.

Another option I've been considering is a paid support model, where I simply make a bare implementation version of the source code available under GPL, and if you have the technical wherewithal to get it working for your environment, you can make the changes to the code to support your particular hardware/setup and profile your films to get results that work for you. If you don't have those skills, then that's where I would come in, but you would need to compensate me for providing technical assistance on an as needed basis. How much it would cost you would depend on how much assistance you wanted to book me for. I could provide a known good set of hardware specifications that are solidly supported by my code "out of the box" that provides really good output, and if you chose to use different hardware, you'd either need to get that working yourself, or again, engage me for paid assistance. I could see that potentially working were it'd be beneficial to the community at large, but allow me to recoup some of my investment.

If this is something that people would be interested in, or if somebody has other ideas for how to potentially make this work, I'm all ears. I know that there's a need that could be filled, and I am in a position that I could fill it, and I've been hearing the requests, but at the same time, I don't want to just give all that time and effort away.
 

dkonigs

Subscriber
Joined
Sep 17, 2009
Messages
372
Location
Mountain View, CA
Format
Multi Format
There has to be a way around this problem, or Negative Lab Pro, ColorPerfect, and the entire commercial software industry wouldn't be able to exist.

This option would work great to get other labs to use your software, and to make money in the process. It would also be good for people like myself. However, I don't think it would help most of the other people on these forums. At least not without someone following up with an effort to build a full product around the code, only for the product to effectively become what you feared from a regular paid model (albeit on firmer legal footing).

I personally love open source software. I also make a point to open-source anything I write, especially if its something I don't intend to directly profit from. However, I've never had anyone actually explain to me a clear way to actually profit from making end-user applications open-source. Every single open source business model I'm aware of tends to involve forms of "support" and "dual licensing" that are great for business customers, but effectively useless for end-user applications.
 

Adrian Bacon

Subscriber
Joined
Oct 18, 2016
Messages
2,086
Location
Petaluma, CA.
Format
Multi Format

The fact that people on this thread are actively probing what NLP does under the covers and how it works in an effort to reproduce what it does outside of NLP says a lot. I don't support or condone that behavior as it has the potential to directly impact the income possibilities for the author. If he's putting the time and effort in, he should be compensated. I predict it will only be a matter of time before somebody reverse engineers it enough to get over 75% of the way there and posts it online somewhere.

The vast majority of the money in the commercial software industry is actually businesses paying other businesses for rights to use the software, and usually includes paid support. The parts that serve end consumers usually reside inside gated communities or on hardware/software platforms that employ DRM so that you can't just copy it from one system to another. If you're going to sell hundreds of thousands, or potentially millions of copies, it's worth it to make that investment into that infrastructure.

In the case of Negative Lab Pro, it's only useful if you already have a paid copy of Adobe's software as it only works in that framework, so the risk is relatively low. In my case, my code is a compiled stand alone application that can be copied and works just fine pretty much anywhere it's copied to as long it's been copied to an operating system that is compatible. Could I invest in putting controls in to prevent copying? Sure, but given the potential market size, is it worth that effort? Or is there a simpler way to do it? Back in the bad old days, software piracy was rampant. With the advent of the internet, companies have pretty much made it a requirement to "call home" so that they can apply DRM if need be. Again, this is something that I could implement, but it boils down to if it's worth that effort for the potential market size.

The other thing I have to take into consideration is my code makes really heavy use of other free/open source code bases, so I have to be somewhat careful about charging and/or bundling/distributing code together for sale. I've not looked at the innards of NLP very closely, or looked at any licensing related to it, but that's something the author should be cognizant of if he's making use of open source code and/or bundling/distributing it, as all it takes is one DMCA take down notice if he's not in compliance to put him out of business at least temporarily.
 

Smaug01

Member
Joined
Jan 27, 2020
Messages
104
Location
Chicago suburbs
Format
Hybrid
Adrian, I'm relatively new to this, but I landed here searching for a solution to scanning color negatives and getting them inverted with colors that are close to what I want. Just CLOSE is close enough. I can make small tweaks to get it perfect. But when it's way off? I just cannot get it there manually.

Right now I'm scanning with my Olympus E-M10 III, then inverting with an app I can't remember the name of right now. For B&W, after the inversion, adjustments to brightness and contrast are pretty easy. For color, it's not that easy.

The software is free, but he asks for donations if you like it. Honor system. I used it for a month, then sent him $20. I'll probably send him another $20 in a couple more months. Of course, lots of people won't bother with that, so I'm not sure you could make your living off of it.

Irfan Skiljan did the same thing with IrfanView. (great basic JPG photo editor)

To me, there are three types of film photographers, these days:

1) Pure analog. Develop & darkroom print themselves. The old guard.

2) Those who love the look and tactile experience of shooting film. They are OK with a hybrid approach, but are not OK with all the fiddling. They want good scans and good prints. These are the guys who pay you to develop and scan their film.

3) The rest of us who are a bit handy. We can develop our own film sometimes. (esp. B&W, trying C-41 or E-6) Maybe we figure out how to scan and invert with a camera + macro lens. We are not the ones who want to pay for scanning and inverting, except for maybe a desperate situation or super important event. Some of this group are also the ones who would post the software online or share the keys, etc. Remember, some cultures just don't have a sense of honor or integrity.

FWIW - I think you should write the software. Make it stable and reasonably user-friendly, but make clear that it doesn't do 100% of the color correcting. Just 90% by inverting and removing the orange tint. Minor corrections would still need to be made by the user. Charge $20-50 and implement the one-time key thing, even if it costs more. It would protect your investment, so it'd be worth it. I don't think you'd lose your customers [ Group 2), above] as they don't want to fiddle with it. They just want it back and looking good and are willing to pay a bit for this convenience. Group 1 will stay where they are until they die.

But Group 3? We're on the lookout and some day soon, with this film renaissance, someone will supply this niche. I'd hate to see you miss the opportunity because you didn't want to take that last step.

To add a bit, here's a shot of mine that was lab-scanned by my local lab. I color-corrected it a bit myself, as it was a bit green:
https://www.flickr.com/photos/182935075@N04/49430338848/in/dateposted-public/
Still the lab scan was close enough; closer than I could get myself.

I have several images that I inverted and color corrected manually after scanning with my camera, but they don't look as good, no matter how much I fiddled. Those were negatives that were lab-developed in the mid-90s. Old family photos. I'll put one up when I think of it.

After I sent him a donation, I emailed the author of my software, asking him to implement an orange mask correction into a future version. He said he'd think about it. (I wish I could remember the name of it right now...)

If it's not clear yet, I'm with the OP. I don't want any more fingers in my paycheck. I want to pay once for a decent app to do one job and be done with it.

Separately, I attached a pic of my scanning rig. It consists of my camera with the Nikon ES-2 digitizer set ($125) a step-up ring to match my macro lens to the digitizer and a cheap ebay light box / tracing light.
 

Attachments

  • Scanning rig.jpg
    179.2 KB · Views: 156

NortheastPhotographic

Advertiser
Joined
Mar 3, 2011
Messages
1,552
Location
Maine!
Format
Medium Format

8x10 Portra 160VC - 300mm 5.6 Caltar - Scanned with an S1R in pixel shift mode and converted with NLP plus minor tweaks.


I really can't believe that this thread is still going. NLP is a fabulous piece of software that gets better with every generation. Maybe someone should start a thread about THEIR prefered way of converting color negative film instead of a thread that is basically anti-NLP for some reason.

I for one have been happily using NLP for months now. It's one of the best things to happen to my scanning workflow, in combination with Negative Supply tools and the Panasonic S1R.
 
Last edited:

MattKing

Moderator
Moderator
Joined
Apr 24, 2005
Messages
54,323
Location
Delta, BC Canada
Format
Medium Format
[
Smaug01 said:
FWIW - I think you should write the software. Make it stable and reasonably user-friendly, but make clear that it doesn't do 100% of the color correcting. Just 90% by inverting and removing the orange tint.
Click to expand...
I just have to comment on this part.
The mask isn't an overall orange tint. If it was, then dealing with it would be trivial.
The mask varies with the colour and intensity of the underlying image. It helps adjust for the non-linear peculiarities of the film dyes that form the image. It helps match the film dyes to the sensitivity of colour photographic paper.
Every approach to the issue - whether Negative Lab Pro's or Adrian Bacon's or Epson's or Nikon's or whomever's - needs to deal with that complexity.
 

Adrian Bacon

Subscriber
Joined
Oct 18, 2016
Messages
2,086
Location
Petaluma, CA.
Format
Multi Format

the fact that this thread is still going is because people are looking for a solution to a need that they have that isn’t quite filled yet.

I can’t speak for other people, but I wouldn’t say I’m anti-nlp. It’s one way of getting there, and it clearly does a good enough job that many people find it useful, which is great, it just isn’t how I went about getting there myself, and there are others where it seems to be less of an ideal fit.
 

Adrian Bacon

Subscriber
Joined
Oct 18, 2016
Messages
2,086
Location
Petaluma, CA.
Format
Multi Format

+1

It’s not that simple. Once you understand what you need to do, it’s relatively straightforward to deal with it, but does take time and effort. Having the raw samples in a non-color managed environment helps a lot, and being able to perform certain operations on the raw samples helps a lot.

the biggest issue that people struggle with is photoshop is not particularly well equipped to do the operations you need to do, and they don’t really understand what is involved in getting there.

I don’t really advertise this, but every time somebody sends in film that I’ve not scanned before, I have to make a scanning profile for it. If it’s something that I can get my hands on, I’ll generally acquire at least a couple rolls of it and make an official profile with my tooling. If it’s not something I can get my hands on, I’ll start with my generic C-41 profile, copy it, and start tweaking it to get to reasonably neutral. Depending on the extent and nature of the changes, this can take anywheres from 20-30 minutes to upwards of a couple of hours. Per emulsion.

I currently have 60+ emulsions in my library, and it’s been growing at a rate of at least a couple per month. This is not an insignificant amount of effort, and this is just for my particular setup. If you go and change the hardware, or the light used, or any number of other things, that affects the output enough that you have to spend some time to true it back in. Going from the 80D to the 90D has caused a fair amount of rework on the profiles I’ve scanned with it so far for me. this is where it’s a real sticking point for me, because as soon as I make it generally available, people are going to want to use their own hardware and want it to work acceptably. NLP (I suspect) deals with this because it’s operating inside of a color managed environment where it’s basically being presented with a relatively uniform input as generated by Adobe’s Camera Raw processor, so the author gets the benefit of Adobes might in dealing with differences in hardware and such. I don’t have that benefit because I’m dealing directly with raw samples.
 

Adrian Bacon

Subscriber
Joined
Oct 18, 2016
Messages
2,086
Location
Petaluma, CA.
Format
Multi Format

I agree with points 1, 2, and 3. I think I'm doing a reasonable job of serving group 2, and I'm not going to bother with group 1 unless they want to have a look-see. Group 3 technically isn't a market as there isn't really any money if nobody wants to pay. I think there's probably a group 4, where they want to shoot film, but keep costs low and would rather pay for software to make that job easier instead of shelling out money for every roll. I find group 3 and 4 somewhat interesting because the cost difference between having me process only and process with a straight scan is literally $5 per roll. How much time and money are you going to spend to get your own rig up and running, and dinking around with it to get reasonable results? Sure, $5 a roll adds up over time, but the reality is a lot of people don't really shoot that much film unless they're a professional primarily shooting film, and then they have a service agreement so they get priority over other work that comes in. The vast majority of other people that send film in to me on a regular basis are generally hard pressed to be sending 2-3 rolls a month, many of them are 2-3 rolls a quarter.

I get that some of group 3 may just be doing it for the joy of it, and enjoy tinkering around with that sort of thing, but that's not really a market as a lot of those would probably just as soon not spend anything.

I totally get that you don't want yet another thing dipping into your paycheck. The reality of the matter is, everything costs something. People will spend money on what they value, so the question is how much do you value that? Is that enough for somebody like me, or the author of NLP to make a living from? The notion of you pay once and never have to pay again is a false truth. The author of NLP is probably having a banner year last year and this year in terms of sales, but since he's not charging for upgrades or support services, at some point in the not too distant future everybody who's going to buy it will have bought it and the money will either stop or slow waaay down. You can't realistically expect him to keep releasing new versions with new features or new versions to support newer operating systems or newer versions of LR for very little new income. He'll either have to come up with something different he can charge for, or release an all new version of the same thing with features that are held back from the "old" version to entice a paid upgrade. If he doesn't, he'll have to do something else for income, and that means that NLP will eventually go by the wayside as it won't be getting much of his attention. This is how it works in a capitalistic society. Companies figured out a really long time ago that if you make toasters that last forever and never break, and are easy and cheap to service and repair, then you'll go out of business because once everybody has a toaster, you won't sell enough to keep the doors open. The only way to stay in business is to make something that gives a reasonable service life, then dip back into that paycheck with a new toaster at an engineered point in time. That is reality.
 

Smaug01

Member
Joined
Jan 27, 2020
Messages
104
Location
Chicago suburbs
Format
Hybrid
The software I couldn't remember earlier is called FastStone Image Viewer. It's great. Intuitive. Just mouse over each button and it says what it does.
 

Marameo

Is it possible with SIlverFast SE 8 to scan raw adn apply Negafix later on?
 

removedacct2

Member
Joined
May 26, 2018
Messages
366
MattKing said:
just have to comment on this part.
The mask isn't an overall orange tint. If it was, then dealing with it would be trivial.
The mask varies with the colour and intensity of the underlying image.
Click to expand...

it varies also because there could be some minor irregular light leaks, in case of 6x6 backs, and in case of chemicals getting on the verge of exhaustion. I have many rolls of 120 where the tint on the border of some frames is not uniform but a gradation... yet it could be that the picture itself, composition, exposure, focus are great, and then it's worth to try to salvage it in the best possible way. Doing the pipette off whitebalancing on such border can be dilemma and auto WB of different softwares work differently. That's where one-click automation doesn't work. I have recently tested some such problematic negatives in NLP and compared with my usual tools until now (imagemagick, dcraw front-ends, gimp), with EpsonScan, with Vuescan, with Silverfast, with ColorPerfect, and the latest NLP v2.1 does still the best job. Sometimes default settings give a bit funky colours but it's fixed in a snap by changing pre-saturation and color model. Former version v2.0 wasn't that good, so when people comment about merits and flaws of NLP it could be good to know what version.
 

GLS

Member
Joined
Apr 29, 2018
Messages
1,729
Location
England
Format
Multi Format
antonio_b said:
Doing the pipette off whitebalancing on such border can be dilemma and auto WB of different softwares work differently
Click to expand...

That's why you don't sample from the border next to the roll edge; you sample the space inbetween frames.
 

BradS

Member
Joined
Sep 28, 2004
Messages
8,126
Location
Soulsbyville, California
Format
35mm


Challenge accepted! I've downloaded the DNG spec from the Adobe website and will start reading today. I've read the TIFF spec...a few time times but will probably need to refresh on that too as it has been a few years.

Incidentally, thank you for sharing so much of your IP. This seems extraordinarily generous on your part. It seems like your posts in this thread have pretty much laid out exactly what your tools are doing. Over the years, I've had any number of projects in industry that were less clearly specified. I've been looking for a new color lab....next time I finish a roll, I think I'll be sending it to you. Thanks again.


I don't know. It seems like some folk...
A) do not want to compensate software developers for their time, effort, products, etc...
or
B) just have to "do it yerself"...even if it does cost more and produce poor results.
Both seem to lack an understanding of basic economic principals.
 

PinkPony

Member
Joined
Aug 11, 2014
Messages
17
Format
35mm

Yes, I don't know how many others would but I definitely would be interested. And with all the talk about NLP, had it not been based on Lightroom I would have bought it already but I'm no longer on the Adobe subscription train. I think there are plenty of people who scan using some kind of copy stand + light source + DSLR. Personally I only do black and white this way at the moment and I know that I invert the wrong way using custom curves in capture one. It's possible to get okay results but it's not the correct way to invert a negative. I'm also a programmer by trade and gave some effort into making my own tool. I even found a well written document explaining much of the process that I don't think I was supposed to find. However in the end I didn't really have the math skills to pull it off.

As long as there is at least some instruction on how to profile ones setup I think a bare implementation would be very valuable.

Another option, although I think the market may be too small for this, is to raise money (via kickstarter or similar) to let you have some financial gain by producing a basic open source version. There may be a problem with expectations though of what one would get out of it (resulting in much complaining ala the Ferrania story). Personally I think I would be fine but that is just because I can cobble code together, not everyone interested in such a tool can do that...
 

removedacct2

Member
Joined
May 26, 2018
Messages
366

you have missed the point or didn't read OP: it's not against NLP itself, but about the fact that NLP runs on top of Lightroom. It's not a standalone software but a plugin which requires LR. Which means, if one is to follow the legal speech and framework of Adobe, that an Adobe license is required. So you want to use NLP but in fact you will not buy just a license for it, but also a license for Lightroom. Get it?
 

BradS

Member
Joined
Sep 28, 2004
Messages
8,126
Location
Soulsbyville, California
Format
35mm


Yup. I get it. Some people do not want top pay for software.
 

PhilBurton

Member
Joined
Oct 20, 2018
Messages
467
Location
Western USA
Format
35mm
For those of us who already use Lightroom, it is a positive, not a negative, that NLP is designed work with Lightroom, not outside of it.

However, I can understand that people who would not otherwise want to use Lightroom object to the fact that NLP is integrated into Lightroom. And it would be especially annoying to have to pay Adobe US $10/month for the subscription. However, it is possible to buy a copy of an older version of Lightroom on eBay. https://www.ebay.com/sch/i.html?_from=R40&_nkw=adobe+lightroom&_sacat=18793&rt=nc&_oaa=1&_dcat=41859 Any Lightroom version up to V6 was a one-time purchase.
 

mohmad khatab

Member
Joined
Sep 8, 2012
Messages
1,228
Location
Egypt
Format
35mm
I strongly support the latter idea.
Kickstarter is the perfect solution.
But the problem lies with the programmer himself.
The programmer (mostly) will not prefer to get the price of his effort in one go, but rather he wants this application to become like a chicken that lays a golden egg for him every morning.,
- If a project was created on the Kickstarter for a quarter of a million pounds, I am sure that the world has people interested in supporting such a project.
But the programmer himself does not want to, and he wants to get 10 million within five years of this application.
The programmer is in itself a big problem, as he overestimates his efforts and their value.
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…