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

[MattKing]

Sure, though I’m in the midst of doing customer work, so it’ll be at least a couple days before I have any of my film scanned in. In the meantime, you can look at my media page, it has a number of images shot on film and run through my process.

I've added a new guide to my website for linearising digital camera RAW files, which can then be used with my CNMY set to invert negatives.

http://eigakai.ro/ps-action/cn-scan-inversion/converting-scans-made-with-digital-cameras

I just wanted to ask:
"Do you have to share the name "Adrian" to be good at this stuff?

 

[Lachlan Young]

Not mere "semantics." Honesty. And btw, where do you see oblong pixels? Certainly not in 20X30 prints.

You see it in cases where the X&Y resolutions differ in practice for various reasons. Overall, it adds up to poorer resolutions than claimed. Effectively you'd see two nominally identical resolution prints, but depending on the extent of the oblong pixels, the one with properly square pixels would look sharper overall.

 

[jtk]

Have you seen that in smallish (eg 13x19") inkjet prints that you've personally made from 35mm ? What scanner?

Have you read the exhaustive reviews that compare Nikon V to various drum scanners? Spoiler: Nikon comes close. I suspect the current Epson 850pro comes close with MF and LF.

 

[Derek L]

Adrian, thank you for your posts in this thread. I've been Googling for C-41 color correction advice for the past two hours and your posts are the only colorimetrically accurate information I've been able to find.

If you’re shooting black and white, once you white balance for your light source, you can treat each sensel directly as if it where monochrome (if you wrote your own code to do this) as you’re taking a picture of a frame that doesn’t contain any color information, so the only real color comes from your light source. Dial that out via white balance and you have effectively shot with no CFA. Doing that *very dramatically* increases the amount of visible fine detail in the photo. If you don’t do that, it’s debayered, then converted to BW, which eats a whole pile of potential resolution you’ve otherwise captured.

This is a fantastic idea. It's so great that I'm going to ignore extremely pressing work to write a script to do it.

The more correct thing to do is digitize it into a positive image of the negative in raw samples, apply gain/multipliers to each channel until the film base plus fog is the same exposure, which will render it as light grey to white, then adjust the gamma of each channel so that it's response is linear relative to the exposure that made the density, then invert that. From there, if you did that correctly, a series of grey cards shot over a range of exposures from -5 EV to +7 EV in full stop increments should render grey at each point with little to no color shifts going on.

Should one really aim for true neutrals, though?

If I understand the rest of your post correctly, (especially the use of the ColorChecker and the sentence, "The look I supply to customers very closely matches what you'd get from a digital camera that was ran through Adobe Camera Raw"), then you've given a recipe for getting (more or less) colorimetrically correct output out from scanned C-41. But if I wanted truly accurate colors, I suppose I'd just shoot digital and make a custom ICC profile with a ColorChecker. It seems to me that a lot of the "film look" is purposely inaccurate but pleasing color changes engineered by the good folks up in Rochester. Personally, I find this desirable and it's the reason I shoot color film instead of digital.

So, trying to think through how to achieve this, I get stuck. Obviously the film base + fog should be neutralized in the same way, but then what? I don't know what I should use as a reference for, say, the Portra look. A good RA4 print? Did these have true neutrals throughout the range?

Does your gloss on Photo Engineer's post on RA4 printing ("I’m interpreting 'speed offset in the layers' as a global multiplier if done digitally, as the net result is the orange cast is effectively dialed out as a result of the multiplication that is effectively happening as a result of the layer speed differences") mean that the way to get what one would get with RA4 is to just do the global multiplier corrections for the film base + fog and skip the individual gamma adjustments for each channel? [Edited to add: this is clearly wrong because the color skews are going to be terrible in the highlights, right? Now I'm just confused.]

Maybe a better question is, what output medium did Kodak scientists use to assess the color fidelity of their emulsions? And did they aim for true neutrals with no color cast?

 

[Adrian Bacon]

Should one really aim for true neutrals, though?

In an ideal world, yes. In reality, there’s always *a little* variation. Once you get close enough and have actual pictures instead of grey cards to look at, it tends to not matter so much because it’s hard to see unless you have large areas of continuous tones or gradations.

But if I wanted truly accurate colors, I suppose I'd just shoot digital and make a custom ICC profile with a ColorChecker. It seems to me that a lot of the "film look" is purposely inaccurate but pleasing color changes engineered by the good folks up in Rochester. Personally, I find this desirable and it's the reason I shoot color film instead of digital.

So, trying to think through how to achieve this, I get stuck. Obviously the film base + fog should be neutralized in the same way, but then what? I don't know what I should use as a reference for, say, the Portra look. A good RA4 print? Did these have true neutrals throughout the range?

Regardless of the emulsion, hues should render relatively accurately. Reds should be red, not orangish red, or pushing to magenta, skin tones should render as nice healthy skin tones not pushing to red or yellow, greens should be green, blues, blue, etc. the Macbeth chart is helpful for that. You don’t have to get it exactly on the money, but it does need to be reasonably close.

Saturation is a different animal, and is primarily where film differs from digital, for example, Ektar punches the daylights out of Portra in the saturation department and makes digital look like a skinny twig.

The way I deal with it is relatively simple. I’ve generated a totally zeroed out profile for almost every color negative I can buy here in the U.S. From that, I generated a generic profile that is an average of all the zeroed out film profiles that I have to date (for hue and saturation) and I use that. Each emulsion will render relative to the generic in its own unique way. Some will have more saturation than the average, some less. Some hues will render a little differently than others, etc. The average baseline is still a reasonably accurate rendering of a Macbeth chart, but each emulsion will play through in its own unique way. I have to imagine a good RA-4 paper is really an average of at least a handful of print films, otherwise you can only really print one emulsion with each paper type.

 

[Derek L]

Regardless of the emulsion, hues should render relatively accurately. Reds should be red, not orangish red, or pushing to magenta, skin tones should render as nice healthy skin tones not pushing to red or yellow, greens should be green, blues, blue, etc. the Macbeth chart is helpful for that. You don’t have to get it exactly on the money, but it does need to be reasonably close.

Is this really the case in practice, though? For instance, I've seen a lot of Ektar images where (white) people have red-pink skin, and I don't think they were improperly exposed or developed. It sure looks like this was an intentional tradeoff to improve how pleasing the color is for other subject matter. Or am I mistaking a change in saturation for a hue shift?

Saturation is a different animal, and is primarily where film differs from digital, for example, Ektar punches the daylights out of Portra in the saturation department and makes digital look like a skinny twig.

So is your claim that, in a Hue–Chrominance–Luminance model, the difference between Ektar and Portra comes down mostly to differences in how they render chrominance and not hue?

 

[Adrian Bacon]

Is this really the case in practice, though? For instance, I've seen a lot of Ektar images where (white) people have red-pink skin, and I don't think they were improperly exposed or developed. It sure looks like this was an intentional tradeoff to improve how pleasing the color is for other subject matter. Or am I mistaking a change in saturation for a hue shift?



So is your claim that, in a Hue–Chrominance–Luminance model, the difference between Ektar and Portra comes down mostly to differences in how they render chrominance and not hue?

The pink-red skin is partly incorrect post processing, and partly saturation. I’ll look through my collection of images to see if I can find some Ektar pictures that I can share when I get some time. In the meantime, the Macbeth chart has two skin tone patches. In the HCL model, yes, most of the differences are in saturation. They don’t render hue exactly the same as each other, but they both do render the hues pretty close to accurate.

 

[Adrian Bacon]

Is this really the case in practice, though? For instance, I've seen a lot of Ektar images where (white) people have red-pink skin, and I don't think they were improperly exposed or developed. It sure looks like this was an intentional tradeoff to improve how pleasing the color is for other subject matter. Or am I mistaking a change in saturation for a hue shift?



So is your claim that, in a Hue–Chrominance–Luminance model, the difference between Ektar and Portra comes down mostly to differences in how they render chrominance and not hue?

Macbeth charts of each Kodak emulsion except Portra 800 and the "new" ProImage 100

Ektar 100

Portra 400

Ultramax 400

Portra 160

Gold 200

Colorplus 200

 

[Derek L]

Macbeth charts of each Kodak emulsion except Portra 800 and the "new"

Oh, wow, this is great data. I'd love to buy you a beer and pick your brain about this stuff. But since you're on the other side of the country, I guess this will have to do.

First, are those charts corrected using your program? Maybe I'm again mistaking hue for saturation, or have undiagnosed abnormal color vision, but I have trouble characterizing the hues as all "pretty close to accurate." For instance, I see the background of the Ektar shot as grey, while the others all have some blue tint, and similarly the column of neutral patches on the left side of the chart in the Portra shot seems noticeably bluer than in the Ektar one. From your description I thought all the neutrals would be pretty much bang-on equal in the RGB channels, but I see variation between them. So it seems the residual variation after applying your "averaged" C-41 curve actually larger than I thought?

Second, why do you think the hue twists are necessary after correcting each channel? If the three primaries are additive and operate on separate layers, shouldn't it suffice to correct each channel individually? I think there are details of the development process that might involve interactions between the layers, but I'm not too sure.

Third, let me open a new can of worms and ask about going the other direction, trying to mimic film with digital. Suppose I take a digital capture with some Nikon DSLR and I decide that I want a "Portra look," for instance, where I mean capturing the general impression of the film and not a complete colorimetrically accurate replication. Assuming I'm happy it with how my raw converter rendered the hues, your analysis implies it would suffice to figure out how to selectively saturate and desaturate the various hues to create the look, right? Is the saturation change mostly a function of the hue, or does it vary in some complicated and nonlinear way depending on hue and lightness?

 

[Adrian Bacon]

Oh, wow, this is great data. I'd love to buy you a beer and pick your brain about this stuff. But since you're on the other side of the country, I guess this will have to do.

First, are those charts corrected using your program? Maybe I'm again mistaking hue for saturation, or have undiagnosed abnormal color vision, but I have trouble characterizing the hues as all "pretty close to accurate." For instance, I see the background of the Ektar shot as grey, while the others all have some blue tint, and similarly the column of neutral patches on the left side of the chart in the Portra shot seems noticeably bluer than in the Ektar one. From your description I thought all the neutrals would be pretty much bang-on equal in the RGB channels, but I see variation between them. So it seems the residual variation after applying your "averaged" C-41 curve actually larger than I thought?

Second, why do you think the hue twists are necessary after correcting each channel? If the three primaries are additive and operate on separate layers, shouldn't it suffice to correct each channel individually? I think there are details of the development process that might involve interactions between the layers, but I'm not too sure.

Third, let me open a new can of worms and ask about going the other direction, trying to mimic film with digital. Suppose I take a digital capture with some Nikon DSLR and I decide that I want a "Portra look," for instance, where I mean capturing the general impression of the film and not a complete colorimetrically accurate replication. Assuming I'm happy it with how my raw converter rendered the hues, your analysis implies it would suffice to figure out how to selectively saturate and desaturate the various hues to create the look, right? Is the saturation change mostly a function of the hue, or does it vary in some complicated and nonlinear way depending on hue and lightness?

Yes, these are generated with simple image tools and the generic C-41 profile, so you should expect to see variations. Our eyes and brains are wonderful at "autocorrecting" so when looking at each of these by themselves with no other reference, a lot of these variations for the most part go away. Looking at them side by side, it becomes obvious that there are variations between them.

The hue and saturation operations are required to conform the image colors to a color space. The color space I chose is ACES 2065-1, so I have to do a per-hue angle hue twist and saturation adjustment so that a given raw RGB color as captured by the emulsion and scanned by me matches up with what that color would be in RGB in the ACES 2065-1 color space. What I'm doing with the hue and saturation operations is actually not unique at all. How do you think Adobe Camera Raw conforms a raw image to a color space? If you look at the DNG spec, or use exiftool to inspect a raw file that has been converted to DNG, you'll discover that it has a per hue-angle hue twist lookup table and per hue angle saturation adjustment lookup table embedded in the file. In simplified terms, that is how it works.

Re Can of Worms: You mimic an emulsion by selectively adjusting the hue and saturation. Mastin Labs (and others I'm sure) does this already. You can buy a bunch of presets from them that will make the generic Adobe Camera Raw look like a given emulsion, or more correctly put, their interpretation of a given emulsion. What it's doing under the covers is selectively adjusting the hue and saturation from the generic ACR baseline look that Adobe supplies. Adobe Camera Raw and Lightroom already provide the tooling to do this with the HSL/Color and Split Toning panels. Those are there so you can adjust the look on top of the generic look that ACR/LR provides. Everybody who sells Lightroom Presets are basically just selling a bunch of presets that tweak the tooling that's already there. Anybody can go in there and create any look that they want. This is why Adobe makes ACR and LR render as many cameras as possible as close to each other as possible. That's the generic baseline look.

 

[Ted Baker]

Should one really aim for true neutrals, though?

Yes, the only reason you don't get this is because your sensor is different to the paper that the negative is designed to be printed on. There is a few way to do this properly.

If your using a DSLR you will need to fudge it some how. I believe Adrian has one such method, that does very well.

 

[Adrian Bacon]

Yes, the only reason you don't get this is because your sensor is different to the paper that the negative is designed to be printed on. There is a few way to do this properly.

If your using a DSLR you will need to fudge it some how. I believe Adrian has one such method, that does very well.

Yep, there's a couple of different ways to do it, mine is relatively simple, but reasonably accurate. There are more accurate ways, but I don't have the resources to go that route, and it's diminishing returns. The way I see it, is if I can be reasonably accurate and preserve as much color fidelity and spatial resolution as I can, I'll be quite a bit ahead of most everybody else who does this sort of thing.

 

[Adrian Bacon]

I just wanted to ask:
"Do you have to share the name "Adrian" to be good at this stuff?

I don't know... I don't consider myself to be particularly good in the sense that I probably couldn't design how any of this works. I know enough to make the existing stuff work reasonably well though, so compared to lots of other people, that's probably pretty good. It's all relative.

 

[Adrian Bacon]

Here's a sample of some more recent tooling I've been experimenting with to get more fine grained control over adjusting hue and saturation. The first image is something that I created in the ProPhoto colorspace in photoshop and printed out on a Canon Pro-1000 printer and Canon's PM-101 Photo Paper Pro Premium Matte paper, which the Pro-1000 has a built in paper profile for. To the naked eye, the print looks just like the image in terms of hue and saturation. I'm trusting that the Pro-1000 with this paper is going to render reasonably colormetrically accurate colors. This may or may not be the case, but I don't have any real way to verify one way or the other right now.

Ignore the inner squares for now. The outer squares are each color in the HSL model placed 22.5 degrees apart and with 50% saturation. I did a fair amount of shopping around and found an LED bulb that is 5000K and 90+ CRI and use that bulb to illuminate the paper print of this image. I then take a correctly exposed picture of the print (so far with a digital camera, film to come as soon as I get everything worked out) and use it to generate a hue twist look up table and saturation adjustment table with 22.5 degree waypoints and the in-between values interpolated between waypoints with basic linear interpolation.

The following image is a picture of the print illuminated with the light described above and taken with a Canon EOS M5 camera and the raw CR2 file ran through Simple Image Tools using the generated hue and saturation tables to create a DNG, then pulled into Adobe Lightroom and exported out as an sRGB JPG.

To measure how accurate my lookup tables are, from Lightroom, I export the image out as a ProPhoto TIFF file and pull it into Photoshop and measure the HSL values of each patch.

And here's the same image (slightly different crop) with no hue or saturation adjustments made, so it's reflecting what the camera sensor is seeing before its conformed to the ACES 2065-1 color space with the hue and saturation tables. By itself, it actually looks reasonably fine, it's not until you measure the colors in photoshop that you realize that some adjustment needs to be made. This is our brains "autocorrecting" for us. If I shot this on film, and corrected the mask out and linearized each channel, we'd be seeing how the emulsion renders each hue and with what saturation (not accounting for the white balance difference between the film and the light I used to illuminate this, I still need to work that part out). From there, I can generate a hue and saturation table for that emulsion, which should render each of the patches to within 1 hue degree and +- 1% saturation. This is what's coming down my image processing pipe in the not too distant future.

Now, a fair amount of people will say something like: "looking at patches is all fine and dandy, but doesn't tell you anything about how real pictures will look" or something along those lines. Basically, they're saying they don't trust this sort of thing, but that's primarily because they don't really understand how it works. I can simply provide an example real image using the above methodology. You can do the same thing with a raw scan or picture of color negative film once you've corrected out the orange mask, inverted, and linearized each channel.

The image below was shot on the same Canon EOS M5 and run through Simple Image Tools using the generated hue and saturation lookup tables from the hue patch squares above to make a DNG file, which was then pulled into Adobe Lightroom and exported out as an sRGB jpg file. It was shot late in the day, so the color temperature is a little warm. This is the stop sign on the street right outside my lab. During the summer, it's great because in one picture, I can get a really well known red, green, and blue. Everybody knows what a stop sign should look like, what the sky looks like, and what tree leaves generally look like because most of us see them in all kinds of lighting conditions all the time and they're actually fairly consistently the same color.

I don't think there's anything to complain about as far as hue or saturation are concerned, though, that's just me, and I'm biased.

 

[jtk]

Vuescan easily does what you want with Nikon, Canon, and Epson scanners. It's great with B&W and it uses infared (unbranded ice) with Win 10...assuming you're using a scanner that implements infared. It doesn't just smudge when dealing with scratches...I think that means you need the most expensive Epson as (I think) the less expensive version doesn't apply Ice for negs, just for reflective.

I hope I'm wrong about that.

 

[Photo Engineer]

Adrian, not too different from what we do, but ours is very very much more detailed with dozens of exposures. We get H&D curves from these.

PE

 

[Adrian Bacon]

Adrian, not too different from what we do, but ours is very very much more detailed with dozens of exposures. We get H&D curves from these.

PE

Yeah, the Macbeth chart is fine, but the patches are too far apart and aren't evenly spaced out (which leaves me with gaps where I could be a lot closer in hue if I had a waypoint there), and the saturation is too high, so it's hard to calculate how much saturation to pull out if a given hue angle is too saturated. Evenly spaced out every 22.5 degrees with 50% saturation is about right for my uses. Tighter than that and it starts to be a bit too fined grained, further apart and you get gaps of accuracy.

 

[Mr Bill]

I did a fair amount of shopping around and found an LED bulb that is 5000K and 90+ CRI and use that bulb to illuminate the paper print of this image. I then take a correctly exposed picture of the print (so far with a digital camera, film to come as soon as I get everything worked out) and use it to generate a hue twist look up table and saturation adjustment table

Hi, I think you're asking for trouble with that led. Since you shopped around so much, I presume you know what the spectral makeup of a white led looks like so I won't harp on that. I think you'd be a lot better off using electronic flash, and yeah, I know they dont generally come in at 5000K. (I presume your printer profile is based on D50, and this is why you selected that lamp?) If you think you HAVE to get closer to 5000K, you can do it with color temperature filters (look up mired shift). But I think you'd probably be fine with whatever the flash puts out with the camera white-balanced for it.

Yeah, the Macbeth chart is fine, but the patches are too far apart and aren't evenly spaced out (which leaves me with gaps where I could be a lot closer in hue if I had a waypoint there), and the saturation is too high, so it's hard to calculate how much saturation to pull out if a given hue angle is too saturated.

In another thread, I think, I referenced the Xrite ColorChecker SG. It's pretty pricey, north of $300, but it will help your problem. I've made hundreds of camera profiles using it or its predecessor, the Color Checker DC, which had a lot more test patches.

Here's an image from the B&H website:

 

[Adrian Bacon]

Hi, I think you're asking for trouble with that led. Since you shopped around so much, I presume you know what the spectral makeup of a white led looks like so I won't harp on that. I think you'd be a lot better off using electronic flash, and yeah, I know they dont generally come in at 5000K. (I presume your printer profile is based on D50, and this is why you selected that lamp?) If you think you HAVE to get closer to 5000K, you can do it with color temperature filters (look up mired shift). But I think you'd probably be fine with whatever the flash puts out with the camera white-balanced for it.




In another thread, I think, I referenced the Xrite ColorChecker SG. It's pretty pricey, north of $300, but it will help your problem. I've made hundreds of camera profiles using it or its predecessor, the Color Checker DC, which had a lot more test patches.

Here's an image from the B&H website:

Hello,

Yes, I selected 5000K because ICC profiles are based on D50, and it's the color temperature you're supposed to use for evaluating prints with. Incidentally, the LED isn't actually for the camera, it's for my eyes because I can't see it with my eyes if I'm using flash. I know there's a lot of trepidation about using LEDs, however, I've taken that same raw CR2 file and ran it through Adobe Camera Raw, then output it to a ProPhoto TIFF file and measured the patches in Photoshop and they all came out to within 1 hue degree, and +-1% saturation, so I very much doubt it's going to cause any meaningful issues for my uses. That exercise also to a degree validated that my printer is outputting chromatically correct colors (at least for my uses) as I've now done a full end to end with the industry standard known good and the colors are where they're supposed to be.

The next step will be to expose the same print using a flash and then measuring what the patches come out to with that. I can then just either go with that, or generate a new print with adjusted patches that come out to the values I want under flash.

I've looked at a fair number of color checker charts, however, due to the method in which my software does it's chromatic adaptation to get raw samples to conform to a known colorspace, my needs are best served with patches that are evenly spaced around the color ring with *a lot* less saturation than what you'd typically see out in real life. Every color checker chart I've looked at either has it's patches way too saturated, or doesn't have patches that are reasonably evenly spaced around the color ring. I have no doubt that these charts are immensely useful if you're going to make a camera profile that uses the known industry standard color frameworks, however, I've opted to not go that route because generating a profile for film has it's own set of challenges unique to it that those color frameworks don't really take into account.

 

[PhilBurton]

Adrian, not too different from what we do, but ours is very very much more detailed with dozens of exposures. We get H&D curves from these.

PE

Who is "We," and "Ours?"

 

[Alan Edward Klein]

Unless you're making images to match the color of a garment for a clothes ad, isn't color pretty much up to the beholder? When I scan Velvia 50 shots I've taken, I don;t try to match the original chrome slide but rather what looks pleasing to me. After all, unless you have both in front of you, who cares if there are differences? Negative film is even harder to match. So, why use the color charts? How do they help you?

 

[Adrian Bacon]

Unless you're making images to match the color of a garment for a clothes ad, isn't color pretty much up to the beholder? When I scan Velvia 50 shots I've taken, I don;t try to match the original chrome slide but rather what looks pleasing to me. After all, unless you have both in front of you, who cares if there are differences? Negative film is even harder to match. So, why use the color charts? How do they help you?

Up to the beholder? Yes and no. Why do you think C-41 has such a crap reputation for getting good colors back from labs? The customer wants to get back something at least reasonably accurate so that they can then apply their aesthetic to it, or have the lab apply an aesthetic on top of that base. Not getting the colors at least reasonably accurate in the first place is not an aesthetic.

With that being said, if you’re developing and scanning your own film at home, then knock yourself out and do whatever you want, but in my experience, if people are paying you to process and scan their film, they expect the colors to be pretty dialed in.

 

[Photo Engineer]

Who is "We," and "Ours?"

Where I used to work!

Eastman Kodak Company Research Labs or KRL to "US".

Sorry.

 

[JWMster]

So to return to the OP query, there is LR + NLP and then there is Adrian's Baconware... which may / may not use Simple Tools... or simply borrows the name. And then for us lesser mortals... we're kind of stuck.
Funny comment a while back: Why is it Vuescan and Silverfast never tied into DSLR scanned images? Clearly a lack of demand?

I've just received my MKI and it's companion hardware 'cause I'm getting tired of waiting on the Nikon LS8000 while it grinds away... and occasionally grinds to a halt. Repair costs aren't bad, just more frequent as the unit ages. And well, hope springs eternal... But I'm a CaptureOne user... so kind of seems like I'm SOL for this stuff.
I've not found the conversions from the DNG all that hard in CaptureOne, but they're do take some time... and so I may have to switch back to LR. Oh well.

 

[Adrian Bacon]

So to return to the OP query, there is LR + NLP and then there is Adrian's Baconware... which may / may not use Simple Tools... or simply borrows the name. And then for us lesser mortals... we're kind of stuck.
Funny comment a while back: Why is it Vuescan and Silverfast never tied into DSLR scanned images? Clearly a lack of demand?

I've just received my MKI and it's companion hardware 'cause I'm getting tired of waiting on the Nikon LS8000 while it grinds away... and occasionally grinds to a halt. Repair costs aren't bad, just more frequent as the unit ages. And well, hope springs eternal... But I'm a CaptureOne user... so kind of seems like I'm SOL for this stuff.
I've not found the conversions from the DNG all that hard in CaptureOne, but they're do take some time... and so I may have to switch back to LR. Oh well.

Adrian Baconware? Let’s be very clear here. I’m the owner and operator of Simple Film Lab, and the author of Simple Image Tools, which is the in-house scanning toolset of Simple Film Lab. There is no Adrian Baconware. It’s Simple Image Tools.