Great comparison ! Its so awesome to see what you can achive with a decent scanning setup and the newest version of nlp. I am currently rescanning film from two years ago and getting way better colours straight out conversion.
Thanks mate! I find that too :-) Although I still get a lot of questions about why they look so different from the lab sometimes, and I think learning the tools to allow manipulation gives that creative freedom.
Hey, I do a lot of scanning at home with nlp and Fuji xt2 and I love your talk in the beginning about the certain film look, it is what we make out of the negative with the setup and vision we have, keep up with the videos !🙌🏻
Thanks for a fantastic tutorial. It is really helpful to see the range of decisions you can choose to make while editing the image alongside the lab scan.
Handy knowing what those the lab glow sliders do but I like the look less blown out than what the lab provides. I’m convinced my lab just puts their frontier scanner on Warm as it has this yellow cast over everything. I think if you want lab results you can get it very close to identical with nlp.
At least the Fuji Frontier scanners are pretty close to the look of their photographic paper. So the way scans look from a well calibrated Frontier is not too far off, from a neg to paper print. Which is, as far as there is such a thing, the native look of film.
Nice tutorial! Any pointers for how to bring out colors on my mom's faded Kodachrome 64 slides? Tried to match my light with my white balance, but I got a feeling there's some magic that needs to happen in the calibration panel? Or maybe HSL or color grading is enough - blown out skies often go yellow/brown while shadows go blue. Maybe I should just apply the opposite. Even tried asking ChatGPT but I think it also thought I wanted to make new photos look like old, so I might have to reverse every recommendation I find
Awesome video as usual! Just so you know, I am scanning with a Nikon coolscan 8000 in negative, DNG without any ajustement at all and I am still having these colored dark road from time to time. I just can't get why they exactly happen. The method you're using is pretty much the best I have heard for controlling them. Thanks!
Great video, I have some problems to enter the reference view in Lightroom. If i start NLP the reference view goes back to one picture view and i am not able to edit the picture with a side by side view to my reference picture. Is there any special setting to do this?
I'm surprised you left the borders (at 9:45 timestamp) while converting. My understanding is that the black border confuses the NLP software and renders inaccurate colors. If you want the borders in the final product first convert without the borders and then add them back in. Am I missing something?
Hey yeah at 9:42 you can see I used the border buffer setting at 10% making cropping unnecessary. What you're referencing used to be applicable to older versions of the software, but this border buffer feature came in a while ago and resolved it. I speak about it more in the main digitisation tutorial that I referenced. But you can probably find a better explanation of the feature on the NLP site.
I've always been curious about the faded blacks and clipped highlights in my scans. My lab uses a Noritsu and they often come back with traits like that and I've always kinda just attributed it to them wanting to give us a "flat" scan to work with on a contrast level... Do you know if it's just a limitation of the scanning software to get those details, or just a matter of the lab techs trying to get through things quickly and not worrying about the details?
Hey Steven, I've experienced this too, but it doesn't seem to be consistent between machines. I think it's more to do with calibration, and maybe the setting of black/white points, or some other variable.
I also recommend inverting the RGB curves manually when NLP messes too much with the image. I have gotten pretty good results doing that in a few simple steps.
@@pushingfilm oh sure it’s time consuming, but I’ve been in the situation that NLP just doesn’t get it right or anything close to it and sometimes it’s worth doing it manually for those special frames
Your biggest problem is that you are working with only 8bit sRGB (web colours) JPG images so you don’t have a full 16bit Adobe RGB (full colour range) TIF image that you can optimise to perfection by pulling out shadow details and multiplying highlights without any banding or excessive noise. David Myers, Kodak Pro Photo CD scanning, Digital Masters Australasia.
Hey! I’ve recently gotten into color grading video, which is a totally different beast than photo. It’s gotten me thinking about how motion pictures are digitally processed. Somewhere along the way this week it sparked an idea for digital negative processing. I put together an alternative Lightroom process to NLP that has been a bit more consistent for me in terms of color, and offers me a stronger workflow in Lightroom panels. Probably just as quick as well + wouldn’t cost any money It’s only fault so far has been working with heavily under or overexposed images. An issue NLP would have as well. Want to do some more experimenting but figured I may publish a video of my own on the process, let me know if you may be interested in hearing it! I want to determine whether it’s truly useful to the public
Hi, I'm interested in knowing what process you're doing, as color grading video has made me think about how to do color in photo here as well. I'd be interested in knowing what you've come up with, thanks a lot.
In fact, Nate (the author of NLP) places great emphasis on leaving no border in the image to be converted in NLP. His CZcams video demonstrates the large effect leaving even a little bit of border causes in conversion colours.
when I hear that particular films do t have a look to them, I often wonder what then is the point of purchasing different film stocks. Why buy portra 400 if I get the same from Kodak Uktra max?
There are still inherent tendencies to different film stocks, just not one universal expected look to the same stock. These tendencies can be more than just the general colour tendencies, but also saturation, sharpness, tonality, grain, exposure latitude, and more.
Different film stocks definitely have a different look to them. It’s apparent when you don’ thave as many ways to manipulate the look, like when you RA-4 print them in the wet darkroom. In digital, you can manipulate to your heart’s content but you still get different results due to film speed, grain, colour shifts, latitude and dynamic range, etc, that are baked into the negative and not availalble to you in the scan.
digital camera scan of his negatives using his Canon 5DIV and a 100mm macro lens. So, he’s basically editing a high resolution macro digital photo of his negative.
@@pushingfilm I have the same setup but I use the negative supply scanning rig. I think I get better scans than using the Epson scanner and even better than lab scans most of the time.
@@vtavares00 Awesome! It's not hard anymore... I remember years ago tinkering with DLSR scanning, and it was still really hard to get consistently great results.
@@pushingfilm yes, negative lab pro and Lightroom makes it easy. Rivals lab scans for sure. And with bigger sensor cameras, better lenses, better software, you can just go back to your negatives and keep making your scans even better over time. Amazing time to be shooting film now.
Why would you digitise a piece of film with your camera without masking out the rest of the light-box surrounds? It’s like shooting into the sun with all the dark areas of the film being degraded!
Just curious, why would you try to match a NLP scan to a lab scan if you already have a lab scan of that frame? I suppose it teaches you a general direction for most scans, but when you only have the NLP scan, its difficult to know where it needs to go to look better.
The main reason I've done it is rescans because the lab scan didn't have enough resolution. In terms of matching though, yes, it's meant to just give an idea of the direction you take things in situations when you don't necessarily have a reference (with "better" being subjective, and there naturally being a lot of variation from lab scanner to lab scanner)
Never use generic scans from a lab because from day one they were sub-par vs. the results from my Epson scanner and then I became adept at maximizing the results from the scanner. It is so straight forward to get the quality required for even the most discerning clients that the only reason I can imagine folks needing all this additional work (requiring software such as NLP 2.3) is they haven't a clue regarding crafting their film originals , lighting and exposures , and film processing . Unfortunately those shortcomings on the front end necessitate the salvage operation on the back end and the results in my opinion are never optimal.
I must disagree a bit, for this video. Indeed you can create the look that you want, but If you take an image of the same subject with Portra, Ektar, Velvia and Kodak E100 there will/must be differences in color , saturation in all the film. If you need to have film & digital look alike this is a good way but … For me I don’t like to use plug-ins or scans done by a Lab technician. I have worked with drum scanner in the ‘90. It’s all about how you interpret an image unless you scan a negative as positive and do a manual conversion in photoshop, this way you keep the ‘film’ look. Take a look at the CZcams channel of Alex Burke how to do this, it’s time consuming but still. For me it’s the way to go.
Wow, amazed at the difference between the lab scan and the NLP, like the detail in the shadow you showed!
That shadow colour recovery series of steps was great! I got some old Kodak E100VS film with a strong cast. This should help me resolve it!
Great comparison ! Its so awesome to see what you can achive with a decent scanning setup and the newest version of nlp. I am currently rescanning film from two years ago and getting way better colours straight out conversion.
Thanks mate! I find that too :-) Although I still get a lot of questions about why they look so different from the lab sometimes, and I think learning the tools to allow manipulation gives that creative freedom.
Have the same experience; some 3 years old negative now gives so much better colors with NLP!
This is also a great video to compare DSLR vs lab scans! Really eye opening and also gives a great idea of what different scanners do. Great video!
Incredibly profound yet modest and chill video. Thanks a ton - instant sub!
Thank you!
Hey, I do a lot of scanning at home with nlp and Fuji xt2 and I love your talk in the beginning about the certain film look, it is what we make out of the negative with the setup and vision we have, keep up with the videos !🙌🏻
Thanks for watching!
This is the best tutorial ever! You explain things so clearly. Cannot wait to try some tips. Thaks a lot for doing this for us!
Happy to help!
I think I liked it just with the exposure adjusted, looks great!
Thanks for a fantastic tutorial. It is really helpful to see the range of decisions you can choose to make while editing the image alongside the lab scan.
Glad it helped!
I'm just getting into film scanning and this was super useful. It's much appreciated, thank you!
Glad it helped!
Great video! I'm really happy that I ended ordering a DSLR scanning setup!
Handy knowing what those the lab glow sliders do but I like the look less blown out than what the lab provides. I’m convinced my lab just puts their frontier scanner on Warm as it has this yellow cast over everything. I think if you want lab results you can get it very close to identical with nlp.
Great! Thank you
Great Tutorial Hashem
At least the Fuji Frontier scanners are pretty close to the look of their photographic paper. So the way scans look from a well calibrated Frontier is not too far off, from a neg to paper print. Which is, as far as there is such a thing, the native look of film.
Nice tutorial! Any pointers for how to bring out colors on my mom's faded Kodachrome 64 slides? Tried to match my light with my white balance, but I got a feeling there's some magic that needs to happen in the calibration panel? Or maybe HSL or color grading is enough - blown out skies often go yellow/brown while shadows go blue. Maybe I should just apply the opposite. Even tried asking ChatGPT but I think it also thought I wanted to make new photos look like old, so I might have to reverse every recommendation I find
I know it's besides the point, but the results straight out of the NLP look way better than the lab scan.
Awesome video as usual! Just so you know, I am scanning with a Nikon coolscan 8000 in negative, DNG without any ajustement at all and I am still having these colored dark road from time to time. I just can't get why they exactly happen. The method you're using is pretty much the best I have heard for controlling them. Thanks!
Great video, I have some problems to enter the reference view in Lightroom. If i start NLP the reference view goes back to one picture view and i am not able to edit the picture with a side by side view to my reference picture. Is there any special setting to do this?
I'm surprised you left the borders (at 9:45 timestamp) while converting. My understanding is that the black border confuses the NLP software and renders inaccurate colors. If you want the borders in the final product first convert without the borders and then add them back in. Am I missing something?
Hey yeah at 9:42 you can see I used the border buffer setting at 10% making cropping unnecessary. What you're referencing used to be applicable to older versions of the software, but this border buffer feature came in a while ago and resolved it. I speak about it more in the main digitisation tutorial that I referenced. But you can probably find a better explanation of the feature on the NLP site.
I've always been curious about the faded blacks and clipped highlights in my scans. My lab uses a Noritsu and they often come back with traits like that and I've always kinda just attributed it to them wanting to give us a "flat" scan to work with on a contrast level... Do you know if it's just a limitation of the scanning software to get those details, or just a matter of the lab techs trying to get through things quickly and not worrying about the details?
Hey Steven, I've experienced this too, but it doesn't seem to be consistent between machines. I think it's more to do with calibration, and maybe the setting of black/white points, or some other variable.
I also recommend inverting the RGB curves manually when NLP messes too much with the image. I have gotten pretty good results doing that in a few simple steps.
I found that method way too time consuming when I used to do it, and NLP's results to be much better. It probably depends largely on other variables.
@@pushingfilm oh sure it’s time consuming, but I’ve been in the situation that NLP just doesn’t get it right or anything close to it and sometimes it’s worth doing it manually for those special frames
@@javixo1997 that makes sense!
Hi, my lightroom does not create positive copy. It is hard to work in reversed sliders... and saturation works like a temperature.
Your biggest problem is that you are working with only 8bit sRGB (web colours) JPG images so you don’t have a full 16bit Adobe RGB (full colour range) TIF image that you can optimise to perfection by pulling out shadow details and multiplying highlights without any banding or excessive noise. David Myers, Kodak Pro Photo CD scanning, Digital Masters Australasia.
Hey! I’ve recently gotten into color grading video, which is a totally different beast than photo. It’s gotten me thinking about how motion pictures are digitally processed.
Somewhere along the way this week it sparked an idea for digital negative processing.
I put together an alternative Lightroom process to NLP that has been a bit more consistent for me in terms of color, and offers me a stronger workflow in Lightroom panels. Probably just as quick as well + wouldn’t cost any money
It’s only fault so far has been working with heavily under or overexposed images. An issue NLP would have as well. Want to do some more experimenting but figured I may publish a video of my own on the process, let me know if you may be interested in hearing it! I want to determine whether it’s truly useful to the public
Nice! I and I'm sure other people would be interested in seeing that video 🙂
Hi, I'm interested in knowing what process you're doing, as color grading video has made me think about how to do color in photo here as well. I'd be interested in knowing what you've come up with, thanks a lot.
You should crop out the little bit of film border after white balancing. That impacts the conversion.
In fact, Nate (the author of NLP) places great emphasis on leaving no border in the image to be converted in NLP. His CZcams video demonstrates the large effect leaving even a little bit of border causes in conversion colours.
when I hear that particular films do t have a look to them, I often wonder what then is the point of purchasing different film stocks. Why buy portra 400 if I get the same from Kodak Uktra max?
There are still inherent tendencies to different film stocks, just not one universal expected look to the same stock. These tendencies can be more than just the general colour tendencies, but also saturation, sharpness, tonality, grain, exposure latitude, and more.
Different film stocks definitely have a different look to them. It’s apparent when you don’ thave as many ways to manipulate the look, like when you RA-4 print them in the wet darkroom. In digital, you can manipulate to your heart’s content but you still get different results due to film speed, grain, colour shifts, latitude and dynamic range, etc, that are baked into the negative and not availalble to you in the scan.
Ngl any film stock can be tuned to have those kodak gold colors and fuji greens
yep
why didn't you crop the negatives before converting them?
Because you don't have to if you enter the border buffer percentage pre-conversion.
Did you scan the negatives yourself? How do the negative scan images have camera metadata in them? Why are they .CR2 files?
digital camera scan of his negatives using his Canon 5DIV and a 100mm macro lens. So, he’s basically editing a high resolution macro digital photo of his negative.
Thanks for answering, Victor!
@@pushingfilm I have the same setup but I use the negative supply scanning rig. I think I get better scans than using the Epson scanner and even better than lab scans most of the time.
@@vtavares00 Awesome! It's not hard anymore... I remember years ago tinkering with DLSR scanning, and it was still really hard to get consistently great results.
@@pushingfilm yes, negative lab pro and Lightroom makes it easy. Rivals lab scans for sure. And with bigger sensor cameras, better lenses, better software, you can just go back to your negatives and keep making your scans even better over time. Amazing time to be shooting film now.
Why would you digitise a piece of film with your camera without masking out the rest of the light-box surrounds? It’s like shooting into the sun with all the dark areas of the film being degraded!
Just curious, why would you try to match a NLP scan to a lab scan if you already have a lab scan of that frame? I suppose it teaches you a general direction for most scans, but when you only have the NLP scan, its difficult to know where it needs to go to look better.
The main reason I've done it is rescans because the lab scan didn't have enough resolution. In terms of matching though, yes, it's meant to just give an idea of the direction you take things in situations when you don't necessarily have a reference (with "better" being subjective, and there naturally being a lot of variation from lab scanner to lab scanner)
Never use generic scans from a lab because from day one they were sub-par vs. the results from my Epson scanner and then I became adept at maximizing the results from the scanner. It is so straight forward to get the quality required for even the most discerning clients that the only reason I can imagine folks needing all this additional work (requiring software such as NLP 2.3) is they haven't a clue regarding crafting their film originals , lighting and exposures , and film processing . Unfortunately those shortcomings on the front end necessitate the salvage operation on the back end and the results in my opinion are never optimal.
Why do you always leave some border? I found it distracting that the 'zoom' on the left and right images were needlessly different
I must disagree a bit, for this video. Indeed you can create the look that you want, but If you take an image of the same subject with Portra, Ektar, Velvia and Kodak E100 there will/must be differences in color , saturation in all the film.
If you need to have film & digital look alike this is a good way but …
For me I don’t like to use plug-ins or scans done by a Lab technician. I have worked with drum scanner in the ‘90. It’s all about how you interpret an image unless you scan a negative as positive and do a manual conversion in photoshop, this way you keep the ‘film’ look.
Take a look at the CZcams channel of Alex Burke how to do this, it’s time consuming but still. For me it’s the way to go.
Curious, wouldn't the scanners used to create the scans include the "differences in color , saturation" found in each film?
so much fluff, get to the point