My preference is to use a scanner for film. The method I use to get the colour right is to begin with a coarse scan over as many frames as possible, including unexposed film between the frames, and, if possible, the totally fogged piece of film at the start of the roll. Then, examine the histogram for the three colour channels, and set the scanner gain controls so that it records each over this range on each channel minus a few percent at the extemeties, and then do the fine scans of each frame. The settings used can be saved or noted and reused for each particular film type, so you only have to do this once. Even with a cheap, old second-hand scanner, this method works well, so the images seem to have an almost realistic 3D quality to them. For photo editing, I use the GIMP and Darktable, which are free, open-source, and very well supported.
I already use the Negative Supply holders and the 99% light source, but I did learn some great tips from this video - thanks so much. The tape to pull through the 120 film was a great one, I have always struggled with getting the film into the second slot. Clipping the corners can also help. Your method to use a grid to ensure consistent placing of the frame and thus allowing batch editing is wonderful - although I have seen often film creep up or down with 35mm making for some inconsistencies - something for me to work on I think. Great video.
What a huge difference between light sources. The images on the left definitely looked much better than the ones on the right. Your subject looked natural.
I tether my A7III to Capture One and digitize that way. Give Capture one a try, the live view and zoom really lets you focus on the grain just like in a darkroom with a grain finder.
I made the same experience. The CRI is crucial. The camera cannot compensate via the white-balance. I tried several WB options. None worked on my Leica M10. I could not believe how much the CRI influences the color rendering. I wanted to build my own light panel with CRI 85, because I could not find higher rates and thought that 85 should be close enough. But this was indeed a delusion. Now I am using a LED lamp with CRI 95 on my very compact copy stand and the colors are rendered correctly. In some of your photos you can see that the color of some cloth details are in one case blue (poor CRI) and in the other case red (appropriate CRI). This is even more significant than the details in the faces. Thanks for the good summary, Robin. It is an important heads-up to be careful with the light source which I can confirm. Greetings from Hannover, Claus
Amazing what a difference the light source makes, I remember noticing the same when I did similar tests, and makes me curious to see how much better it could still get with something like the NS light pad. Thanks for the shout outs btw!
Nice timing, I just finished to build my copy stand setup to scan my negatives and the results are INSANE compared to my epson flatbed. I bought a monitor stand with a glass panel base so I can put the light panel + the film holder on it and it's easier to slide the whole light panel to take multiple shots. I bought a 50mm enlarger lens for durt cheap (20€) and with some extention tubes the results are very good. My digital camera has a full frame sensor so I can scan negatives with a 1:1 magnification ratio which is good enough for small prints or social media. If I want to blow-up a frame I can extend the tubes and take 4 shots which gives mindblowing details, it really choked me, haha. Now I'm waiting for my essential film holder to arrived 😆
I shoot B&W almost exclusively and mostly scan my own film, but I'm hoping to branch out into color soon so I'll definitely keep this in mind. Thanks for the demonstration.
Wow Robin, great video and very good and and useful explanations. I've been to the companies sites that you mentioned in the video and although the quality of their gear is outstanding it is also pricy if you don't scan a lot !! I also like the quality Carmencita delivers for the money they charge. It's hard to find the right point between send it out or do it yourself. Thanks for sharing.
With my canon DSLR I tether directly to lightroom classic and via a preset make the raw file a positive one so that I can edit the scan easier. I should say that I don't use negative lab pro and instead use a combination or lightroom and Photoshop in order to colour correct when I scan c41 film.
I'm curious if you have tips for finding the right height for scanning. I'm using a Fuji X-E4, with the Nikon 55mm f3.5 Ai, but the rest of the setup is the same.
What's really frustrating is, Kodak developed color accurate ways to digitally scan film in the early 90s. But because of their exit from scanning and developing, the process is now lost to the public. They could get color accurate scans every time. Now we are stuck fumbling with curves or Negative Lab pro and its still not accurate.
There’s really no such thing as “color accurate” scans or prints from color negative film. It’s all up to interpretation, and always has been, even back in the dark room days. Slide film, totally different story.
@@joeltunnah Well, removal of the color mask color cast is probably what the OP was talking about. Modern flatbed's and other dedicated neg scanners scan color negatives with a longer exposure time of the blue and green channels. This acts as an analog glass filter at the lens, and the longer exposure boosts the blue and green components and reduces the orange. If Kodak figured out how to do this with software, it is very sad it never made it to the public. Getting consistent results with a DSLR for color film is still a lot of trial and error and usually gives you a cooler (blue) image that you really can't 100% correct for.
@@patrickjclarke I gave up dslr scanning a long time ago, for many reasons. A dedicated scanner like the Plustek 8200i is far superior, consistent, and ultimately less time consuming (unless you don’t mind or even want dust all over your images).
You need to use a macro lens or convert a normal lens with extension tubes. I like extension tubes for macro photography where sharpness at the corners is not critical, but macro is much better for film scanning.
How could be better scanning an negative with a SLR camera, if the best Sony sensors have around 25bit color depth, while an entrylevel Epson scanner have 48bit color depth and 16bit grayscal depth... Look the photos made with some scanner cameras, the colors a true natural colors, no interpoletion like on area sensors
I'd like to see a 35/120 film scanner that's as quick and accurate as doing it using a camera. I think there's room in the market for them to do that. And, if film keeps getting more popular, maybe they might even start to make some more films. Bring back some oldies...anyone for Kodachrome??!!
What really turns me off is how much of a rich kid's hobby film photography seems to become, all these accessories seem so ridiculously overpriced. I mean, I get that these are often small companies making small batches and the quality is probably good, but damn, $300 for a mechanical film holder is as much as a decent film scanner costs, and that's not counting all the other things you need, like the light source, stand, $99 for Negative Lab, and of course a digital camera with a good macro lens. As much as I'd like to be able to scan a little faster using this method, pricing is such a downer that I'd rather invest in another film scanner and enjoy the slow scanning time with podcasts and audiobooks. I hope someday some Chinese companies start making cheaper knockoffs of these negative scan supplies, I mean no hate for the small companies who keep the hobby alive but I simply cannot afford it and leave it to others to support them.
Camera scanning isn’t faster if you have to spend an hour cloning out dust specks and scratches. Not to mention the inversion/conversion process time. A $500 dedicated scanner pays for itself after about 30 rolls, vs lab scans.
@@joeltunnah I completly agree. Get a Plustek scanner and you are way better off then with a camera setup. The Plustek has an IR channel for automatic dust removal. I have bought used Minolta scanners because I was not happy with the scen results of the labs.
Thanks for sharing this setup, but I'm hoping more companies will come to market to offer a more desk friendly solution as this setup is quite big and requires a lot of investment to begin with
No automatic dust removal is a deal-breaker for me. Camera scanning is more time consuming, and ultimately more expensive than a dedicated scanner, when you add up the holder, light source, negative lab pro, and a decent macro lens. For 35mm I recommend a Plustek 8200i ai. It really beats any lab scans I’ve gotten, and costs less than just the macro lens you need.
I have a 8200i ai as well and, although being pretty good, it's nowhere near lab quality. It cannot match it in terms of detail. If you want better dust removal too, get a Prime Image XAs. I had one and the dust removal was lab quality. Unfortunately it had these streaks through the images I couldn't get rid of so sent it back and got the 8200.
@@doozledumbler5393 I’ve done the direct comparison against Noritsu lab scans using the same negatives. The 8200i is superior in resolution, color, and dynamic range. I won’t mention the lab name, but they are in the top tier online, and the scans cost me $24 per roll for their top service and resolution. There are many possible reasons why you’re not getting better results. The software, the scan settings, the post processing, etc. I used Silverfast 8.8 with auto unsharp, 3600dpi, 24bit. Someday I’ll make a blog post on my website about the comparison.
@@joeltunnah I scan using Vuescan and convert using Neg Lab Pro. The results I get are better merely because I have control over the process and can choose certain things like contrast levels etc. But I still think lab scans on a Frontier SP3000, which my lab uses, end up having more detail. I pay $14 per roll for low res scans. In any case, it's hard to tell since my lab slathers their scans with a buttload of noise reduction.
@@doozledumbler5393 lots of variables in there. I’ve seen comparisons between Vuescan and Silverfast, and I have to say it wasn’t even close to my eyes, Silverfast produced better more detailed results. People seem to prefer the interface of. Vuescan, but if you spend a little time with SF it all makes sense. If you’re saying your 8200i scans are worse than low res Frontier lab scans, then something is definitely wrong on your end.
@@joeltunnah Vuescan is just easier because you can get raw DNGs exported with IR cleaning. Silverfast can't do cleaning on it's raw files. I'd use Silverfast if it did. It cleans TIFFS but then Negative Lab Pro is a bit cumbersome with TIFFs.
My preference is to use a scanner for film. The method I use to get the colour right is to begin with a coarse scan over as many frames as possible, including unexposed film between the frames, and, if possible, the totally fogged piece of film at the start of the roll. Then, examine the histogram for the three colour channels, and set the scanner gain controls so that it records each over this range on each channel minus a few percent at the extemeties, and then do the fine scans of each frame. The settings used can be saved or noted and reused for each particular film type, so you only have to do this once. Even with a cheap, old second-hand scanner, this method works well, so the images seem to have an almost realistic 3D quality to them. For photo editing, I use the GIMP and Darktable, which are free, open-source, and very well supported.
I already use the Negative Supply holders and the 99% light source, but I did learn some great tips from this video - thanks so much. The tape to pull through the 120 film was a great one, I have always struggled with getting the film into the second slot. Clipping the corners can also help. Your method to use a grid to ensure consistent placing of the frame and thus allowing batch editing is wonderful - although I have seen often film creep up or down with 35mm making for some inconsistencies - something for me to work on I think. Great video.
What a huge difference between light sources. The images on the left definitely looked much better than the ones on the right. Your subject looked natural.
I tether my A7III to Capture One and digitize that way. Give Capture one a try, the live view and zoom really lets you focus on the grain just like in a darkroom with a grain finder.
I made the same experience. The CRI is crucial. The camera cannot compensate via the white-balance. I tried several WB options. None worked on my Leica M10. I could not believe how much the CRI influences the color rendering. I wanted to build my own light panel with CRI 85, because I could not find higher rates and thought that 85 should be close enough. But this was indeed a delusion. Now I am using a LED lamp with CRI 95 on my very compact copy stand and the colors are rendered correctly. In some of your photos you can see that the color of some cloth details are in one case blue (poor CRI) and in the other case red (appropriate CRI). This is even more significant than the details in the faces. Thanks for the good summary, Robin. It is an important heads-up to be careful with the light source which I can confirm. Greetings from Hannover, Claus
Extremely useful. No waffle at all. I learnt so much, thank you! BTW your street walks are interesting too, but this video was educational.
Amazing what a difference the light source makes, I remember noticing the same when I did similar tests, and makes me curious to see how much better it could still get with something like the NS light pad. Thanks for the shout outs btw!
Thanks for this very comprehensive interesting and informative video
Yeah I’ve been using my iPad as a light source and I think it’s finally time to step up to a dedicated source.
Nice timing, I just finished to build my copy stand setup to scan my negatives and the results are INSANE compared to my epson flatbed.
I bought a monitor stand with a glass panel base so I can put the light panel + the film holder on it and it's easier to slide the whole light panel to take multiple shots. I bought a 50mm enlarger lens for durt cheap (20€) and with some extention tubes the results are very good. My digital camera has a full frame sensor so I can scan negatives with a 1:1 magnification ratio which is good enough for small prints or social media. If I want to blow-up a frame I can extend the tubes and take 4 shots which gives mindblowing details, it really choked me, haha.
Now I'm waiting for my essential film holder to arrived 😆
I shoot B&W almost exclusively and mostly scan my own film, but I'm hoping to branch out into color soon so I'll definitely keep this in mind. Thanks for the demonstration.
Wow Robin, great video and very good and and useful explanations.
I've been to the companies sites that you mentioned in the video and although the quality of their gear is outstanding it is also pricy if you don't scan a lot !!
I also like the quality Carmencita delivers for the money they charge.
It's hard to find the right point between send it out or do it yourself.
Thanks for sharing.
Quality video dude!!! Very useful!
With my canon DSLR I tether directly to lightroom classic and via a preset make the raw file a positive one so that I can edit the scan easier.
I should say that I don't use negative lab pro and instead use a combination or lightroom and Photoshop in order to colour correct when I scan c41 film.
Nice video. Would be great if capture one jumped on the film bandwagon and offered conversion while tethered. Not holding my breath, though.
With the Sony remote app, I hit the 1 key to fire the shutter. Makes import much faster as I don’t need the 2 sec delay.
Really informative .
Thanks .
I think nikon D850 has a negative conversion preview option.
What about automatic dust removal?
I thought NLP recommends you crop out the border after colour correction and before conversion as that affects the conversion quality?
I'm curious if you have tips for finding the right height for scanning. I'm using a Fuji X-E4, with the Nikon 55mm f3.5 Ai, but the rest of the setup is the same.
What's really frustrating is, Kodak developed color accurate ways to digitally scan film in the early 90s.
But because of their exit from scanning and developing, the process is now lost to the public.
They could get color accurate scans every time. Now we are stuck fumbling with curves or Negative Lab pro and its still not accurate.
That’s sad to know.
Yeah it’s basically just random at this point, which feels a bit lame haha but oh well
There’s really no such thing as “color accurate” scans or prints from color negative film. It’s all up to interpretation, and always has been, even back in the dark room days.
Slide film, totally different story.
@@joeltunnah Well, removal of the color mask color cast is probably what the OP was talking about. Modern flatbed's and other dedicated neg scanners scan color negatives with a longer exposure time of the blue and green channels. This acts as an analog glass filter at the lens, and the longer exposure boosts the blue and green components and reduces the orange. If Kodak figured out how to do this with software, it is very sad it never made it to the public. Getting consistent results with a DSLR for color film is still a lot of trial and error and usually gives you a cooler (blue) image that you really can't 100% correct for.
@@patrickjclarke I gave up dslr scanning a long time ago, for many reasons. A dedicated scanner like the Plustek 8200i is far superior, consistent, and ultimately less time consuming (unless you don’t mind or even want dust all over your images).
Does anyone has a link to that kind of copy stand (in EU) the only thing I can find is not as basic, overkill and heavy...
It's important own a macro lens? Or I can use a normal lens to scan?
You need to use a macro lens or convert a normal lens with extension tubes. I like extension tubes for macro photography where sharpness at the corners is not critical, but macro is much better for film scanning.
How could be better scanning an negative with a SLR camera, if the best Sony sensors have around 25bit color depth, while an entrylevel Epson scanner have 48bit color depth and 16bit grayscal depth...
Look the photos made with some scanner cameras, the colors a true natural colors, no interpoletion like on area sensors
I'd like to see a 35/120 film scanner that's as quick and accurate as doing it using a camera. I think there's room in the market for them to do that. And, if film keeps getting more popular, maybe they might even start to make some more films. Bring back some oldies...anyone for Kodachrome??!!
Is it possible to use Leica Q in Macro mode for scans?
Good question ⁉️
Hallo Robin! Off-topic but do you have any plans to resume the podcast? I really enjoyed it.
What really turns me off is how much of a rich kid's hobby film photography seems to become, all these accessories seem so ridiculously overpriced. I mean, I get that these are often small companies making small batches and the quality is probably good, but damn, $300 for a mechanical film holder is as much as a decent film scanner costs, and that's not counting all the other things you need, like the light source, stand, $99 for Negative Lab, and of course a digital camera with a good macro lens.
As much as I'd like to be able to scan a little faster using this method, pricing is such a downer that I'd rather invest in another film scanner and enjoy the slow scanning time with podcasts and audiobooks. I hope someday some Chinese companies start making cheaper knockoffs of these negative scan supplies, I mean no hate for the small companies who keep the hobby alive but I simply cannot afford it and leave it to others to support them.
Agreed. A Sony A7III costs about $2,500 just for the body. Lol. Who has one of those just laying around?
Camera scanning isn’t faster if you have to spend an hour cloning out dust specks and scratches. Not to mention the inversion/conversion process time.
A $500 dedicated scanner pays for itself after about 30 rolls, vs lab scans.
@@joeltunnah I completly agree. Get a Plustek scanner and you are way better off then with a camera setup. The Plustek has an IR channel for automatic dust removal. I have bought used Minolta scanners because I was not happy with the scen results of the labs.
Your negatives weren't focussed properly if you had to change their plane angle in lightroom.
Thanks for sharing this setup, but I'm hoping more companies will come to market to offer a more desk friendly solution as this setup is quite big and requires a lot of investment to begin with
A linear scan from a proper film scanner and ColorPerfect is still the best.
No automatic dust removal is a deal-breaker for me. Camera scanning is more time consuming, and ultimately more expensive than a dedicated scanner, when you add up the holder, light source, negative lab pro, and a decent macro lens.
For 35mm I recommend a Plustek 8200i ai. It really beats any lab scans I’ve gotten, and costs less than just the macro lens you need.
I have a 8200i ai as well and, although being pretty good, it's nowhere near lab quality. It cannot match it in terms of detail. If you want better dust removal too, get a Prime Image XAs. I had one and the dust removal was lab quality. Unfortunately it had these streaks through the images I couldn't get rid of so sent it back and got the 8200.
@@doozledumbler5393 I’ve done the direct comparison against Noritsu lab scans using the same negatives. The 8200i is superior in resolution, color, and dynamic range. I won’t mention the lab name, but they are in the top tier online, and the scans cost me $24 per roll for their top service and resolution.
There are many possible reasons why you’re not getting better results. The software, the scan settings, the post processing, etc. I used Silverfast 8.8 with auto unsharp, 3600dpi, 24bit. Someday I’ll make a blog post on my website about the comparison.
@@joeltunnah I scan using Vuescan and convert using Neg Lab Pro. The results I get are better merely because I have control over the process and can choose certain things like contrast levels etc. But I still think lab scans on a Frontier SP3000, which my lab uses, end up having more detail. I pay $14 per roll for low res scans. In any case, it's hard to tell since my lab slathers their scans with a buttload of noise reduction.
@@doozledumbler5393 lots of variables in there. I’ve seen comparisons between Vuescan and Silverfast, and I have to say it wasn’t even close to my eyes, Silverfast produced better more detailed results. People seem to prefer the interface of. Vuescan, but if you spend a little time with SF it all makes sense.
If you’re saying your 8200i scans are worse than low res Frontier lab scans, then something is definitely wrong on your end.
@@joeltunnah Vuescan is just easier because you can get raw DNGs exported with IR cleaning. Silverfast can't do cleaning on it's raw files. I'd use Silverfast if it did. It cleans TIFFS but then Negative Lab Pro is a bit cumbersome with TIFFs.