Idea: a great big film stock, emulsion and scanner test

Karim D. Ghantous · February 13

There’s an idea which I have been thinking about for a couple of years: imagine a free resource where people can access DPX files of all the available film products, so that they can compare them against each other, exposed in different ways.

This test will include three gauges: 8, 16, and 35. It will include all motion stocks and photographic emulsions. Films will be scanned on different scanners, including a few photo scanners. Films will be underexposed, overexposed, pushed, pulled, and scanned. This test will include colour, b&w, negative, reversal, and 2383.

All 35mm stocks and emulsions will be shot as 4-perf S35, although the emulsions may have to be pulled through the camera one frame at a time, due to perforation differences.

The final product will be a fairly sizeable collection of files (mostly DPX), ungraded and uncorrected. This resource could help DPs and directors save time and money. It might even catalyse enthusiasm for film.

There is a very good chance that I’ll be able to fund it later this year. I presume that the total cost of this project will start at $20,000. I wouldn’t be surprised if it’s higher. The one thing I cannot do is make a promise at this point.

This exercise will require camera rental, lens rental, talent (with different skin types), at least one DP and one assistant, the film, the processing, the scanning, the locations, some lights, some travelling, and hopefully some PR in conjunction with Kodak.

So, I need three things: expressions of interest, opinions of viability/utility, and a cost estimate. All opinions welcome.

P.S. I'm aware of @Florian Noever's upcoming results of his resolution test.

Jeff Bernstein · February 13

Hi there. Great idea, KDG. Shall we call you The Celluloid Evangelist? Cinematography.com should set up a forum specifically for this resource. Also, it would be awesome if other cinematography.com artists contribute to the effort. In fact, what you have come up with sounds like an ultimate collective cinematography.com project. Artists joining together to do something positive for everyone's encouragement and inspiration -- just a great idea all round. Celluloid film is like true love -- accept no substitutes!

Andrew Trost · February 13

I love this idea and am very interested! The more information there is available to people the better. What are your plans for what you put in front of the lens? And which lens(es) were you considering using, knowing that there will be some kind of bias in terms of contrast, color, sharpness, etc?

The planning/prep phase would be so important to ensure consistency. I would love to help in any way, as I'm often looking for examples of different "cocktails" of exposure, developing, scanning, etc. This is right up my alley!

Tyler Purcell · February 13

I've been working on a similar test, but more technical, using density testers, measuring MTF, using all 3 major formats (s8,16,35) and multiple scanners; Arri Scan XT, Scan Station and Scanity. Kinda showing the difference between the formats, but also how they deal with post workflows. Then we are going to also compare digital cinema cameras with the same crop factor as the film, tho this would be probably S16 and S35mm, rather than S8 as well. The final part of the test will be projecting Ektachrome test patterns on 16mm and 35mm, so we can see how much resolution humans can deduce from the formats. Sadly even one light prints of negative reduce resolution quite a bit.

The final video will be on our channel (Now named Narrow Gauge Films LLC) and we'll be in production shortly. We've already started the S16mm vs Blackmagic 12k G1 (cropped to S16mm) resolution test. Sadly the Blackmagic whips film off the block, even when cropped to S16mm. We are going to do another test because the first one was so bad for film, we need to potentially use sharper lenses, but that's easy to figure out.

I think the key with any resolution test is to first look at what Steve Yedlin has done and try not to repeat it. He's already done a fantastic job at showing the difference between digital and film with 35mm. He even did 2 videos and they're both excellent. Spending a lot of money doing exactly what he did, isn't going to further anything.

Anyway, good luck! Looking forward to seeing your results! I'll for sure post mine when done, probably over the summer.

Jon O'Brien · February 13

2 hours ago, Tyler Purcell said:

... Sadly even one light prints of negative reduce resolution quite a bit ... Sadly the Blackmagic whips film off the block ...

I like the look of a film print. Digital resolution can be self-defeating past a certain point. I feel the same way with paintings, too. I prefer an impressionist look for the overall impression of the subject and the scene and the meaning, and not the painting of every single leaf, as it were. That's a conscious choice.

I'm not surprised the digital camera whips film with resolution. I'd rather shoot film though.

Edited February 13 by Jon O'Brien

Tyler Purcell · February 14

2 hours ago, Jon O'Brien said:

I like the look of a film print.

Me too, just hard to measure MTF and resolution with a one light off the negative.

Jack Jin · February 14

That would be amazing! May I suggest a test with different extreme colored gelled lights? To test extreme saturated colors and to see any gamut clipping and color response between different filmstocks and scanners too!

Geffen Avraham · February 14

This would be great! Please include big color charts, I am planning on making an open-source program soon for matching different cameras and stocks based on color charts, this would be a huge resource!

Florian Noever · February 15

Very interesting idea!

Such a test would be an immensely helpful resource, especially for people who don't have the opportunity to test these things before shooting.

My test is nowhere near as extensive and was conducted with the tools I had at my disposal.

Karim D. Ghantous · February 15

On 2/14/2025 at 1:18 AM, Jeff Bernstein said:

Hi there. Great idea, KDG. Shall we call you The Celluloid Evangelist?

You can. But I'd be guilt of stolen valour! There are people on this forum, like Tyler, who do a far better job, and who have seen more feet of film than I ever will.

On 2/14/2025 at 4:15 AM, Andrew Trost said:

What are your plans for what you put in front of the lens? And which lens(es) were you considering using, knowing that there will be some kind of bias in terms of contrast, color, sharpness, etc?

The best performing lens that fits in the budget. Maybe Summicron C's? Or Master Primes? There's no room for Super Speeds or anamorphics or Primos. That would distract from the exercise. If/when I can guarantee funding, I'll contact you. This test will probably be conducted in America. I'm just not sure whether it's going to be east coast or west coast.

On 2/14/2025 at 6:32 AM, Tyler Purcell said:

I've been working on a similar test,

I very much look forward to seeing that. You're right though - digital is way too good as far as resolving power is concerned. There's room for film to win the DR contest, but nobody's doing anything.

On 2/14/2025 at 12:51 PM, Jack Jin said:

That would be amazing! May I suggest a test with different extreme colored gelled lights? To test extreme saturated colors and to see any gamut clipping and color response between different filmstocks and scanners too!

Great idea. Have you noticed how digital takes a huge resolution hit when a scene is lit by primary light sources?

On 2/14/2025 at 3:49 PM, Geffen Avraham said:

This would be great! Please include big color charts, I am planning on making an open-source program soon for matching different cameras and stocks based on color charts, this would be a huge resource!

That's a good idea too. Using charts costs the same as not using them. I'll include them even with the b&w films. I'll also include white point and black point chips.

Abdul Stone Jackson · February 15

On 2/13/2025 at 8:21 AM, Karim D. Ghantous said:

The final product will be a fairly sizeable collection of files (mostly DPX), ungraded and uncorrected. This resource could help DPs and directors save time and money. It might even catalyse enthusiasm for film.

This is an excellent idea. Once the work is done initially, with the right support, it can grow into a website that lists the test and the related file download. More people are shooting film, so the test results will be valuable for film shooters of all experience levels. This could be the start of a film test repository. I think we can all learn something from everybody who has meaningful tested efforts. Examining results from Ghantous, Purcell, Yedlin, Noever, and others who may be inspired to shoot their own tests (Me included) with short-ends or unused leftover stock would be a great thing to have in a single place someday.

Echoing Mr. Tyler, we should pay attention to what was already done to further expand the work. Mr. Ghantous, you’re taking on a big project that I think we all will need and use.

Daniel D. Teoli Jr. · February 15

OP, commendable project!

But you got $20k+ to burn...you can become a world leader in open content 8mm film archives.

1) You buy an overpriced PictorPro 8mm scanner from FilmFabriek for $20K-ish

film%20lot%208mm%20eBay%20D.D.Teoli%20Jr

2) You buy pig in a poke 8mm and S8mm lots off of eBay for a few bucks a reel.

3) You scan worthwhile films and upload to the Internet Archive.

4) After scanning you put the same lots of film you bought and scanned on eBay to sell and recoup some of your $.

5) You keep going, scanning and recycling the films.

Personally, I'm not interested in 8mm. Too low Q for me. Before getting into 16mm, I was into 8mm. Eventually I sold off most of my 8mm film collection, but I still have 400-500 8mm reels to scan someday. Hence, my interest in a PictorPro.

Now, getting back to your original project...

Ask for donations of material for your reference collection. You can do both projects. You just won't be able to do your original project in the depth you envisioned. I don't know how much material you will get with donations. People nowadays are pretty shitty with donations when you request material, but you may be luckier than me. And it is not like I am asking for them to give material to me, I'm usually asking for scans or to let me scan material. I've been an archivist off and on going back to the 1980s and have seldom if ever got any donations of material from people when requested. I have to buy all my material. I don't care that I have to buy it, that is my work and my life. I only care if I don't have the money to buy it.

Well, whatever direction you head in...good luck with it all!

<><><><>

Daniel D. Teoli Jr. Archival Collection
Daniel D. Teoli Jr. Small Gauge Film Archive
Daniel D. Teoli Jr. Advertising Archive
Daniel D. Teoli Jr. VHS Video Archive
Daniel D. Teoli Jr. Popular Culture Archive
Daniel D. Teoli Jr. Audio Archive
Daniel D. Teoli Jr. Social Documentary Photography

Tyler Purcell · February 15

9 hours ago, Karim D. Ghantous said:

I very much look forward to seeing that. You're right though - digital is way too good as far as resolving power is concerned. There's room for film to win the DR contest, but nobody's doing anything.

Yea we did a DR test with 5219 in 2023 that basically put it worse than our Blackmagic 12k, at around 7 stops. The 12k was 8 stops. Some people think this is due to the scan, so we will probably do an HDR scan in our next test to kind of help eliminate that issue. But the film had 5 stops of usable latitude above middle grey and the digital shined in the under middle grey.

Tyler Purcell · February 15

3 hours ago, Daniel D. Teoli Jr. said:

But you got $20k+ to burn...you can become a world leader in open content 8mm film archives.

IDK how you'd make money doing that. The 8mm folks, generally don't want to pay anything.

Abdul Stone Jackson · February 15

1 hour ago, Tyler Purcell said:

IDK how you'd make money doing that. The 8mm folks, generally don't want to pay anything.

Indeed.

Dan Baxter · February 16

On 2/14/2025 at 6:32 AM, Tyler Purcell said:

I've been working on a similar test, but more technical, using density testers, measuring MTF, using all 3 major formats (s8,16,35) and multiple scanners; Arri Scan XT, Scan Station and Scanity.

I'll send you an email, I'd like you to send some test film over to a friend of mine to compare ScanStation-to-ScanStation sometime. If you have any 17.5mm film that would also be of interest for a test.

On 2/14/2025 at 12:21 AM, Karim D. Ghantous said:

There is a very good chance that I’ll be able to fund it later this year. I presume that the total cost of this project will start at $20,000. I wouldn’t be surprised if it’s higher. The one thing I cannot do is make a promise at this point.

The scanning itself should be relatively cheap, if not almost free for this type of project - the costs involved would more be in getting the film sent out. Unless you're doing an unrealistic volume of film for these tests of course - ideally you collect all your test film including SMPTE resolution film, colour density film, etc and then make sure it's clean and then send it out. You should also have something in rough condition that will benefit from wetgate scanning. Send me a DM please because I'll let you know some good companies abroad who might be interested in participating.

Where that might not be the case is where you have older model scanners. Now, personally I think you should look to include scans from the JAI 5K ScanStation and the JAI 10K Directors if you can find any and the companies will participate (heck even the CCD ScanStations as they still exist too). There's an incorrect perception by the public that all the companies that advertise they have a LG, Arri, or DFT all have the latest models, however that's often not the case.

I think $20,000 would be enough to get what you're wanting. Reciprocal access to the test scan DPXs and DNGs would go a long way to showing the companies how their own scans compare to others. If we need to build a website, that's easy and doesn't cost much.

On 2/14/2025 at 12:21 AM, Karim D. Ghantous said:

The final product will be a fairly sizeable collection of files (mostly DPX), ungraded and uncorrected. This resource could help DPs and directors save time and money. It might even catalyse enthusiasm for film.

Any of the Bayer scanners should look to scan to camera raw or to DNG. You can scan to both DPX and DNG at the same time on a ScanStation it's just a higher data flow, that way you have LaserGraphic's debayering in the DPX and then the DNG that can be debayered separately. There are companies that use their own in-house debayer algorithms too.

Karim D. Ghantous · February 16

15 hours ago, Daniel D. Teoli Jr. said:

2) You buy pig in a poke 8mm and S8mm lots off of eBay for a few bucks a reel.

3) You scan worthwhile films and upload to the Internet Archive.

4) After scanning you put the same lots of film you bought and scanned on eBay to sell and recoup some of your $.

image.png.87d967d109c40346b44a67c3b88ad94d.png

But seriously, asking for donations (i.e. already exposed film) could be part of the test. I'd like to compare well exposed 5298 to modern films, just as one example. I might end up doing free scanning for people who want their films digitised, but can't afford it. Once I can guarantee that I can fund this project, maybe we can put the word on the street, so to speak.

4 hours ago, Dan Baxter said:

I think $20,000 would be enough to get what you're wanting. Reciprocal access to the test scan DPXs and DNGs would go a long way to showing the companies how their own scans compare to others. If we need to build a website, that's easy and doesn't cost much.

Any of the Bayer scanners should look to scan to camera raw or to DNG. You can scan to both DPX and DNG at the same time on a ScanStation it's just a higher data flow, that way you have LaserGraphic's debayering in the DPX and then the DNG that can be debayered separately. There are companies that use their own in-house debayer algorithms too.

I see. Isn't DPX some kind of standard though? There's no harm in offering both.