Perry Paolantonio

We've been doing Double 8mm scans on our ScanStation recently - you don't need to slit the film, and we can do 4k scans of the film frame itself. That saves a fair bit of money because you can process the film wherever you'd normally process 16mm, without having to deal with someone who has a mechanical slitter. We just ask that you have your lab keep each reel separate, so that we can scan both sides and keep them together (longer, consolidated reels require editing after the scan to get the two halves of each reel back together. -perry

Hi Mark, You can PM me here on the forum or send me an email through our web site -- that goes directly to me. Thanks! -perry

The bigger issue with the BMD scanner is not that it's a bayer sensor (our ScanStation has one and the results are outstanding). Lack of color information isn't typically an issue with bayer sensors. But poorly implemented bayer sensor designs mean lots of fixed pattern noise, something but the BMD scanner definitely has (unless they've specifically addressed this recently). In the BMD scanner, the sensor is fixed relative to the film gate, so only 35mm is scannable at UHD (the machine's max resolution). 16mm is less than HD as you've surmised. The GoldenEye is a superior scanner to the BMD in almost all respects, so if you need to do the work in Australia, that's the way I'd go (though, PM me if you want a quote for doing this in the US. we work with lots of international clients, including some down under). It does have issues with splice bumps, however. And because it's a line scanner with a continuous motion transport (like the Spirit), it's susceptible to streaking artifacts if the sensor gets dirty. 16bit is way overkill. you just don't need it, and you're going to be fighting with the files if you go that route. They're massive and very few systems can actually play them correctly. 10bit DPX, ProRes 422HQ (10bit), ProRes 4444 (12bit) are the way to go, depending on your workflow. -perry

That's a bummer. I still use the speed checker I bought from him about 20 years ago!

Thanks, Drew. Glad everything worked out, and we're looking forward to seeing the final product! -perry

I don't see how it can do anything but help them. On the one hand, if their film isn't as good looking, it'll push them to improve. On the other, it will just mean more options available, and potentially more competition thus lower prices for filmmakers, thus more people using it/more film purchased. Win-Win.

I would agree somewhat with Simon - There are two issues I see immediately: One is that it wasn't scanned on a scanner that can correct for the lateral perf movement problem on Super 8. That's why your frame jerks back and forth like that, and it's *really* distracting. I wouldn't consider that to be one of the charms of Super 8, I'd consider it to be a defect, and one that distracts from the story. The other thing is that by showing the perfs, you're also making it clear to the viewer that you're flipping shots, which is equally distracting. At about 2:30, the perf is suddenly on the right, not the left, then it jumps back. It's superfluous to the story, but suddenly becomes part of it, because the viewer is left thinking about why you chose to flip that shot. If the whole thing was cropped to just inside the frames, that wouldn't be an issue. -perry

Post-striping was common on S8. Super 8 Sound, which eventually became Pro8mm, used to do this in Cambridge. From earlier in this thread: It's not an ideal workflow, and is probably hugely expensive. Not to mention there aren't as many devices that can play mag striped 16mm as there are those that can play optical. But it's certainly possible.

That's a really weird distinction. For one thing, I could scan a film without any optical pin registration and arrive at an identical result using pretty commonly available desktop software to align the frames in post. This is precisely what the ScanStation does, only it's doing it in real time, saving you a step. But it's the same basic algorithm: find the perf, put it in a fixed location relative to the file size, vertically. Find the film edge, put it in a fixed location laterally. Crop to output size. Not exactly rocket science. You'll need to take my word for it. The problem with the perfs being placed inconsistently relative to the film edge causes a lateral jerking motion, IF you use the perf to "pin register" the frame. The perf is held in place, but then the image jerks back and forth. By using the perf for vertical registration and the film edge for horizontal, it's effectively acting like the camera does, where an edge guide or some similar mechanism is holding the perf-opposite side in place. This fully corrects the lateral jerking motion. I personally worked with Lasergraphics (along with at least one other ScanStation owner) to test this and make sure it's good. It is. There remains some slight vertical jitter, but this seems to vary from camera to camera. Some have suggested that the perf next to the film is not the one to use for registration, but I don't see a way around that, at least not in the design of most modern scanners. The vertical jitter is almost imperceptible once you crop the frame lines (it's more apparent when you can see the frame lines moving, but you barely notice it when cropped. And again, different cameras have different levels of vertical jitter. Well, there is. The format was designed for consumer use, and the perfs were never meant to be used as registration references. In most cameras and projectors, the pulldown claw is tiny, and doesn't behave like a registration pin because it doesn't fill the whole perf. In some cameras, the pulldown claw does act like a registration pin. In most cases you don't see this problem in projection. Within the system it was designed for, it's a perfectly good design. You're expecting it to be like 16mm. it isn't. You need to get past that. They're unlikely to change the mechanical specifications, this late in the game. I do agree they need to work on branding, but I don't expect them to change the way the film is made. For one thing, I wouldn't be surprised if perfect perf placement is a major technical issue. The variance from one perf's position to the next is a fraction of a millimeter. It's not until you start applying precision hardware to the film in camera and in scanning that this problem appears.

Scanners like the ScanStation already compensate for this by doing stabilization in-scanner correctly. You're saying that the fact that it's done "electronically" is worse than mechanically? In fact, optical stabilization is *better* than mechanical because it's both more flexible and more accurate. According to Kodak, the perfs aren't "made wrong," they're that way by design. This has been hashed out on this forum and others over and over again, but the basic idea is that pin-registered cameras and pin registered scanners expose the underlying issue, which doesn't come up with the generally sloppier tolerances of typical camera/projector combinations. It's a problem only in the sense that 50 years ago, Kodak didn't conceive of modern film scanners or pin-registered cameras. There are simple workarounds, which in the days before the internet, nobody outside of a lab would have known or even cared about. The end result of scans done on properly shot film with cameras like these is really impressive. -perry

Wrong. Please read what I wrote. The rocking is because it was stabilized incorrectly in post.

Pre-coffee post. bad idea. What I meant to say was "unless they're using a scanner that's doing (rotational) stabilization on the (perf)," not the frame. As soon as you add in rotational stabilization and you're using a perf that's in the middle of the frame, you're opening yourself up to the rocking motion. If you notice, the worst rocking on these examples, (like on the Logmar video), is always on the opposite end of the film from the perf. That's because using a small single object for rotational stabilization will cause things farther away to have a more extreme movement. Basically, the perf becomes the pivot point for a fairly wide swing. the tiniest imperfection in the perf edge relative to the film edge will cause the film to swing up and down on the right side. So the right way to do stabilization post-scan, is to use either the frame edge (if you can see it - dark shots are harder), or a stationary object in the shot, or some combination of the two -- and to use multiple reference points. This is one case where gate hairs can be your friend...

This is incorrect. The problem is not the scanner (unless they're using a scanner that's doing stabilization on the frame, which is kind of crazy to try to do in-scanner (because you can't always see the edges of the frame). See my post above. This is a problem of post-scan stabilization. The perf issue is not limited to color, we've seen it happen on Black and White as well. It's also completely untrue that it would be better with old film. We've scanned film going back to the 60s that exhibits the same problem.

The "rocking boat" syndrome is due to post-scan stabilization. It has little to do with the camera. The logmar footage that I scanned for Friedemann Wachsmuth, when he was testing the early Logmar cameras, shows the same rocking. It was stabilized because of the problem of Super 8 Perfs not being consistently placed relative to the edge of the film. When the film is shot in a pin-registered, or very stable camera (it happens with Canon, Nikon, Beaulieu and other cameras as well), then scanned with a pin-registered scanner like the ScanStation, you see the problem appear. That rocking motion WAS NOT in the film or the scan, it was from stabilization that was done after the scan was delivered. Since then, the Lasergraphics ScanStation software has a new feature that corrects for the horizontal jerking motion, by using the film edges to simulate a spring-loaded edge guide. In any case, this is not a good example of the quality of the camera, and I don't think we should judge it by this.

True, but even if I didn't I'd still shoot on S8.

yes. Far as I know, all of us do.

That is some twisted up logic right there. I am one person. I speak for me and only me. I'm shooting Super 8 home movies because I like the way they look and I know that 50 years from now, they'll be viewable. I have little interest in projection because projection damages the image. Please, pray tell, enlighten us on how the 6000+ photos I've taken with my DSLR are any different than scanned films? Digital archiving is just a different mindset than film archiving. Shooting the home movies on film means I have a worst case scenario backup by default (the film). But what I'd show people are the files I've edited, graded and prepared for viewing. Those are the digital files. You'd have to be a fool to stick them on a hard drive and think you'll be able to view them in 10 or 20 years. We tell our clients who scan home movies to plan on getting two more drives, making copies and sending them around to family members. Then every 5-10 years, moving the contents of those drives onto whatever the current format is. Digital archiving means keeping the film in constant motion from format to format. If you're not doing that, you're going to lose the files. It's that simple. I would neither do, nor advise that.

None of this is new. If every film student who got a BA in filmmaking became an actual filmmaker, there would be a glut of filmmakers out there. I have fairly close contact with most of the people I went to school with 25 years ago. Few are making films. Life happens, people's interests change, etc. Of the 25 or so people in my program, I think maybe 5-6 are still involved in the film world in some capacity (two teach - one at an art school, one at a more traditional film school, one is the tech for the animation program at the school we went to, one is working here, scanning film as I type, others are making small films in their spare time). Everyone else went on to different careers, some related to film, but most not. This is not unique to film. (though I have to say that the fact that 25 years on, at least 20% of my graduating class is still working in the field is pretty impressive, all things considered).

We scan film for students almost every day of the week. There are three or four colleges with film programs that still shoot film around here. Some are annoyed that they have to use it for classes. Most are totally into it, and once they've shot film they come back and shoot more, either for their subsequent classes (where it's not required), or on their own. Several of them have graduated and moved away but still send us their film for processing. Your blanket statements about Super 8 are tiring, as is your dismissal of the format. It's not for you. So don't shoot it, but stop disparaging it because it doesn't fit your aesthetic and it's not what you would use. You are in LA, which is going to have a different mindset than, say, Boston, NY, Austin, San Francisco, Portland, cities like that. LA is the center of the US commercial film industry, and your approach reflects that. Your priorities are your priorities, but they're not shared by many other filmmakers. We do a *ton* of scanning work for artists, documentary filmmakers, students, wedding filmmakers, bands, etc. and the volume is about equal: Super 8 and 16mm. We're even seeing a (tiny) resurgence in regular 8mm lately. Yes, the market is measurably smaller than it used to be. Of course it is. Nobody is arguing that. But so what? It's still big enough to sustain a lot of companies like ours, and the point here is that as Kodak comes online with their own processing service, they will be incentivized to promote the format more, which only helps everyone else.

To be clear, I'm not saying it's not more expensive than it used to be. (20 years ago I was paying about $12 to have 100' of 16mm processed. those days are long gone) but that price increase has (only a) little to do with the price of silver, it's econ 101: supply and demand. There simply isn't the kind of demand for Super 8 that there used to be, when it was artists/filmmakers AND the home movie shooting masses buying the stuff. Kodak isn't making it in the quantities they used to, so the price goes up because it costs more to make. Now the S8 market just a small percentage of that old market, and the vast majority of the old customers will never come back because, well, why would they? I shoot my home movies on film, but that's because I want them to last. I also shoot a lot on my iPhone because it's in my pocket all day long and the video looks pretty damned good in good lighting. My point above is that while the market for S8 has shrunk dramatically in the past 25 years, it isn't as small as Tyler seems to be suggesting, and it's not limited to a(s) tiny (a) niche (as it once was). Case in point: we just recently scanned to 4k the S8 film for an upcoming feature. I think there's a lot of room for growth, and adding more labs and Kodak's backing (assuming they don't do what they do best and screw it up), should only help that market to grow.

I don't think kodak is looking to "reign (sic) anything in." I think they're doing what any business would do - seeing and exploiting a market opportunity. That said, Processing costs are not the problem with S8, in terms of economics. Let's break this down: There are 3 labs I can think of that will process S8 Neg in the US: Cinelab, Spectra, Pro8mm (I don't think Dwaynes does neg). Only one of those labs is "exorbitant" in its pricing. Cinelab: $18/roll Spectra: $17/roll Pro8mm: $20/roll The cost of raw film stock is about $24 if you buy it from Kodak and ask nicely for a discount (student or otherwise). The cost of a scan is roughly the same. So for Kodak to make this viable, they're going to have to do high quality scans, and they're going to have to do it for under the current low price of about $65 if you send it around yourself (assuming you're sending film from lab to lab relatively cheaply, such as Priority Mail rather than FedEx or another Express service). Given that Pro8mm, which is the defacto standard most people base pricing on, charges $98 for a film/process/2k package, my guess is that Kodak will come in somewhere around $65-$75, in order to make it competitive and still make a profit. I don't actually think what Kodak is doing is going to have much of an effect on most labs, all of which do other gauges as well as offering a variety of post production services. It's in Kodak's best interest to keep local labs in business, to make it easier for people to get into film, quite the opposite of what you're suggesting is their motive.

There are very good reasons to shoot at higher resolutions even if you're delivering at HD - oversampling, not having to scale up for future use on higher resolution displays, quality of available cameras, etc. And unless you've been under a rock, well, here: https://www.amazon.com/Samsung-UN40KU6300-40-Inch-Ultra-Smart/dp/B01DUTL4OI/ref=sr_1_1?s=tv&ie=UTF8&qid=1481570064&sr=1-1&keywords=4k+television That'd be a sub-$500 4k television. In a couple years HD will be the new SD.

Not always. Dismissing "art" film as less relevant is something that doesn't do anyone any good. And that's the sense I get from what I can understand of your argument. The point here is that all films are used for a "look" whether you think of it that way or not. Whether it's super 8, 16, 35mm, is beside the point. There are a lot of people shooting S8 for a look, and a lot shooting it because they like it (the format). That's great. As I said, anecdata.

Actually they weren't. I went to a 70mm screening last year here in Boston, and the print we watched was one of the ones from the 90s. I was disappointed because I was led to believe that it would be the new print. Instead it was the same one I saw in the 90's. Some of the reels were good, some were scratched quite a bit, and the presentation at the Somerville Theater (not my favorit) was sub-par. The screen is too damned small for 70mm, but they're one of the few places in town that regularly presents in this format. Also, the mag audio track had an annoying repetitious click throughout a good chunk of the movie, from someone who screwed it up at some point in the past 20 years. But prints from the 90s exist and are in active circulation.