Edgar Nyari

Hi everyone,

I did some tests with a stroboscope light for an upcoming shoot, and found it extremely difficult to replicate a strobing light effect on film due to shutter and stroboscope sync issues. The perfect solution would be to use some sort of camera-strobe synchronization, which is out of the question. The camera is a 35mm film camera, and it's impossible to sync it in any way with the device I'm using.

Is there a set of "magic" strobe frequencies that work best with 180 degree shutter on 24 fps, to give a nice flickering effect without too much skipping?

Considering this is my only light source in this scene, and total darkness should be between flashes, should I perhaps shoot the scene in full light, then just edit in blank frames at any desired frequency?

Is this ever done in editing?

thanks

Edgar

Arri 35-III MOS camera.

Here is some Kodacolor 200, developed in the local photo minilab, and scanned on a really cheap flatbed. Hand-registered. Didn't use sharpening because of the noise. Makes for a really interesting hobby. :)

LOL...I'm getting to where I'm having my doubts too and sorry about your film. Sounds like people here aren't biased about TVs and I'll get the real scoop...you'd be surprised at what goes on in the TV Audio Video forums.

 

Basically I'm thinking that Movie Theaters set the standard for how a movie should look...I mean it's a movie that Directors agreed to...generally speaking and I'd like to adjust my TV settings to get close to what's showing in a Theater...assuming the Theater's setup properly. I normally leave the settings at the default movie/cinema mode. The default backlight is close to 20 which is max...looks nice and sharp...very crisp with a lot of detail and on the bright side of things.

 

There's a websight that has my TV calibration settings with the backlight set at 5. It's a good bit darker and not as sharp and doesn't show as much detail but it does look exactly like the movies I've seen in Theaters like Black Panther, Ready Player One and Pacific Rim Uprising. I went to these movies only to see how they looked...to get an idea of how they look in Theaters.

 

I don't mind the darker and less sharp and less detail if that's how it's supposed to look which is about what I'm seeing in Theaters. So does the lower backlight setting sound about right as far as picture quality goes? I'm judging how movies should look on TV by how they look in Theaters...generally speaking and would like to hear what others think and feel. Thanks.

I would agree that theater projection is the closest you can get to the filmmaker's vision, without actually being involved in some way in the filmmaking process. There are variations there too of course, but as far as I know DCP projectors don't use crazy "enhancement" filters. I think the backlight is only an issue if the TV has a weak black, in which case setting the backlight too high would make the blacks milky. Even in theaters a high contrast ratio is a "good thing". This was the debate back in the days when digital projection was first taking over. Many argued that the early projectors didn't have the contrast ratio that film projection does, but now that problem is solved. The dynamic properties of the display device, be it a PC/MAC monitor, TV or movie projector, are the lesser part of the problem IMO. Image "enhancement" algorithms are the real enemy here. I would just figure out how to turn EVERYTHING OFF, even "cinema/movie" profiles, and simply force the device to display the material without any image processing whatsoever (if possible) :)

I don't know much about current TV technology, but I had a pretty unpleasant experience once, when I attempted to show someone my short film at their home. I prepared a nice 1080P file and played it from an USB stick. To my horror I realized that their TV use some sort of motion interpolation at work, making my 24 fps footage look like 60 fps. There was also some sort fo filter on that brightened everything, and sharpened all the film grain in some scenes, while completely filtering it out in others. It all looked like noisy video footage in ultra-sharp high definition. I don't really trust TVs.

I once had a bad roll that flickered due to uneven x-ray damage. I've also seen footage flickering caused by a faulty camera that couldn't maintain a steady speed.

As most of us here, I primarily shoot with digital cameras. Occasionally I get a chance to shoot a project on film.

 

One thing I still love about film is the traditional film print. The quality of the image is intimate and naturalistic. There is also a distinct quality to the light on the screen, a kind of "3D effect", highlights glow and the blacks are deeper than most digital projectors (with the exception of laser). Colors can be truly gorgeous. For instance, I watched "The Love Witch" and I loved how colors popped off the screen.

 

I know how to achieve this with traditional printing (e.g. high printer lights, exposing a dense negative, costume design etc.)

 

However, in most cases, even when I get to shoot on film, I won't be finishising on film. The reasons are obvious: cost, titles, VFX, color grading limitations, most festivals accept only DCP's, general audience can't tell the difference, it doesn't sell more tickets or make the story better, etc.

 

So if I shoot on film but finish digitally, what are some ways to emulate the look of a traditional print?

 

One idea I have, but never tried, is to print from the camera negative, then digitally scan that. In other words, scan the print instead of the negative.

I'd love to see those results.

Tarantino did something similar on Death Proof, at least for the first segment. The only difference is that they probably had to copy the prints back to negative stock, in order to intercut with the rest of the film. When you look at the first segment on bluray, I'm not sure how many generations is that removed from the original print (you might be looking at a telecined duplicate negative of the print, or maybe even a positive made from the duplicate negative), but it has an interesting hyper-analog look to it. Contrary to what stories might circulate in the media, all the damage and splices are REAL; they are not a digital effect.

I have been thinking about doing the same thing that you propose. The only problem for me is cost.

This is a really interesting topic. It seems to me that you can probably draw a parallel between various generations of technology they used, rather than actual emulsions. For example, when T-grain came out, it was probably use in the Kodacolor emulsions of the early 90s. Some of the advances in "Vision" stocks were probably also reflected in the Gold films of 2000s, and also Portra and Supra. Then came the two-electron sensitization inovation of Vision2. Some time after that Kodak came out with a brand new generation of Portra films and Ultra Color. I suspect they intergrated some of the advancements of V2 in those new still emulsions.

As far as the general "flavour" goes of each emulsion, I think they are quite different. Portra has an odd color palette, which renders skintones in a certain pleasing way, but looks a bit "thin" in the exteriors. Ultra Color is on the other end of the spectrum, and also has no parallel in MP world, except maybe Eterna Vivid. Gold was I guess "normal", but still more contrasty than a MP emulsion I think.

That all being said, I plan to play a bit with still films in my movie camera soon. I ran some tests last night (coincidentally), to see how still film behaves in a motion picture camera. I made some measurements, and concluded that I can get about two and a half seconds from a roll of 36 exposure still film. It could make for an interesting (and cheap) hobby. I suspect there might be some problems like pressure plate reflections due to the lack of rem-jet, but we'll see.

I do remember that someone used Kodak Gold 1600 in the early 90s for some nature documentary footage shot in low light. This is the only instance that I know of where someone used still film in a motion picture camera.

This is great! Thanks. As a user of short-ends, I wish there were more tests like this.

Is there any reliable information as to when does x-ray damage happen? I mean film is shipped on a regular basis everywhere. If you buy fresh stock from various companies, they often ship it to you by mail. So what happened to that roll of 800ASA?

I'm not really sure if it is allowed to advertise crowdfunding campaigns on this forum, so I won't post any links yet. Before II do that, can some of the mods please comment on this issue? And if it is allowed, which forum section would be best suitable?

It's about a short film called Instar which I directed and DP'd a couple of weeks ago shot on a multitude of formats: mostly standard 35mm, with segments in 16mm negative and even some 16mm reversal. Film itself as a medium is part of the story, so different looks and formats will play into it.

The film is currently in the stage of postproduction and since it's self-financed by me and a couple of friends, we are running an Indiegogo campaign to help cover the costs of scanning. The negatives are still at the postproduction house awaiting the end of their holidays, so I don't have any stills or footage yet.

If you get about 3-4 MP with a still camera, then it must be a low quality lens or a low quality scanner in question.

 

It happens badly on my zoom lens.. I would never shoot at f22... its the law,s of physics .. but i guess if you didnt see it in the view finder should be ok..

It's hard to see anything at f22 in my groundglass even with the sun blasting on a summer noon. All I see is groundglass grain and a dark bronish image :)

With color negative film you have a lot of overexposure latitude -- someone once went out to shoot a second unit shot for me on a low-budget feature and he accidentally used 500T when he thought he was using 50D, so the footage was a bit over 3-stops overexposed and blue-ish as well. This was a project that was answer printed and even though the printer lights were near the max at 50 points, the final image was more or less correct, just a bit hazy.

 

With b&w films you want to expose more carefully...

Yes that's what I figured, that it can tolerate two stops without any significant problems. The highlights can be a bit compressed though, which I don't mind. I'm more worried about that small aperture. I have never shot anything on f22, and I did hear about the diffraction issue, but I have never seen to what extend that happens.

It's funny that this thread is on top today, as I had to deal with this exact problem today.

I was shooting some "B-roll "material, which featured some railroad scenery in bright sunlight, on Eterna 500T.

It just happened to be the sunniest day in the whole year (21st of June), and I just happened to NOT have any sort of ND filter at the moment,

so I just shot it at f22 and hoped for the best. I caught a couple of moments when the sunlight got a bit hazy, so the light came down to about f32 (usually it reads f45...),

but that was metering for the sun and not shadows, so I should be fine.

I think I've read that some restorers combine a scan of the original negative, which has color fading problems but is the sharpest and finest-grained source, with scans of the b&w YCM separations made for archiving, which have the original color information if done correctly but often have grain and contrast problems. Not sure how they combine those elements digitally.

Maybe they use something like a YUV color system, where they can use the luminance from a finer grained and less-contrasty negative and combine it with color information (U/V) from the YCM separations, then convert back to RGB?

If it was down to "Eastmancolor" then all films from that era would look yellow. No matter what it says: "color by technicolor", "color by deluxe", metrocolor etc. or explicitly color by eastmancolor, it's always eastmancolor stock, unless specifically identified as something else (like gevaert).

I forgot that the earlier stocks were all tungsten,yes. It's only with 5245 where daylight stock became the norm for daylight exteriors.

A correctly exposed daylight color negative film should in daylight always produce natural and neutral results. I don't think there was ever a color negative product created for motion pictures that was intentionally designed to make everything look yellow, or anything other than "normal". The only intentional looks were a slightly desaturated softer look of Kodak "expression" stock in the 2000s and Eterna Vivid which was made to look a bit more saturated. But all eastmancolor products from the 50s till today were designed to look "normal".

And most films from that time (in both Europe and US), were shot on Eastmancolor anyway. There did exist alternatives (Ferrania color up to about mid 60s, AGFA Gevaert, some Russian stocks, Fujicolor) but Hollywood and most of European productions almost exclusively used Kodak stock (Eastmancolor).

What I suspect is that some people remember prints to look a bit warmer, and commented on how the Bluray looks more neutral than original prints. But like Mark said, it's a matter of color grading.

To be fair, the Harryhausen thing does look like a bit of slightly clunky stop-motion, which is just as fake as the CGI.

 

Neither is ideal, and they both look like the British prime minister.

 

P

I still find stop motion (or animatronics, or masks for that matter) more scairy than CG. The reason for that is because my brain registers it as a real object. Sure there is something "wrong" with it; either it's jerky or not quite organic enough etc. but it's a physical object. With most of the CG, my brain is fighting to decide whether it's a physical object or not. There are exceptions of course.

If the filmmakers were going to do such an extreme creative grade, it's sensible to assume that is going to happen in the DI suite from the film scan, not in a telecine transfer from the IP for home video release. Don't you think?

Yes it is :D I was just reacting to his intention to immediately submit an edit to the imdb page based on just visual observations.

Well the second one does look a bit contrasty but you can't just assume it's DI because of that. It could just be that the transfer to video looks like that because the director wanted it that way.

Actually, now I'm really curious what you guys think about color grading.

 

Is it an art in its own right? Or is it best seen as a completion of the cinematographer's work? Should there be a 'Best Color Grade' Oscar?

 

 

That's a good point. There probably should IMO, because it came a long way from setting 3 values for printer lights. On the other hand I prefer films where color is manipulated in front of the camera using light, art direction and production design.

I thought your argument was that you could see the inherent look of the original camera film stock by watching a Blu-Ray?

Not in any absolute way. I feel there is a common denominator that comes through which I can view in a large number of materials shot with a single stock, transfered and graded differently and say: Ok I see where this is going. Like for example every single film shot on 5294 I've seen has this certain rough feel to it compared to every single film shot on Vision stocks. But that judgement is nowhere near that specific as it might have sounded to you when I first stated my point.

Right, but again, how much of the 'look' is actually the lighting, production design, wardrobe, locations, lensing, filtering, lab work, optical work, etc.? Not to mention, your particular display device, the digital workflow chain, the type of scanner used, choices made in terms of sharpness enhancement, noise reduction, and compression.

 

It's all so mixed together that there's no way to say for sure that you 'can see the look of the camera original filmstock' from a Blu-Ray. If you're not watching the film on a calibrated P3 monitor in a dark room, then what you're seeing is likely already quite different from what the colorist saw in their suite.

 

The closest you can probably get to that would be to compare an original print off the o-neg to a new print struck from the the same source, off the same projector back-to-back. But of course, you have to take into account fading and damage to the film, and the difference in available print stocks. The print stock was always part of the look.

 

Also realize that films which intended to change the look of the o-neg through post techniques like different print stocks, bleach-bypass at various print stages and various optical tricks won't look the same when the original negative is scanned because those elements of the look were added later in the chain, and would have to be re-created digitally.

 

For example, on 'Se7en', Dariusz Khondji intended to use ENR Technicolor process on all the print stocks, which basically desaturates the colors, adds grain, and crushes the blacks by leaving some silver in the print. It also makes the image darker. So he also used a Panaflasher and lots of fill light on set to make sure shadow detail didn't completely disappear.

 

When scanning the o-neg, the filmmakers added that some of that contrast back in, but instead of just desaturating they went for more of a modern color pallete. Consequently, there are more 'Fincher greens', cold blues, dirty yellows, and vivid reds than there were in the film originally.

 

So - when you watch a Blu-Ray, how can you sure that the green of Kim Novak's dress in 'Vertigo' is the product of the camera film stock and not one of one of these other processes? Honestly, I don't think you can. If you watched a projected print side-by-side with the Blu-Ray, that would be the best you could do. But how many of us have the chance to do that?

 

The bottom line for me is, if you want to appreciate film and 'the film look', watch projected prints at revival theaters. Digital display is something else entirely. Great for enjoying movies in your own home. But it's a fundamentally different experience from watching actual film projection...

I don't see anything here I'd disagree with. I'm not sure I was trying to assert anything other than one can see scans of original negatives on bluray; that you aren't necessarily seeing a contact or optical copy of these materials (like an IP). And yes, you can scan a negative on many scanners and grade it differently in many ways; but some qualities of the film image always come through except when some sort of heavy processing is used for whatever reason.

If this is 'how the negative looks', then why does every new transfer look different in color and contrast from the previous one?

 

You could also point to examples of old (and not so old) films that have been newly rescanned and digitally graded to look completely different than the answer prints made for the original theatrical release. 'Star Wars', 'Blade Runner' and 'Se7en', to name just a few.

These are stylistic differences that go in the domain of grading. When I use the word "look", I use that word to mean; underlying image characteristics of a certain medium. So when I see the old and new edition of Graduate on bluray; what I'd say is: the transfer looks very different and the color grading is very much different; but I can still see the "look" of old filmstock (and lenses) in both versions.