Jump to content


Photo

Super 16mm versus 35mm


  • Please log in to reply
10 replies to this topic

#1 bragis chut

bragis chut
  • Basic Members
  • PipPip
  • 46 posts
  • Director

Posted 19 June 2006 - 04:14 PM

Hi all:

I've been pricing out the difference in cost between shooting my short on super16mm versus 35mm. Obviously, 35mm is more expensive film stock and processing but if I shoot super 16mm, in order to get a decent blowup to 35mm I've been told that you really have to do an HD transfer which is VERY expensive (about 400$ an hour). Further costs involve outputting from HD back to 35mm. Whereas if I simply shoot on 35mm and do a standard definition transfer to digibeta, I would actually save money in the long run.

I know that my final format will be A) a film print, and B) a DVD. Any advice from the pros? I don't think the HD transfer is hugely necessary for me, is it? The only part of this that still confuses me is that I have quite a few FX shots and compositing shots. Does this mean that I will still need to do an HD transfer regardless of whether I shoot super 16mm or 35 to give my FX team?

Any help would be greatly appreciated.

Thanks!

--Bragi
  • 0

#2 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 19 June 2006 - 05:02 PM

It sounds like you're going to have transfer at least the effects material to some digital format, so that expense becomes the minimum criteria. Budget out the rest of the film assuming you have to either do a film-out of the completed effects shots (and make a subsequent contact print of 35 or optical blow-up of 16), or transfer the rest of the film to a digital format like HD.

The first thing I would do to save yourself a lot of grief is to outline your priorities. Are you trying to get your film done for the cheapest price? Are you trying to get the best quality? Is a 35mm print mandatory, or is that negotiable? If you can't pick only two main criteria and be willing to bend on the others, you'll just drive yourself nuts with positives and negatives (pun intended) :P

Remember that you can also do an HD "conform" of selects after you've created an edit list from a cheaper SD "dailies" transfer (you don't have to transfer all your footage to HD). Don't forget post costs (renting decks or time at facilities) for all the workflows in consideration.

There are several pieces to this puzzle that will swing the costs this way and that, depending on the length of your program. So it's really tough to advise exactly the best process without knowing the details and your priorities.

Just offhand I'd say the two biggest points of negotiation here are:
1) Are the effects shots absolutely necessary, and how many of them are there?
2) Do you absolutely need a 35mm print, or will an HD master be good enough for the forseeable future?
  • 0

#3 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 19 June 2006 - 05:21 PM

Shooting ratio is also a factor in this.

Forgetting about the efx work for now, generally I'd say that the cheaper route towards a single 35mm print is to shoot in 4-perf 35mm for contact printing in a standard sound release format (spherical for standard 1.85 projection or with anamorphic lens for scope projection) -- forget 3-perf, forget Super-35, etc.

Shoot a low shooting ratio, edit the movie in a way that allows straight cuts combined with standard-length lab A-B roll effects (fades and dissolves.) Anything that makes your post as simple as possible, as long as you don't spend too much on the camera, stock, processing, and telecine.

But as the volume of footage rises, then Super-16 becomes cheaper to shoot IF you do an old-fashioned neg cut and blow-up using an optical printer and IP/IN dupes. If you have to do the transfer to 35mm using a D.I., then the shooting ratio has to be high enough that the savings over 35mm pays for the D.I. costs.

But if your film has A LOT of visual effects, it would be probably simpler (and thus maybe cheaper ultimately) to treat it all digitally and then record it out to 35mm. In that case, shooting in HD, doing the efx in HD, and recording out to 35mm may be the cheapest route, followed by Super-16-to-HD, followed by 35mm-to-HD. However, if the amount of efx is small, it may be cheaper to just shoot in 35mm (again, assuming a low shooting ratio) do a traditional film finish, cutting into the conformed negative the efx shots that were recorded out to a 35mm I.N.

If you do a traditional film post to create a 35mm print, transferring the footage cheaply to standard def video only for NLE purposes to generate an EDL, you do have to factor in additional costs later of making a color-timed IP after answer printing is over, and then doing a telecine transfer of that to whatever format you want.

In other words, you have a lot of careful budgeting work to do to answer this question...
  • 0

#4 Laurent Andrieux

Laurent Andrieux
  • Sustaining Members
  • 1527 posts
  • Cinematographer
  • France

Posted 19 June 2006 - 05:47 PM

A few questions here come to my mind... Though I shot a lot in both super 16 and 35 mm, I never went the SD/HD transfer route on a neg to pos process...

"I've been told that you really have to do an HD transfer."

That's how it's said when someone can't just say : "you need a HD transfer"...

1) isn't digibeta lower in quality than Super 16 ?... So that if a 35 mm transfer to digibeta is good, why a S16 shouldn't be ?

A print from a digibeta is really of a poorer quality than either a contact print (35), an optical print (S16) or a digital export from a computer straight on a negative. I'm talking of a theater release, here.

Then I guess this would be for the DVD release... So what's the use for HD, digibeta is good enough for a DVD, unless you plan to release an HD-DVD before anybodys' got such a deck...

2) What's the use of such tape transfers for the positive print route ? I've always seen the computer export straigth to a 35 mm negative, whatever your source was S16 or 35 mm... Is it that your post house can't do that ? 'm not a post prod man, but I don't remeber I've seen the need of an intermediate tape... Did I miss something ?

3) If your print is for a theatrical release, even an HD transfer (if not 4K) of a 35 mm neg might be under the quality of a blown S16 to 35 whatever route... Well, ok different but certainly not better... The fact you pass by an HD transfer at one point - unless 4 K - might mess it all up.The only reason I'd see is a important use for SFX...

I'd ask for a demonstration of the different solutions. They must have that in your lab... I'd go S16 export straight from the computer on a 35 mm neg, no need for HD, for the theatrical release, and have a export on a digibeta tape for the SD DVD, even if shot Super 8 it would be sufficient. What would they do before HD exists ? That was not so long ago... right, this is why they need money back on their HD decks, by now ?...
  • 0

#5 Hal Smith

Hal Smith
  • Sustaining Members
  • 2280 posts
  • Cinematographer
  • OKC area

Posted 19 June 2006 - 05:56 PM

Shooting ratio is also a factor in this.

Forgetting about the efx work for now, generally I'd say that the cheaper route towards a single 35mm print is to shoot in 4-perf 35mm for contact printing in a standard sound release format (spherical for standard 1.85 projection or with anamorphic lens for scope projection) -- forget 3-perf, forget Super-35, etc.

In other words, you have a lot of careful budgeting work to do to answer this question...

Assuming one has a 35mm projector (which I have), what do you think of a workflow where one shoots negative film, gets an one-light print (dailies), projects the one-light while taping with a miniDV camera, edits the miniDV in an NLE, generates an EDL, pays a professional to cut the negative, then gets an answer print? Sound's another question but not for this forum.
  • 0

#6 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 19 June 2006 - 06:05 PM

I'd go S16 export straight from the computer on a 35 mm neg, no need for HD, for the theatrical release, and have a export on a digibeta tape for the SD DVD, even if shot Super 8 it would be sufficient. What would they do before HD exists ? That was not so long ago... right, this is why they need money back on their HD decks, by now ?...


The Super-16 image has to exist in some digital format in this scenario, so if you're not saving it onto an HD tape, you're storing it in a computer as 2K RGB data probably.

So you're talking about a D.I. basically... but what about editing and color-correction? Are you suggesting scanning all the camera rolls at 2K, storing the data at the lab, and then having them make a downconversion to standard-def tape for offline editing?

That could be more expensive at some facilities here in the U.S. than a transfer to HD, just depends.

Or it may be cheaper to just transfer originally to standard def video, edit offline, generate an EDL, and then go back and transfer selects to 2K data, do an auto-conform session, etc. Basically a 2K D.I. instead of an HD D.I. Just depends on whether the facility likes to park a lot of data in their computers, or copy the data to computer tapes like LTO's, or store it as HD on tapes. And that also depends on the workflow -- if the scanning happens first to camera originals before the editing process, then they will want to shift that data to something that can be shelved cheaply, either computer tapes or HD tapes.


Assuming one has a 35mm projector (which I have), what do you think of a workflow where one shoots negative film, gets an one-light print, projects the one-light while taping with a miniDV camera, edits the miniDV in an NLE, generates an EDL, pays a professional to cut the negative, then gets an answer print? Sound's another question but not for this forum.


I doubt videotaping a projected workprint image is going to have the resolution to see the edgecode / keycode information necessary to generate a frame-accurate EDL... you have to be able to correspond timecode on a tape to edgecode / keycode on the negative. Plus your videotaped image won't have video frames that correspond to film frames in any sort of regular pattern, unlike a telecine transfer with a 3:2 pulldown to convert 24 frames to 60 fields.
  • 0

#7 Laurent Andrieux

Laurent Andrieux
  • Sustaining Members
  • 1527 posts
  • Cinematographer
  • France

Posted 19 June 2006 - 06:14 PM

So you're talking about a D.I. basically... but what about editing and color-correction? Are you suggesting scanning all the camera rolls at 2K, storing the data at the lab, and then having them make a downconversion to standard-def tape for offline editing?

That could be more expensive at some facilities here in the U.S. than a transfer to HD, just depends.

Or it may be cheaper to just transfer originally to standard def video, edit offline, generate an EDL, and then go back and transfer selects to 2K data, do an auto-conform session, etc. Basically a 2K D.I. instead of an HD D.I. Just depends on whether the facility likes to park a lot of data in their computers


Yes exactly. But I'm sure I've seen such a route allowing editing and timing and SFX with a Flame. On super 16 digitally blown to 35.

This is a short movie, not a feature length.

I think the 2K DI route without any tape on the way would be the best... unless one has access to a 4 K HD decks... May be the case, but as you say, it depens on how the manage their space.

I think the DVD only may need a tape intermediate, but SD should be enough, then.

And... if there is a tape intermediate for the theater release, HD will always be better tha SD... and native 35 mm better than S16...
  • 0

#8 bragis chut

bragis chut
  • Basic Members
  • PipPip
  • 46 posts
  • Director

Posted 19 June 2006 - 06:24 PM

The Super-16 image has to exist in some digital format in this scenario, so if you're not saving it onto an HD tape, you're storing it in a computer as 2K RGB data probably.

So you're talking about a D.I. basically... but what about editing and color-correction? Are you suggesting scanning all the camera rolls at 2K, storing the data at the lab, and then having them make a downconversion to standard-def tape for offline editing?

That could be more expensive at some facilities here in the U.S. than a transfer to HD, just depends.

Or it may be cheaper to just transfer originally to standard def video, edit offline, generate an EDL, and then go back and transfer selects to 2K data, do an auto-conform session, etc. Basically a 2K D.I. instead of an HD D.I. Just depends on whether the facility likes to park a lot of data in their computers, or copy the data to computer tapes like LTO's, or store it as HD on tapes. And that also depends on the workflow -- if the scanning happens first to camera originals before the editing process, then they will want to shift that data to something that can be shelved cheaply, either computer tapes or HD tapes.
I doubt videotaping a projected workprint image is going to have the resolution to see the edgecode / keycode information necessary to generate a frame-accurate EDL... you have to be able to correspond timecode on a tape to edgecode / keycode on the negative. Plus your videotaped image won't have video frames that correspond to film frames in any sort of regular pattern, unlike a telecine transfer with a 3:2 pulldown to convert 24 frames to 60 fields.


Wow... a lot of information to digest here. Thanks for all the feedback. The effects load is not insignificant. There are three sky replacements, and a stage shoot which involves quite a bit of greenscreen. However, I really haven't been very impressed with the look of HD imagery and this project certainly require a film look. That said, the other factor is that my lead is a little girl (8 years old) and I worry that this may take a toll on my shooting ration. Currently, I've budgeted approximately 7,000ft of Super16mm or the quivalent in 35mm (approx.17500) which gives me a shooting rationm of about 15:1. A little less actually, as there will be some high-speed cinematography.

The other factor is that we are looking for a deal on a camera package and this seems more likely with a super16mm package as opposed to 35mm. From what I've seen online, 35mm packages generally rent for quite a bit more than super16mm.

However, if we shoot super16mm, and do an SD transfer and create an EDL and then go back and transfer selects to HD, I will incure the cost of a negative cutter which may be roughly equal to what it would have been to transfer the entire batch of footage to HD in the first place. And then I still have the transfer costs from HD back to film for my 35mm print to look forward to... ARGHHHH!

-- PS -- David -- Thanks for the DP referral. I'm expecting a reel either today or tomorrow.
  • 0

#9 Michael Most

Michael Most
  • Basic Members
  • PipPipPipPip
  • 765 posts
  • Other

Posted 19 June 2006 - 06:49 PM

The effects load is not insignificant. There are three sky replacements, and a stage shoot which involves quite a bit of greenscreen. .........However, if we shoot super16mm, and do an SD transfer and create an EDL and then go back and transfer selects to HD, I will incure the cost of a negative cutter which may be roughly equal to what it would have been to transfer the entire batch of footage to HD in the first place. And then I still have the transfer costs from HD back to film for my 35mm print to look forward to... ARGHHHH!


First, regarding the visual effects. 16mm is not an ideal format for green screen work, for a number of reasons. For that, you'd be much better off with 35mm, or, quite frankly, HD video if it's a good flavor of it. You won't need a negative cutter for retransfers, as they can be cued from the original camera rolls by keykode. As for the film recording, that's a subject for more discussion. If you're thinking about an electronic path back to a film print (aka, a DI of some sort), as it appears you are, you might consider mixing different formats, as this is something that a DI allows you to do, with little cost penalty (albeit with a slight bit of complication in post production, but on a short, that won't be much of an issue) - say, 16mm for the bulk of the narrative work and the overcranked scenes, and either 35mm or HD video for the green screen work.

Where are you located? If you contact me off list, I might be able to help you with this project.
  • 0

#10 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 19 June 2006 - 07:32 PM

However, if we shoot super16mm, and do an SD transfer and create an EDL and then go back and transfer selects to HD, I will incure the cost of a negative cutter which may be roughly equal to what it would have been to transfer the entire batch of footage to HD in the first place.


If you transfer selects from the camera rolls using the EDL as a guide, you aren't cutting the negative -- you're basically going to have to assemble the selects into a edited digital master, more like an online session. So there are no neg cutting costs, just the costs of the digital conforming to get the select scans to match the offline edited version.

I agree with Mike that you should consider doing the efx elements in 35mm or HD, even if the rest of the project is in Super-16. You can always add a grain effect to the finished composite to match the Super-16 footage. But it depends on the effect -- if you are just painting over a shot, like doing sky replacements, then you probably can deal with a stabilized Super-16 frame. But for layering chromakey elements and background plates, you may be better off starting out with a finer-grained, sharper, steadier image.
  • 0

#11 Hal Smith

Hal Smith
  • Sustaining Members
  • 2280 posts
  • Cinematographer
  • OKC area

Posted 19 June 2006 - 09:10 PM

I doubt videotaping a projected workprint image is going to have the resolution to see the edgecode / keycode information necessary to generate a frame-accurate EDL... you have to be able to correspond timecode on a tape to edgecode / keycode on the negative.


No/bad keycode! That definitely shoots the idea down. Thank you - I didn't want to hear that answer but I needed to.

Hal
  • 0


rebotnix Technologies

Technodolly

Glidecam

Paralinx LLC

Tai Audio

The Slider

Broadcast Solutions Inc

CineTape

Media Blackout - Custom Cables and AKS

Ritter Battery

Visual Products

CineLab

FJS International, LLC

Abel Cine

Gamma Ray Digital Inc

Opal

Aerial Filmworks

Willys Widgets

Wooden Camera

Metropolis Post

Rig Wheels Passport

Opal

CineTape

Willys Widgets

FJS International, LLC

rebotnix Technologies

Broadcast Solutions Inc

Visual Products

Aerial Filmworks

Media Blackout - Custom Cables and AKS

Tai Audio

Metropolis Post

Gamma Ray Digital Inc

Ritter Battery

Glidecam

CineLab

Abel Cine

Wooden Camera

Technodolly

Paralinx LLC

Rig Wheels Passport

The Slider