Jump to content


Photo

Telecine Ratio for "Selects?"


  • Please log in to reply
16 replies to this topic

#1 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 17 May 2007 - 01:06 PM

Telecine Time Ratio for ?Selects??

I am trying to budget for post production telecine, probably 16mm to HD, but I am getting some rather confusing information from film labs. I am doing a low budget feature.

My original idea was to transfer to SD, do a very rough edit on my regular Final Cut Pro Mac, then use the EDL to go back and do a scene-by-scene transfer of only the footage I want to work with in HD.

I have done scene-to-scene transfers before, but never with ?selects?. Being very efficient, I was always able to make it a 3 to 1 ratio, so I was calling to find out what the ratio would be for working with selects.

I was told that doing ?selects? will be more costly then a scene-to-scene of all 15 hours of footage at once (100 min film). I am shooting at a 9:1 ratio which means that if I am as efficient as before, would be looking at 45 hours of lab time if transferring all the footage.

Even if the selects take me twice as long as my regular scene to scene (6 to 1 lab time), that only equates to 10 hours lab time, so how in the world are they saying it is going to be cheaper to transfer all the footage scene-to-scene? If this were the case, I don?t understand why anyone would even do selects (other than a small benefit of less tape used)

I understand that there are different factors that affect this process, but what have you found to be the average lab time ratio for select (supervised) transfers?

If there are any legitimate factors they may be putting into this equation (or is this absurd?), please let me know how I can eliminate this from the process.
  • 0

#2 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 17 May 2007 - 01:35 PM

Telecine Time Ratio for ?Selects??

I am trying to budget for post production telecine, probably 16mm to HD, but I am getting some rather confusing information from film labs. I am doing a low budget feature.

My original idea was to transfer to SD, do a very rough edit on my regular Final Cut Pro Mac, then use the EDL to go back and do a scene-by-scene transfer of only the footage I want to work with in HD.

I have done scene-to-scene transfers before, but never with ?selects?. Being very efficient, I was always able to make it a 3 to 1 ratio, so I was calling to find out what the ratio would be for working with selects.

I was told that doing ?selects? will be more costly then a scene-to-scene of all 15 hours of footage at once (100 min film). I am shooting at a 9:1 ratio which means that if I am as efficient as before, would be looking at 45 hours of lab time if transferring all the footage.

Even if the selects take me twice as long as my regular scene to scene (6 to 1 lab time), that only equates to 10 hours lab time, so how in the world are they saying it is going to be cheaper to transfer all the footage scene-to-scene? If this were the case, I don?t understand why anyone would even do selects (other than a small benefit of less tape used)

I understand that there are different factors that affect this process, but what have you found to be the average lab time ratio for select (supervised) transfers?

If there are any legitimate factors they may be putting into this equation (or is this absurd?), please let me know how I can eliminate this from the process.



We recently did this on several projects so here is my take on this:

Transfer all of your film to Dvcam (beta, Dvcpro, etc.) one tape clean and one tape with keycode, timecode, Foot+Frame numbers, lab roll, etc. burnt in and make your edit in FCP, Avid, etc. The rates for a SD keycode transfer should be around $0.14 -$0.16 per foot. The finish edit will have the lab roll (flat) with foot and frame and keycode numbers so you organize the lab rolls and figure the shots you need based on Keycode and foot + Frame numbers and go through the rolls in a "finish" telecine session with the TK house running their keycode reader and transfer just your selects with maybe 5 frame "handles" for each shot. I would figure around $800 per hour for the supervised HD session YMMV.

We did this for Chris Burke (who posts here) we did his SD keycode transfer and National in Boston (www.nationalboston.com) did a HD transfer to disk from his selects. This process could be considred a didital negative cut and is standard industry practice and certainly should be less expensive than doing a scene to scene on all your raw footage.

Check out Pixelharvest.com in LA too they do 2K pin registered scans (selects) we did a 35mm job with them a few months back (they do 16mm too) but these are ungraded so that may push it out of your price range...

-Rob-
  • 0

#3 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 17 May 2007 - 02:23 PM

I don't fully understand all the details in this post, but it does not seem to answer the central questions which are 1) What ratio should be average for the time it takes to transfer selects and 2) What more would I need to do (if anything) to use the lab time more efficiently (I think they understood that my SD edit was going to use Keycode, etc).
  • 0

#4 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 17 May 2007 - 02:30 PM

I don't fully understand all the details in this post ("ungraded", etc), but it does not seem to answer the central questions which are 1.) What ratio should be average for the time it takes to transfer selects? & 2.) Am I missing something about the factors time it takes to do selects? I was told 50 hours. I don't think I should need 50 hours to transfer the less than 2 hours of footage I would need from my SD edit (I think they understood that my SD edit was going to use Keycode, etc).
  • 0

#5 Michael Collier

Michael Collier
  • Basic Members
  • PipPipPipPip
  • 1262 posts
  • Gaffer
  • Los Angeles, CA

Posted 17 May 2007 - 02:38 PM

I think the difference comes in with the supervision. If you telecine everything at once, then your only paying them to set up the machine and send it through. If you do selects, then they have to charge you at a supervised rate, regardless of if you need any scene to scene correction or not. With the labs I looked at, the numbers usually work out in favor of a onelight of everything. I was told a ratio of 3:1 is a good estimate. 3 hours for every 1 transfered.

Why would someone do selects? I suspect that is an outgrowth of the printed dailies model. If you have to print rather than telecine, the savings associated with selected prints will exceed the cost of the extra man-hours to do it. Since telecine is relativley cheap compared to printing, a full telecine is not much more expensive than selects (a full one-light is much cheaper per foot transfered than a supervised session) you might find it to be a better option than selects. Now if you plan on printing dailies then selects will reduce the amount of print footage you'll pay for and be cheaper.

Thats just a comparison based on research and experience from some of my projects. I have never printed material, just telecined and did an HD match-back from the original keycode.
  • 0

#6 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 17 May 2007 - 03:00 PM

I am not doing a one light, nor am I doing an unsupervised transfer.
If anyone can address my questions, that would be great.
  • 0

#7 Michael Collier

Michael Collier
  • Basic Members
  • PipPipPipPip
  • 1262 posts
  • Gaffer
  • Los Angeles, CA

Posted 17 May 2007 - 03:34 PM

I guess I don't fully understand your question. I reread your posts, and you want to know: 1. how much to time to budget for; and 2. what can go into making the ratio higher or lower? I'd say about 3:1, but anything can change that ratio. Its really more how many shots you have. If they only need to do a few very long takes then it will go down. If your asking them to pick out a ton of 2 second reaction shots it will go up. But in general 3:1 is what I have been told, and in my experience its about right.

You also said you don't understand how it will take 50 hours to transfer 2 hours. I don't fully understand how your going to transfer just 2 hours runtime to complete a 100min movie. from what I understand this telecine is your first (and only?) telecine on this film? How do you plan to transfer just 2 hours to make a complete film, esp. when you shot 15 hours?

I think at least I need clarification on your workflow. Your selected takes only account for 2 hours of the 15? I could only see that if you shot only masters/long takes and did multiple retakes of each setup. Otherwise if there are other camera angles you'd have to transfer certain portions of the movie at least twice (select master take, select CU) If you only have 2 hours, I would think that 6-10 hours should cover you, and I don't see how they would say 50 hours, that seems a little out there too. Maybe they were basing that off scene to scene on all 15 hours?

Here are your options I guess:

1. one light SD for EDL purposes: 15 hours (33,000ft) @ .16/foot = $5,280 +
2 hours supervised HD matchback @ 500 hr x 3 = $3000 = $8,280

2. HD supervised for first edit: 5 hours of footage (your 3:1 ratio of selects) x 3 (lab ratio) x $500/hr = $7500

3. HD telecine of everything: 15 hours footage x 3 (labtime ratio) x 500/hr = $22,500

Obviously you aren't intending to do three (I think thats what they must of assumed when they told you to budget for 50 hours, its the only thing I can think of that would take 50 hours) It sounds like you want to do #2, but want to only transfer 2 hours of footage? That doesn't seem very practicle if you haven't yet done an edit (unless have you?) I assume you would need at least one of every 3 takes. At a 15 hours, that aprox. 5 hours to transfer, and then multiply by the hours it would take to do that work (3:1) you have to pay for 15 hours at the HD rate.

Your other option not fully explored here is a onelight of the selects at the lower rate, then a matchback in HD, though you probably wouldn't see much difference in costs between that and the first two options. Those costs are assuming 16mm. Option 1 might be a bit more if you were shooting 35mm. Also keep in mind that the less you transfer initially, the more open you are to added costs. If something needs to be cut around and the editor doesn't have the shot he needs, then additional shots must be pulled to get something for him to work with. One light it all for SD edit and HD matchback and your price is more or less locked. Telecine just 5 hours in HD and you still have 10 hours of footage, any bit of which might be needed at any point, adding costs.

I guess if this doesn't answer your questions, I would ask you to explain your workflow up to this point, and how you plan to procceed. 50 hours is too much for what your talking about, they must have been confused in what you wanted.
  • 0

#8 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 17 May 2007 - 03:49 PM

I don't fully understand all the details in this post ("ungraded", etc), but it does not seem to answer the central questions which are 1.) What ratio should be average for the time it takes to transfer selects? & 2.) Am I missing something about the factors time it takes to do selects? I was told 50 hours. I don't think I should need 50 hours to transfer the less than 2 hours of footage I would need from my SD edit (I think they understood that my SD edit was going to use Keycode, etc).



A scan made on a pin registered scanner is (or should be) a full dynamic range representation of the original negative at the chosen resolution (2k, 4k, 8k, etc.), and these are made in RGB color space not YUV like hd the scans are usually individual .DPX files for each frame and the color fidelity and dynamic range of the file is much higher than a HD transfer.

The issue is that the scans are not color timed and if you just looked at them direct you might think they looked washed out and somewhat flat. The typical workflow would be to put these scans into a color finishing system like a Lustre or baselight, etc. and do all of your color timing and film assembly in a calibrated system with a calibrated projector to match what the finished film will look like on a 35mm print at 70' wide. This is a typical feature film DI workflow a transfer to a HD video format is just that a video format not data.


-Rob-
  • 0

#9 Dean Gold

Dean Gold
  • Basic Members
  • PipPip
  • 12 posts
  • Director

Posted 17 May 2007 - 04:01 PM

check out Filmmaker's Handbook by Stephen Ascher, might help

Edited by Dean Gold, 17 May 2007 - 04:04 PM.

  • 0

#10 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 17 May 2007 - 04:09 PM

I don't know how I can make this any clearer. I don't need info on pricing, only ratios.
There will be two telecines.
1) I will have a one light done on Standard Def. I will edit the film with that footage (all of the footage). That means I will have an EDL for the 100 min film plus some other footage to be safe.

2) I simply want to telecine the footage correlating to my rough edit (and then a little more to be safe). The total footage transferred for the second telecine should only be about two hours.

Edited by Devon Green, 17 May 2007 - 04:13 PM.

  • 0

#11 Brad Grimmett

Brad Grimmett
  • Sustaining Members
  • 2660 posts
  • Steadicam Operator
  • Los Angeles

Posted 17 May 2007 - 04:41 PM

Devon,
Why don't you just call the lab and ask them these questions? I understand that you're looking for tips here, but it seems like the question of price is lab specific. If you don't know why it's more expensive, just ask. Maybe you can make adjustments that will save you money based on what they tell you.
  • 0

#12 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 17 May 2007 - 04:59 PM

I don't know how I can make this any clearer. I don't need info on pricing, only ratios.
There will be two telecines.
1) I will have a one light done on Standard Def. I will edit the film with that footage (all of the footage). That means I will have an EDL for the 100 min film plus some other footage to be safe.

2) I simply want to telecine the footage correlating to my rough edit (and then a little more to be safe). The total footage transferred for the second telecine should only be about two hours.



The ratio for the HD telecine will mostly depend on how many flats of film you have and how randomly distributed the shots are. Also a highly organized edl list and an idea of where the shots are on each flat plus a list for each flat will aid tremendously in moving through the film. I do not think anybody could give you an exact ratio until the organization of the original negative is understood. I have found that the most film you can get through in a hour is 44 minutes with roll changes and basic timing this would be 1600' of 16mm or 4000' of 35mm. Figure that you have 1600' in 2 800' flats and you have 6 shots on these flats that have a total run time of 15minutes I would think that between scanning through, finding the shot, timing it and laying it down you could probably go through the full 1600' in 45 minutes maybe less maybe more.I can guarantee that the selects scan will be less than a scene to scene on all the footage.


-Rob-
  • 0

#13 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 17 May 2007 - 07:18 PM

I don't know how I can make this any clearer.

Unfortunately, these processes aren't simple, there are many, many different ways of going about the job, and the answers posted so far seem generally to be trying to show you the options that you haven't already closed off in your question.

As I understand it, your will edit from your first SD transfer, then go back with an EDL for a retransfer at HD. You simply want to transfer the required shots in HD so that you can then do a final conform in HD. (If I'm correct, you already know this - I've just restated it to check, and for the benefit of the other readers.)

In 35mm, the common method used to be to cut the negative (full takes) from the EDL, and simply retransfer everything on the select rolls - so the neg cutter was physically doing the work of sorting and locating the takes you need - in a cheaper mode than sitting at telecine doing it. That's not a good idea in 16mm, as the physical handling of the negative takes a bigger toll on the image cleanliness.

So your EDL will need to be sorted into source roll order so that you can put each lab roll of negative (flat) up on the HD telecine, and have it find the shots you want so they can be transferred. If you are doing an HD transfer they will need to be graded. In some cases you will only have one or two shots to come from a complete lab roll - so the time taken lacing up and running through the neg to find that shot will be pretty long in comparison with the actual transfer time.

Fifty hours does seem excessive - but it's not unusual for the final video mastering transfer of a high end feature to take that long - and that is transferring from a final cut version, on an IP. It's all down to how long you expect to take grading.

Are you sure the lab/post house hasn't given you a "go away" price?
  • 0

#14 David Sweetman

David Sweetman
  • Basic Members
  • PipPipPipPip
  • 757 posts
  • Student

Posted 17 May 2007 - 11:16 PM

That's not a good idea in 16mm, as the physical handling of the negative takes a bigger toll on the image cleanliness.

...how big a toll? unacceptably big? even if messiness would help the story? and even at a post house specializing in neg cutting?
  • 0

#15 Devon Green

Devon Green
  • Basic Members
  • PipPip
  • 26 posts
  • Director

Posted 18 May 2007 - 02:15 AM

Dear Dominick,

Finally some understanding! As for the "go away price" it was funny they acted like it was just normal to give me a high end feature rate for a low budget indie, a statement which doesn't really come close to the absurdity of the cost difference between my vision and theirs (20 to 1 versus 3 to 1). I called back and got the impression they wanted to keep their cards close , so that they can claim whatever time they waste with the process is normal. Maybe they thought I was some newbie who hadn't done telecine before.

That brings me to a pressing question. If I know that I can supervise the transfer and it should take no more than 4:1 and they insist it is going to be, say 10 to 1, is there anything I can do to make sure I don't get ripped off by the lab? I don't want to agree to let a lab transfer the feature if they are purposely going to draw out the time it takes. I've supervised before and it shouldn't be any more than 5 to 1 at the most.
  • 0

#16 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 18 May 2007 - 03:49 AM

...how big a toll? unacceptably big? even if messiness would help the story? and even at a post house specializing in neg cutting?

Well, ONLY at a post house specialising in neg cutting. Don't think of allowing anyone else to handle your original 16mm negative. Also the post house would know how to reassemble the material and re-log it so that your edl still had meaning.

As for messiness helping the story, I can't envisage that in any circumstances. Call me old-fashiond if you like, I won't change my mind.

With care it is certainly possible to have all the shots you need in 16mm extracted without damage. Te cost of doing that would probably be less than the cost you save in telecine time at the re-transfer, so it's a win - so far. But if you do find the dirt level has increased, you will be up for dirt removal costs: either done automatically at the expense of image resolution, or digitally at the expense of your budget. Or you accept it as is. You could be lucky, or not - it's hard to say which.
  • 0

#17 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 18 May 2007 - 04:00 AM

, is there anything I can do to make sure I don't get ripped off by the lab?

Make sure that you only have to pay for the time you actually take. You are supervising. You will control how long you spend on grading each shot. Book the telecine suite for the time you think you will need. If you run out of time, you will have to re-book, and they will have been proved right.

Or you could get an estimate from another place to compare, if you really don't trust this one. In fact, ask yourself if you really don't trust them, why are you proposing to do business with them?
  • 0


Media Blackout - Custom Cables and AKS

Gamma Ray Digital Inc

Metropolis Post

CineLab

Visual Products

FJS International, LLC

Willys Widgets

Tai Audio

Abel Cine

Ritter Battery

Broadcast Solutions Inc

Wooden Camera

Paralinx LLC

Rig Wheels Passport

Aerial Filmworks

CineTape

Opal

rebotnix Technologies

The Slider

Glidecam

Technodolly

Media Blackout - Custom Cables and AKS

Gamma Ray Digital Inc

Glidecam

Wooden Camera

FJS International, LLC

The Slider

Willys Widgets

Broadcast Solutions Inc

Technodolly

Visual Products

Abel Cine

Paralinx LLC

Rig Wheels Passport

Tai Audio

Aerial Filmworks

Metropolis Post

Opal

CineLab

Ritter Battery

rebotnix Technologies

CineTape