Jump to content

Roll number


takev

Recommended Posts

Hello everyone,

So a couple of weeks ago I saw the video about the CAS meeting for non-linear workflows. Roll numbers caused a heated discussion.

The mixers don't have a use for roll numbers at all.

The archive people would like a roll number for each day, so they propose the date, which fits inside the 6 digit roll number, they don't care about multiple media to have the same roll number.

But the telecine house doesn't like this because they would like a roll number for each medium you provide; you can not attach an extra digit to the date because the roll can not be larger than 6 digits.

So I just had a tought. How about a 5 digit days-since-january-1st-2000. followed by a 1 digit media number for that day?

For example the roll number of the first media for today would be: 024810

The roll number for the first media for tomorrow would be: 024820

The second media for tomorrow would be: 024821

The limitation for this way of working, is that you can not get more than 10 media per day and it stops working in the year 2273. By then we would probably have moved on to other things.

How wrong am I with this?

Cheers,

    Take

Link to comment
Share on other sites

  • Replies 55
  • Created
  • Last Reply

Top Posters In This Topic

Take,

During that seminar we proposed that they stop using the TAPE or Roll field as a primary organizational field in Pro tools and in the AVID since they are meaningless in a non-linear world that doesn't record on Tape or divide recordings into Rolls.

We all seemed to agree on the shoot date and Time Code.  If AVID, FCP, Pro-Tools and any other software based synchronization tools would just use the USER BITS in the time code arranged as the date + Media/disk Number or Take Number mm:dd:yy-tt that field is available in the current BWF spec and the User bits are carried along with the Time Code in LTC and VITC. No metadata is even required and the User bits of SMPTE Time Code have been Standardized for more than 15 years.

----Courtney

Link to comment
Share on other sites

So If one completely eliminated roll numbers, has it been discussed how one knows what sound report goes with what disc/Cf card, folder, or what have you???

Why is there a movement underway to eliminate sound rolls?  I feel that they still have relevance, at least the way I work.  I find them to be more of a convenience than an inconvenience.

I'd be very appreciative if someone would please fill me in on the history of this.

Thanks,

Darren

Link to comment
Share on other sites

Darren,

The end result would be that all audio files would be loaded into a server, with all the metadata for this audio in a database.

Simply execute a search on the database and retrieve the audio.

No more sound reports, at some point probably no more media either, just send the audio over the internet connection at the location or studio.

The same is going to happen with video, except the amount of data for video is still a bit much.

From what I understand is that some studio archives already have all audio on one big server.

Cheers,

    Take

Link to comment
Share on other sites

Darrin,

The problem is not with the Roll Number or Disk number you write on a DISK or put in the upper corner of your sound log.

The Problem is putting that roll number into the metadata of each file on that Disk.

With many machines you may not know what disk a file is going to end up on until the end of the day.(if it ends up on a disk at all)  Plus, with all the backups, some on Hard drive some on CF, the files may not end up on the same media and Roll becomes meaningless in terms of locating the files.  By the time it gets to post, all the files may end up in different folders on a server somewhere.  So using a Roll number to sort or locate them becomes problematic. Problematic because the sound mixer does not know how to determine a Roll number when the files are being recorded on 3 different media with different capacities simultaneously.

The Roll or Tape number was a holdover from the days when all dailies were digitized off Daily Tape Rolls. Either analog or digibeta from the Telecine transfer.

With file based ingest,  Roll or Tape numbers become confusing or meaningless. EDLs need to evolve to reflect this paradigm shift.  They have always had problems because they always assumed contiguous time code on each roll.  That is the problem with crossing midnight using time of Day code. Which is why the editors always specified each tape roll to have REC RUN time code. Great if you are only shooting one camera, but not good as a means of auto-syncing up multiple cameras and double system sound.

----Courtney

Link to comment
Share on other sites

Courtney,

Thank you for the explanation.  Personally, I use a "ROLL-SCENE-TAKE" file naming scheme.  I suppose it's a shame that, while some of them can, not all of the recorders out there are able to encode the Roll Number into their metadata.

I find that, at the very least,with the possible exception of the Sound Devices recorders, rolls can be quite easily organized on any media using folders that are named after that particular roll.  That being said, I certainly do not have experience with all of the NLR's out there, so this may be an issue with other recorders as well.Thanks again Courtney,

Darren

Link to comment
Share on other sites

So If one completely eliminated roll numbers, has it been discussed how one knows what sound report goes with what disc/Cf card, folder, or what have you???

Why is there a movement underway to eliminate sound rolls?  I feel that they still have relevance, at least the way I work.  I find them to be more of a convenience than an inconvenience.Darren

I think there is too much going on here with the whole roll number thing. First off, I think that ALL the non-linear recorders do provide some place to put an identifying number which is quite prominent in the primary header or metadata, and this number I believe should still be referred to as a "roll number" but could, of course, be called when referred to, anything you want to call it. Also, all TC systemes in use that I know of make provision for the DATE to be in the user bits and as readilly accessible as the TC. Most jobs, but not all, will only require 1 "sound roll" for each day, particularly if this sound roll resides on a high capacity hard drive with variable folder size (here we call the sound roll a "folder") or a large hard disk partition. So, no problem so far but when it comes to the DELIVERY media to post I do realize that some of the points Courtney and others have brought up become relevant. What I don't understand, though, is let's say this sound roll actually has to be delivered on 2 disks (because of the capacity of the optical disks vs. the basically unlimited capacity of the hard drive) what is the problem? The same "sound roll" number is part of ALL the files that are on these disks or carriers, along with the date (sort of like the file creation date which incidentally also most probably exists in the file), so that seems to be more than sufficient to identify all the sound takes no matter how many disks were needed to carry them to post.

I think all the proclamations about how meaningless the SOUND ROLL is in the non-linear world are somehwat off the mark --- certainly some adjustments to procedures, file naming and so forth may be required to accomodate all the systems presently being used, but retiring the sound roll number I don't think is one of them.

Regards,  Jeff Wexler

Link to comment
Share on other sites

In the commercial world I live in, I call roll 1 of a job an abrieviation of the company, today I'm working with Smuggeler, so I call my folder SMG1, next day, SMG2,etc. This way I don't have multiple roll 1's if I need to revisit a folder on my drive. For my world this is better than dates as folder names. Besides the date is already in the user bits. On the dvdram and the jewel case I lable them as folderSMG1 I also say day 1, the date, product, 24/48k/30ndf/b.wav.p. So far this has worked well enough for me in that no one has complained, But I would do it differently if asked to do so. We still call movies, films no matter what medium it was shot on so I guess it really doesn't matter if we call them rolls, folders, or todays load of BS, but we have to call them something.

old school

Link to comment
Share on other sites

I think there is too much going on here with the whole roll number thing. First off, I think that ALL the non-linear recorders do provide some place to put an identifying number which is quite prominent in the primary header or metadata, and this number I believe should still be referred to as a "roll number" but could, of course, be called when referred to, anything you want to call it. Also, all TC systems in use that I know of make provision for the DATE to be in the user bits and as readily accessible as the TC. Most jobs, but not all, will only require 1 "sound roll" for each day, particularly if this sound roll resides on a high capacity hard drive with variable folder size (here we call the sound roll a "folder") or a large hard disk partition. So, no problem so far but when it comes to the DELIVERY media to post I do realize that some of the points Courtney and others have brought up become relevant. What I don't understand, though, is let's say this sound roll actually has to be delivered on 2 disks (because of the capacity of the optical disks vs. the basically unlimited capacity of the hard drive) what is the problem? The same "sound roll" number is part of ALL the files that are on these disks or carriers, along with the date (sort of like the file creation date which incidentally also most probably exists in the file), so that seems to be more than sufficient to identify all the sound takes no matter how many disks were needed to carry them to post.

I think all the proclamations about how meaningless the SOUND ROLL is in the non-linear world are somewhat off the mark --- certainly some adjustments to procedures, file naming and so forth may be required to accomodate all the systems presently being used, but retiring the sound roll number I don't think is one of them.

Regards,  Jeff Wexler

While many are getting creative in what they put in the Tape/Roll Field,  as long as you have a workflow established and communicate what you are putting in the ROLL filed it can work. The problem is Roll used to mean something and we all knew what to put in it.   It also represented the primary means of locating the original sound recordings.  Now that we have digital file based recording and the ability to Clone those files ,there is no need to return to the actual original production media.  So now Roll not only has no meaning,(because there are no actual Rolls) it's usefulness is rendered moot. So my point was why does it remain the primary sort field and media locator in AVID and Pro Tools.  Some Recorders allow you to enter a "Tape or Roll #", some don't use the field at all  and some automatically fill the Tape field with a number (usually the Date) which the user can't alter. (Tascam and SD machines do this) So it may be beyond the control of the mixer to make sure that there are no duplicate "Tape" or Roll fields (a source of confusion come time to edit)  Deva didn't have a Roll field in it's metadata until the last firmware upgrade. Why not just use the Date (either in the User Bits or in the Date Stamp in the Bext Chunk) as the primary sort field. No extra thought or special communication necessary. and All machines that record Broadcast WaV files have this data built in.

---Courtney

Link to comment
Share on other sites

The problem is Roll used to mean something and we all knew what to put in it.

---Courtney

I think the Roll number still has meaning but you are quite right that it no longer represents the primary means of locating thec original recordings --- and, as you point out, even the meaning of "original recording" and "master recording" has changed fundamentally with file based workflow.

quote - So now Roll not only has no meaning,(because there are no actual Rolls) it's usefulness is rendered moot. So my point was why does it remain the primary sort field and media locator in AVID and Pro Tools.

Both Avid and ProTools have a long history that pre-dates all of the file based production sound deliveries, and as you pointed out, everything was brought into these systems in a linear tape based manner (where the sound roll really was some sort of roll of something). The fact that it is still called a sound "roll" is quite similar to editorial systems still referring to "bins" (as I said before, real film bins have not been used since the days of the Movioloa).

quote - Some Recorders allow you to enter a "Tape or Roll #", some don't use the field at all  and some automatically fill the Tape field with a number (usually the Date) which the user can't alter. (Tascam and SD machines do this) So it may be beyond the control of the mixer to make sure that there are no duplicate "Tape" or Roll fields

This is a problem --- anytime the recorder will only do things one way and you do not have control over the numbers and where they are being put. As for the Deva, a Roll Number has always been available as 3 of the primary numbers right in the file header. This was true with the Deva I but of course what we were really naming a sound roll at that point was a dedicated hard disk partition that "held" all the files that were that given sound roll.

Regards,  Jeff Wexler

Link to comment
Share on other sites

With many machines you may not know what disk a file is going to end up on until the end of the day.(if it ends up on a disk at all)  Plus, with all the backups, some on Hard drive some on CF, the files may not end up on the same media and Roll becomes meaningless in terms of locating the files.  By the time it gets to post, all the files may end up in different folders on a server somewhere.  So using a Roll number to sort or locate them becomes problematic. Problematic because the sound mixer does not know how to determine a Roll number when the files are being recorded on 3 different media with different capacities simultaneously.

I usually solve this problem by putting the files in a folder marked "Roll #X".  I can do this on all of the media and on the hard drives that the files end up on.  Recorders like the 744T automatically put the files into the folder as they are recorded.  It makes it a lot easier to find a file later if editorial calls with a question about a certain take or wild track or whatever -- I can look at the sound report for a *roll* and quickly find the file.  Even with advanced search functions, something tells me it would be more confusing if I had to sift through an entire show to find something.

Link to comment
Share on other sites

While many are getting creative in what they put in the Tape/Roll Field,  as long as you have a workflow established and communicate what you are putting in the ROLL filed it can work. The problem is Roll used to mean something and we all knew what to put in it.   It also represented the primary means of locating the original sound recordings.  Now that we have digital file based recording and the ability to Clone those files ,there is no need to return to the actual original production media. 

No, but you still need the ability to find material on whatever volume you deliver, whether it's a hard drive or optical disc.  In that sense "roll" does what it always did -- tells everyone after you working with your tracks where to find them, regardless of whether or not it is the "original" media.  No matter where the media is, it's useful to have it organized in separating folders.

Why not just use the Date (either in the User Bits or in the Date Stamp in the Bext Chunk) as the primary sort field. No extra thought or special communication necessary. and All machines that record Broadcast WaV files have this data built in.

The problem with that is when you shoot splits after midnight, then your files go into the next day's "roll" or folder.  Let's say you shoot from 12 PM to 1 AM Oct 20 and this is day #15 of the shoot.   The next day, day #16 of the shoot (Oct 21), you go from 3 PM to 4 AM.  The production "day" is the date you showed up for work and where post will expect to find your files -- they are not going to keep track of your hours.  What happens in this scenario is that in the "roll" or "folder" or search header for day #16 (Oct. 21), you have files from day #15 that were done after midnight, and files from day #16 that were recorded between 12 and 4 AM during that production day are missing from the folder or "roll" (they end up in the Oct. 22 folder).

Now, let's say someone asks you to burn a DVD of all of the files from Day 16.  Now you're going to have to dig through two day's folders, add and subtract files, hunt and poke around with the time stamp of what time of day the file was recorded, etc etc.  I had to deal with this when Sound Devices' early firmware only placed files in "Daily Folders" based upon the date, and it was a pain in the ass, in my experience.  Now that they allow for use of "Scene folders" it becomes much more straightforward.

.02

Link to comment
Share on other sites

Noah,

I think you are confusing the point. This thread is about entering the TAPE field in the Metadata. It is not about how you organize your folders.  The ROLL or TAPE field in the metadata must be entered at the time the file is recorded. It is used in a database to locate files.  Those files can be in any folders with any names. Those folder names that you or the machine uses are of no consequence further down the workflow where all the files end up in different folders organized however other people see fit.  There is no reference in the file's metadata that says what folder you used when recording.  So whether you use the DAILY folder in the SD machines or the SCENE folder it doesn't change the value entered in the TAPE fieldof each file's metadata.

Using the Date and Timecode as a locator allows you to locate files without any possibility of duplication, and without regard to what folders were used originally. The files will always be arranged chronologically and files will be arranged in the order shot.

IF you are shooting a dramatic film where you may shoot 5 to 10 scene numbers during the production day and not necessarily shot in numerical order, it can be confusing to use the SD Scene Folder option. It makes it difficult to track which folders have been burned to disk and it makes mid day  break-off a bookkeeping chore. So the Daily folder is currently the only viable option for folder organization on anything that you want to enter the Scene and Take Numbers into the Metadata.  In the Current 744T firmware you cannot name a folder something like ROLL#1 and then record in it.  You can only rename folders after they are recorded and you can't record in a folder after its name is changed.

----Courtney

Link to comment
Share on other sites

Noah,

I think you are confusing the point. This thread is about entering the TAPE field in the Metadata. It is not about how you organize your folders. 

The SD "Date" folders business was an analogy, since it works (and has the same pitfalls) as the system you suggest (using date, TOD, etc).

The point is that a "roll" number that references one day's work gives the files an easy-to-find location, whether that's done in metadata, on the original media, or both.  If you take that roll number away and rely only on the time and date, you introduce a lot of errors and confusion due to the fact that just about every movie has at least a few split shoots, where the date no longer references the production day and the time of day code becomes confusing because it goes from 23 hours to 0 hours (or 22 hours to 2 hours, or whatever it is -- based on how long it takes that particular movie to do another shot).

The ROLL or TAPE field in the metadata must be entered at the time the file is recorded. It is used in a database to locate files.  Those files can be in any folders with any names.

Eventually perhaps, but first telecine has to find them.  Then the assistant editor has to find them.  Then, twelve or twenty weeks later, different members of the sound editorial tieam have to find them to do the conform -- and here's where it's very critical, as I've seen the OMF from the Avid used many times when sound editors lose patience for tracking down elusive files with full bit depth and iso tracks.

Those folder names that you or the machine uses are of no consequence further down the workflow where all the files end up in different folders organized however other people see fit.

Maybe not "further down the workflow", yes, but they are of important consequence at the head of each stage of post production (dailies, pix ed, sound ed) for reasons stated above.  Because of this, they are useful.

  There is no reference in the file's metadata that says what folder you used when recording.  So whether you use the DAILY folder in the SD machines or the SCENE folder it doesn't change the value entered in the TAPE fieldof each file's metadata.

No...again, we're talking about two different things here.

Using the Date and Timecode as a locator allows you to locate files without any possibility of duplication, and without regard to what folders were used originally. The files will always be arranged chronologically and files will be arranged in the order shot.

Better hope that post puts in a production time code burn in, then, and the assistant editors diligently list the production TC in the EDL for each cut.  I find this to be *very* rare these days.  Without that, there's no reference tying your TOD TC or date to the cut.  In your date search suggestion, the files will be arranged chronologically as to the order shot, yes -- but as you note in reference to the folder tangent, the order scenes are shot in has zero bearing on a picture edit.  The movie is edited, obviously, in scene order -- not in chronological shooting order.

IF you are shooting a dramatic film where you may shoot 5 to 10 scene numbers during the production day and not necessarily shot in numerical order, it can be confusing to use the SD Scene Folder option. It makes it difficult to track which folders have been burned to disk and it makes mid day  break-off a bookkeeping chore.  So the Daily folder is currently the only viable option for folder organization on anything that you want to enter the Scene and Take Numbers into the Metadata.  In the Current 744T firmware you cannot name a folder something like ROLL#1 and then record in it.  You can only rename folders after they are recorded and you can't record in a folder after its name is changed.

Again, this is a second subject, but there is a way to do what you say can't be done.  My way of doing it is to call the "Scene" the Sound Roll number.  Then every file on the sound roll is recorded directly to that folder.  I simply have to drag and drop one folder when it's time to send something out.  I keep a sound report where every file is numbered (sort of like "sound" numbers way back when) and that number references the scene information (23A take 5 on camera roll A215).  On the 744T this is input as the "take" number. 

Yes, you lose the benefit of having the scene and take information input into metadata right at the recording stage, but it makes doing paperwork and organization on set (as well as film breaks, as you note) much better, quicker, and easier with the 744T, and at least until recently, it was very, very difficult and time consuming to correct the inevitable errors when slates changed at the last minute or things were misslated on camera and so forth.  So far there have been no complaints, and I've been told on several projects that the assistant editor has an easy time inputting the scene and take information into the metadata (maybe with your software, or maybe with something on the avid -- I never asked but I can find out).  They only have to do it for takes used, as opposed to everything shot.  This is a much more efficient process, considering out of the hundred or so takes we shoot a day, only about twelve of them end up in the cut and the rest go in the trash.  It's also an on-set burden I'm happy to pass on as it's apparently easier to do in the edit room and has less room for error there.

Regards,

Noah Timan

Link to comment
Share on other sites

The SD "Date" folders business was an analogy, since it works (and has the same pitfalls) as the system you suggest (using date, TOD, etc).

The point is that a "roll" number that references one day's work gives the files an easy-to-find location, whether that's done in metadata, on the original media, or both.  If you take that roll number away and rely only on the time and date, you introduce a lot of errors and confusion due to the fact that just about every movie has at least a few split shoots, where the date no longer references the production day and the time of day code becomes confusing because it goes from 23 hours to 0 hours (or 22 hours to 2 hours, or whatever it is -- based on how long it takes that particular movie to do another shot).

All telecine operators I have talked to have told me that putting time code across midnight in the same folder and in some cases on the same disk is to be avoided.  You are proposing putting all the shoot day's files in a single folder regardless of time of day (or night).

This is what creates problems in telecine it doesn't solve them.  The DV-40 puts all the files in the folder into a virtual timeline based on the time code. Then it chase-locks that virtual timeline to the SMPTE edit controller.  If you put takes before midnight in the same folder as those after midnight they will be arranged on the timeline in reverse order.  putting them in separate folders based on the Date will prevent this problem.  I'm sorry I don't see any pitfalls in using the Date as a folder name as long as you consider the Date as the primary locator then the Time Code as the secondary.  There will never be any confusion especially when "Split" days are involved.

What was a production day (from call to wrap) is not now(in a non-linear workflow) useful information in locating material.  It was when there was no metadata possible and things were stored on Linear Rolls that were marked sequentially.  And of course you can emulate that old workflow using file based recorders, but it involves the old method of logging everything  and referral to logs to manually locate things. it  mandates storage and retrieval of the original recording media in order to use those logs. It also relies on "Burn in TC" on the dailies to locate and cross reference the audio manually.   Today's Editing workstations are designed to store and use the metadata within the files to locate them and align them on the timeline automatically. No manual syncing or cross referencing to logs should be necessary provided the correct metadata is contained in the files. The software should be able to find all the iso channels from the Date and Timecode in the metadata. If not, then from the scene and Take in the metadata to find the appropriate tracks, then use  the embedded timecode to align them to picture.  This is the whole concept of the automated workflow that everyone is trying to achieve.  However without a workable standard of metadata across platforms and machines, this will not become an attainable goal.  This is why storing the Roll number in the Metadata Scene field and the Date in the Roll number field, may work if you are doing things the old way and you use a hand-written log to cross reference things.  It however throws a wrench in any possibility of using machine automation to handle the storage and syncing of all the elements of a production.

  ---Courtney

Link to comment
Share on other sites

All telecine operators I have talked to have told me that putting time code across midnight in the same folder and in some cases on the same disk is to be avoided.  You are proposing putting all the shoot day's files in a single folder regardless of time of day (or night).

This is what creates problems in telecine it doesn't solve them.  The DV-40 puts all the files in the folder into a virtual timeline based on the time code. Then it chase-locks that virtual timeline to the SMPTE edit controller.  If you put takes before midnight in the same folder as those after midnight they will be arranged on the timeline in reverse order. 

That's why it's not wise to use TOD code, as we discussed in another thread.  TOD code offers no significant advantages and offers all of these problems.  When you start your code at zeros at the beginning of the day, this problem magically vanishes.

  I'm sorry I don't see any pitfalls in using the Date as a folder name as long as you consider the Date as the primary locator then the Time Code as the secondary.  There will never be any confusion especially when "Split" days are involved.

What was a production day (from call to wrap) is not now(in a non-linear workflow) useful information in locating material.

I've already illustrated why I feel this IS useful information -- maybe not way down the line in sound edit, but certainly at telecine and on our end to prepare files for telecine.  The telecine operator doesn't want to sift through metadata chunks -- he/she wants a regular sound report with a list of what's on the file.  Ditto the assistant editor who has to log all of the files in to put them in the editor's bin.  Maybe someday there will be metadata-employing software that will do all of this for all of these people, but in my day to day working experience this is not yet a reality in the working world.  I'll end this paragraph here as it appears we'll just have to agree to disagree on this one -- either I'm not expressing it right or you just don't want to hear it, but for whatever the reason I don't seem to be getting through with an expression of the concept.

  It was when there was no metadata possible and things were stored on Linear Rolls that were marked sequentially.  And of course you can emulate that old workflow using file based recorders, but it involves the old method of logging everything  and referral to logs to manually locate things. it  mandates storage and retrieval of the original recording media in order to use those logs.

No it doesn't...so long as the file and roll are referenced (the roll in the metadata, and the file in the filename) one can use any subsequent clone of the original recording on whatever media.

It also relies on "Burn in TC" on the dailies to locate and cross reference the audio manually. 

Now you've lost me.  How does having or not having a sound roll referenced or not have anything to do with whether or not you need "burn in" TC?  A) The TC is in the metadata regardless of whether or not you use the roll number, and B) even if you're referring to not using scene & take info in the metadata, the TC timestamp is STILL there.  I don't know of a DAW or editing controller, though, that can sync to a cut without any TC reference in the cut based upon the metadata entries of the files used in the cut...do you or does anyone?  My experience has been VERY contrary -- in fact, most prominent and careful editors here are loading in all of the tracks seperately from telecine because what is used is an OMF most often.  If it was a simple matter to auto-conform the cut via metadata, do you think they would still be doing this?

Today's Editing workstations are designed to store and use the metadata within the files to locate them and align them on the timeline automatically. No manual syncing or cross referencing to logs should be necessary provided the correct metadata is contained in the files. The software should be able to find all the iso channels from the Date and Timecode in the metadata.

There are a lot of "should"s here but as stated, I personally do not know of any system that performs this kind of auto-conform to an EDL, timeline or whatever based solely on metadata values and can reference these metadata values from the cut.  If I've been living in a cave, someone please enlighten me, but again, none of the editors I've worked with in the last few years have had such a system at their disposal.  Quite contrarily, there have been a lot of multi-day discussions about solving many issues in a much more complex way BECAUSE no such system exists.  I have to wonder aloud how much of this comes from experience and how much of this is theoretical and based on assumptions of NLEs and DAWs having functions that they do not yet have.  I'm not accusing you of the latter, but it comes as a surprise since it it so contrary to any experience I have had or even heard talked about.

use  the embedded timecode to align them to picture.

How to do this if there is no TC reference in the cut?  No burn in, nothing in the EDL.  Metadata again?

This is the whole concept of the automated workflow that everyone is trying to achieve.  However without a workable standard of metadata across platforms and machines, this will not become an attainable goal.  This is why storing the Roll number in the Metadata Scene field and the Date in the Roll number field, may work if you are doing things the old way and you use a hand-written log to cross reference things.  It however throws a wrench in any possibility of using machine automation to handle the storage and syncing of all the elements of a production.

I apologize, but I still do not understand, even giving you the benefit of the doubt that such machine-automated workflow exists in this day and age as you describe, how having a "tape" or "roll" number in or not in the metadata alloted field for it would screw up this automation.

If I and the editors and assistants I have worked with in recent years have missed the boat completely and all of this auto-conform software is a reality in practice, apologies and I am more than happy to publicly eat crow over it.  But I just haven't seen it.  Please enlighten if so.  There's a seminar coming up next weekend here in NYC being hosted by one of the bigger telecine houses and one of the audio dealers in town -- I'll try to bring up these issues and see what I/we can learn.

Regards,

Noah Timan 

Link to comment
Share on other sites

Noah,

All of what I described is doable today with existing software and properly entered metadata in BWF files.

I hope they have some good speakers at the seminar in NY.  If you want to view the video of the CAS seminars from last June where Scott Wood from Digi demonstrated may of the fieatures I described in Pro-Tools 7.2. check out the CAS flash video here: http://www.cinemaaudiosociety.org/seminars/workflow3.php

  Although it hadn't shipped when the video was recorded, Pro-tools 7.2 is available now.

There is resistance to entry of metadata in proper fashion, as is shown by several comments in the seminar. However the technology is there if people are willing to learn how to use it and agree on a few common methods for entering and working with the metadata.  If you are using modern recording tools but using a workflow that you have used for years when syncing and editng 35MM Mag or multitrack linear audio, you can always make it work, but you are not using the tools to their maximum capability.

---Courtney

Link to comment
Share on other sites

Hi Courtney,

I'll check out the CAS seminar video as well as attend the upcoming NYC seminar and report back.

In the meantime, if you have evidence that auto-conform based solely upon metadata entry and today's workflow is available, inquiring minds want to know.  Maybe it's all present in the link you presented, which I will look at tomorrow. Of course, it'd probably be helpful for the readership if there's more simple evidence of this than watching a long video presentation (which I will do, but I'm not certain everyone who could benefit from the answer to this question will).

Regards,

Noah Timan

Noah,

All of what I described is doable today with existing software and properly entered metadata in BWF files.

I hope they have some good speakers at the seminar in NY.  If you want to view the video of the CAS seminars from last June where Scott Wood from Digi demonstrated may of the fieatures I described in Pro-Tools 7.2. check out the CAS flash video here: http://www.cinemaaudiosociety.org/seminars/workflow3.php

  Although it hadn't shipped when the video was recorded, Pro-tools 7.2 is available now.

There is resistance to entry of metadata in proper fashion, as is shown by several comments in the seminar. However the technology is there if people are willing to learn how to use it and agree on a few common methods for entering and working with the metadata.  If you are using modern recording tools but using a workflow that you have used for years when syncing and editng 35MM Mag or multitrack linear audio, you can always make it work, but you are not using the tools to their maximum capability.

---Courtney

Link to comment
Share on other sites

Hi Courtney,

I'll check out the CAS seminar video as well as attend the upcoming NYC seminar and report back.

In the meantime, if you have evidence that auto-conform based solely upon metadata entry and today's workflow is available, inquiring minds want to know.  Maybe it's all present in the link you presented, which I will look at tomorrow. Of course, it'd probably be helpful for the readership if there's more simple evidence of this than watching a long video presentation (which I will do, but I'm not certain everyone who could benefit from the answer to this question will).

Regards,

Noah Timan

The proper forum for those specific questions and for more authoritative answers are the digidesign User Conference at http://duc.digidesign.com/

Today's workflows are a superkludge because people are naturally resistant to change. They know how to do something one way that has worked successfully for them in the past and feel no need to learn a different way. As a result of technology advances we find people tyring to apply their old methods of work to the new technology and ignoring new features or automation that was designed to streamline work but ends up being a roadblock  because we don't understand it's design concept or are unwilling to give up old practices.  Workflows are also a Kludge because the designers of the various tools have come from different disciplines.  Post sound tools came from both the multi-track Music studio recording industry and the Film post world where synchronization with an image was a primary concern and things were traditionally kept lined up based on mechanical means. (sprockets and sync blocks). And those disciplines are filtered through yet a third and fourth technology of Video and Computer software both of which have a history of how things were done in their respective worlds.  Guiding all these divergent technologies to a common goal is not a simple task but I think through communication and education workable standards are attainable.

Let me know how the seminar in NYC turns out.  Also if anyone is putting it on tape for viewing online please post that information here and on RAMPS.

---Courtney

Link to comment
Share on other sites

I will look further at the digidesign forum and at the CAS video.  I'm curious about real-world success stories of this workflow, though.  I understand thoroughly your well-put reasoning of how difficult it is to implement industry-wide workflows, and there are even more complications than the ones you mention.  For example, picture editorial departments with varying degrees of education and experience with non linear sound (and sound in general) is certainly a big one.  Post houses not willing to spend fractions of what we constantly spend to become more future-progressive is another.  (Look how much work it was/continues to be to get facilities to buy a $5K DV-40). 

Unfortunately, even though there is potential for -- or maybe even already developed and existing -- tools to streamline and organize the process, it becomes a bit of a "tree falling in the forest" argument if no one is going to implement those tools and procedures as standard in today's world.  The only way one can change this from our standpoint, beyond generally furthering and educating and so forth (a slow process that will take years), is somehow individually command the direction of the workflow for a project that one is working on.  Obviously, this is not likely to happen in all but the most rare cases, since even if the post sound department is hired before picture is wrapped (not often), they might think it presumptuous to have the production mixer dictate their workflow.  The producers are often not technically competent enough at that level to participate in these discussions with even if said discussions yield an advantage to budget or movie quality.  Thus, you are led to a position where you know the right way to do things but are powerless to change things being done the wrong way, even as that process unfolds before your eyes.  It's not uncommon for many facets of the filmmaking process.  We're not an exception.

So, as much as we want to advance things and realize potential, we also have to make our current projects work and our clients satisfied.  Because of the aversion to change in our industry that you clearly note, we will, however, need more than a few real-world successful models of your suggested workflow (a system where conforms are accomplished by metadata entries, thereby doing away with all of the traditional methods of using time code references in EDL and picture).  When I think of experiences like my most recently completed project, where the editorial team required that all of my wild tracks be transferred first to DVCAM for importation because their Avid was not up to date and not upgradable and it couldn't even process *BWF* files properly -- let alone metadata -- your suggestions seem like they are very far away indeed.  (This was a fairly major studio project, BTW, not a penny pinching indie -- and it's not the first time I've had to deal with edit stations running old OS 9 Avid versions that get awfully flummoxed when you try to import a broadcast wave file).

If we can point to real-world examples of the metadata-based conform process that you reference -- preferably several, to properly be able to sway the producers out of their usual process and dangle the carrot that there is a better and more efficient way that benefits all -- then we may be going where you want.  But if we can't, it's a bit silly to sit around squabbling amongst ourselves about whether or not what is entered in the "roll" field of the metadata or whether or not the sound mixer or assistant editor enters scene and take information.  As someone who has developed software to easily edit the metadata chunk, you above all are aware of how simple this can be done at any stage.  But if nobody's currently using that metadata for if anything, more than a convenient reference as opposed to a tool to conform -- it doesn't yet matter who does it or if it is done at all.  It's a bit like arguing about whether or not your tie is properly tied for a business meeting where the other party you are meeting with is not going to show up.

Let me know how the seminar in NYC turns out.  Also if anyone is putting it on tape for viewing online please post that information here and on RAMPS.

OK, will do.  There's a flyer for the seminar on Gotham Sound's website, who is hosting the seminar along with PostWorks:

http://www.gothamsound.com/main.shtml

There doesn't seem to be mention of it on the website, but I know they are gearing up to stream video of the seminar over the internet live, and I assume they will also provide a recorded version that can be played back later.  The conference seems to be emphasizing telecine issues, but I am sure that the discussion will continue and involve further post production process.  I'll try to post more detailed links as they emerge, if Peter Schneider at Gotham does not do so himself.

nvt

Link to comment
Share on other sites

Correct me if I'm wrong, but, I believe that there is a very popular automated software solution that uses editorial's EDL's and the metadata from the sound files to re-conform the sound to picture.  It's Titan.  I know that this has been used with  success on at at least the last 4 or 5 projects that I have worked on.

It is definitely in the sound post house's best interest to have an automated conformation process.  Conforming can come out of their time and budget.  They would rather get the conforming done in a couple of minutes instead of a couple of days so that they can get on with the real work and meet their ever tightening deadlines.  Most of the sound houses that I have dealt with are well aware of this.

I know that Titan reads the iXML info, and it may well read the Bext info as well.

If I'm wrong about this, please accept my apologies.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...