Film-Tech Forum ARCHIVE: ID Frames and their effect on Sound

This is topic ID Frames and their effect on Sound in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=004504

Posted by Michael Brown (Member # 890) on 01-31-2003, 06:23 PM:

From 'The Recruit' in Feature Info.

quote:
Brad Miller Wrote: Chop away ID frames between reel 2 and 3 and you will have a jump in the analog audio playback.

Why only the analog track? [Confused]

Posted by Brad Miller (Member # 2) on 01-31-2003, 06:39 PM:

Frequently 2 missing frames will not be noticeable when playing in digital, dts especially due to their crossfade. However since the digital audio is not recorded directly alongside the analog track, it would take some calculation to figure out if cutting them would create a noticeable jump or not. Just looking at the analog soundtrack and seeing the waveform go from very loud to almost nothing is easy to see.

Posted by Daryl C. W. O'Shea (Member # 1303) on 01-31-2003, 06:41 PM:

Of course, if you don't know how to properly make a splice the gigantic pop will distract the audience from the lost audio. [thumbsdown]

Posted by Michael Brown (Member # 890) on 02-01-2003, 05:16 AM:

quote:
However since the digital audio is not recorded directly alongside the analog track

Oh, is digital not recorded 22 frames ahead of the image like the analog track?

Posted by Leo Enticknap (Member # 534) on 02-01-2003, 12:15 PM:

No. I don't know the offsets from memory, but it's a long way ahead. The delay is set manually, in numbers of perforations, on the processor. That way gives you flexibility as to where in the film path you position the reproducer.

Posted by Stephen Furley (Member # 1277) on 02-01-2003, 12:44 PM:

Dolby Digital is printed 26 frames in advance of picture, I can't remember the others.

Posted by Michael Brown (Member # 890) on 02-01-2003, 04:17 PM:

Wouldn't it make more sense to have the same offset for all the soundtracks?

And why did they choose 22 for analog as opposed to 24?

Posted by Jeffry L. Johnson (Member # 453) on 02-01-2003, 06:39 PM:

Read the threads "Digital Offsets" and "Audio Offset Documents" in this forum.

35mm analog is 21 frames ahead of picture.

CDS is offset zero frames.

Dolby Digital is 26 frames ahead of picture.

DTS is offset zero frames.

SDDS P side is offset zero frames. SDDS S side is 17 frames behind picture.

Posted by Steve Kraus (Member # 476) on 02-01-2003, 06:53 PM:

Dolby was smart because even though they started with penthouse playback they anticipated their system would lend itself to basement readers that retrofit existing soundheads so they went with an advance offset. 26 frames rather than 20 or 22 gave them a bit of flexibility in how such equipment would be designed.

BTW, what sort of big jump in audio in The Recruit will there be with the aforementioned two frames removed and what's in that 1/12 second of audio that somehow makes it less jarring? Would it not qualify as fair use for scientific research to digitize a couple seconds on either side of the join then edit 1/24 second out on both sides to illustrate the difference with and without ID frames?

Posted by Brad Miller (Member # 2) on 02-01-2003, 07:05 PM:

It goes from almost silence to very loud. (I may have that backwards.) The buildup is over those 2 frames that most people have removed. It would be like snipping the beginning part of a song or word of dialogue. It would be noticeable. Had I had my digital camera with me, I would've taken a pic.

Posted by Steve Kraus (Member # 476) on 02-01-2003, 11:03 PM:

Well, I'm not saying you're wrong but I think it's pretty hard to judge by looking what the result of the snips would be. Tracks often go from very quiet to very loud. It depends on what exactly is in those two 1/24 second bits whether it matters at all. Either cut or uncut it's still going to be a big jump in level almost instantly and one presumes it makes sense in context. The lost frames certainly cannot be containing a gradual buildup. So how much of a difference they make remains an open question.

Maybe they are important and maybe they are not. The visual evidence that a big change in level takes place there does not in itself prove the importance of the two frames worth of track.

Posted by Brad Miller (Member # 2) on 02-01-2003, 11:38 PM:

quote:
The visual evidence that a big change in level takes place there does not in itself prove the importance of the two frames worth of track.

Yes it does! Anyone who has ever edited audio and viewed the waveform can understand why it would have a jump/chopped sound. All it takes is 1/24 of a second for a jump to not sound right. Cut that down to 1/12 of a second and if the audio happens to be at the beginning or end of a word, or even just on the strike of a drumbeat in music...and it's obvious.

As an example, if you were to record the sound of your voice saying "WHAT???" in a quick, stern tone and then went back and chopped out 2 frames worth at the beginning or end of it, you would hear it. Anyone would hear it.

Posted by Steve Kraus (Member # 476) on 02-02-2003, 10:50 AM:

Yes I agree but you cannot know if that is the case simply by looking. You're giving an example of where it makes a difference and I can give you all sorts of examples of sounds where a 1/12 second loss won't matter. What if it's a scene with quiet room tone up until an alarm clock goes off or a phone rings or there is a car honking outside? What if the sound change corresponds to a scene change 22 frames up in picture, going from quiet indoors to loud street scene? And on and on, ad infinitum. The point is not that you are wrong because you might be right. The point is that you cannot know just by looking.

Posted by Michael Brown (Member # 890) on 02-02-2003, 10:55 AM:

I think that Brad means that the 2 frames contain the fade up.

Posted by Brad Miller (Member # 2) on 02-02-2003, 11:20 AM:

Yes that is what I am trying to explain, Michael. Without the 1 or 2 frame "fade up" of the sound, it would have a chopped effect to it. Go find a waveform of a car honking and lay it into a non-linear editor. Now edit out the first frame or two of the point where the honk starts. It would be obvious to anyone that something is missing or "skipped".

Posted by Darren Briggs (Member # 1071) on 02-02-2003, 05:06 PM:

Analogue sound is 20 frames ahead of the picture.
Just count the frames on your head leader, from any number laced in the gate to the diamond printed on the film which lines up with the optical sound head.

This should always be checked when lacing a film (im sure you should already know this) if not you may be lacing the film up with out of sync sound. This is a basic rule of using a film projector.

If you laced it up 2 perforations out of sync and the film was printed with sound 2 perforations out of sync, your sound could be a whole frame out. You should check the Picture and Sound sync marks on the head leaders of the film. And is the film was printed out of sync, you could then lace up to compensate for this.
Digital basement readers would also be out of sync if laced incorrectly. Penthouse readers depend on the correct top loop.

Darren

Posted by Steve Kraus (Member # 476) on 02-02-2003, 05:34 PM:

There is a difference between the official standard and the normal practice as far as analog sound pullup is concerned.

As for the ID frames in question the only way we'll ever know for sure would be to try it both ways or digitally edit a copy of the non cut version. Even if Brad's assertion turns out to be true for this join it still represents poor editing that is ignorant of 75+ years of sound on film experience as regards where to put the breaks when balancing reels and it remains an open question as to how much the projectionist needs to try to accomodate sloppy editing. For example, while this situation won't affect me as a reel to reel operator there are others that do, namely dialog too close to the beginning or end of a reel resulting in part of a syllable being lost on changeover. But I have no intention to start using a platter just accomodate crappy editing. (Although I did build part of Russian Ark the last time I ran it but that was a special situation. Yes, I left ID frames even there.)

Posted by Brad Miller (Member # 2) on 02-02-2003, 07:45 PM:

Alright Steve, since you just don't believe me, here is a quick sloppy example of why platter people should not be chopping ID frames without looking at the track first to make sure they will not be permanently cutting out audio continuity. I hope this is simple enough for everyone doubting to understand. If not, I'll start collecting reel change examples over the next couple of months and make a huge page in the tips section showing real life examples.

Still I definitely agree with you that this entire argument of whether to cut or not is all because of sloppy reel change placement. However that is out of our hands and anyone running a platter, tower, or building to large reels should respect the film enough to make sure that they are not causing damage by chopping off 2 frames when they build it up (1 per leader). If this wasn't such a somewhat common problem, I wouldn't have a problem leaving an ID frame in tact on the leaders.

Click here

(I couldn't post these pictures on this particular thread due to the size of them.)

Posted by Joe Redifer (Member # 3) on 02-02-2003, 08:13 PM:

Well this isn't even an issue at all if you don't normally cut ID frames from a new print.

But yes, two frames is a lot and having edited 5.1 and 2 channel audio myself I can tell you that looking at the waveform DOES represent whether sound will be smooth or jumpy. Sure, you have to look a bit closer at 35mm analog audio since it is so small, but other than the size the way it works is exactly the same.

Posted by Steve Kraus (Member # 476) on 02-02-2003, 09:18 PM:

Hehehe...nice demo but let's see what we have there. You've taken a piece of audio/video where it is known in advance that all the major modulation is narration, snipped out two frames at the start of some modulation and predictably ended up cropping a spoken word. That's pretty much preordained given the material and doesn't really address anything except that cropping words is bad, not whether some random place where modulation increases is necessarily equally bad. Like I've been saying all along, I don't claim you're wrong only that you don't know.

To make your point you'd really have to look at random points in a random, unknown track and make snips at the start of a major rise in modulation and then listen to and see how often the snip causes a noticable disruption. And even that may not represent a totally fair test since one presumes that for the most part editors still do tend to respect the reel change areas. A big modulation change may mean they didn't but then again it might not.

Oh, and despite the nature of the demo it wasn't all that much of disruption. The small pop occurring at the typical unblooped splice is worse.

Posted by Brad Miller (Member # 2) on 02-02-2003, 09:26 PM:

Steve, ANY snip of a rise in modulation like that is noticeable. Go ask any professional audio editor. I picked that piece of dialogue since it was something I had convenient and because I knew it would be a demo that anyone could hear on crappy computer speakers. It does not have to be dialogue though.

I'm sorry you cannot comprehend how audio works, but this is the way it is. Do you really think audio editors look at the waveforms of the audio because it looks cool? No! Do manufacturers of software put waveforms on their programs because it makes it look more high tech? No again! The whole point is for the editors to be able to visually read the waves so they can better do their job. This is no different.

It is one thing to say that you do not personally find such an interruption distracting, but that is your opinion and you are trying to pass it off as fact. Jumps like that are very distracting to me, again...my opinion. However the fact remains that if you lose 2 frames worth of audio at a point where it is rapidly getting louder or softer (over a 2 frame period), it WILL be noticeable. Again, fact.

Posted by Steve Kraus (Member # 476) on 02-02-2003, 09:28 PM:

I'm sorry you feel the need to slight my knowledge and experience. My background is in post production.

Posted by Brad Miller (Member # 2) on 02-02-2003, 09:29 PM:

In this specific area of post production audio editing, I'm afraid it is lacking a bit. Your post is every bit as insulting to me as my post appears to be to you. I have spent much time in post production as well. I guess it all depends on how picky someone is. Blips like that are unacceptable to me. Heck, hairline scratches on the picture are acceptable to a lot of people, but I'll bet neither one of us would be too happy about it.

You asked and I gave you a quick example. I think we are done discussing this with each other. Fortunately you run a changeover theater and do not cut frames, so this really doesn't even apply. [Wink]

Posted by Steve Kraus (Member # 476) on 02-02-2003, 09:37 PM:

Insulted? Are you really insulted when someone respectfully disagrees with you?

Posted by Brad Miller (Member # 2) on 02-02-2003, 09:41 PM:

Pick whatever word you want Steve. Am I really angry about it? No. I used that word because that is how I interpreted you as interpreting my post before. If you want to nitpick words newsgroup style, email me as this is a waste of space here since we've both already stated our arguments.

Posted by John Pytlak (Member # 331) on 02-03-2003, 09:04 AM:

Darren Briggs wrote: "Analogue sound is 20 frames ahead of the picture.Just count the frames on your head leader, from any number laced in the gate to the diamond printed on the film which lines up with the optical sound head."

NO! The analog sound is printed 21 frames ahead of the corresponding picture frame, per standard SMPTE 40. The 20 frame offset shown on the threadup leader gives a 1/24 second advance to the sound, assuming it takes that long for the sound to travel from the screen speakers to the center of the audience.

Posted by Frank Angel (Member # 248) on 02-03-2003, 10:54 AM:

.....and that spec was written way back in the day when even neighborhood single screen theatres were huge compared to most of today's multiplexs. Nowadays, if you are threading for anything from a screening room to a modern, moderatly sized cinema, lipsync would be accurate with zero advance.

Frank

Posted by Mark J. Marshall (Member # 1409) on 02-03-2003, 11:51 AM:

When you do changeovers, how is this compensated for, do processors fake a slight fade between the two sources, or does the sound jump right from one reel to the next? It seems to me that even while doing effective changeovers is easy to do, getting them to be as EXACT as splicing reels together would be extrordinarilly difficult given power fluctuations, motor position, ramp up speed etc.

Posted by Dustin Mitchell (Member # 372) on 02-03-2003, 11:54 AM:

Not wanting to get in the middle of the Steve Kraus vs. Brad Miller ID Frame Deathmatch I'll just repeat the advice I give every time this thread comes up.

If you do leave ID frames, only leave one on the head of every real. There is no reason to have an ID frame on each side of the splice or each end of the reel to identify it, one is sufficient.

Posted by Leo Enticknap (Member # 534) on 02-04-2003, 02:27 AM:

Jeffry L. Johnson writes:

quote:
DTS is offset zero frames.

SDDS P side is offset zero frames. SDDS S side is 17 frames behind picture.

Then presumably that means that DTS and SDDS reproducers have to be positioned a minimum distance in front of the gate (i.e. the film passes the reproducer a minimum of X frames before the gate), and that without redesigning the encoding protocols used, a combined basement reader such as are available for Dolby digital and analogue will never be possible.

I wonder what the logic behind that decision was, given that it inherently limits the flexibility you have in installing the hardware?

Posted by Mathew Molloy (Member # 666) on 02-04-2003, 02:39 AM:

Mark Wrote (When you do changeovers, how is this compensated for, do processors fake a slight fade between the two sources, or does the sound jump right from one reel to the next?)

The sound does changeover from one reel to the next at exactly the same time the picture does. Unlike a splice with missing frames and the sound jumps nearly a second later. Therefore, even a mildly poorly accomplished reel change will appear natural.

As far as ramp up speed and timing; it's all based on how well you know your machines and how the cues are 'timed'. What I mean is that a film that is cut right has a certain beat to it when the reel change happens - you can feel when you need to hit the c/o button and voila! [Big Grin]

As much as I love running changeovers, a plattered presentation missing ZERO reference frames is the way to watch a movie. [eyes]

Posted by Duncan Smith (Member # 1556) on 02-04-2003, 01:16 PM:

I run 35mm changeover and also from a tower. With the amount of films I have moving through my projection box I would be completely lost without an ID frame at each end of the reel. I can't tell you how many times I have received a print with the wrong leaders attatched and in the wrong cans. If I assumed they were correct then I would have had far worse problems to contend with than 2 frames lost; more like 15 minutes while I try and put the film back in the right order following advice from a luckless customer who has seen the film before. If I don't find ID frames on the leader I call the distributor to get an ID on each reel, this has paid off more than a few times. I know it's regretable to lose any amount of image or sound content but the alternative is far worse. No-one ever asked for a refund or had a show cancelled because of a couple of frames at the end of a reel were missing (I hope not anyway).

Unfortunately this is the real world we live in where some people do not consider the full implications of what they do when making up a feature onto a cakestand or tower and those of us who proudly run changeover are left hanging.

It would be great if practises were agreed upon and maintained to everyones advantage. (I'll take off the rose tinted glasses now).

Posted by Steve Kraus (Member # 476) on 02-04-2003, 02:02 PM:

Analog changeover switching is done by the pre-amp card of the processor. I've never heard of one that does anything other than an instantaneous change but I suppose such could exist but I don't think a crossfade would be desirable since it's supposed to be just like a straight cut edit. By instaneous I mean instant when signalled. That signal may coincide with with the picture changeover but it need not be. My booth has separate buttons for picture and sound changeover so while they are normally operated together it is possible to change picture while hanging on the old reel's sound to finish a syllable.

Posted by John Pytlak (Member # 331) on 02-04-2003, 02:18 PM:

In some older installations (like the Grandview Drive-In where I worked as projectionist 1967-1970), the sound changeover was done by switching between the exciter lamps on the two projectors. The lag of the heavy filament exciter lamps gave a natural "cross-fade" between the reels.

Posted by Mathew Molloy (Member # 666) on 02-04-2003, 02:18 PM:

My Seattle booth (The Egyptian) had just that, seperate buttons for the audio c/o and the dowser. We also had a three-position switch for the sound format for each machine. (This was run as a festival booth quite often and shorts packages would go from one format to the next, to the next, to the next.)

I remember a couple times prefering to do one ahead of the other, but I don't remember why now. Perhaps I was just being 'artsy'.

I do remember an instance where I realized at the end of one short I started the next short a little too early (during the end credits of the current short) so I changed the sound over to catch the opening of the next short followed by a 'crossfade' with the hand dowsers. Looked quite nice, actually.

Posted by Scott Norwood (Member # 30) on 02-04-2003, 02:19 PM:

Don't forget places that use exciter-lamp changeover systems, where the sound does a quick fade-out/fade-in at each changeover. Some places feed a very small amount of current to the "idle" exciter to keep the filament warm enough to just barely glow, which allows it to come to full brightness rather quickly, making the sound dropout less noticeable.

Having separate picture/sound changeover controls is nice, but that seems uncommon in the New England area.

Posted by Brad Miller (Member # 2) on 02-04-2003, 02:28 PM:

Missing frames when building prints is definitely more of an issue than running on changeovers. The guys at the labs who are QCing the prints don't ever buildup the films and only see clean starts and stops. This is the same problem with so many policy trailers, where the audio starts in "dirty black" or in the countdown leader. No one ever remembers that easily over 95% of all films will have their leaders removed for projection, and in turn they never pay attention to the fact that the audio is ahead of the picture.

Posted by Steve Kraus (Member # 476) on 02-04-2003, 04:04 PM:

I remember once long ago being confused by the way an exciter lamp C/O circuit was wired in an ancient booth, that is, until I realized how elegantly simple it was. One side was a SPDT switch (regular electrical hardware type, sold as "3-way") with the DC fed to the pole and the two "throw" lines eventually leading to one and the other bulb...send power down one line or the other to select the projector. But first the two wires pay a visit to the other position where there is a DPDT switch wired in what might be called a polarity reversing configuration (electrical hardware switches expressly designed to do this are sold as "4-way" switches). This has the effect of either passing the power along on the same wire or flip flopping the lines to invert the selection made by the other switch. So flipping either switch changed which exciter was on. No muss, no fuss, no relays.

Posted by John Wilson (Member # 269) on 02-04-2003, 04:52 PM:

Just ran sneaks of Far From Heaven last weekend. The change from 2 to 3 has a guy talking going right through the changeover. If ID frames had been removed on building the print, there's no way that reel change would have worked.

When building and not leaving ID frames, I make sure the print is well identified by black marker on the edges of the film on both the print and the leader. It's tricky getting such small writing happening, but you get used to it.

I used to leave an ID frame, but then I tried it without... I'd never go back.

Posted by Karen Hultgren (Member # 395) on 02-04-2003, 07:37 PM:

Because DTS matches analog frame to frame, we have superior editing capabilities. Any size edit is seamlessly accomidated by DTS (as long as all discs with the needed digital audio are loaded). That is why DTS is most often used during the Academy sound "bake-offs" (for sound academy award nominations).

We have looked at "basement" readers and found them not feasible. You never know, though, we might incorporate them in future products.

Karen at DTS
khultgren@dtsonline.com

Posted by Scott Norwood (Member # 30) on 02-04-2003, 08:57 PM:

Would DTS even really benefit from a basement reader? The only advantage that I can see for SRD is that threading the cat. 699/700/701 takes a few extra seconds. The DTS reader is so easy to thread (and nearly impossible to thread incorrectly), that I don't really see the point, except maybe for installations that use enclosed reel magazines (and which aren't likely to have DTS, anyway).

Posted by Mark J. Marshall (Member # 1409) on 02-04-2003, 09:59 PM:

quote:
If ID frames had been removed on building the print, there's no way that reel change would have worked.

I'm thinking that in cases like that, a change over would be almost impossible to pull off correctly.... consistantly. Maybe I'm wrong, but it just seems like way too many things (some of which are out of your control) that have to be in EXACTLY the right place at the right time.

[ 02-06-2003, 08:11 AM: Message edited by: Mark J. Marshall ]

Posted by Ray Derrick (Member # 1001) on 02-09-2003, 06:12 AM:

I agree with Steve AND Brad. You can't tell by looking at it AND chopping a frame off the front of an audio "rise" will usually be audible. In fact you can clip substantially less than a frame and still hear a noticable cut-in on most audio material.

Regarding analog audio offset, if a film is correctly laced it will be perfectly in sync for audients sitting precisely 45.83333 feet back from the screen speakers (that's just under 14 metres for those living in enlightened countries). And anyone sitting 91.6666 feet back will hear the audio exactly one frame behind.

Our brain is less critical of an audio lag than a lead, so a one frame lead and two frame lag is about the limit of our tolerance for lip sync. So even in the largest indoor theater, the sound will appear in sync at the very back (provided the film is correctly laced). But the important thing is that the bigger (longer) the theater, the more critical you must be of the picture-to-audio lacing offset.

Posted by Steve Kraus (Member # 476) on 02-09-2003, 08:25 AM:

Let's say Moses supposes he wants to thread up non-erroneously and suppose that Moses takes into account the size of his theatre and a 1 frame-early sync built into the track but how do you suppose Moses should determine the picture position? Just before pull-down, just after, or in the middle of the stationary dwell period?

Posted by John Hawkinson (Member # 1135) on 02-09-2003, 09:28 AM:

That's of course a function of how Moses threads: the position of the intermittant at the time the constant-speed sprocket driving the film through the soundhead starts moving.

Let's assume that Moses threads by setting the intermittant to be right at the start of the dwell period ("just after pulldown"), because that's how people thread. Also assume that the "sound of one frame" should be heard starting 1/2-way through the pulldown to the frame, throughout the frame, and then ending 1/2-way through the pulldown to the next frame. We can conclude Moses is 1/2 of a frame (2 perfs) further along than he should be, so to compensate for his threading he should tighten the sound loop by two perfs relative to the position of the diamond.

Of course, if Moses threads out-of-frame and then cranks the framing knob...well, someone's gotta deal with that red sea runoff. D'you suppose he filed an EIS?

--jhawk

Posted by Gordon McLeod (Member # 33) on 02-09-2003, 09:52 AM:

If memory serves it is supposed to be in the centre of the dwell cycle that moses suposes that pix start is in the apperture block

Posted by Ray Derrick (Member # 1001) on 02-10-2003, 12:58 AM:

Reference to SMPTE 40-1997 states that there is +/- 1/2 frame tolerence on that 21 frame offset, so in reality the offset isn't all that accurate. Furthermore, Annex A of the same document suggests that the projectionist is at liberty to place the picture/sound synchronisation point anywhere in the cinema by varying the length of the threading path in the projector.

So much for standards.

Posted by Leo Enticknap (Member # 534) on 02-10-2003, 05:22 AM:

quote:
Would DTS even really benefit from a basement reader? The only advantage that I can see for SRD is that threading the cat. 699/700/701 takes a few extra seconds. The DTS reader is so easy to thread (and nearly impossible to thread incorrectly), that I don't really see the point, except maybe for installations that use enclosed reel magazines (and which aren't likely to have DTS, anyway).

The real advantage would be the ability to supply a pickup which reads all four formats (SVA/SRD/DTS/SDDS) in one unit, and which could be supplied as standard with a projector mechanism. However I can't think of any reason why this couldn't be above the gate, except that you'd need a means of setting a delay for the SVA.

Posted by Don Bruechert (Member # 1538) on 02-10-2003, 08:45 AM:

We got a letter in the mail a few days ago from some "Marshall" guy associated with Chicago - I forget the details. He was asking projectionsists to remove the absolute minimum of frames when building up the print because there are audio sequences that flow across the reel change. A bit too late for us as we already did our standard single reference frame when we built it up last week, but maybe a heads up to some of you other folks.

Posted by John Pytlak (Member # 331) on 02-10-2003, 08:56 AM:

That's Rob Marshall, the Director of the film:

http://us.imdb.com/Title?0299658

Posted by Christopher Dreiling (Member # 1596) on 02-11-2003, 01:18 PM:

So, is it acceptable to have an ID frame on the head and tail leader, but preferable to only have a ID frame on the head leader?

Chris

Posted by John Pytlak (Member # 331) on 02-11-2003, 02:03 PM:

The debate over ID frames goes on...and on...and on. [Roll Eyes]

In a perfect world, every frame of picture would be shown. In reality, ID frames provide verification of reel ID if leaders are not properly reattached. I favor a single ID frame at both head and tail, if it helps prevent later reel mixup, and no frames end up being discarded over the life of the print.

Posted by Don Bruechert (Member # 1538) on 02-11-2003, 02:22 PM:

Moses supposes the projectionist knowses to build up the new print adjudiously. But Moses he knowses the projectionist s'poses he will get away with it eventually. So Moses he goeses and thumbs his noses at all them that build up unethically. And Moses he knowses that that's how it goeses at the low end of the projection industry.

OK, sorry... I just couldn't resist... we just did that play last year even tho I can't remember the name of it.

Posted by Dustin Mitchell (Member # 372) on 02-11-2003, 04:57 PM:

Christopher,

While the standard practice is to leave an ID frame on the head and tail of a reel I see no reason for there to be one on the tail. You only need an ID frame on one side of the splice to positivly identify it and you only need one at one end of the reel to determine which reel it is. Losing too frames (one on the head and one of the tail) at each splice is completely unecessary.

Posted by Scott Norwood (Member # 30) on 02-11-2003, 06:13 PM:

But what if a print arrives with the reels wound tails-out? Then, each reel has to be rewound before one can determine that the leaders are indeed attached correctly.

Personally, I think that this whole topic is about splitting hairs--the much bigger issue in my mind is to ensure that frames don't get chopped by every theatre that plays the print. I've seen too many prints that were completely missing the original second set of cue marks (which would mean at least 22 missing frames) to be concerned with the loss of two reference frames per reel by the first theatre to play the print.

Posted by Frank Angel (Member # 248) on 02-11-2003, 06:39 PM:

"He was asking projectionsists to remove the absolute minimum of frames when building up the print because there are audio sequences that flow across the reel change"

Yeah....and whose fault is that, pray tell?

Like the tale of, who was it said that when they ran a preview of Ed Harris's directorial stint with the film POLLOCK, he sat in the booth constantly telling the projectionist to rack the framing up a bit then down a bit, never seemed satisfied with where the image was centered.

My take on that was, if he was so damn concerned about framing, then why didn't he hard-matte the thing in the release print exactly where he wanted it, instead hassling the projectionist for two hours? Did he think in the real world there would be even a living person in the booth to tweek the framing knob (which, as we all know, Brad and Joe have already welded it to the casing)? [Big Grin]

Frank

Posted by John Hawkinson (Member # 1135) on 02-11-2003, 10:01 PM:

So what if he screwed up; doesn't seem appropriate to criticize him too much for at least trying to make the best he can of an unfortunate situation.

--jhawk

Posted by Dustin Mitchell (Member # 372) on 02-11-2003, 10:33 PM:

If the print arrives tails out its going to need to be rewound anyways to be platterized or put on house reels for c/o booths.

And besides, there are only two kinds of theatres that would send prints out tails up:

-C/O houses, where either the leader hasn't been removed anyway or it is on correctly (one should hope so at leas), or
-Platter houses who tear down from the platter onto 6k's and then bring them to the bench. I would hope that if a theatre cares enough to inspect their prints on the way out (the only reason I can think of for tearing down at the bench instead of directly onto shipping reels from the platter) they would be able to re-attach the leaders properly

Posted by Daryl C. W. O'Shea (Member # 1303) on 02-11-2003, 11:23 PM:

Or a third way...

Disorganized c/o house that rewinds the print, forgetting that it needs to be broken down, then breaks down resulting in tails out.

Or a fourth way...

A c/o house with a limited number of 6000' take-up reels that rewinds the print so they can use their floating hub take-ups before the print gets broken down. Then breaking the print down, resulting in tails out.