I thought a post about stereo was called for because there are some serious problems with stereo that many of you are probably unaware of. Some context and explanation: Many of you on this forum will at some stage be looking to enter film festivals. Most film festivals, certainly all the more major ones, take place in commercial cinemas and commercial cinemas virtually always support a range of channel formats:
1.0 Dolby, Mono (C)
2.0 Dolby, Stereo (L, R)
3.0 DCP, Dolby (L, C, R)
4.0 DCP, Dolby Stereo/Surround (L, C, R, S)
5.0 DCP (L, C, R, Ls, Rs)
5.1 DCP, Dolby Digital, DTS, SDDS (L, C, R, Ls, Rs, LFE)
Bigger cinemas (major film festivals), often also support the formats:
6.1 DCP, Dolby EX, DTS ES, SDDS (L, C, R, Ls, Rs, Cs, LFE)
7.1 DCP, Dolby 7.1, SDDS (L, Lc, C, Rc, R, Ls, Rs, LFE)
A couple of questions: 1. Of all these channel formats which do you think works the least well and has never been considered a film format? 2. Of these formats, which do micro/no/low budget indie filmmakers use most commonly? I'll give you a clue, the answer to both questions is the same and is also in the title of this post!
Before anyone starts screaming "BS", I'm not saying stereo is terrible format, it's a very clever and very good format for many audio applications, just not for films destined to be screened in a cinema. The reason for this lies in how stereo manipulates the way we perceive sound (psychoacoustics) to create the illusion of a sound-field. The way we are able to tell where a sound is coming from is quite complex but a simple explanation is: When, for example, a sound occurs to the left of us, the sound waves hit our left ear a tiny fraction of a second before they hit our right ear. Our brain subconsciously works out this arrival time (phase) difference and converts it into the perception of position. So even with our eyes shut, we can tell where a sound is coming from to an accuracy of about half a degree. Of all the formats I listed, Stereo is the odd one out, it's the only one which doesn't have a centre (C) channel. When you pan a sound to the centre in stereo, what you are actually doing is playing the same sound out of both speakers simultaneously. Providing you are sitting equidistant (in the middle) of the two speakers, the sound from the left speaker will hit your left ear at the same time as the sound from the right speaker hits your right ear and as far as the brain is concerned, the only way the same sound can hit both your ears simultaneously is if the sound is coming from directly in front of you (in the centre). This illusionary centre position is called the Phantom Centre.
All very clever and effective but there's an obvious problem; if we are sitting closer to one speaker than the other, won't the sound arrive earlier in the ear closest to that speaker and break or distort the illusion? The answer is "yes" but when we are listening to our home stereo system, TV or computer speakers, the speakers are only a few feet apart and we are only sitting a few feet away, so the stereo image isn't very wide to start with and the distortion of the illusion probably won't be particularly dramatic. However, as the room gets bigger and the speakers are positioned further apart, this positional distortion becomes more pronounced. Take a look at this diagram:
If we imagine the room in the diagram to be a cinema, the speakers may be 50ft or more apart. For various reasons, including the fact that the characters talking tend to be in the centre of frame, dialogue is almost always panned to the centre. If you were sitting in a cinema in the position of the Blue person (or anywhere on the same vertical plane), the perceived position of the dialogue would line up with the visual position of the character talking. But our Red audience member has a problem, he/she is going to hear the character's dialogue coming from about 20ft to the left of where they expect (the character's on screen position). This will sound quite off-putting and is likely to disengage them from the film, a filmmaking disaster! There are only two solutions: 1. Only allow 10 or so people at a time into the cinema to watch your film and make sure they all sit along that vertical plane in the middle or 2. Add a speaker to the centre of the screen. With a centre speaker, no matter where you are sitting, the sound will always be perceived to be coming from the centre and will always line up with the character's position and audience expectation. Obviously, only the second solution is sensible and that's why all the film channel formats have a centre speaker and why stereo is not considered to be a film format!
There's a few downsides to this centre speaker solution though. The first and most obvious is the extra cost of another output channel and speaker. The second problem is how to deal with ambiances, sound FX, reverbs and music which are recorded in stereo. If we just output these stereo sounds to the left and right speakers aren't we going to have the same problem we had with the two speaker stereo system? And, if we want to pan a sound to a position other than the centre, left or right, don't we have to rely on a phantom position again? The answer to both these questions is "yes" but there are tools which provide a solution when we have more than two channels to play with. Unfortunately though, you need knowledge, experience and a monitoring environment the same as a cinema, in order to judge when and how much of these tools you need to apply. The final consideration is that with the added cost and mixing complexity of 3.0, we're already over half way towards all the advantages and the worldwide audio standard for cinema, full 5.1 surround.
So what can the indie filmmaker do when screening at a film festival in a cinema? There's only 5 possible options, in order of preference they are:
1. Get the budget together for a proper professional 5.1 surround mix.
2. Get a slightly smaller budget together for a professional 3.0 mix.
3. Buy yourself a 3.0 system, learn to use divergence tools and hope you are applying them appropriately.
4. Carry on working with your stereo system, acknowledging the problem and trying where possible to minimise it's effects.
5. Pretend you never read this post or that I don't know what I'm talking about and completely ignore the whole thing!
I realise that for the vast majority, options 1, 2 and 3 are not viable options but at least after reading this post you are aware of the problem, will consider it's implications and investigate/experiment with ways to minimise it a little, thereby taking option 4 rather than option 5.
Hope this has been useful? Ask away if you have any questions.
G
EDIT: Changed the listing above for Mono and Stereo. DCP only supports 3.0 formats and higher.
1.0 Dolby, Mono (C)
2.0 Dolby, Stereo (L, R)
3.0 DCP, Dolby (L, C, R)
4.0 DCP, Dolby Stereo/Surround (L, C, R, S)
5.0 DCP (L, C, R, Ls, Rs)
5.1 DCP, Dolby Digital, DTS, SDDS (L, C, R, Ls, Rs, LFE)
Bigger cinemas (major film festivals), often also support the formats:
6.1 DCP, Dolby EX, DTS ES, SDDS (L, C, R, Ls, Rs, Cs, LFE)
7.1 DCP, Dolby 7.1, SDDS (L, Lc, C, Rc, R, Ls, Rs, LFE)
A couple of questions: 1. Of all these channel formats which do you think works the least well and has never been considered a film format? 2. Of these formats, which do micro/no/low budget indie filmmakers use most commonly? I'll give you a clue, the answer to both questions is the same and is also in the title of this post!
Before anyone starts screaming "BS", I'm not saying stereo is terrible format, it's a very clever and very good format for many audio applications, just not for films destined to be screened in a cinema. The reason for this lies in how stereo manipulates the way we perceive sound (psychoacoustics) to create the illusion of a sound-field. The way we are able to tell where a sound is coming from is quite complex but a simple explanation is: When, for example, a sound occurs to the left of us, the sound waves hit our left ear a tiny fraction of a second before they hit our right ear. Our brain subconsciously works out this arrival time (phase) difference and converts it into the perception of position. So even with our eyes shut, we can tell where a sound is coming from to an accuracy of about half a degree. Of all the formats I listed, Stereo is the odd one out, it's the only one which doesn't have a centre (C) channel. When you pan a sound to the centre in stereo, what you are actually doing is playing the same sound out of both speakers simultaneously. Providing you are sitting equidistant (in the middle) of the two speakers, the sound from the left speaker will hit your left ear at the same time as the sound from the right speaker hits your right ear and as far as the brain is concerned, the only way the same sound can hit both your ears simultaneously is if the sound is coming from directly in front of you (in the centre). This illusionary centre position is called the Phantom Centre.
All very clever and effective but there's an obvious problem; if we are sitting closer to one speaker than the other, won't the sound arrive earlier in the ear closest to that speaker and break or distort the illusion? The answer is "yes" but when we are listening to our home stereo system, TV or computer speakers, the speakers are only a few feet apart and we are only sitting a few feet away, so the stereo image isn't very wide to start with and the distortion of the illusion probably won't be particularly dramatic. However, as the room gets bigger and the speakers are positioned further apart, this positional distortion becomes more pronounced. Take a look at this diagram:
If we imagine the room in the diagram to be a cinema, the speakers may be 50ft or more apart. For various reasons, including the fact that the characters talking tend to be in the centre of frame, dialogue is almost always panned to the centre. If you were sitting in a cinema in the position of the Blue person (or anywhere on the same vertical plane), the perceived position of the dialogue would line up with the visual position of the character talking. But our Red audience member has a problem, he/she is going to hear the character's dialogue coming from about 20ft to the left of where they expect (the character's on screen position). This will sound quite off-putting and is likely to disengage them from the film, a filmmaking disaster! There are only two solutions: 1. Only allow 10 or so people at a time into the cinema to watch your film and make sure they all sit along that vertical plane in the middle or 2. Add a speaker to the centre of the screen. With a centre speaker, no matter where you are sitting, the sound will always be perceived to be coming from the centre and will always line up with the character's position and audience expectation. Obviously, only the second solution is sensible and that's why all the film channel formats have a centre speaker and why stereo is not considered to be a film format!
There's a few downsides to this centre speaker solution though. The first and most obvious is the extra cost of another output channel and speaker. The second problem is how to deal with ambiances, sound FX, reverbs and music which are recorded in stereo. If we just output these stereo sounds to the left and right speakers aren't we going to have the same problem we had with the two speaker stereo system? And, if we want to pan a sound to a position other than the centre, left or right, don't we have to rely on a phantom position again? The answer to both these questions is "yes" but there are tools which provide a solution when we have more than two channels to play with. Unfortunately though, you need knowledge, experience and a monitoring environment the same as a cinema, in order to judge when and how much of these tools you need to apply. The final consideration is that with the added cost and mixing complexity of 3.0, we're already over half way towards all the advantages and the worldwide audio standard for cinema, full 5.1 surround.
So what can the indie filmmaker do when screening at a film festival in a cinema? There's only 5 possible options, in order of preference they are:
1. Get the budget together for a proper professional 5.1 surround mix.
2. Get a slightly smaller budget together for a professional 3.0 mix.
3. Buy yourself a 3.0 system, learn to use divergence tools and hope you are applying them appropriately.
4. Carry on working with your stereo system, acknowledging the problem and trying where possible to minimise it's effects.
5. Pretend you never read this post or that I don't know what I'm talking about and completely ignore the whole thing!
I realise that for the vast majority, options 1, 2 and 3 are not viable options but at least after reading this post you are aware of the problem, will consider it's implications and investigate/experiment with ways to minimise it a little, thereby taking option 4 rather than option 5.
Hope this has been useful? Ask away if you have any questions.
G
EDIT: Changed the listing above for Mono and Stereo. DCP only supports 3.0 formats and higher.
Last edited:
