This post was first published in Feb 2014 at projectstudiohandbook.com/PSHforum
Bi-polar activity
Postby FairFax » Fri Mar 14, 2014 5:52 pm
Understanding why something is the way it is can be rewarding, but not necessarily useful.
A common misconception amongst audio novices when looking at waveform diagrams, is to assume that increases above the central horizontal time axis represent increases in amplitude, and those below represent decreases.
This is an easy mistake to make because we are used to reading diagrams that show a single property increasing upwards and decreasing downwards. In fact we use diagrams that show exactly this when we want to look at audio compression or EQ filter action diagrammatically.
Open up your favourite EQ plug-in and compare it to the visualisation of any .wav or .aiff file in your wave editor, and you will see that this is so.
Waveform diagrams employ a bi-polar axis to show amplitude over time. The three axis are therefore ..
1) Horizontal axis (left to right) = time
2) Vertical access above the time axis = increasing pressure (compression) and increasing amplitude
3) Vertical axis below the time axis = decreasing pressure (rarefaction) and increasing amplitude
Not until you understand that sound is changing air pressure energy, oscillating between higher than normal and lower than normal atmospheric pressure, does this make sense. Silence is, of course, when there is no pressure change. Whenever air pressure change is happening, those inner ear hairs move back and forth and fire off signals to our brains.
This means that when we edit digital audio files, not only must our edit points be at zero crossing points (where the changing pressure crosses the time line at silence), but also that we cannot join two files together if at the edit point, the preceding file it is returning from compression, and the following entering compression.
Fortunately, our audio editing software is smart now and compensates automatically by moving the edit points to appropriate points in the cycles. Some of us remember when this wasn't the case and it took some time to get things right and avoid clicks. I think the Akai S900 sampler may have been the first device to feature a (hit and miss) automation button to analyse the waveform when you were trying to loop a short sample.
The bi-polar nature of amplitude also has significant implications for phase relationships between multiple waves of the same frequency, such as occur when we make multi-mic'd or stereo recordings, and there are powerful tools to analyse and correct problems. It can't be too long before we have software that does this without even letting us know.
It interesting to see how advanced technology can make problems seem as though they don't exist. Perhaps it doesn't matter that we don't know what's going on the background (providing the NSA aren't involved). That is of course, unless you hate the feeling that you aren't in control. Which brings us back to where we started.
Cycle end point reached.
Thanks for reading.
Bi-polar activity
Postby FairFax » Fri Mar 14, 2014 5:52 pm
Understanding why something is the way it is can be rewarding, but not necessarily useful.
A common misconception amongst audio novices when looking at waveform diagrams, is to assume that increases above the central horizontal time axis represent increases in amplitude, and those below represent decreases.
This is an easy mistake to make because we are used to reading diagrams that show a single property increasing upwards and decreasing downwards. In fact we use diagrams that show exactly this when we want to look at audio compression or EQ filter action diagrammatically.
Open up your favourite EQ plug-in and compare it to the visualisation of any .wav or .aiff file in your wave editor, and you will see that this is so.
Waveform diagrams employ a bi-polar axis to show amplitude over time. The three axis are therefore ..
1) Horizontal axis (left to right) = time
2) Vertical access above the time axis = increasing pressure (compression) and increasing amplitude
3) Vertical axis below the time axis = decreasing pressure (rarefaction) and increasing amplitude
Not until you understand that sound is changing air pressure energy, oscillating between higher than normal and lower than normal atmospheric pressure, does this make sense. Silence is, of course, when there is no pressure change. Whenever air pressure change is happening, those inner ear hairs move back and forth and fire off signals to our brains.
This means that when we edit digital audio files, not only must our edit points be at zero crossing points (where the changing pressure crosses the time line at silence), but also that we cannot join two files together if at the edit point, the preceding file it is returning from compression, and the following entering compression.
Fortunately, our audio editing software is smart now and compensates automatically by moving the edit points to appropriate points in the cycles. Some of us remember when this wasn't the case and it took some time to get things right and avoid clicks. I think the Akai S900 sampler may have been the first device to feature a (hit and miss) automation button to analyse the waveform when you were trying to loop a short sample.
The bi-polar nature of amplitude also has significant implications for phase relationships between multiple waves of the same frequency, such as occur when we make multi-mic'd or stereo recordings, and there are powerful tools to analyse and correct problems. It can't be too long before we have software that does this without even letting us know.
It interesting to see how advanced technology can make problems seem as though they don't exist. Perhaps it doesn't matter that we don't know what's going on the background (providing the NSA aren't involved). That is of course, unless you hate the feeling that you aren't in control. Which brings us back to where we started.
Cycle end point reached.
Thanks for reading.
No comments:
Post a Comment