Volume vs. Pitch : seeing what is heard
Posted: Mon Jan 04, 2016 9:56 pm
Hi Eric,
I am using Magic to create images of recorded songs. I want to get the most accurate image of what I hear as the music plays. I am using a waveform to drive several parallel threads of geometry, each the same except for frequency band. The result is an image that changes with the audio spectral content.
Each thread starts with a scale module, driven by the waveform output, which feeds the rest of that thread. I have two programs, each the same except for the way the scale module is set-up. The first program uses the same scale factor for each of the different frequency bands. The second program adds a multiplier in each scale module that is different for each frequency band. The multipliers are chosen to follow the log response of the ear, similar to the PITCH function, in that the low frequencies have a small multiplier vs, high frequencies with a high multiplier. The result is a significant visual difference in the images produced by the two programs using the same song. As you would expect, the straight scale factor program produces images with much greater low-frequency visibility, while the "ear compensated" program produces mainly mid & high frequency emphasis. So, is it correct to say that the first, straight-scale-factor, program creates an image that reflects the physical sound pressure level in the air, whereas the second, ear compensated, program creates an image closer to what is (theoretically) heard? OR...am I misunderstanding something?
If you like, I'll send you the set-ups, but I'd rather not post them.
Thanks!
/Jim
I am using Magic to create images of recorded songs. I want to get the most accurate image of what I hear as the music plays. I am using a waveform to drive several parallel threads of geometry, each the same except for frequency band. The result is an image that changes with the audio spectral content.
Each thread starts with a scale module, driven by the waveform output, which feeds the rest of that thread. I have two programs, each the same except for the way the scale module is set-up. The first program uses the same scale factor for each of the different frequency bands. The second program adds a multiplier in each scale module that is different for each frequency band. The multipliers are chosen to follow the log response of the ear, similar to the PITCH function, in that the low frequencies have a small multiplier vs, high frequencies with a high multiplier. The result is a significant visual difference in the images produced by the two programs using the same song. As you would expect, the straight scale factor program produces images with much greater low-frequency visibility, while the "ear compensated" program produces mainly mid & high frequency emphasis. So, is it correct to say that the first, straight-scale-factor, program creates an image that reflects the physical sound pressure level in the air, whereas the second, ear compensated, program creates an image closer to what is (theoretically) heard? OR...am I misunderstanding something?
If you like, I'll send you the set-ups, but I'd rather not post them.
Thanks!
/Jim