Hi,
Earlier I questioned in this forum how to compute the graphics for a 2D-spectrogram most efficiently
http://www.opengl.org/discussion_boards/ubb/ultimatebb.php?ubb=get_topic;f=2;t=020999
Now I would like to display the spectrogram in 3D so that the audio input is drawn as a moving landscape (in the frequency domain that is); intially in a grid/wire fashion(guess
thousands of people have done this before). So my question is how to do the best way in terms of computational complexity.
The FFT procesing I already have running - so now I have to take care of the drawing.
Initially I thought that maybe I should draw the whole 3D landscape for every arriving FFT frame, but then I thought
that it could be done similar to the spectrogram problem (which was solved by applying an texture offset to the time direction such that it could imitate a circular buffer)
Is it therefore possible to allow the offset in the time direction to increase in infinite time so that I only have to draw the lines from the last FFT frame. What will happen when
the offset variable wraps around- can I control this somehow?
How I’m making myself clear
Thomas