Standing waves are generated when two sound waves travelling in opposite directions, i.e. a sound wave and its reflection, are superposed. In a room, they only occur between parallel and opposite walls (to keep it simple). At the wall a phase change of 180 degrees occurs so when the incident and reflected waves (pink and blue) are superposed amplitude cancellation and enhancement occurs:
The physical location in the room of the nulls and peaks of the black wave does not change (see red dots), that's why the (black) wave is called a standing wave.
When a room dimension is equal to multiples of half the wavelength, we have a resonance which means that at both walls pressure maxima are present so that the sound pressure of this particular standing wave or room mode is the highest possible and higher than the sound pressure of an ordinary standing wave.
Standing waves and room modes exist at all frequencies throughout the audio range. The number of room modes per third octave, for instance, increases with frequency and above a threshold frequeny (commonly called Schroeder frequency) is that high that the individual mode is no longer perceivable as such.
So at a particular location in a room you will measure a pressure peak of the 800 Hz standing wave (I suppose you do mean 800 and not 80, but for the 80 Hz wave the same applies). The physical distance between peaks and nulls is one quarter of the wavelength, so at 800 Hz that is 10.7 cm (4.22 inch). It's possible that you can distinguish peaks and nulls at this frequency when walking through the room but I doubt it.
When a sound wave travels toward a wall at an angle, say the side wall, the reflection from that wall arrives at measuring position some time after the initial wave. The two waves (initial wave + reflection) are superposed and because of the time delay, again, cancellations and enhancement occurs, a structure resembling a comb is generated:
The frequencies at which the peaks and nulls of this structure occur depend of the time delay T (in milliseconds). Peaks occur at 1000 x n/T, nulls occur at 1000 x (0.5n/T). The amplitude of the peaks and nulls depends on the level of the reflection. Comb filters may cause audible coloration, but not necessarily so: thresholds of audibility exist so the mere fact of comb filter structures appearing in a measurement does not say anything about their audibility. The more reflections there are and the less regularly they are distributed on the time scale, the less audible is the coloration.
Unless there is a solid correlation between what can be measured and what is actually perceived, I would not worry too much about peaks and valleys in graphs, as long as the music sounds fine to your ears. If on the other hand you do perceive a problem when listening to music, measurements will help you to identify the problem area.
Klaus
