Difference between revisions of "Frequency analysis"

From Hydrogenaudio Knowledgebase
Jump to: navigation, search
m (Updated Cool Edit link to Adobe Audition)
m (Minor fixes. See diff)
Line 1: Line 1:
 
;Why Shouldn't We Use Frequency or Spectral Graphs to Compare Audio Codecs?
 
;Why Shouldn't We Use Frequency or Spectral Graphs to Compare Audio Codecs?
  
# You can't know precisely what is audible and what is not audible just by looking graphs.
+
# You can't know precisely what is audible and what is not audible just by looking at graphs. ''(Remember: It's '''music''' that we are talking about.)''
 
# [[FFT]] [[frequency]] response graphs are rough and can't tell practically anything about the temporal (short time) accuracy.
 
# [[FFT]] [[frequency]] response graphs are rough and can't tell practically anything about the temporal (short time) accuracy.
 
# All psychoacoustic models are different. What can look good in graphs, doesn't necessarely sound too good and vice versa.
 
# All psychoacoustic models are different. What can look good in graphs, doesn't necessarely sound too good and vice versa.
# Graphical analysis is often useful, but only when used together with listening testing. The most useful view is the sound spectra view (spectral view in [[Adobe Audition]]).
+
# Graphical analysis is often useful, but only when used together with a listening test. The most useful view is the sound spectra view (spectral view in [[Adobe Audition]]).
# Blind listening testing ([[ABX]]) is the only method that can really be justified when testing quality. Blind testing, when enough propability is reached, is not affected by anykind of bias or placebo.
+
# Blind listening testing ([[ABX]]) is the only method that can really be justified when testing quality. Blind testing, when enough probability is reached, is not affected by anykind of bias or [[placebo effect]].
# Difference/substraction/inverse-mixing method is not conlusive either
+
# Difference/substraction/inverse-mixing method is not conclusive either.
  
* [http://www.hydrogenaudio.org/index.php?act=ST&f=1&t=4734 More discussion]
+
 
* [http://www.hydrogenaudio.org/index.php?act=ST&f=1&t=754 More discussion]
+
;See the following links for more discussion:
* [http://www.hydrogenaudio.org/index.php?act=ST&f=15&t=7834 More discussion]
+
* [http://www.hydrogenaudio.org/index.php?act=ST&f=1&t=4734 Why subtracting one waveform from another is not a good analysis of encoding]
 +
* [http://www.hydrogenaudio.org/index.php?act=ST&f=1&t=754 Why merely comparing waveforms is not really effective]
 +
* [http://www.hydrogenaudio.org/index.php?act=ST&f=15&t=7834 Why waveforms cannot be used to deduce quality, but may be useful to know what the encoder is doing]

Revision as of 21:09, 23 March 2006

Why Shouldn't We Use Frequency or Spectral Graphs to Compare Audio Codecs?
  1. You can't know precisely what is audible and what is not audible just by looking at graphs. (Remember: It's music that we are talking about.)
  2. FFT frequency response graphs are rough and can't tell practically anything about the temporal (short time) accuracy.
  3. All psychoacoustic models are different. What can look good in graphs, doesn't necessarely sound too good and vice versa.
  4. Graphical analysis is often useful, but only when used together with a listening test. The most useful view is the sound spectra view (spectral view in Adobe Audition).
  5. Blind listening testing (ABX) is the only method that can really be justified when testing quality. Blind testing, when enough probability is reached, is not affected by anykind of bias or placebo effect.
  6. Difference/substraction/inverse-mixing method is not conclusive either.


See the following links for more discussion