Comparison of MLSSA Resolution

Introduction

The test is to compare and find out - what resolution is sufficient - not too much or too little. Presented below are 12 graphs, 1st 6 are smooth using 1/20 octave method. While the last 6 is not smoothed.

65536 = 16 bits                                                32768 = 15 bits

16384 = 14 bits                                                8192 = 13 bits

4096 = 12 bits                                                  3030

Not smoothed, above smoothed using 1/20 octave

65536 = 16 bits                                                32768 = 15 bits

16384 = 14 bits                                                8192 = 13 bits

4096 = 12 bits                                      3030

It is noticed that smoothing is required for easy viewing, while non-smoothed curve simply produce too much detail, even when resolution is down to 3030 points.

Comparing 3030 points vs. 65536 points. The questions now would be - what is the frequency of interest under investigation. If the overall performance is required, and knowing the product / loudspeaker does not fluctuate at low frequency, therefore 3030 is high sufficient and produce very detailed information down to 200Hz.

However is sub-woofer studies / critical performance is under investigation, then MAX sampling frequency may be reduce and maintain around 3030 points of resolution. This would be more ideal use of computing power efficientcy.

Conclusion

3030 points of resolution is sufficient. Infact more, however it allows the screen to automatically adjust to show fom 20Hz to 20kHz, therefore it is used to safe post-processing studying time.

ACOUSTICS SECTION             AUDIO SECTION             CAR SECTION

HOME - Technical Website for Acoustics, Audio and Car

  Malaysia Boleh