pitch jnd's at low pitches

[Roy Patterson <rdp1@xxxxxxxxx>]

Before getting too concerned about how accurately you can measure fo in 
the lower octaves, perhaps you would like to consider how accurately one 
can hear notes in the lowest octaves.

The just-noticeable-difference for melodic pitch is rising rapidly in 
the 30 Hz region. So whatever pitch you are hearing, you are hearing it 
much less accurately at 30 Hz than at 60 Hz. As a rough guide

at 16 Hz, the jnd is over 6% (i.e., greater than a semi-tone)
at 32 Hz, the jnd is about 3% (greater than a quarter tone)
at 64 Hz, the jnd is under 1.5%.

So, around 32 Hz, you might be 90% correct when asked if a note is  B, C 
or C#, but you would not be 100% correct, and you would not be able to 
say whether the instrument was tuned accurately, if by that you mean to 
within 1%.
And with the lowest organ notes, around 16 Hz, although you may hear a 
pitch, you would be embarrassingly bad at saying whether it was B, C or C#.

We did some experiments on the lower limit of melodic pitch a few years 
ago, upon which I base these comments.

Krumbholz, K., Patterson, R.D. and Pressnitzer, D. (2000). The lower 
limit of pitch as determined by rate discrimination. J. Acoust. Soc. Am. 
108 1170-1180.

Pressnitzer, D., Patterson, R.D. and Krumbholz, K. (2001). The lower 
limit of melodic pitch. J. Acoust. Soc. Am. 109 2074-2084.

I can provide pdf's if anyone needs them.

Regards Roy P

--
Roy and Karalyn Patterson
17 Woodlands Road
Great Shelford, Cambridge, CB22 5LW

Phone  +44  1223  842310  (International format)
email	
rdp1@xxxxxxxxx
karalyn.patterson@xxxxxxxxxxxxxxxxx