hello I understand that our ears perceive midrange sounds as louder than very low or very high ones. I am wondering if there is some sort of simple method or formula or object that compensates for the perceived difference in volume, so that if i have say a patch randomly generating osc~ frequencies between 50 and 15000 hz, the volume [*~] of each frequency can be adjusted so that they all have the same overall presence? thanks in advance. S
There's this: https://en.wikipedia.org/wiki/Equal-loudness_contour But I tend to think of massive sinewave polyphony as a kind of "artifact". To me it's a bit like modeling Felix the Cat in 3d. Still, I'd be interested in hearing the difference, esp. in a patch that toggles the curve on/off. -Jonathan
On 08/24/2015 05:29 PM, school shoes wrote:
hello I understand that our ears perceive midrange sounds as louder than very low or very high ones. I am wondering if there is some sort of simple method or formula or object that compensates for the perceived difference in volume, so that if i have say a patch randomly generating osc~ frequencies between 50 and 15000 hz, the volume [*~] of each frequency can be adjusted so that they all have the same overall presence? thanks in advance.
there's [mtx_phon_curves] (in iemmatrix) which uses an approximation formula for the phon curves. being a matrix object¹, it takes a vector of sinusoidal components and outputs the relative gains.
gfmrsa IOhannes
¹ i have no clue why this is in the iemmatrix toolbox, and not a general scalar object.
great thanks for the leads
To: pd-list@lists.iem.at From: zmoelnig@iem.at Date: Tue, 25 Aug 2015 10:57:50 +0200 Subject: Re: [PD] perceived loudness
On 08/24/2015 05:29 PM, school shoes wrote:
hello I understand that our ears perceive midrange sounds as louder than very low or very high ones. I am wondering if there is some sort of simple method or formula or object that compensates for the perceived difference in volume, so that if i have say a patch randomly generating osc~ frequencies between 50 and 15000 hz, the volume [*~] of each frequency can be adjusted so that they all have the same overall presence? thanks in advance.
there's [mtx_phon_curves] (in iemmatrix) which uses an approximation formula for the phon curves. being a matrix object¹, it takes a vector of sinusoidal components and outputs the relative gains.
gfmrsa IOhannes
¹ i have no clue why this is in the iemmatrix toolbox, and not a general scalar object.
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
IOhannes m zmölnig -
Jonathan Wilkes -
school shoes