arnyk wrote: 
> More substance?
> 
> http://www.aes.org/journal/online/JAES_V65/5/#paper1
> 
> Modeling Perceptual Characteristics of Loudspeaker Reproduction in a
> Stereo Setup
> 
> http://www.aes.org/tmpFiles/JAES/20170602/JAES_V65_5_PG356hirez.pdf

Hi Arny!

Sorry to be late responding but your reference paper was a tough read.
And I found it unconvincing - the 3 esteemed academics started off with
6 subjective measures of loudspeaker performance that they wanted to
measure, dumped 2 of them & then combined 2 more into 1, leaving 3 of
the original 6!

There was an awful lot of fiddling about before they arrived at their
"results" based on a panel of 11 (odd number in both senses - did they
dump 1 of those as well?) "experienced" listeners, & a caution that
their experiment could only be performed by "experts".

Outlying results appear to have been arbitrarily disregarded - if you
get rid of the data points that DON'T fit your regression curve, you
WILL improve the correlation coefficient. That was 'O' Level statistics
back in my day.

A massive number of reference papers, not explained & no doubt equally
dry, & no actual DATA. They casually mention the loudspeakers chosen but
don't indicate which made it to the "evaluation" stage, let alone
identify the data plots.

The measured frequency response plots for the loudspeakers are so
similar as to suggest that the room acoustic was dominant, whatever a
ITS2464 or whatever room is. I know that I haven't got one...

Masses of DSP & then they suggest that they've found the Holy Grail of
loudspeaker measurement. Yawn!

I realise that the paper appeared in the May issue, but I actually
checked more than once that it wasn't actually dated 1st April!

Repeatable experiment? Not in my book...

Dave :cool:


------------------------------------------------------------------------
Golden Earring's Profile: http://forums.slimdevices.com/member.php?userid=66646
View this thread: http://forums.slimdevices.com/showthread.php?t=106519

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to