NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/rec/audio/19135

Text File | 1993-01-22 | 4.1 KB | 82 lines

Newsgroups: rec.audio
Path: sparky!uunet!timbuk.cray.com!walter.cray.com!daw
From: [e-mail] (Dave Wright)
Subject: Re: Direct vs Reflecting speaker designs
Message-ID: <[e-mail]>
Lines: 71
Nntp-Posting-Host: nothing.cray.com
References: <[e-mail]> <[e-mail]>
Date: 22 Jan 93 10:04:22 CST

In article <[e-mail]> [e-mail] (Gabe M Wiener) writes:
>In article <[e-mail]> [e-mail] (Jim Nicholsen) writes:
>>
>>
>>The preceding would seem to say that room reflections are indeed a
>>"bad thing". However, increased reflections are listed as a "good thing"
>>in a couple of well regarded speakers (e.g. Definitive Tech "Bi-polar"
>>designs, some Magneplanar ads). Where does the truth (tm) lay?
>>
>
>It really depends on a number of things.
>
>If you design a speaker system that works via reflections...say The
>Definitive Tech bi-polar (as you mentioned), or the Miarge M-1, then
>you want a good deal of reflectivity in the live end of the room,
>because the design depends on that.  Ian Paisley up a API recommends
>that, in the ideal setup, the side walls should be reflective from the
>live end back to 2/3 of the way between the listener and speaker, so as
>to get the first reflection but no more.
>
>HOWEVER....when you're working with DSP, you generally do not want your
>room to be too reflective.  This is out of a need for *control*.  You can
>simulate reflectivity electronically.  You *cannot* make a reflective room
>sound dry.  Therfore, if you have t err in a DSP setup, you always want
>to err on the dry side.
>
>Some professional DSP units, such as the SigTech processor used for
>location-recording work, etc., is rated up to 50 ms....in other words,
>its DSP effect will work in a room with up to 50 ms of natural delay.
>After that, the unit can't exert enough control.

Of course a room with a 50ms Initial Time Delay Gap (ITDG) would give a very
clear echo and terrible to listen in. 50ms would translate to a travel
path of over 50ft more than the direct sound. A large room indeed.

THere are metrics for the quality of speech and music that take into
account the direct vs reflected sound.

For speech it is called Articulation Loss of Consonants %AL(cons)
that determines what
percentage of words at a listening position would be unintellegible.
Sound System Engineering (Don Davis) lays all this out.

For Music, the energy content is measured from 0-20ms, 20-80ms, and 80
to 120ms for metrics labeled C20, C80, C120  (Anyone know where papers
on this are?) I've looked in JAES and ASA but I understand it is European
in origin. With computers and 3D visualization metrics looking at the
direction/intensity of the sound in what is often refered to as ballon data.

Some of what is said above corresponds to the Live End/Dead End concept of
giving the listening room a ITDG longer the the recording so the ambience
in the recording will not be masked by the room. THis is one reason that
moving the speakers can have effects on imaging.

Being to refelctive is not the issue, the issue is the intial time delay
gap of the room. the 2/3rds rule mentioned is ok for that 1st order
reflection but then it should come in at least 20ms after the direct sound.
In studio control rooms like LEDE, this is done with what were refered to
as Haas Kickers. BUt after the 1st order reflections you then want the
diffuse sound field to come in. This means the back of the room should be
live not dead. The front of the room was dead to extend the ITDG so you
can here the recording but also so early reflections don't destroy the
frequency response of the speakers. The old Westlake compression ceiling
was a common culprit to bringing in refelctions to early and putting wide
notches in the frequency response at the listening postion. Finally to get
the spaciousness of the room to sound larger, Diffusors such as RPG
Diffusion was added to break the 1st order reflections coming from the back
of the room into multiple order reflections creating a diffuse sound field
such as you would have in a concert hall.

Dave W