[icecast] automatic gain control

Sean /The RIMBoy/ sean at rimboy.com
Wed Nov 14 07:41:21 PST 2001



On Wed, 14 Nov 2001, Tim Pozar wrote:

> On Wed, Nov 14, 2001 at 08:36:36AM -0800, William Goldsmith wrote:
> > >From what you describe, your comp/limiter can't possibly be working
> > correctly. It should be the last unit in line before the sound card, and
> > needs to be adjusted properly. You also need to balance the levels on your
> > mixing board (so that the correc t level comes at predictable place on the
> > slider). It might be worthwhile to find someone with some sound-mixing or
> > radio engineering experience to help you out. Sounds like you're missing
> > some very basic stuff.
> 
> I have to agree with Bill on this.  Worst case your compressor /
> limitor should be getting things in the ball park.  If you need
> some help with this, drop me a line.  (I am a broadcast engineer.)

I agree also.  Furthermore, I think the original author mentioned why
cannot he get something like the ACG on a Minidisc?  

Couple of issues with that.  At the end of the day, ACG really sucks, no
matter how good you get it.  Compressors suck to a certain extent, but ACG
gives the end user no real control, it's just there.  Furthermore,
Minidiscs employ the ACG so that you'll get better sounding
MD's.  Afterall, you're throwing away quite a bit of info since you're
doing compression (ATRAC I believe, and it took about 2 revisions of MD
before it got good enough for people to stop laughing, hard).  Since
you're thowing away info, you really want it normalized and at the peak
level so that when the compression of the format kicks in you're not
loosing too much (headroom).  

Again, it's implemented in hardware.  It's fast and dedicated.  With
computers, they might be fast, but they're a jack of all trades.  You're
trading away specific design purpose for flexibility.  Thus, you have to
code up what you need to get the job done.  

See below:

> > Software dynamics processing is more involved than you might think. And
> > you'd be trying to use software to solve what is basically a hardware/user
> > problem.

Hit the nail on the head.  What is happening in the point of sale system
is starting to happen in the audio industry, why design specific hardware
when all we really need is a hardware base, we'll take care of the rest
(interface, actions, functions) in software. 

Unfortunately for audio that often means the digital domain.  For some
people that's fine.  For other, the analog chain is needed to keep from
sucking the life out of a work.

> > Furman also makes total crap.

With the exception of their rackmount power distro gear, but I'm sure
someone has some opinions on that too.  At the time it was just what the
doctor ordered for a number of people.

Furman is a relative newcomer to the effect processing area, are they
not?  It's been awhile since I was gear drooling.  I really need to start
doing that again. :/

> You'd be much better off with one of the newer
> > Behringer units. Cheap but very functional. Though that doesn't negate the
> > need to take care of the basics mentioned above.
> 
> I have had some minor success with the Behringer units.  I bought

I've unfortunately not had any experience with the Behringer's.

> I may have mentioned this on the list in the past, but Aphex makes
> some rather nice boxes that do very well by having a wide dynamic
> range and very little artifacts.  You can usually find the Compellor
> (compressor) and Dominator (limiter) on Ebay for about $300 and
> $500 respectively.

They also make some other really neato effects gear.

IIRC, I enjoyed using the dbx 170's we had in the studio, but I think
those are more geared towards a studio / esoteric uses than
broadcast.  I've seen the Aphex's in a number of radio stations.  Overall
I think they'd be a wise coice.  

Sean...


-- 
Believing I had supernatural powers, I slammed into a brick wall.
	--Paul Simon
_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
KG4NRC
www.rimboy.com		<-- Your source for the crap you know you need.

--- >8 ----
List archives:  http://www.xiph.org/archives/
icecast project homepage: http://www.icecast.org/
To unsubscribe from this list, send a message to 'icecast-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.




More information about the Icecast mailing list