Hi Guys, 

I'm just finishing my Schroeder reverb unit now and have added several
standard features e.g. high/low cut, predelay, time and density. I now want
to add room size. I however am not quite clear what the room size function
does. Sure, the reverb might last longer, but this is already handled buy
reverb time. I have attached my csd. file so you can see how I've designed
it. 

Please see if you can tell me how I would go about implementing a room size
function. Something that can be controlled via k-rates please though, as you
can see I'm making it user definable.

Thanks!!
Doug  http://www.nabble.com/file/p21563641/Reverberation.csd
Reverberation.csd 
-- 
View this message in context: http://www.nabble.com/Reverb-Unit-Room-Size-tp21563641p21563641.html
Sent from the Csound - General mailing list archive at Nabble.com.

I guess there is no definite answer to how to implement room size with
a Schroeder reverb, as the reverb itself is "artificial", that is, it
does not actually simulate a physical room. The freeverb opcode has an
argument kRoomSize, but afaik this controls reverb length (not
actually size" as you also describe). I guess you could modify the
delay times of the alpass and comb filters, making the delay times
longer should give a more sparse echo pattern like you would have in a
larger room.

best
Oeyvind

2009/1/20 basshead6949 :
>
> Hi Guys,
>
> I'm just finishing my Schroeder reverb unit now and have added several
> standard features e.g. high/low cut, predelay, time and density. I now want
> to add room size. I however am not quite clear what the room size function
> does. Sure, the reverb might last longer, but this is already handled buy
> reverb time. I have attached my csd. file so you can see how I've designed
> it.
>
> Please see if you can tell me how I would go about implementing a room size
> function. Something that can be controlled via k-rates please though, as you
> can see I'm making it user definable.
>
> Thanks!!
> Doug  http://www.nabble.com/file/p21563641/Reverberation.csd
> Reverberation.csd
> --
> View this message in context: http://www.nabble.com/Reverb-Unit-Room-Size-tp21563641p21563641.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>

Oeyvind Brandtsegg wrote:
> I guess there is no definite answer to how to implement room size with
> a Schroeder reverb, as the reverb itself is "artificial", that is, it
> does not actually simulate a physical room. The freeverb opcode has an
> argument kRoomSize, but afaik this controls reverb length (not
> actually size" as you also describe). I guess you could modify the
> delay times of the alpass and comb filters, making the delay times
> longer should give a more sparse echo pattern like you would have in a
> larger room.
> 
> best
> Oeyvind
> 

The classic Schroeder reverb (at least as described in  Moorer's paper 
"About This Reverberation Business) does include  the use of "early 
reflections"  in the form of a multiple-tapped delay line (implemented 
as a FIR filter), feeding into the dense reverb network. The general 
distance between the delays, together with the overall duration before 
the dense reverb takes over, naturally relate to room size. However, 
there seems to be a general rule of thumb that early reflections for 
'typical" rooms take place between 30 and 80 msecs. The length of the 
"pre-delay" before the first early reflection can itself relate to room 
size.

With a small room, there may be a very small number of taps, with larger 
room, proportionately more, increasing in density. Moorer advocates a 
19-tap FIR (modelled on Symphony Hall Boston) of 79msecs overall, but 
also demonstrates a 7-tap form where computation is at a premium 
(unlikely these days!).  The tricky part is defining the delays, for 
which an algorithm (e.g. from a simple, room model) is usually better 
than manual guesswork. It is essential that the delays are mutually 
random, so that there is minimum danger of ovbious periodicities (unless 
yrou want that effect of course!).

Richard Dobson

Thats what I thought, ok i'll do that maybe setting up radio buttons for room
size. Thanks Oeyvind

Oeyvind Brandtsegg-2 wrote:
> 
> I guess there is no definite answer to how to implement room size with
> a Schroeder reverb, as the reverb itself is "artificial", that is, it
> does not actually simulate a physical room. The freeverb opcode has an
> argument kRoomSize, but afaik this controls reverb length (not
> actually size" as you also describe). I guess you could modify the
> delay times of the alpass and comb filters, making the delay times
> longer should give a more sparse echo pattern like you would have in a
> larger room.
> 
> best
> Oeyvind
> 
> 2009/1/20 basshead6949 :
>>
>> Hi Guys,
>>
>> I'm just finishing my Schroeder reverb unit now and have added several
>> standard features e.g. high/low cut, predelay, time and density. I now
>> want
>> to add room size. I however am not quite clear what the room size
>> function
>> does. Sure, the reverb might last longer, but this is already handled buy
>> reverb time. I have attached my csd. file so you can see how I've
>> designed
>> it.
>>
>> Please see if you can tell me how I would go about implementing a room
>> size
>> function. Something that can be controlled via k-rates please though, as
>> you
>> can see I'm making it user definable.
>>
>> Thanks!!
>> Doug  http://www.nabble.com/file/p21563641/Reverberation.csd
>> Reverberation.csd
>> --
>> View this message in context:
>> http://www.nabble.com/Reverb-Unit-Room-Size-tp21563641p21563641.html
>> Sent from the Csound - General mailing list archive at Nabble.com.
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 
> 

-- 
View this message in context: http://www.nabble.com/Reverb-Unit-Room-Size-tp21563641p21573385.html
Sent from the Csound - General mailing list archive at Nabble.com.