Do we want more inter slot busses? |
No thanks, 4 is enough. |
|
12% |
[ 3 ] |
Hmmm... I'd like some more audio busses please. |
|
4% |
[ 1 ] |
Those control signal busses sound like a good idea... yes please! |
|
25% |
[ 6 ] |
I could do with more audio AND new control signal busses! |
|
58% |
[ 14 ] |
|
Total Votes : 24 |
|
Author |
Message |
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Sun Sep 25, 2005 9:39 am Post subject:
More inter slot busses please. |
|
|
I have to admit that for the average user, this wish may fall on deaf ears, but I feel that the experienced user may drool at this idea...
If we had 8 audio busses, that would be enough to send a stereo signal through each slot and back again. This would enable us to make full use of the available DSP to either create monster sounds, complex full tracks, or in my case, better sound design capabilities in Performance mode.
For instance, you could have a stereo signal from each slot going to one reverb in one fx section. Or you could create complex stereo FM between sounds in different slots.
Even better though, maybe even essential, are inter slot control signal busses... again 8 would be good. If we had dedicated control signal busses, we could do away with having to use a quantise module (after an audio buss to get the rate back to control signal level). Furthermore, when composing arrangements, we would be free to send clock signals between slots without the DSP disadvantage of using the audio rate.
So what do you think... more busses? _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Sun Sep 25, 2005 10:04 am Post subject:
|
|
|
RE more audio busses. There are indications (from Clavia sources) that what we have now is about the limit, or maybe even just over the limit already (although I didn't run into it yet).
RE control busses : I've not tested yet the precision of inter slot (area) MIDI control, but I was thinking that MIDI that is not leaving the synth could maybe have full resolution. There are of course some bandwidth problems with the internal MIDI channels ...
Otherwise blue busses seems like a good idea, I'd think (but this would probably be prone to more or less the same quantity limitations now present for red busses).
More in general it would be nice to have better control over signal rates without resorting to explicit type casts (i.e. note quantizers).
But, yes, I'd like to have more busses for sure :-) _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Sun Oct 16, 2005 6:43 am Post subject:
|
|
|
Come on guys, don't forget to vote... 62 views and only 4 votes!??!? _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
arturmeinild
Joined: Sep 12, 2005 Posts: 19 Location: Denmark
G2 patch files: 4
|
Posted: Fri Oct 21, 2005 2:37 pm Post subject:
|
|
|
Technical question: Wouldn't it be possible to route four blue 24 kHz signals through one red 96 kHz wire??? Has this been done, if yes where can I find it? |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Oct 21, 2005 4:21 pm Post subject:
|
|
|
I'll not say impossible, but 4 times 24 kHz is challenging.
The idea is that you use a multiplexer in one slot and a demultiplexer in another. In order though to be able to reconstruct the original signals you'll also need the control signal from the multiplexer to be able to demultiplex correctly. For this it is easiest to send it over another audio bus, but this could maybe cleverly be encoded into the multiplexed signal as well.
To do it at 24 kHz you would need a 24 kHz control signal for the multiplexer, this should be a saw shaped signal, I don't know how to make such a signal at that rate.
When however you'll allow for some water int the wine, you could use an 8->1/1->8 set of mpx/de-mpx modules that you control with the highest rate sawtooth an audio osc will give you, so you'll end up with 8 signals at just under 2 kHz each, or so.
Well, should end up with, I didn't try it actually :-) _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
arturmeinild
Joined: Sep 12, 2005 Posts: 19 Location: Denmark
G2 patch files: 4
|
Posted: Sat Oct 22, 2005 8:55 am Post subject:
|
|
|
Yeah, I figured it would be pretty tough - maybe the 2 channels in 1 wire patch that already exists is the best you can get.
I sat down and read in the manual and I came to the same conclusion as you Blue - that you would need a control signal at the sample rate (24 kHz minimum, 96kHz preferred). And even then you would definitely loose a number of samples from each signal in the coding...
EDIT: Ok, found out how to generate a 48 kHz signal. Now this has to be converted into a saw shape, and we need to find a way to embed a 48 kHz logic signal into the control stream... Will give this a try...
http://www.electro-music.com/forum/topic-5385.html |
|
Back to top
|
|
|
arturmeinild
Joined: Sep 12, 2005 Posts: 19 Location: Denmark
G2 patch files: 4
|
Posted: Sun Oct 23, 2005 8:54 am Post subject:
|
|
|
Found a way to create a 48 kHz signal to trigger the MPX - with a logic inverter and a sequencer. Now this concept works under the assumption that an identical clock generator placed in both areas will send out the same clock signal - therefore there is no need to send the control signal through the bus. 2 files are attached - a low speed file that shows the concept (0) and a high speed file with a 48 kHz clock (1). Neither are tested with actual control signals, but please let me know what you think...
Actually, only 7 channels are available since you need 1 channel to set the MPX to 96 kHz mode... In theory, this should give you 7 control channels each in 6 kHz. If synchronization doesn't work this way I guess you have to use an entire bus channel for the sync signal, but this still gives you 3 x 7 control channels...
If anyone can find a smart way to embed the sync signal in the MPX, please step forward!
EDIT: The working version is now up here... http://www.electro-music.com/forum/topic-8355.html
Description: |
|
Download |
Filename: |
8cont1wire_conc0.pch2 |
Filesize: |
1.68 KB |
Downloaded: |
1504 Time(s) |
Description: |
|
Download |
Filename: |
8cont1wire_conc1.pch2 |
Filesize: |
1.59 KB |
Downloaded: |
1552 Time(s) |
|
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Mon Oct 24, 2005 7:06 am Post subject:
|
|
|
Thanks for the private message Artur. Your discoveries are indeed interesting, but I'm not sure if it will work as I think that the busses are mono... I seem to remember having difficulties sending several clock signals through one buss in a similar fashion to this.
Even so, this is a wish list, so wouldn't it still be better to have the busses without the hard work?
I must say that I am quite surprised that some people have voted that 4 busses are enough... I can't really see any down side to having more, but everyone has they're own ideas, which is good. I'm thinking that serious 'patchers' would want more, and 'part timers' wouldn't?
It's good that more people have started voting too - it makes it more interesting! _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
Posted: Thu Mar 09, 2006 5:12 pm Post subject:
Inter-DSP Audiobus probably maxed out on G2 |
|
|
As I remember vaguely from my Motorola 563xx DSP programming praxis, there is a fast serial channel on the chip which could be put into
some kind of network/bus mode... they are fast, but cannot carry
more than a few audio channels at 96kHz for sure,
so probably no luck here...
selvmarcus |
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Thu Mar 09, 2006 5:35 pm Post subject:
|
|
|
That's a good point... but I'd still prefer the option of having eight 48kHz audio busses or four 96kHz audio busses. A 48kHz mode would probably allow longer delays and cheaper reverbs, etc. too!! _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
Posted: Thu Mar 09, 2006 6:26 pm Post subject:
|
|
|
Yes, these 8 busses 48KHz would be nice.
What about just two new mux/demux modules for that purpose?
Like merging by interleaving four logic signals (24KHz) or choicewise
two audio signals (by downsampling to 48kHz) into one audio signal
(96KHz) and vice versa.
Then they could be used for the delay lines ..., too.
Would be very easy to implement systemwise (this bit of global sync
should be available at the system level already).
On the downside, it may look "too special"
to be implemented.
Sidenote:
Several years ago, I was in a development team, where we
were designing a multi-DSP harddisk-based sampling system.
We had 256 channels with 48kHz on a bus, just to shuffle
64 voices from disk to DSPs to mixing to audio-out...
The effort was needed because highest fidelity in the sampling
rate conversion (pitch transposing) was requested.
A real monster...
So I know it´s so much easier if you can just use
the on-chip serial channels.
selvmarcus |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Thu Mar 09, 2006 7:42 pm Post subject:
|
|
|
sheridan wrote: | A 48kHz mode would probably allow longer delays |
You can already do that of course, again by using some multiplexing. And the good news is that its easier to solve than the bus problem ;-) _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Thu Mar 09, 2006 7:59 pm Post subject:
|
|
|
Interslot midi has its limitations in the actual g2 versions.
its not fast enough for high datarates... I however get along quite well with the 4 busses by trying to split things over fx and va area with the fx bus whenever possible...
I rather would like to see the fx busses bi directional...
this maybe wouldnt be so expensiv but i allready had a few cases where it would have been usefull |
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
|
Back to top
|
|
|
deknow
Joined: Sep 15, 2004 Posts: 1307 Location: Leominster, MA (USA)
G2 patch files: 15
|
Posted: Fri Mar 10, 2006 6:52 am Post subject:
|
|
|
i assume that if you are only using stereo out, and not processing external sounds, you could patch 2 inputs to 2 outputs and have 6 busses?
deknow |
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Fri Mar 10, 2006 9:30 am Post subject:
|
|
|
That's very clever Selvmarcus... I don't quite understand it, but very clever... but if you remember the original problem, it was about getting signals from one patching area to another, and even though you have successfully got two signals going down one cable, you also need another cable going to the Pan Mod input... so if the 1-Demux-2 Module in your example was in a seperate patching area, then you would still need to send two signals to it: one audio, one control.
That's also a good idea Deknow... I just tried it and it also works. It might not be too user friendly for sharing patches though? Who knows?
Quote: | --------------------------------------------------------------------------------
sheridan wrote:
A 48kHz mode would probably allow longer delays
You can already do that of course, again by using some multiplexing. And the good news is that its easier to solve than the bus problem |
How Jan? _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Mar 10, 2006 3:39 pm Post subject:
Re: Two audiochannels in one |
|
|
selvmarcus wrote: | Sorry, if this is an old hat! |
There have been some tries on this before yes, but no need for worry or sorry. You just invented a novelty high speed oscillator type
(the "usual" one is a logig inverter fed back on itself and then made orange by connecting an audio signal to the unused inverter's input, which then can be removed again - little trick to fool the optimizer; your solution is cheaper and gives a symetric output signal !). _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
Posted: Sat Mar 11, 2006 5:49 am Post subject:
|
|
|
This experimental perf is transmitting two different sinewave signals over
one bus from slot A to slot B by means of ultra-fast (48 kHz)
multiplexing.
They are then output to Out 1/2, respectively.
(I made only one patch for both slots, sender is switched to ON in slot A,
receiver OFF in A, ON in B)
On receivers end, some circuit was added to refill the
resulting gaps (every second sample) for each channel.
I did not use any special sync communication, just a copy of my
clock-oscillator (a feedback level-inverter) is working in both slots.
It has not been out of sync so far
(it would exchange L/R in my patch then), so it seems to me the
patches/slots are always started in sync by the system.
If this will be a problem, I would try using the shared midi clock logic
(glockgen module) to fire things up.
Another possibility: Sender transmits a short pulse at startup on one channel,
receiver checks where it comes and adjusts accordingly
i.e. by exchanging L/R if nessecary. Startup mode.
To go further:
With controlsignals or audiosignals band-limited(lowpass-filtered) to
12 kHz it should be possible to transmit even four signals over one bus
just by cascading this.
(4->2->1--bus-->1->2->4)
Marcus
(This is nice community. Happy to be here)
Description: |
Experimental performance ransmitting two different sinewave signals simultaneously over one bus by means of ultra-fast (48 kHz) multiplexing |
|
Download |
Filename: |
2audio-over-bus1.prf2 |
Filesize: |
4.14 KB |
Downloaded: |
1323 Time(s) |
|
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Sat Mar 11, 2006 6:56 am Post subject:
|
|
|
Wow!
This kind of stuff is all a bit deep for me to fully understand, but these patches certainly seem to work. Well done!! _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
|
Back to top
|
|
|
Fozzie
Joined: Jun 04, 2004 Posts: 875 Location: Near Wageningen, the Netherlands
Audio files: 8
G2 patch files: 49
|
Posted: Sat Mar 11, 2006 11:13 am Post subject:
|
|
|
This is just soooo coooool!!! Thanks Selvmarcus for the mux technique & Jan for the longdelay (it also seems to work on temposynced delays ) |
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Sat Mar 11, 2006 2:44 pm Post subject:
|
|
|
Funny, the two way multiplexer works ok for me, but for the higher level ones I get channel bleed and dropped channels.
In the 4 channel one I tried to switch off all oscillators except one to see which output it would end up on. The first two oscs ended up both on two outputs, the 2nd two oscs behaved ok.
For the eight channel one I just altered some pitches, and had an osc that didn't come through. Furthermore the sound changed when I changed things in the patch, that made me worry a bit :-)
Exactly as to what channels are affected seems to be partly random. I tried to improve by sending the clock over a bus line as well, but that didn't solve it. I didn't see an obvious thing wrong either. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
selvmarcus
Joined: Feb 08, 2006 Posts: 121 Location: Berlin, Germany
G2 patch files: 39
|
Posted: Sat Mar 11, 2006 3:01 pm Post subject:
|
|
|
Hello Jan,
the eight channel version got at least one error.
The samplerate is only 48 kHz/8.
I´m just experimenting with this.
(Excited and sweating)
With the four channels I don´t know.
I´ll check it.
May be the crossfaders as DIY sample and hold
behave not so well, sometimes. Will try this.
One could exchange the other ones for valswitches and use the
"standard" logic inverter as clock, as they didn´t work
together with my level converter clock for some reason.
Hell, what´s going on here...
Marcus |
|
Back to top
|
|
|
|