Author |
Message |
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
|
Back to top
|
|
|
mother misty
Joined: May 13, 2004 Posts: 681 Location: Ghent / Belgium
Audio files: 82
G2 patch files: 130
|
Posted: Mon May 15, 2006 1:27 am Post subject:
|
|
|
Hi rob,
Nice sequencer, but I've got some questions if you don't mind.
What are those 2 yellow modules at the bottom (s&h and constant switch) suposed to do,
I see the s&h isn't connected?
You didn't connected the last output from the binary counter, is there a reason for this?
greets,
misty. |
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Mon May 15, 2006 5:10 am Post subject:
|
|
|
mother misty wrote: | Hi rob,
Nice sequencer, but I've got some questions if you don't mind.
What are those 2 yellow modules at the bottom (s&h and constant switch) suposed to do,
I see the s&h isn't connected?
You didn't connected the last output from the binary counter, is there a reason for this?
greets,
misty. |
Aah, those two unused modules are there for test purposes: when disconnecting the Ctr input on the SeqCtr1 modules and connecting it to the S&H1 output it is possible to select a certain step on the SeqCtr1 module manually, by setting a value on the ConstSwitch1 module. This I used to check if the rhythmic divisions were indeed correct. The S&H1 makes sure the change to another division starts at the beginning of a bar. And another little curiosity is that the selection range of a SeqCtr module is from 0 up to but not including +64, whcih can be checked with that constant module. This SeqCtr conrol range thingy is also the reason why that Clip1 module is there, it clips a possible +64 value coming from the SeqLev module to a value of +63.5, so the SeqCtr won't park.
The idea is to have a lookup table that stores possible division rates (theSeqCtr1 module) and have another sequencer (SeqLev1) define an address in this table. Then the table value is used to let a programmable counter counter the number of 1/96 clockticks selected from the table. The SeqCtr1 module is used as the lookup table where the amount of upwards counts is stored before the BinCounter1-DAConv1-CompSig1 combination (the programmable counter ) is reset.
The D7 input on the DAConv module is actually the sign bit, it negates the DAConv output value as defined by the other seven inputs. Basically the DAConv module can output a range between -64 and +64 in half steps, but for this particular application only 16 values in a range between 1.5 and 36 are needed, and these are all positive values. So, basically I let the BinCounter1-DAConv1 module combination count upwards from 0 to +64 in half steps, and not upwards from -64 to +64 in half steps.
(The reason for this SeqCtr Ctr input range is that this way two of those modules can easily be 'chained' to give a 32 locations table, for this all one has to do is add 64 units to the Ctr signal and feed it to the Ctr input of a second SeqCtr module which now reacts to the negative range of the Ctr signal. See attached patch, where two SeqCtr modules are used in a table lookup configuration to create rhythm variations.)
/Rob
Description: |
Patch where two SeqCtr modules are used to create a 32 location lookup table. In this patch that lookup table is used to create rhytmic variations. Of course there is some other interesting stuff here, like the patent pending Kassen-Rob digitizing brickwa |
|
Download |
Filename: |
32LocationLookupTable.pch2 |
Filesize: |
4.12 KB |
Downloaded: |
1632 Time(s) |
|
|
Back to top
|
|
|
mother misty
Joined: May 13, 2004 Posts: 681 Location: Ghent / Belgium
Audio files: 82
G2 patch files: 130
|
Posted: Tue May 16, 2006 1:55 am Post subject:
|
|
|
Thanks for the explination Rob, much appreciated! |
|
Back to top
|
|
|
W.T.
Joined: Jul 11, 2004 Posts: 272 Location: The Netherlands
Audio files: 1
G2 patch files: 11
|
Posted: Fri Oct 20, 2006 3:06 am Post subject:
|
|
|
very interesting! can't wait to get home and explore this patch.
Thanks in advance |
|
Back to top
|
|
|
|