Nova3 Generative Sequencer - bugfix version, 0.8

[ToneCarver]

I'm not sure if this is a bug or I just don't understand the concept correctly...

For obstacle probability, the readme says "the probability of an obstacle being applied to a rover. 1.0 means obstacles will always affect rovers. 0 means all obstacles are effectively bypassed."

If I set the obstacle probability to 0 and have one rover follow a few squares behind another rover, the second rover does not collide with the second rover after the first rover bounces off a wall and meets with the second rover (which is what I figured would happen). However, if I follow a third rover a few squares after the second rover and the third rover meets the first rover after the first rover bounces back from the wall, there is a collision, sending the third rover in a perpendicular direction.

Rover collisions are determined by first moving each of the rovers then treating the ones that ended up in the same cell as a collision. This means that rovers that are an even number of cells apart from each other (for example, point to point, or 2 empty cells between them) will not collide, they will pass right by each other. Rovers that are an odd number of cells apart (facing each other with 1,3,5, etc empty cells between them) will eventually end up occupying the same cell resulting in a collision.

The obstacle probability controls the likelihood that an obstacle will affect a rover. For example, set up a rover moving east, and somewhere to the right of it put a horizontal mirror, and set obstacle probability to 1.0. The rover will reverse direction when it encounters the mirror. Now move the obstacle probability to 0. The obstacle will have no effect on the rover, it will pass right through the obstacle. Values between 0 and 1 set the probability that obstacles will affect the rovers.

I don't know if anyone else is having this problem, but when I set up a pattern starting on beat one (which seems to be with the rover two squares from a wall), I would like to be able to "rewind" the pattern so that it will reset to that position. Currently what I am doing is letting the pattern play so that I catch the pattern in the start position.

You could use the pattern capture slots for this. Set up the pattern you want to rewind to then select a capture (downward triangle) on one of the pattern slots at the bottom of the gui. When you want to reset the pattern click the restore button (little square with triangle on top) to restore the matrix to the pattern that was captured.

[ToneCarver]

Thanks for the detailed information about the missing notes and the extra note off messages. There was definitely a problem with Nova3 being too generous with the note off messages.

There is a new version of Nova3, version 0.5, HERE that fixes that problem and includes a couple of tiny bugfixes for some issues mentioned in the forum thread. It is tested in Reaper 4.0 beta9 and energyXT26 Beta 18-nov-10 on win7 x64. If you were having trouble with version 0.4 please try the new version and let me know how it works for you.

Changes in v0.5:
+ bugfix: increase line length of .ini file
+ bugfix: corrected velocity off by one
+ bugfix: remove misleading comment in .ini file
+ bugfix: removed incorrect and excessive note-offs

Have fun.

[Juanjo]

Hey that was quick... I cannot reproduce my missing note-ons issue anymore, thanks for the fix!

Unfortunately, before there were too many note-offs, but apparently now there are too few of them Check out this picture:

http://img856.imageshack.us/img856/5109/novaissue2.gif

This is after a few bars. In the midi monitor, note how the rover in the first column (D2) doesn't generate note-offs, most probably because the rover in the second column (A2) generates its note-offs right after the D2 note-ons and Nova mistakenly thinks that the D2 notes are finished. The audible result is that the D2 notes are hanging.

[DarkStar]

Changes in v0.5:
+ bugfix: increase line length of .ini file: FIXED
+ bugfix: remove misleading comment in .ini file: FIXED

Using v0.5
--- I am missing Note_Offs here too. In fact with one "up" rover in the first grid position and no obstacles I get no Note-Offs at all
-- I also see occasional leading Note_Offs too (but in few output sessions than in v0.4)
-- velocity off by one: still off by 1 here (can be seen in Juanjo's screenshot too)

HTH

[ToneCarver]

Sorry, moving too fast and quality suffered. I pulled v0.5 until the note issues get sorted correctly. The link is still valid if anyone feels adventurous but I'd recommend waiting for the proper update.

[ToneCarver]

Here, for the brave at heart, is another update that should be better behaved about note offs and such.

Link: http://members.cox.net/bdavies9/Nova3_alpha05a.zip

I won't make this an official release until we get good results.

Also, I'm thinking it might be better to set up a group of beta-testers and early-lookers so we can iron out the bugs offline, rather than hogging the instrument forum by bumping this thread. If you'd like to be on the list just PM or email me to let me know.

[ToneCarver]

-- velocity off by one: still off by 1 here (can be seen in Juanjo's screenshot too)

If you are loading a saved project try disabling the sequencer or dragging all the bars to the full top position. It should be correct if you load a brand new instance of the plugin. Let me know if not.

[terriandralph]

I'm not sure if this is a bug or I just don't understand the concept correctly...

For obstacle probability, the readme says "the probability of an obstacle being applied to a rover. 1.0 means obstacles will always affect rovers. 0 means all obstacles are effectively bypassed."

If I set the obstacle probability to 0 and have one rover follow a few squares behind another rover, the second rover does not collide with the second rover after the first rover bounces off a wall and meets with the second rover (which is what I figured would happen). However, if I follow a third rover a few squares after the second rover and the third rover meets the first rover after the first rover bounces back from the wall, there is a collision, sending the third rover in a perpendicular direction.

Rover collisions are determined by first moving each of the rovers then treating the ones that ended up in the same cell as a collision. This means that rovers that are an even number of cells apart from each other (for example, point to point, or 2 empty cells between them) will not collide, they will pass right by each other. Rovers that are an odd number of cells apart (facing each other with 1,3,5, etc empty cells between them) will eventually end up occupying the same cell resulting in a collision.

The obstacle probability controls the likelihood that an obstacle will affect a rover. For example, set up a rover moving east, and somewhere to the right of it put a horizontal mirror, and set obstacle probability to 1.0. The rover will reverse direction when it encounters the mirror. Now move the obstacle probability to 0. The obstacle will have no effect on the rover, it will pass right through the obstacle. Values between 0 and 1 set the probability that obstacles will affect the rovers.

Sorry about the confusion on my part. There were certain types of repeated note patterns I was trying to get the program to play, and obviously I wasn't using the correct feature. However, I think the wrap feature might help me get close to the kind of effect I was looking for.

I don't know if anyone else is having this problem, but when I set up a pattern starting on beat one (which seems to be with the rover two squares from a wall), I would like to be able to "rewind" the pattern so that it will reset to that position. Currently what I am doing is letting the pattern play so that I catch the pattern in the start position.

You could use the pattern capture slots for this. Set up the pattern you want to rewind to then select a capture (downward triangle) on one of the pattern slots at the bottom of the gui. When you want to reset the pattern click the restore button (little square with triangle on top) to restore the matrix to the pattern that was captured.

Actually, using a pattern capture slot for the rewind works better than I envisioned, because now it gives me 8 pattern rewinds!

[Juanjo]

Here, for the brave at heart, is another update that should be better behaved about note offs and such.

Link: http://members.cox.net/bdavies9/Nova3_alpha05a.zip

I won't make this an official release until we get good results.

Just tested 0.5a in the same situation that I reported before (see my previous post), recreated from scratch with a fresh instance of Nova, and I still get the same results (hanging notes due to missing D2 note-offs)...

[DarkStar]

-- velocity off by one: still off by 1 here (can be seen in Juanjo's screenshot too)

If you are loading a saved project try disabling the sequencer or dragging all the bars to the full top position. It should be correct if you load a brand new instance of the plugin. Let me know if not.

Yep, that's OK with a brand new project I was reloading a project saved with v0.4. And dragging the sequencer bars upwards also worked.


Talking about the Sequencer:
(a) I wouldn't mind some keyboard Modifiers to allow me to draw smooth velocity ramps
(b) also vertical divider markers (say every 4 steps), as in:



Big pic: http://img710.imageshack.us/img710/6316 ... idiers.png

Uploaded with ImageShack.us

[ToneCarver]

Bump for new version. Version 0.5e is available HERE

This version has the following changes and addresses the missing notes and stuck notes introduced in a previous release:

+ bugfix: increase line length of .ini file
+ bugfix: corrected velocity off by one
+ bugfix: remove misleading comment in .ini file
+ bugfix: removed incorrect and excessive note-offs
+ bugfix: collision handler updated to rotate all rovers in a collision
+ misc bugfixes and gui tweaks

One thing to highlight - the collision handler was changed in this version to fix a bug where one of the rovers in a collision was not being rotated. That behavior has been corrected with this release so if you restore a saved project its patterns with collisions are likely to mutate differently than before.

Have fun.

[maxdis]

thank you for this great plugin!
I have an issue (or maybe I'm just missing something...): very often a rover starts going around the matrix perimeter and never stops, in other words it plays a continuos arpeggio...I dont's know how it could happens, don't should the rover invert it's directions when it hits a wall? Instead it keeps going around the matrix perimeters, playing arpeggios and never stopping...If I try to put an obstacole, to make the rover change its direction, the rover change its direction for a cell or two, then it goes back to the matrix perimeter...in other words, it seems magnetically attracted to the matrix perimeter...
I'm using Reaper 3.75

[ToneCarver]

thank you for this great plugin!
I have an issue (or maybe I'm just missing something...): very often a rover starts going around the matrix perimeter and never stops, in other words it plays a continuos arpeggio...I dont's know how it could happens, don't should the rover invert it's directions when it hits a wall? Instead it keeps going around the matrix perimeters, playing arpeggios and never stopping...If I try to put an obstacole, to make the rover change its direction, the rover change its direction for a cell or two, then it goes back to the matrix perimeter...in other words, it seems magnetically attracted to the matrix perimeter...
I'm using Reaper 3.75

That can happen when a group of rovers clump together into a sort of permanent collision group. Are you seeing a square in the cell? That would confirm that the collision is happening. Internally the rovers are switching directions but being near the wall makes some of them switch directions again leaving all the rovers in the collision group facing the same direction every cycle.

There are 2 ways to break up collision groups. Both techniques use some form of random redirection to break up the group.

One way is to increase the Limit value above 0. The limit value tells the number of cycles that a rover may belong to a collision group before selecting random directions for the rover to attempt to shake the rover out of the group. A value of 0 means that no limit is applied.

The other way is to place obstacles that introduce randomization, (spin, bounce, wormhole) into the path of the collision group. For me this is more useful as an interactive technique (place and remove) just to break up a group when I don't want a randomizing obstacle permanently on the matrix.

[ThomasHelzle]

Hi ToneCarver,

just wanted to drop you a note and thank you for Nova!
I tried a lot of generative tools but Nova so far is the most fun.

When I saw Batuhan Bozkurts Otomata I was amazed by the complexity from such a simple basic concept and even thought about recreating it in processing, but you have driven it to a whole new level.

Thank you so much, it's really wonderful and inspiring!

Best regards,

Tom

[maxdis]

thank you for this great plugin!
I have an issue (or maybe I'm just missing something...): very often a rover starts going around the matrix perimeter and never stops, in other words it plays a continuos arpeggio...I dont's know how it could happens, don't should the rover invert it's directions when it hits a wall? Instead it keeps going around the matrix perimeters, playing arpeggios and never stopping...If I try to put an obstacole, to make the rover change its direction, the rover change its direction for a cell or two, then it goes back to the matrix perimeter...in other words, it seems magnetically attracted to the matrix perimeter...
I'm using Reaper 3.75

That can happen when a group of rovers clump together into a sort of permanent collision group. Are you seeing a square in the cell?

yes, it's a square. Thank you for your suggestions , I'll try them the next time it will happen on me