Posts: 3.679
Threads: 123
Joined: Jul 2017
Reputation:
215
(10-03-2019, 05:42 PM)ural Wrote: (10-02-2019, 11:00 AM)XORadmin Wrote: About the CV input and recording. I was experimenting and searching for the problems i had while recording. Test keyboard was the Arturia Keystep, which i thought would be perfect to test. But the results were very bad, while the trigger was right, the notes often arrived one step too late.
I now found out that this is a problem with the keystep, it has quite a slew between notes. I researched for it and directly got to the information that it is not only my unit:
https://www.muffwiggler.com/forum/viewto...82#2283582
That means the Keystep is causing the problems that i have. What i will do is to add a adjustable delay for CV recording, so also here the correct notes are filled in on the correct steps. However, i am not sure how tricky this is going to be, as the sequencer can already be on another step when the voltage is stable....they are probably the same issues with the BSP.
I will do my best to get the release ready soon. Some more surprises are in there... Once I used CV input of Nerdseq to record a piece from another sequencer (Cirklon). Everything went even better than I expected - notes and triggers were in place. What was little bit difficult - to find "right" Nerdseq groove setting (time signature) what fits correctly together with Cirklon clock. At the end of the day I have tried different Cirklon clocks to get the right division and leave Nerdseq groove as it was.
Yes you need of course matching clocks if you want to record 1:1 and the best solution is most probably a clock with 24ppqn on both, while nerdseq being just 6 ticks/step groove setting.
But the main problem is when using either the Keystep or the BSP as CV source and try to record that. (i didn't test the BSP as i don't have one, but checking different forum items on MW or Arturias Forum show me that it has the same issues) The fact that the CV voltage needs 3-5 milliseconds until it reaches the note, while the trigger happened already before the CV is even starting to move to the new note is actually not accepteable and forces me now to add some kind of CV delay to be sure the correct note is being recorded (and not too late then).
I expect good recording results from good sources, and average ones from bad sources. So it'll be a best possible implementation.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help! 
Posts: 38
Threads: 3
Joined: Mar 2019
Reputation:
0
(10-03-2019, 07:18 PM)XORadmin Wrote: (10-03-2019, 05:42 PM)ural Wrote: (10-02-2019, 11:00 AM)XORadmin Wrote: About the CV input and recording. I was experimenting and searching for the problems i had while recording. Test keyboard was the Arturia Keystep, which i thought would be perfect to test. But the results were very bad, while the trigger was right, the notes often arrived one step too late.
I now found out that this is a problem with the keystep, it has quite a slew between notes. I researched for it and directly got to the information that it is not only my unit:
https://www.muffwiggler.com/forum/viewto...82#2283582
That means the Keystep is causing the problems that i have. What i will do is to add a adjustable delay for CV recording, so also here the correct notes are filled in on the correct steps. However, i am not sure how tricky this is going to be, as the sequencer can already be on another step when the voltage is stable....they are probably the same issues with the BSP.
I will do my best to get the release ready soon. Some more surprises are in there... Once I used CV input of Nerdseq to record a piece from another sequencer (Cirklon). Everything went even better than I expected - notes and triggers were in place. What was little bit difficult - to find "right" Nerdseq groove setting (time signature) what fits correctly together with Cirklon clock. At the end of the day I have tried different Cirklon clocks to get the right division and leave Nerdseq groove as it was.
Yes you need of course matching clocks if you want to record 1:1 and the best solution is most probably a clock with 24ppqn on both, while nerdseq being just 6 ticks/step groove setting.
But the main problem is when using either the Keystep or the BSP as CV source and try to record that. (i didn't test the BSP as i don't have one, but checking different forum items on MW or Arturias Forum show me that it has the same issues) The fact that the CV voltage needs 3-5 milliseconds until it reaches the note, while the trigger happened already before the CV is even starting to move to the new note is actually not accepteable and forces me now to add some kind of CV delay to be sure the correct note is being recorded (and not too late then).
I expect good recording results from good sources, and average ones from bad sources. So it'll be a best possible implementation.
perhaps you could implement a kind of trigger input delay for recording so it will go in sync with CV?
that's a way we doing when using CV to change slices or samples in Sample Drum.
(I'm using Sample Drum internal trigger delay setting (1ms), but I noticed later there are trigger delay function also in Nerdseq.)
Posts: 3.679
Threads: 123
Joined: Jul 2017
Reputation:
215
(10-04-2019, 05:07 PM)ural Wrote: (10-03-2019, 07:18 PM)XORadmin Wrote: (10-03-2019, 05:42 PM)ural Wrote: (10-02-2019, 11:00 AM)XORadmin Wrote: About the CV input and recording. I was experimenting and searching for the problems i had while recording. Test keyboard was the Arturia Keystep, which i thought would be perfect to test. But the results were very bad, while the trigger was right, the notes often arrived one step too late.
I now found out that this is a problem with the keystep, it has quite a slew between notes. I researched for it and directly got to the information that it is not only my unit:
https://www.muffwiggler.com/forum/viewto...82#2283582
That means the Keystep is causing the problems that i have. What i will do is to add a adjustable delay for CV recording, so also here the correct notes are filled in on the correct steps. However, i am not sure how tricky this is going to be, as the sequencer can already be on another step when the voltage is stable....they are probably the same issues with the BSP.
I will do my best to get the release ready soon. Some more surprises are in there... Once I used CV input of Nerdseq to record a piece from another sequencer (Cirklon). Everything went even better than I expected - notes and triggers were in place. What was little bit difficult - to find "right" Nerdseq groove setting (time signature) what fits correctly together with Cirklon clock. At the end of the day I have tried different Cirklon clocks to get the right division and leave Nerdseq groove as it was.
Yes you need of course matching clocks if you want to record 1:1 and the best solution is most probably a clock with 24ppqn on both, while nerdseq being just 6 ticks/step groove setting.
But the main problem is when using either the Keystep or the BSP as CV source and try to record that. (i didn't test the BSP as i don't have one, but checking different forum items on MW or Arturias Forum show me that it has the same issues) The fact that the CV voltage needs 3-5 milliseconds until it reaches the note, while the trigger happened already before the CV is even starting to move to the new note is actually not accepteable and forces me now to add some kind of CV delay to be sure the correct note is being recorded (and not too late then).
I expect good recording results from good sources, and average ones from bad sources. So it'll be a best possible implementation.
perhaps you could implement a kind of trigger input delay for recording so it will go in sync with CV?
that's a way we doing when using CV to change slices or samples in Sample Drum.
(I'm using Sample Drum internal trigger delay setting (1ms), but I noticed later there are trigger delay function also in Nerdseq.)
Yep thats the plan to do. But it would still mean: Best effort.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!
Posts: 1
Threads: 0
Joined: Dec 2017
Reputation:
0
(10-02-2019, 11:00 AM)XORadmin Wrote: About the CV input and recording. I was experimenting and searching for the problems i had while recording. Test keyboard was the Arturia Keystep, which i thought would be perfect to test. But the results were very bad, while the trigger was right, the notes often arrived one step too late.
I now found out that this is a problem with the keystep, it has quite a slew between notes. I researched for it and directly got to the information that it is not only my unit:
https://www.muffwiggler.com/forum/viewto...82#2283582
That means the Keystep is causing the problems that i have. What i will do is to add a adjustable delay for CV recording, so also here the correct notes are filled in on the correct steps. However, i am not sure how tricky this is going to be, as the sequencer can already be on another step when the voltage is stable....they are probably the same issues with the BSP.
I will do my best to get the release ready soon. Some more surprises are in there...
You probably have tried the worst case with the KeyStep. Once solved, everything else should be perfect. I tried CV recording with an SH-09 and the results were better than what you described. I only had double triggers sometimes from the worn out key contacts in my SH-09, which tend to send a tiny spike at the release of a key. Would it make sense to have CV recording delay and also a minimum trigger length?
Posts: 3.679
Threads: 123
Joined: Jul 2017
Reputation:
215
(10-09-2019, 04:38 PM)MatzeMillion Wrote: (10-02-2019, 11:00 AM)XORadmin Wrote: About the CV input and recording. I was experimenting and searching for the problems i had while recording. Test keyboard was the Arturia Keystep, which i thought would be perfect to test. But the results were very bad, while the trigger was right, the notes often arrived one step too late.
I now found out that this is a problem with the keystep, it has quite a slew between notes. I researched for it and directly got to the information that it is not only my unit:
https://www.muffwiggler.com/forum/viewto...82#2283582
That means the Keystep is causing the problems that i have. What i will do is to add a adjustable delay for CV recording, so also here the correct notes are filled in on the correct steps. However, i am not sure how tricky this is going to be, as the sequencer can already be on another step when the voltage is stable....they are probably the same issues with the BSP.
I will do my best to get the release ready soon. Some more surprises are in there...
You probably have tried the worst case with the KeyStep. Once solved, everything else should be perfect. I tried CV recording with an SH-09 and the results were better than what you described. I only had double triggers sometimes from the worn out key contacts in my SH-09, which tend to send a tiny spike at the release of a key. Would it make sense to have CV recording delay and also a minimum trigger length?
I got the keystep under control now. And i think your issue should be solved as well. I introduced 2 additional new recording possibilities, which samples the note after a trigger or a gate. The sample delay can be set from 0 to 9 milliseconds. If sampling with gate, then it filters out peaks as well which should help in your situation.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!
Posts: 13
Threads: 4
Joined: May 2019
Reputation:
2
Is it possible to do step recording in the NerdSeq? Like when the record mode is active that pressing a note on a midi keyboard is recorded into the current highlighted step?
Posts: 3.679
Threads: 123
Joined: Jul 2017
Reputation:
215
(10-09-2019, 11:17 PM)modbang Wrote: Is it possible to do step recording in the NerdSeq? Like when the record mode is active that pressing a note on a midi keyboard is recorded into the current highlighted step?
As mentioned here in the release notes on top of the thread:
Quote:- added cursor/step edit from midi or from CV inputs. (Input/Record settings) Insert a note when enabled and if on note column in pattern screens. Hold key will remove the note again. Sets a gate for note on and a trigger if released on same step, else a note off if released on another step for CV.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!
Posts: 13
Threads: 4
Joined: May 2019
Reputation:
2
(10-10-2019, 07:29 AM)XORadmin Wrote: (10-09-2019, 11:17 PM)modbang Wrote: Is it possible to do step recording in the NerdSeq? Like when the record mode is active that pressing a note on a midi keyboard is recorded into the current highlighted step?
As mentioned here in the release notes on top of the thread:
Quote:- added cursor/step edit from midi or from CV inputs. (Input/Record settings) Insert a note when enabled and if on note column in pattern screens. Hold key will remove the note again. Sets a gate for note on and a trigger if released on same step, else a note off if released on another step for CV. My man! Very nice work! 
Posts: 37
Threads: 5
Joined: Oct 2018
Reputation:
0
Sorry if this has already been covered, but I have a bit of quick feedback that is hopefully easy to fix for the next release.
Launchpad Integration:
- No way to stop a pattern from playing via the Launchpad.
- Pressing an empty pattern below a non-empty pattern, in an attempt to stop the pattern, queues up the pattern above it.
- Default color for each track should be set to a dim gray, so it shows up on the launchpad by default. I had mentioned this earlier, but am just jotting it down again.
Sequencer:
- Pressing OK on an existing pattern, replacing it with a blank, should be disabled while playing. I have accidentally clicked ok a couple times during a set and had to scroll back and find what pattern was there before.
Launchpad Question:
- I am using multiple channels of MIDI out on my nerdseq, and noticed if I have the launchpad set to channel 1, channels 1-5 should also not be used as outputs as they interfere with the UI. Not sure what to do about this, maybe it is a documentation thing. Should these channels be avoided?
Posts: 3.679
Threads: 123
Joined: Jul 2017
Reputation:
215
10-13-2019, 07:10 PM
(This post was last modified: 10-13-2019, 07:14 PM by XORadmin.)
(10-13-2019, 04:28 PM)gromler Wrote: Sorry if this has already been covered, but I have a bit of quick feedback that is hopefully easy to fix for the next release.
Launchpad Integration:
- No way to stop a pattern from playing via the Launchpad.
- Pressing an empty pattern below a non-empty pattern, in an attempt to stop the pattern, queues up the pattern above it.
- Default color for each track should be set to a dim gray, so it shows up on the launchpad by default. I had mentioned this earlier, but am just jotting it down again.
Sequencer:
- Pressing OK on an existing pattern, replacing it with a blank, should be disabled while playing. I have accidentally clicked ok a couple times during a set and had to scroll back and find what pattern was there before.
Launchpad Question:
- I am using multiple channels of MIDI out on my nerdseq, and noticed if I have the launchpad set to channel 1, channels 1-5 should also not be used as outputs as they interfere with the UI. Not sure what to do about this, maybe it is a documentation thing. Should these channels be avoided?
Thanks for the feedback...and just in time before my next (and hopefully final) release candidate which i wanted to upload tomorrow.
About your info:
Quote:- No way to stop a pattern from playing via the Launchpad.
Hmm, are you sure? The third button from the Bottom combined with a track should mark the track to stop. Doing it twice stops the track directly. That works with the other launchpads and i don't know why it shouldn't work with the LP Pro. Shift+STOP+a track, stops the sequencer.
Quote:- Pressing an empty pattern below a non-empty pattern, in an attempt to stop the pattern, queues up the pattern above it.
This is not a stop pattern situation. I don't remember that it would queue up, but i will test it tomorrow. I think it is a general question how it should behave. In my opinion i want to keep it mostly the same as nerdSEQ own behaviour.
Quote:- Default color for each track should be set to a dim gray, so it shows up on the launchpad by default. I had mentioned this earlier, but am just jotting it down again.
I will give used patterns a DIM grey, but only if they didn't get a colour yet. I still think and say that it only makes real sense if using with colours.
Quote:- Pressing OK on an existing pattern, replacing it with a blank, should be disabled while playing. I have accidentally clicked ok a couple times during a set and had to scroll back and find what pattern was there before.
Thats not the behaviour in the new version anymore. There is a thread about it and i took over the ideas so this can't happen anymore. I think the new approach is now much better and these issues are solved with it. But also, i advise to disable edit when playing a set. This also brings the direct jump function then which i always use intensively when playing live.
Quote:- I am using multiple channels of MIDI out on my nerdseq, and noticed if I have the launchpad set to channel 1, channels 1-5 should also not be used as outputs as they interfere with the UI. Not sure what to do about this, maybe it is a documentation thing. Should these channels be avoided?
I don't know. As mentioned before, i don't have a Launchpad Pro so i can't confirm the behaviour. (Still waiting for Novation to provide me one, so i can also give support for that one) The only thing the NerdSEQ does is to use Midi Channel 1 which seem to be channel-fixed in the Launchpad to control it. All other Midi channels i don't know about and you might refer the Launchpad midi integration manual for it to find out. With the LP integration i totally depend on what a LP can do and what not and keep in mind that this is still Beta and will also be beta until i make a proper USB expander for it.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!
|
