Some thoughts on future possibilities

[Zentrumsounds]

So first :

I've used the Nerdseq for a week in a system with Monome Teletype, ER 301 and other modules. But I've been working with it non-stop preparing for a festival concert in 10 days and I've even hit the pattern maximum in creating my set.

However I've a few thoughts and have been taking notes throughout resulting in the following. 

First simple suggestions:


1. Page Up / Page Down button behaviour: Could this be the same as cursor behaviour? Long press cursor works (data changes fast and navigation changes fast) but long press Page Up Page Down does not have the same behaviour. 

2. Can we have Basenote option in MIDI FX?

3. Can we have more information in Current Values? Basenote would be very good to have there. 

4. Copy Colours with Mark (copy paste etc.). This currently leaves colours behind and the only way I see to move colours with patterns in a sequencer window is to delete empty positions to move the patterns AND their colours together. 

5. Visualisation of Probability % in the empty space at the bottom right of the Pattern window. The No.1 issue I have when working with probability is troubleshooting why something is not playing or not changing in some way. Its almost always probability percentage but if I could see FX % value in a similar way to the note activity in the Sequencer window that would be so helpful. There is space!

6. Mute / Solo behaviour in Sequencer window is Great! Can we also have the same behaviour in the four MIDI note column? 

Bigger ones:

7. I've mentioned this before but the copy and paste approach using MARK is for me very slow. I also know that this is likely to be evolved soon! 
I really think I'm actually finding the Mark system challenging and sometimes wonder am using it correctly. Look forward to seeing next update approach! 
But anyway... The approach that would work faster for me is: 


Shift Copy -> Shift Copy (simply puts entry into memory for a paste command using Shift copy again)
 
Shift Delete -> (removes entry but also places it into memory for paste command using Shift Copy)

8. Clock multipliers that allow for different tupelos (triplets, maybe even quintuplets, septuplets. etc... but definitely triplets. I've searched forum for answers to this and have not found anything that works using multipliers. Groove for me is quite a slow solution.

9. Okay this one may be impossible... BUT. How about an EVERY or MAX function that can supplement the generative side of things like probability? For instance, EVERY 6 would only play THAT entry every 6th pass. Or Maybe every even EVERY 0606 is every 6th pas and EVERY 0406 is a repetition between four and six times on every pass (just add a random between 0 and 2 to 4 every pass to determine when the next event takes place. Similarly a MAX entry could state that an entry occurs a maximum of N times (So MAX 0009 would mean the entry happens 9 times and then stops). 

All the above being stated within the context that Nerdseq Eurorack technology at its best. I'm really looking forward to spending the next number of years exploring its potential further.  

Jürgen

[mgd]

8. Clock multipliers that allow for different tupelos (triplets, maybe even quintuplets, septuplets. etc... but definitely triplets. I've searched forum for answers to this and have not found anything that works using multipliers. Groove for me is quite a slow solution.

I do triplets by SCLK /3 immediately (I think it is SCLK 103) or SCLK *3 immediately (SCLK 10D) and then using half or double the number of steps in the pattern. I got back to normal (SCLK 109) as soon as the triplet is finished.

This requires some extra care (and careful computation) when there are several pattern playing in parallel but it works. Though you may consider this too complicated.

9. Okay this one may be impossible... BUT. How about an EVERY or MAX function that can supplement the generative side of things like probability? For instance, EVERY 6 would only play THAT entry every 6th pass. Or Maybe every even EVERY 0606 is every 6th pas and EVERY 0406 is a repetition between four and six times on every pass (just add a random between 0 and 2 to 4 every pass to determine when the next event takes place. Similarly a MAX entry could state that an entry occurs a maximum of N times (So MAX 0009 would mean the entry happens 9 times and then stops).

IIRC such loop counters have been requested repeatedly. My memory tells me the biggest problem with them is, that there is not enough spare memory to store these and no infrastructure to keep track of the loop counts. After all would that count for single patterns or would you want to allow repetition of islands? What happens when you jump between patterns. Does that reset the counter? Etc. pp.

Don't get me wrong:
I had asked for such counters myself. But I see that apart from the trivial cases they actually are quite complicated and I think Thomas more than once stated, that he does not see a way to implement them.

Kind regards,
Michael

[Zentrumsounds]

8. Clock multipliers that allow for different tupelos (triplets, maybe even quintuplets, septuplets. etc... but definitely triplets. I've searched forum for answers to this and have not found anything that works using multipliers. Groove for me is quite a slow solution.

I do triplets by SCLK /3 immediately (I think it is SCLK 103) or SCLK *3 immediately (SCLK 10D) and then using half or double the number of steps in the pattern. I got back to normal (SCLK 109) as soon as the triplet is finished.

This requires some extra care (and careful computation) when there are several pattern playing in parallel but it works. Though you may consider this too complicated.

Aha moment. Yes that's not hard at all. My XOR forum search of "triplets" just discussed some elaborate groove manipulation. This I can work with! (perhaps I just keep forgetting about the Direct version SCLK 109 versus the non direct version SCLK 009) Thanks.  

9. Okay this one may be impossible... BUT. How about an EVERY or MAX function that can supplement the generative side of things like probability? For instance, EVERY 6 would only play THAT entry every 6th pass. Or Maybe every even EVERY 0606 is every 6th pas and EVERY 0406 is a repetition between four and six times on every pass (just add a random between 0 and 2 to 4 every pass to determine when the next event takes place. Similarly a MAX entry could state that an entry occurs a maximum of N times (So MAX 0009 would mean the entry happens 9 times and then stops).

IIRC such loop counters have been requested repeatedly. My memory tells me the biggest problem with them is, that there is not enough spare memory to store these and no infrastructure to keep track of the loop counts. After all would that count for single patterns or would you want to allow repetition of islands? What happens when you jump between patterns. Does that reset the counter? Etc. pp.

Don't get me wrong:
I had asked for such counters myself. But I see that apart from the trivial cases they actually are quite complicated and I think Thomas more than once stated, that he does not see a way to implement them.

Kind regards,
Michael

Yes I had a strong suspicion that this one is impossible due to memory... hence my opening line on that one. There is no doubt already a massive memory commitment in the Nerdseq. But from a musical perspective an EVERY command would be SO useful. Otherwise it means perhaps sending out triggers to something like a Teletype and performing the EVERY and other logic memory hogs there. 

In fact do you know how one might dynamically change probability using the CV inputs? Can I route to FX 1 %?  

Many thanks,

Jürgen 

[XORadmin]

Just woke up and 2 things to start with, the rest later and *** spoiler alert *** much is possible already.

Cv-> probability fx. Use the automators for that. That can also be an lfo -> probability and much more

Triplets, they are ratcheting features, use triggers E0 to E8
Or use the tables for more extensive ones.

[XORadmin]

1. Page Up / Page Down button behaviour: Could this be the same as cursor behaviour? Long press cursor works (data changes fast and navigation changes fast) but long press Page Up Page Down does not have the same behaviour.

Yep, requested before and will do that at some point.

2. Can we have Basenote option in MIDI FX?

There is Basenote for all track types

3. Can we have more information in Current Values? Basenote would be very good to have there.

Basenote is not a value that stays. It is only valid until the pattern is finished. For now the current values screen doesn't have any scrolling and you can see the most important values. If it is gonig to scroll then you don't have that direct view anymore. But I'm going to check that in the future when more functions are included.

4. Copy Colours with Mark (copy paste etc.). This currently leaves colours behind and the only way I see to move colours with patterns in a sequencer window is to delete empty positions to move the patterns AND their colours together.

This is very specific. One would not like these to be copied since they mostly belong to 'blocks' for them while others would like your approach, where you give patterns a specific color. I got to see what I'm doing with it, but that at a very low priority.

5. Visualisation of Probability % in the empty space at the bottom right of the Pattern window. The No.1 issue I have when working with probability is troubleshooting why something is not playing or not changing in some way. Its almost always probability percentage but if I could see FX % value in a similar way to the note activity in the Sequencer window that would be so helpful. There is space!

Yep, definitely usefull and this space need to be crowded.

6. Mute / Solo behaviour in Sequencer window is Great! Can we also have the same behaviour in the four MIDI note column?

Could be possible.

7. I've mentioned this before but the copy and paste approach using MARK is for me very slow. I also know that this is likely to be evolved soon! 
I really think I'm actually finding the Mark system challenging and sometimes wonder am using it correctly. Look forward to seeing next update approach! 
But anyway... The approach that would work faster for me is: 


Shift Copy -> Shift Copy (simply puts entry into memory for a paste command using Shift copy again)
 
Shift Delete -> (removes entry but also places it into memory for paste command using Shift Copy)

What you should consider is that this is a 3 button option to do all this. With a regular computer you got more buttons to archieve these things.
You will for sure get used to it and it allows all kind of copy/paste situations. And in the new coming version it is copy/cut and paste.
I agree that the name copy is maybe not the perfect chosen word, but that is now what it is and in the end it does copy ;-)
Do you know that there is also a mark all option in patterns? (SHIFT + DOWN)

I am not going to change the behaviour as it is right now. The new coming version got some improvements but that is all I will do with it.

8. Clock multipliers that allow for different tupelos (triplets, maybe even quintuplets, septuplets. etc... but definitely triplets. I've searched forum for answers to this and have not found anything that works using multipliers. Groove for me is quite a slow solution.

Answered that already.

9. Okay this one may be impossible... BUT. How about an EVERY or MAX function that can supplement the generative side of things like probability? For instance, EVERY 6 would only play THAT entry every 6th pass. Or Maybe every even EVERY 0606 is every 6th pas and EVERY 0406 is a repetition between four and six times on every pass (just add a random between 0 and 2 to 4 every pass to determine when the next event takes place. Similarly a MAX entry could state that an entry occurs a maximum of N times (So MAX 0009 would meanthe entry happens 9 times and then stops).

That has been discussed as well. As much as I would like to have the conditional stuff, to get the functionality every cell would need it's own counters and there is simply no memory for that.


I got to say that these packets of feature requests are very hard to find afterwards.
I can't read all the threads deeply to see what was in there when I am adding features. (Sometimes I need to find time to develope stuff ;-)  ) The ones which point out that one function in the subject got a much higher chance because I recognize it fastly.

[Zentrumsounds]

Yes... sorry about the package of requests (please think of them more as suggestions!). I will parse in future. 

2. Can we have Basenote option in MIDI FX?

There is Basenote for all track types

I must look harder. I don't see it. There are 43 FX for base notes but BASE is not there on my machine when looking at the MIDI pattern.  I mean SAMPLE AUDIO Pattern (to manipulate FM voice).

3. Can we have more information in Current Values? Basenote would be very good to have there.

Basenote is not a value that stays. It is only valid until the pattern is finished. For now the current values screen doesn't have any scrolling and you can see the most important values. If it is gonig to scroll then you don't have that direct view anymore. But I'm going to check that in the future when more functions are included.

So Current values is only for pervasive values not for shorter term ones? I understand the limitation regarding scrolling. Perhaps a second page that focuses on non-pervasive values?

4. Copy Colours with Mark (copy paste etc.). This currently leaves colours behind and the only way I see to move colours with patterns in a sequencer window is to delete empty positions to move the patterns AND their colours together.

This is very specific. One would not like these to be copied since they mostly belong to 'blocks' for them while others would like your approach, where you give patterns a specific color. I got to see what I'm doing with it, but that at a very low priority.

Aha. So quite a different intended use case. I am more thinking about how perhaps someone would use colours to designate an important pattern in a piece. For instance as a warning that changes need to be made in configuration during a live set then. Of that a pattern is in fact empty and is only there as a way of muting the previous pattern. If I move these in my arrangement then these details are however lost. 

7. I've mentioned this before but the copy and paste approach using MARK is for me very slow. I also know that this is likely to be evolved soon! 
I really think I'm actually finding the Mark system challenging and sometimes wonder am using it correctly. Look forward to seeing next update approach! 
But anyway... The approach that would work faster for me is: 


Shift Copy -> Shift Copy (simply puts entry into memory for a paste command using Shift copy again)
 
Shift Delete -> (removes entry but also places it into memory for paste command using Shift Copy)

What you should consider is that this is a 3 button option to do all this. With a regular computer you got more buttons to archieve these things.
You will for sure get used to it and it allows all kind of copy/paste situations. And in the new coming version it is copy/cut and paste.
I agree that the name copy is maybe not the perfect chosen word, but that is now what it is and in the end it does copy ;-)
Do you know that there is also a mark all option in patterns? (SHIFT + DOWN)

I am not going to change the behaviour as it is right now. The new coming version got some improvements but that is all I will do with it.

All clear!

9. Okay this one may be impossible... BUT. How about an EVERY or MAX function that can supplement the generative side of things like probability? For instance, EVERY 6 would only play THAT entry every 6th pass. Or Maybe every even EVERY 0606 is every 6th pas and EVERY 0406 is a repetition between four and six times on every pass (just add a random between 0 and 2 to 4 every pass to determine when the next event takes place. Similarly a MAX entry could state that an entry occurs a maximum of N times (So MAX 0009 would meanthe entry happens 9 times and then stops).

That has been discussed as well. As much as I would like to have the conditional stuff, to get the functionality every cell would need it's own counters and there is simply no memory for that.

Also clear. Acceptable given the power of the system. I'll send triggers into something like teletype (via i2C perhaps in future) and route from there. 

I got to say that these packets of feature requests are very hard to find afterwards.
I can't read all the threads deeply to see what was in there when I am adding features. (Sometimes I need to find time to develope stuff ;-)  ) The ones which point out that one function in the subject got a much higher chance because I recognize it fastly.

All good see above.