when will come the TR seq?

[xarolium]

hello

i want to make a seq for drum like a tr808 or 606 and i want to know if nobody have ever make it ?

thank you

[stryd_one]

It's on the way. A few of us will be working on this based on a framework which Bill has started. You're welcome to help out!

[xarolium]

I'm sorry but can't help you because a frog is more intelligent than Me ;D ;D

i don't how to make a program for PIC

sorry

but i'm waiting your excellent  work

and i want just to know if the seq will be so easy to use like the tr808 seq?

thank you

[Jaicen]

I'd like to help out, but I'm not sure how much use i'd be!

I really need a TR style sequencer for my upcoming analogue drum clones. Something simple and low cost is the order of the day if you ask me.

I'd like to see something in a similar style to the TR-606. A row of 16 step buttons (With LED's to denote accent level. More brightness = more accent). There should be a permanent 'accent' (velocity I guess) as well, with a rotary to select the accent/velocity level. Each step should have a default level which can be varied up or down using the pot. This does away with having to assign an accent to each step, making it more intuitive in my opinion.

Perhaps a row of bank and pattern buttons for changing on the fly. I think it would also be nice to have a rotary encoder for changing tracks on the fly, ie go from editing the kick patterns to the snare.

I guess a shuffle knob wouldn't go amiss either, but I think that's all that i'd be looking for from a drum sequencer. Oh, and a Mute button for killing instruments on the fly would be cool.

I'd need to be able to sequence around 8 tracks at once, but I guess that's not really important.

If there's anything I can do (that doesn't involve programming!) let me know!

[nebula]

It's on the way. A few of us will be working on this based on a framework which Bill has started. You're welcome to help out!

Stryd,

Any forum posts or wiki pages on this?

I'm still trying to figure out the best way to approach a 'schaltwerk' -type sequencer, which is really just like a tr-style sequencer with multiple rows of buttons and LEDs.

[stryd_one]

There are bits and pieces around the place but not much. Bill mentioned that he has already started on this so I will wait until I've seen that before I say what's next...

[DrBunsen]

The new DrumStuff wiki page links through to the original threads

[sonicwarrior]

What's with the MulletRhythm project?

http://www.mulletronic.com/diy/mr9090/

It is also PIC-based and it looks like the developer could need some help

(last new from 30.3.2006).

Also Open Source but has IMO nothing to do with the MIDIbox stuff.

[TK.]

What I don't like on this project is, that he is using Trevor Page's source code, and replaced the copyright by his own. And he uses a license which is not intended for software.

Best Regards, Thorsten.

[audiocommander]

And he uses a license which is not intended for software.

I have to aks about that, because I belive CC is a great thing and I also often use this type of license, esp. on the code-snippets I uploaded on the Wiki (BY-NC-SA, which means: name the name, non-commercial, share as you like - incl. right to change or alter, but then under the same conditions).

And although it has only 6 letters, it should have juristical success if it would ever come to some sort of trial...

What type of license would you wish to see instead (esp. for contributions to the community)?

Cheers, ;)

Michael

[Twin-X]

What I don't like on this project is, that he is using Trevor Page's source code, and replaced the copyright by his own. And he uses a license which is not intended for software.

Best Regards, Thorsten.

Hey i did not know that.

It was discussed a few times on the mail group i don't think Trevor has a problem with this.

[TK.]

If Trevor has no problems with this, it's ok. Nevertheless, I would mention the original name in each re-used file, regardless if (small) modifications have been made or not. It's just good style

To the CC - in the FAQ http://creativecommons.org/faq#Can_I_use_a_Creative_Commons_license_for_software? you can read:

Can I use a Creative Commons license for software?

Creative Commons licenses are not intended to apply to software. 
They should not be used for software. 
We strongly encourage you to use one of the very good software licenses available today. 
[/code]

What type of license would you wish to see instead (esp. for contributions to the community)?

either GNUPL if it's ok for you if somebody uses the code commercially (e.g. sells your projects on Ebay for profit), or no "copyleft" license at all, but just something like this hint, which can be found in most of my projects in the meantime:
[code]
;  Copyright <date> <name>
;  Licensed for personal non-commercial use only.
;  All other rights reserved.

Best Regards, Thorsten.

[bill]

i want to make a seq for drum like a tr808 or 606 and i want to know if nobody have ever make it ?

Hello,

please, could you write a list of features a good old tr seq should have ?

(resolution, interface, number of voices ...)

and i want just to know if the seq will be so easy to use like the tr808 seq?

please explain us how easy it is, i dont own a tr808  :'(

also describe the machine interface, and write it on a wiki page.

(please dont do complicated)

I did this page for it : http://www.midibox.org/dokuwiki/doku.php?id=trsequencer

You do not need to learn ho to code for this, and it could help the project a lot !!!

Hope to read your descriptions soon  :)

[sonicwarrior]

please, could you write a list of features a good old tr seq should have ?

(resolution, interface, number of voices ...)

Features:

- 16 trigger tracks (I think this is what you call "voices") + Accent,

this would allow to do a fully featured TR-808 clone (which had 16 instruments)

- Shuffle

- For the resolution I think 96 ppq would be OK.

- At least 32 steps per bar as maximum

- Editing while playing

- Syncable via MIDI Sync (Master or Slave)

Interface:

- A dedicated tempo poti

- At least 16 buttons with LEDs for the Steps,

I would personally prefer a solution with 32 buttons: 16 bigger ones and 16 small ones

so that the machine wouldn't be blown up to much in size.

It would be the first machine with direct access to 32 steps.

With a shift button the step buttons could be used to select the track/instrument/voice

or however you like to call it.

- Patterns could be selected also with the step buttons and another shift button

- Transport buttons: Stop, Pause/Play

That would be the base.

[TK.]

I would personally prefer a solution with 32 buttons

I second that, 32 buttons/LEDs are very important, the additional LEDs not at least to display the current sequencer position correctly.

A roll function (selectable like the accent for each step seperately) would be usefull as well

Best Regards, Thorsten.

[xarolium]

if we can do a sequencer like the korg ER-1 with 64 step

but it's a good id to put 32 switch for the live playing

[cheater]

if we can do a sequencer like the korg ER-1 with 64 step

but it's a good id to put 32 switch for the live playing

I like the idea with the small buttons. However...

How about just chaining sequencers for more? Then you could put one next to the other to have 64, 96, 128, 256, 512, 1024, 2048, 4096 or more steps ;)

What I'd like would be a led matrix which would show which step for each voice is on.

Accent: you have to realize there are two kinds of "accent". Some TRs had global accent - a single "accent" track where you selected "accent" (in the selector where you selected between instruments) and each step where it was turned on, it accentuated everything - the most common 16-step seq that did that is TB303. Some had accent for each voice separately (it was like a third state of the step - "off"/"on"/"accent"). I think the second idea is better. However, perhaps instead of having to click twice to select "accent" for a step, and have to edit in a menu what velocity that would be, a more hands-on approach would be nicer - where you'd have a knob that would select what velocity you would be setting the current steps to. You could easily make rolls that way, brushing your fingers over the buttons while turning the knob up - but that's just one idea.

Shuffle: personally, I'd *love* something like "groove clips" in acid 5 mutated into shuffle. Those groove clips work like this: you open up a normal-looking bar of, lessay, 4/4 metrum. Each 1/4th or 1/8th has a vertical line representing it. You can drag those lines around, in such a way that those "bends" are kindof smooth. You can make notes earlier or later. So what it actually would mean for the sequencer would be: if the sequencer is 96 ppqn, a normal straight beat would mean "play each step 24 pulses after the previous one" (we're assuming the resolution is the same as midi resolution here, for simplicity). A groove clip would then be able to say, "the 3rd note is after 75, instead of 72 pulses; and the 11th note is after 260, instead of 264 pulses". To save PIC processing time, an algorithm (in a PC app for example) could compile such lookup tables for each (of lessay 32 possible) "shuffle" knob positions. Those table-sets could then be uploaded into a PIC.

Those tables could also contain other info, like gate time or some CC info. It would be like a wavetable of wavetables really.

Thanks for reading all that! Cheers :)

[Jaicen]

However, perhaps instead of having to click twice to select "accent" for a step, and have to edit in a menu what velocity that would be, a more hands-on approach would be nicer - where you'd have a knob that would select what velocity you would be setting the current steps to. You could easily make rolls that way, brushing your fingers over the buttons while turning the knob up - but that's just one idea

I'm fairly sure I suggested the same thing in another thread, but I thought everyone else deemed it a bad idea ;)

I think the easiest way to implement this is to just have the knob select velocity for each note event, rather than having a specific 'accent' control. Most modern drum machines (even the 505!) can respond to velocity, even if they don't send it.

I really like your idea of shuffle 'tables' as well. The way I see it working would be to have a number of presets which you can the assign to each 16-bar pattern, I think that would be the easiest implementation.

Another way to do it would be to have the shuffle work as a positive going LFO with variable waveshape depth and speed, that way you could have a continuosly variable shuffle which might be nice.

Finally, i'd be looking to keep the user interface fairly stripped down. 16 step buttons is plenty, perhaps with variations (ABCD). I'd have just the 16 buttons for note on (trig) then a pot for each step to set the velocity value (which would default at around 70). It would be nice if the LED got brighter to indicate velocity level, but that could be complex.  The only other controls i'd like would be transport controls and pattern selectors (ABCD 1-8 for 32 different patterns per song). Everything else I'd control via an LCD, such as writing songs (chaining patterns)

[sonicwarrior]

Perhaps we should set up a vote for 32 step button solution (with the half being smaller ones).

In modern music styles 32 steps per bar are very important.

That's why I think it should be possible to directly edit them

and not via chaining.

[Twin-X]

I vote for 2 rows of 16 buttons for the step.

Where the lower or upper section can be switched for other functions. This saves buttons and room on your layout.

[cheater]

However, perhaps instead of having to click twice to select "accent" for a step, and have to edit in a menu what velocity that would be, a more hands-on approach would be nicer - where you'd have a knob that would select what velocity you would be setting the current steps to. You could easily make rolls that way, brushing your fingers over the buttons while turning the knob up - but that's just one idea

I'm fairly sure I suggested the same thing in another thread, but I thought everyone else deemed it a bad idea ;)

I think the easiest way to implement this is to just have the knob select velocity for each note event, rather than having a specific 'accent' control. Most modern drum machines (even the 505!) can respond to velocity, even if they don't send it.

But that's exactly my idea as well: the knob sets the velocity of notes you'll click :)

I really like your idea of shuffle 'tables' as well. The way I see it working would be to have a number of presets which you can the assign to each 16-bar pattern, I think that would be the easiest implementation.

Another way to do it would be to have the shuffle work as a positive going LFO with variable waveshape depth and speed, that way you could have a continuosly variable shuffle which might be nice.

I was thinking of it more performance-wise - set a track to a shuffle "table" and then, once it's set, change the amount of shuffle with a knob while performing your song - but it could be more than "shuffle amount", the shuffle wouldn't need to linearly apply to the pattern. It would be like this: knob at 0% - no shuffle. going from 0% to 50% - you add shuffle to every 4th step (3, 7, 11, 15). Going from 50% to 100% - you add shuffle to steps 2, 6, 10, 14 while the shuffle to 3, 7, 11, 15 stays how it was.

Can you dig it? :)

[Jaicen]

Yeah I dig it!

Actually what would be unbelievably cool is to have a feature to mimic the way a drummer plays a real kit. For example, if they're playing a straight beat on the snare, then go into a roll ending on a cymbal, it takes time to reach across to the crash and back to the snare, which slightly delays the first beat of the following bar.

Perhaps an algorithm can be developed to add slight delays between adjacent 'hits' depending on which drum is being played.

I don't know if this is a new idea, but I think it would be cool, especially when you're sequencing real drums sounds.

[sonicwarrior]

I don't know if this is a new idea, but I think it would be cool, especially when you're sequencing real drums sounds.

We're talking about a TR-style sequencer.

For sequencing real drums you should IMHO use software like FXPansion BFD

and also a software sequencer. That gives you more control about things

like "groove quantize" etc. .

[Jaicen]

I know that, i've been following this thread for some time. I just think that the TR sequencer has a really nice interface so any similar design should not be limited to one intended use, especially for those that want to play live without a PC/Laptop.

[cheater]

I know that, i've been following this thread for some time. I just think that the TR sequencer has a really nice interface so any similar design should not be limited to one intended use, especially for those that want to play live without a PC/Laptop.

Stay focused.

A does-it-all is pointless and often useless in live situations.