Magic MIDI Delay

[ultra]

you got me thinking about this:

get into the delay menu, select which track you want to edit directly with the corresponding.  track select button.

the three main delay parameters for each of the four delays shows up on the lcd above the encoders.  this would take 12 of the 16 spaces available on that immediate screen.

space 13 could have a space, and space 14-16 could have the option to include layer a, b, or c in the delay!

is anybody at all interested in lifting the mmd code and adding a menu to the seq?  unfortunately when i look at asm code my head spins.

[dcer10]

Thats a nice panel layout, would also be good to include holes for 2 panel mount LED's for an LTC module so you can see the in/out as well as have a thru. I think it would be good to have a few (or 4)of these in one rack with some sort of routing if they were not in the SEQ itself (which would me more handy as maybe the mixer layouts could include some saved delay configs and it could run from the master clock and benifit from the divider settings in the seq?).

I have been finding that you can get interesting results having the input of the seq have the delay in it for recording from a master keyboard, then placing it over one of the IIC outputs afterwards to split out over some different channels with a different kind of delay. Having it realtime record the input with the delay can make for some interesting patterns. I wonder if it would fit into the sequencers firmware and agree with the concepts of the firmware if someone actually had the time to do the programming?

[stryd_one]

The current firmware does not seem to do this, you can manually select the BPM but nothing more in terms of the sync. Perhaps someone could develop this?

Grab the clockbox sync code and strap it into mmd_clock.inc

Piece of cake (yeh right)

Maybe for you guys it might even be easier to recreate the app in C?

I doubt that the seq has enough resources to build this into the same core...

[dcer10]

Maybe for you guys it might even be easier to recreate the app in C?

I think I would have a better chance of walking on water  ;)

I doubt that the seq has enough resources to build this into the same core...

What about on a seperate core(s)?

[stryd_one]

seperate core/display/buttons etc, yeh.... Sorry I shoulda been more specific. Essentially another box. Sorry bro :(

Uhm, anyway I broke two of my own golden rules. Shoot me.

1) I didn't copy and paste the filename. It was mmd_timer.inc. Shoot me.

2) I was not patient. If I was, I would have read the code, and seen that it already has MIDI sync built in. Set the BPM to 0, and it's driven by midi clock. You can start bowing to TK.... now. And shoot me.

:D

Tasty code snippets:

	;; if BPM is zero, the delay will act as MIDI slave. Stop timer in this case
	movf	MMD_BPM, W
	bz	MMD_Timer_BPM_Stop


USER_MPROC_NotifyReceivedByte
	;; if MMD_BPM is not zero (not in MIDI slave mode), ignore incoming clock
	movf	MMD_BPM, W
	skpz
	return

	;; check if MIDI Clock, jump to MIDI clock handler if true
	movf	MIOS_PARAMETER1, W
	xorlw	0xf8
	skpnz
	call	MMD_CORE_ClockHandler

	return

	

[ultra]

this wouldn't be too hard of an app to recreate as long as i can figure out timing.

i like the idea of this, especially delaying/repeating control changes.

however, how do you recommend it behaves, dcer?

it could take a cc at any moment and repeat it a set time later, but would it return the parameter to the starting value, or stay put?  how would you determine what is starting value?

what do you think dcer10?  could you give me a description of how delaying cc's could look?

[ultra]

i swear i read your post but missed the part about it syncing when the bpm is set to 0.

i'm gonna build this as-is :D

i can live with just sequencing cc's from the mb-seq and delaying notes from the magic midi delay.  in fact, i wouldn't exactly call it "live with", this is gonna be awesome. :)

[stryd_one]

Well, you could probably mod it to delay CC's quite easily too ;)

[dcer10]

If you did you might want to "transpose" them to other cc numbers so that they never happened at the same time else you will glitch your synths out (maybe desired??). Would be nice to be able to delay sysex too... Im thinking of a situation where Im using my Yamaha TX81Z (or any other FM synth needing sysex control) and to be able to delay the notes as an echo, but to send some to other channels (in performance mode with other voices happening) and then to finally delay some sysex to control the cutoff freq in a tempo sync manner. The possibilities are endless with this thing. Its quite simple as is, but im sure it could be dressed up a lot, especially with the developments in ram on the chips since its invention.....

[ultra]

so i've finally come back to this project and i think i can quickly get it done now.

i've replaced the 30 buttons with 15 encoders.  although i'm not sure if this is completely practical or not.  do any of the buttons work as "modifiers" where you have to hold it for another function?  or are all buttons just a quick press, and can be replaced with encoders?

thanks,

ultra

[dcer10]

I think encoders would work just fine as its pretty simple and there isnt any double pressing involved from what I remember of it. I would put in a button for "slave" which sets the BPM to zero and puts it into slave mode if you can work out how to do it. Constantly pushing the bpm down button on boot up is a bit annoying :)

Good luck , post some photos when you get it done!

[ultra]

thanks dcer10.  unfortunately i went ahead and ordered a schaeffer panel so no slave button.

however, once i learn about how it all works, i'd like to redo this one in c because i think some cool modifications could be done and i don't know assembler.

stryd and i were talking about cc sequencing a bit and that might be a fun add-on.  but at the very least, i want to re-write it anyway so i can learn a bit about timing, time delays, and that sort of thing.  and i could also have it store last used settings to a bankstick.

i'll post some pics soon.  it should be done in a couple weeks.  i'm just waiting on schaeffer. :D

[dcer10]

Let me know how you go, esp with sequencing by cc's... Very interesting....

Got a pic of your panel design handy??

[ultra]

yeah here it is

[dcer10]

Nice. Much cleaner with encoders!!!

[ultra]

yeah i like it.  it's basic but it was easy to wire up a proto board and the panel was pretty cheap.  it fits into a pactec pt-8 case.  :)

[tashikoma]

hi

is it a delay like an analog delay for midi or a function who define one time for one event?

thanks

[Smithy]

Here a short demo tune:

http://www.midibox.org/mios/magic_midi_delay_demo.mp3

Reminds me of the McGyver intro for some reason!

hi

is it a delay like an analog delay for midi or a function who define one time for one event?

thanks

Im guessing its a "midi delay" i.e. it copies the incoming notes and plays them back a second time, after a chosen time interval, to achieve a delay like effect.

but thats just a total guess, i havent looked into the application at all yet.

Edited February 14, 2010 by Smithy

[tashikoma]

thanks Smithy

i was thinking that it doesn't play the original event but delaying back it!!!

it will be cool to have a feedback fonction (controlled by button or encoder) for repeating more than one time.........!!!!!!

(two times three times .....six times....)

exactly like audio delays

one time to hold.........!!!! so it could be very simple!!! a killing grooving machine to midi looper.......

i have no idea to program that!!!!!!! :frantics:

if someone can do that i will build it!!!!!

Edited March 18, 2010 by tashikoma

[Smithy]

Knowing TK, he may have added an option where you can choose between Enabling or disabling the original notes!