MIDI CLOCK ticks and bea(s)ts

[cimo]

hi

i need to know how to use the midi clock

so far i ve understood that every beat will send/receive F8 while every tick (24 per beat) will send a syxex: is this correct? which syxex to be sent?

thanks simone

[stryd_one]

Close.... The way it works is that you have a master, and that sends a midi message of one single byte, 0xF8, which it sends 24 times per beat... Everything else will 'slave' to that master, by receiving those clock ticks, and counting them in a way that allows them to calculate the BPM.

What are you trying to do?

[audiocommander]

Hi Simone,

just a quick mentioning:

0xF8 is no SysEx

SysEx is a SystemCommon Message, whereas the Clock is a SystemRealtime Message ;)

http://www.midibox.org/dokuwiki/doku.php?id=midi_specification

http://www.midi.org/about-midi/table1.shtml

http://www.midi.org/about-midi/table2.shtml

and of course:

http://www.borg.com/~jglatt/tech/midispec.htm

Maybe a clear table helps you understanding all these strange numbers like it helped me :)

Regards,

Michael

[cimo]

hi

how does the slave understand which one of the 24 ticks is the "beat" one?

it s about a bpm controller for ableton (we re all enslaved to ableton!! and that s not midi clock)

sorry i have no time right now but i ll be back soon with our idea (me and alogic had a long chatting yesterday night)

now gotta go i ve met a girl who s asked me if i knew amon tobin and venetian snares, i am definitely in love!

simone

thanks

[audiocommander]

There are also START, STOP and CONTINUE messages ;)

Amon Tobin sounds very nice...

[cimo]

still... is there a specific message saying to the slave that a specific tick is the "beat" one or they are all considered the same by the slave an it will start counting from the first received? it can t be because drum machine will sync to the beat and no to the tick.

so far this is the plan:

let s take a plastic disc 1 cm bigger in radius than a normal vinyl, let s drill 64 small holes at regular distance among each other all the way round, let s play the turntable at 45 RPM and use a phototransistor with a diode, or any infrared device to read through the holes.Let s connect the phototransistor to a DIN and let the CORE send out the fabulous F8, there you go: control the BPM of a sequencer, vinyl style, BUT no STOP,no START,no scratch,etc, just BPM control.It good enough for me.I have open DINs on my controller with mini jack connection so i will connect it directly to my MB.

other considerations..

i want around 120 BPM at pitch 0, 45 RPM, 24 ticks per beat: that makes 64 dots around the disk

let s switch to 33 RPM and you get 88 BPM (play some down beat!!)

a technincs 1200 has 16% range on pitch shift that makes from 101 to 139 BPM at 45 RPM and from 74 to 92 BPM at 33 RPM

not enough? we ll make 2 different plastic disks, each with a different amount of dots to cover the whole range, or a disk with 2 rows of dots and 2 phototransistors and a "range switch", switching between the two

no place for a vinyl in your setup? then take a slip mat (??? one of those soft disks you put between the vinyl and the turntable) and place some kind of reflecting material at each dot (tick) (aluminum foil from your last falafel) and use an infrared device (again, or LED-PHOTOTRANSITOR) to detect the movement

i would add 3 more buttons and an encoder: a kill button to stop counting, 2 buttons for BPM up/down (cd player style) and an encoder to freak out with 999 BPM on your crazy sessions

it sounds sooo easy is it possible do you see anything i am missing?

best

simone

[stryd_one]

There's one major flaw with your concept - the BPM is determined by the record, not the turntable, so this would not have any advantage over just using a few buttons... Seems a lot of trouble to go to for no real benefit...but maybe I missed something :)

Even still, you could retrofit the turntable so that you could get the setting of the RPM and pitch adjust directly - no need to do crazy things with LEDs ;)

If you must use the optical encoder method you described, you would do better with two sets of holes around the outside like this:

x__x__x__x__x__x__x__

_x__x__x__x__x__x__x__

The asymmetrically staggered rows mean that the encoder can really see where it's going, otherwise you can get some strange patterns happening.

Another thing you might want to consider is the relationship between the size and number of holes, and the circumference of the disk, to the rate at which you would need to sample the encoder.

Anyway, as for midi clock, it works just like a metronome or a ticking watch... It just goes tick, tick, tick, tick, tick..... So there is no 'beginning' or 'end' or step, just a pulse. You are correct that it just starts counting from the tick at the 'beginning', but the ticks are all the same, so how does it know when to start? MIDI Start message (FA).

it can t be because drum machine will sync to the beat and no to the tick.

When you hit play on your seq, it sends a message to say "The song starts on the next clock tick", and then the drum machine starts to wait for a tick (F8). Next, your seq sends clock ticks as it plays. When the drum machine gets the first tick, it starts to play it's own sequence, and counts each tick as it arrives. It knows that there are 24 F8 ticks per quarter note, and so when it gets to the 24th tick, it knows that a 1/4 note has passed and it should play the next step.

If you don't start from the beginning of the song, it will use SPP to set the location within the song, and then continue from there.

I HIGHLY recommend that you read this entire site, but as a minimum, this article is a MUST : Syncing Sequence Playback

Hope that helps :)

[cimo]

There's one major flaw with your concept - the BPM is determined by the record, not the turntable, so this would not have any advantage over just using a few buttons... Seems a lot of trouble to go to for no real benefit...but maybe I missed something Smiley

well that s what i wanted i want to go round with the disk and the controller not my personal turntable

The asymmetrically staggered rows mean that the encoder can really see where it's going, otherwise you can get some strange patterns happening.

Another thing you might want to consider is the relationship between the size and number of holes, and the circumference of the disk, to the rate at which you would need to sample the encoder.

why should i need a double ring? i am not going to detect the backward spinning

thanks for the infos about MIDI CLOCKing it s more clear now to me, a buttons sending FA is what i need then to sync at the beat..

simone

simone

[Ruff-G]

one question in the meantime...

is it possible to build an external standalone midiclock with several midi-outputs?

regards

ruff-g

[audiocommander]

Yes, the Clockbox v 1.1. can do that:

http://www.midibox.org/forum/index.php?topic=5691.msg48932#msg48932

(and ucapps.de -> download)

Regards,

Michael

[Ruff-G]

thank you for this helpful link :)

[Ruff-G]

i have another question...

can i set a different clock delay to each midi output?

regards

ruff-g

[stryd_one]

If you mean clock speeds, yes :) If you mean a delay as in an offset, no