Hello everyone!

I recently stumbled upon a PuraData based step sequencer for my QuNeo MIDI controller, the so called Rythmboy:
http://subtersonic.weebly.com/blog/rhythmboy-a-step-sequencer-for-kmi-quneo

I want to use it as a step sequencer to use with the remix decks in Traktor, so I can add drums to the tracks I play.
Unfortunately, the Rythmboy has an internal MIDI clock, so I can't sync it with an external clock. I already tried to replace the BPM rotary in the patch with "MIDIrealtimein" and the stuff that's also needed, but I can't get it to work.

I managed to do all the other stuff that is needed to work with Traktor (changing the note output, optically mirroring the rotaries, limiting the notes ranges etc.), so I tried, but I don't know how to work with an external midi clock. Another problem that may occur is that the Rhythmboy then needs two MIDI inputs, #1 is the QuNeo and #2 is the external clock (Traktor or my Xone4D mixer).

This is the part of the Rhythmboy that controls the MIDI clock:

The blue box is a tap button to tap in the tempo. This is not needed.
The red box converts a rotary MIDI input between 0 and 127 into a BPM value between 60 and 240 bpm and gives out that value. This is where the changes need to happen I think. I need something that changes the MIDI clock in into a numerical BPM value.
The green box gives output to the rotary LEDs to show the current BPM. This is also not needed.
The other stuff lets a LED blink a metronome, I think.

So, would one of you please tell me what to put in instead of the red box, to make the magic happen?

Best regards,

Spatz

EDIT: I updated the [midirealtimein] patch to be correct in my earlier post.

@rjp9 :
This is the third time I rewrite this post, as I am digging deeper and deeper into the Rhythmboy, to understand how it works. And I am learning a lot.
Lesson 1: The Rhythmboy is deeply rooted into using [receive metro_set], which is a float number, and represents the length of a quarter note in milliseconds. Changing this is nearly impssible, so I need to live with that, and work around it.
Lesson 2: The sync between the Rythmboy and an external clock seems a bit jittery. When I use the bangs of the MIDI clock to trigger a timer, the value jumps the a lot, for example between 499.2 ms and 500.2 ms for 120 bpm (500 ms). When I raise [mod x], the accuracy rises, but also the inertia, meaning the timer will react slowly to suddenly changing bpm. When I lower [mod x], it reacts quickly, but the values jump like crazy.

So after trying a lot, I think it's time to do something else... I just got some kind of "sync" up and running, at least with the restraints mentioned above, and it sounds terrible. So, off to new projects! Thanks for your help anyways!

@Spatz I guess I'm just curious why you are using [timer]. Think of it this way: if you are trying to sync your tempo to an external midi stream, you need to replace your timing mechanisms. Instead of using [metro] to output bangs, you are deriving quarter note bangs (or whatever division you want) from the [midirealtimein] patch I made. This means REPLACING everything linked to metro.

I asked above why you need to display bpm in your patch. This is because if you are syncing your tempo to an external midi stream, you will not be able to set your tempo from within your patch.

For example, here's what you would have to do to make this patch work. I downloaded the RhythmBoy patch and took a look around. Pretty messy, but I found what the [receive metro_set] is for:

Essentially, it's for determining the subdivision if you ever want to change it. If you are using metro, you need to know the exact bpm in order to do this. Essentially you take your original bpm (let's say 120) and multiply by some number if you want to work with smaller subdivisions. 120*2=240bpm which is eights notes at 120 bpm. 120 * 4 = 480 bpm which is 16th notes at 120 bpm and so forth.

When using [midirealtimein], you have to think about everything backwards. The midi message "Continue" (which you derive your timing info from) is sent at some crazy number like 200 times a second. I'm sure there's someone more knowledgable than me who knows the actual number!

In any case, in order to derive tempo and subdivisions, you simply select how large of a "cycle" you want to represent your beat. This is accomplished with modular arithmetic (aka the [mod] or [% ] objects. So with [midirealtimein], you have to start with your smallest beat division and work from there like so:

midihelp.pd

In any case, if you want your to modify you patch to be able to sync up eternally, it is definitely possible. It would take a lot of work because you would have to swap out the whole [metro] mechanism for something like the above.

Hey there!

Thanks a lot for you help! I think I understand where you want me to go, and I will look into it.
BUT:
I found another problem with the interaction between Rhythmboy and Traktor that may or should be solvable, but right now it is big enough to make me stop going further.
The problem is as follows:
Rhythmboy outputs a certain note with a certain volume and length. This length is internally set to 125 ms. Traktor then plays every sample only an 8th of a second, which is okay for kicks and snares, but problemativ for longer samples like cymbals, rides or vocals.
If I set the length longer, to 5000 ms for example, to cover loooong cymbals and vocals, shorter notes won't get triggered properly.
I need to convince the Rhythmboy to trigger the note correctly, otherwise some sounds that should be triggered won't play in Traktor.
Setting the timing different and dependend on the triggered notes won't work, as one can only tell the time between two notes after the second note was played, and maybe the rythm is shuffled or not that monotone, so you can't just use the time of the two notes before. Rhythmboy really needs to send another note.
Maybe one workaround would be to send the notes always with a velocity difference of -1 (127, 126, 127 and so on).
What do you think? Is this problem solvable? If yes, how?
Otherwise it wouldn't make sense to dig deeper into the MIDI syncing bits.

Bye,

Spatz

PS: I already checked if Traktor treats CC messages differently, but they are time dependent, too...

PPS: I also checked checked with MIDI-OX wether the problem is on the Traktor- or on the Rhythmboy-side, and it is the Rhythmboy output that is faulty...

@Spatz said:

Rhythmboy outputs a certain note with a certain volume and length. This length is internally set to 125 ms. Traktor then plays every sample only an 8th of a second, which is okay for kicks and snares, but problemativ for longer samples like cymbals, rides or vocals.
If I set the length longer, to 5000 ms for example, to cover loooong cymbals and vocals, shorter notes won't get triggered properly.
I need to convince the Rhythmboy to trigger the note correctly, otherwise some sounds that should be triggered won't play in Traktor.

This is hard for me to imagine...(since I don't have traktor and can't test it out, but I'm guessing your Rhythmboy patch isn't "polyphonic", that is, it can't handle multiple instances of noteon/noteoff messages. In other words, the patch can only handle generating one note at a time.

Maybe one workaround would be to send the notes always with a velocity difference of -1 (127, 126, 127 and so on).
What do you think? Is this problem solvable? If yes, how?

I like the intuition you've got here. You've recognized that there has to be some sort of variation between notes. I think the answer can be much simpler (and which doesn't require artificially messing with velocities).

Before I answer, though, I would like to know how serious you are about continuing with this patch. Would you be willing to design your own from scratch? I'd be more then happy to help you through the process, but I don't really feel like changing the very fundamentals of the patch you've already downloaded. You'd be able to customize it exactly how you want (plus it would help you learn more imo).

I'm pretty serious about getting this patch up and running. Until now, while discovering the patch and finding out how it works, I always discovered new problems that made me want to quit and go on without a step seqeuncer, but I think that now I have found all the "bugs" I need to get rid of to make this thing work with Traktor. And the idea of using my QuNeo, which would else be rather unused, in my actual setup is very tempting.

But, to be honest, I'm not that much into programming (I only know about pd for one week), and I think I could not program my own sequencer from scratch, at least in an acceptable amount of time and work. Plus, the Rhythmboy already has already 90% of the work I need done, and I'm a lazy person, in the techy meaning of lazy, which means not to do any work that is already done. For example, the story of this guy made me smile.

So yes, I'd love to fix the Rhythmboy, and put some work and learning into it, but I don't think I'd build a new step sequencer from scratch. Yet, I know that your help here already gave me a lot of ideas to work with, and I'm not one of those guys who expect someone else to solve the whole problem for me. I love to find my own solutions if someone points me into the right directions, which you already did, and my main goal is not a perfect program, but something that at least works as it should.

Now for the technicals: Yes, the Rhythmboy isn't polyphonic, but neither are the remix decks in Traktor. The remix decks in Traktor are multiple cells in multiple slots, that can be triggered by a MIDI command, and can be a loop or a one shot. So in other words, it's a very simple sampler.

I got the MIDI sync up and running! Testing now!

Yep, it works as it should! Thank you for showing me the way!

Now the problem with the note length remains, but you said you maybe have an easy solution for that?

Oh, and one more thing, but this is only a detail:
I want the "reset" and "play" button only to trigger on a downbeat. I tried this by filtering the button signal through a [spigot] with the MIDI clock and [mod] and [sel 0] on the right inlet, but this doesn't work as intended, no signal passes through then. Do you have any idea? To be more exact, I want to store the "play" and "reset" triggers until the next time the [mod 24] from the MIDI clock is 0.

Bye, Spatz

@Spatz said:

Now the problem with the note length remains, but you said you maybe have an easy solution for that?

I'd be glad to help. Could you describe this problem in a little more detail? I think I have an idea (that I hinted at above), but I want to make sure. Could you post an example if possible?

The samples in Traktor only trigger with the length of the note that comes out of the Rhythmboy, so especially for longer samples, the length should be high enough, something like 5000 ms.
But if the length of one note is longer than it takes to play the next note, the next note will not be played, or at least it seems like some notes are randomly skipped, whereas the chance to be faulty rises with the length:
10 ms -> everything fine, but even the base drum is just some ticktickticktick
5000 ms -> the base drum is a whoompf, as it should be, but is only triggered once per every few sceonds
So I need to make sure that the note length is always long enough to play the sample as long as it should (sample is over, or same sample triggered again), but still short enough to not disturb the triggering of the next sample. The problem is that we can't use fixed or calculated lengths, because sometimes the rhythm could not be absolutely straight, and we need the exact time at the moment the first note ist played, and then we don't know yet when the next note will be triggered.
Because of this I think it is better to concentrate on properly retriggering the notes.

Rhythmboy modded.pd

Here's the modified patch, my modifications are mainly on the first layer (MIDI clock in, metronome, syncing "play" and "reset" with the clock), and on the track layers (the part you already made a screenshot from, adapted to the midiclock, the [send subdivision_track] part is only responsible for triggering the lights and thus is not that time critical).
Furthermore, I've flipped the rotaries upside down (the QuNeo sends volumes like a watch, 0 and 127 is at the top, not like a normal rotary, where both is on the bottom. I flipped this), and changed the output notes and ranges to work with Traktor.
Maybe I didn't not find the cleanest solutions, but at least it works.

@Spatz Cool. Now what does your Traktor set up look like? What are you doing to actually play your samples in your DAW?

@rjp9: My setup basically is the following:
I have an Allen&Heath Xone4D mixer, which has an built in audio interface and some MIDI controllers. The mixer also sends a clock signal to sync up all decks and the Rhythmboy/QuNeo. The Xone is the main mixer, and the midi controllers control the first two decks in Traktor, which are normal decks to play tracks like usual.
The other two decks are set up as remix decks, which each offer 4 columns with 16 cells each (so 2 * 4 * 16 = 128 cells total), which can play one-shots or loops. These are controlled by the Rhythmboy.
My goal is to not only play normal tracks in Traktor, but to enhance them with my own samples and drums, and even do some live remixing by playing/looping non-electronic tracks on the first two decks and backing them up with my own drum loops from the QuNeo/Rhythmboy, to give my DJ sets a special touch that not everyone offers.

The Xone is the new component in this setup, until now I did all mixing in Traktor (on 2 normal decks) with my QuNeo. Now that I have this new mixer, which has enough buttons/faders etc. to render the QuNeo useless, I still want to use the QuNeo in my setup in some way. This is why I looked for step sequencers for the QuNeo and found the Rhythmboy. And the rest of the story happened more or less here.

I hope this answers your question...

Have you looked at my modifications to the rhythmboy? What do you think? Sorry for not commenting, I hope you still get what does what...

Bye, Spatz

@rjp9
Could you maybe tell me what solution you were thinking about. Just a hint, which command to use or what to do? I will then have it easier to find out for myself. Thanks!
I've already tried the flipping velocity thing, but this doesn't help.

@Spatz said:

Could you maybe tell me what solution you were thinking about.

Take a look at this sample patch and tell me if this is sort of the problem you are experiencing. As I understand it, when you are triggering the same sample multiple times, the sound is getting cut off. Is that correct?

polyhelp.zip

@Spatz said:

Have you looked at my modifications to the rhythmboy? What do you think?

As far as I can tell, it looks like you've made the right changes to use an external MIDI clock signal as the source of your timing information. Is everything working for you?

@rjp9
I'll have a look at the patch tonight. I have to work soon.

So far, the MIDI sync works as it should. I've added a [+ x] to the MIDI clock counter to fine tune the timing, so the beats of the Rhythmboy and Traktor are in sync.

The last bug I have to fix are some performance problems: When I switch pages (tracks 1-4 or 5-8) the redrawing of the LEDs causes the Rhythmboy to stutter for a moment. You can see how the beats are delayed, for example when you look at the result of the timer that measures the time between two beats. Usually it is around 125 ms for 120 BPM, but when I switch pages, the value drifts a lot for on second.
I've already tried to add small delays to the redrawing of the different tracks, but this didn't help. The redrawing only occurs of the MIDI output to the QuNeo (the LED output) is activated, else it doesn't stutter.
Do you have any idea how to reduce this stutter?