BrainModular

Hi,
I have crackles when I use my pedal board to send midi in and out of Usine. It happens only when I send audio (voice in a microphone) in another track.
It becomes obvious when I use a pedal (range fader 0>127).
I've checked everything : it seems it's because I've wired my MIDI IN module pin "receive out" to the "create in" pin of a CREATE MIDI module.
Is it a wrong way to create MIDI messages ?
Thanks

not really the classic way , but depending of what you want to achieve of course....?

here's a capture. I've simplified the patch to have a clear vision :

To be simple, Usine gets my pedal board midi messages, and use them to "midi create" other messages to be sent out to my hardware (guitar modeling effect and voice harmoniser).
I've made midi learns on switches and faders to match my pedalboard and use them for the values (0>127).

That's weird - unless your ASIO load is reaching is limit and the extra activity is too much for the machine. Do you have crackles when using the vocals and moving the fader with the mouse instead of from the pedal?

Come to think of it, it might have something to do with the MIDI signals interfering with the audio - possibly in the analogue domain. If you can, you might try to switch around MIDI and audio in- and outputs ans see if that helps.

and no problem with your patch...

Hi,
I've checked my hardware and nothing seems wrong but I have a clue : I send PC in another patch, on the same MIDI out. When I disconnect it, the crackles disapear. I'm going to join together the two patches to use only one MIDI out per channel and see what happens. Does the pass if change works on MIDI pins ?
Thanks for your help.

La Tenaille wrote:Does the pass if change works on MIDI pins ?

Haven't checked, but I think it will pass MIDI as well. However, I don't think it's of much use, as MIDI messages can be likened more to triggers from buttons rather than values from eg faders or switches. A MIDI output is populated with X number of messages per block in an array. If nothing is sent in the block, the array is empty (length =0) which can be considered as a "non-event" (aka NIL or NULL). Similarly, the button module has in fact now got an option for the off to be a NULL event - ie nothing is sent instead of the off value.

Ok Bsork,
I'll continue my investigation tomorrow, but the fact is I have two midi wires from two create MIDI modules (one for PC, one for CC) that feed the same midi out, and when I delete the PC wire, the CC messages stop generating crackles...

Well, then I don't have any suggestions other than it seems to me that there might some problems when combining several MIDI streams into one input. Any comments, Senso?

Hi,
I haven't found the solution yet. I have a clue : when I select the midi wire that goes in the MIDI out (data displays on screen), the crackles increase a lot, a crackle with each event.
Perhaps a ground loop ? Nobody has had this problem before ?
Thanks

no solution but curious what midi interface your using, is it usb or built in soundcard?
im experiencing crackles too but for me its when scrolling around in interface browser,, cpu spikes, then causes crackle,,
do you see spikes in your activity monitor/task manager when you hear the spikes?
for me i see the spike causing drop out,, its definately coming from usine, checking dpc latency,, no spikes there,,

Hi,
I'm using my built-in MIDI Fireface 400 interface, no spikes monitores in Win XP task manager.

Perhaps loveandelectrik your noises are caused by a ground loop. The simplest way to check this is to temporarily "cut" the ground of your laptop.

To isolate my problem I've made this patch in a new workspace :



MIDI in unpacked, sending CCs with two expression pedals to stress the midi messages.

The sampler plays a Usine demo file (lagoon).
With one MIDI out, no crackes (or too few crackles to be heard). They appear when I select the MIDI wire.

The crackles become obvious with the 3rd midi out, and terribles if I select a wire.
No crackles if I select the audio wire.

I've tried to cut the ground but it doesn't change anything.

... still stuck.

ok, I think I've found the reason: I've increased the buffer size of my sound card (latency was 48 samles, increased to 128) and it seems to work now.
With my simple patch above, when stressing the midi like a mad with my two pedals, the audio started slowing. I didn't imagine there was a link between MIDI activity and sound card buffer size !
I'm going to test this today to close this thread if my solution is concluding

good to see u fixed it,, 48s pretty small buffer, im rocking 128 with my fireface and latency is nice and low,,
my issues are different than yours, already using groundlift, but its in another thread so ill leave it there for now