Im noticing that the calibration of my sequencer and marbles do not match.
I can tell this since, when marbles is quantising already quantised (from sequencer) pitches - Im getting slight different frequencies out of Plaits. (about 25c sharp)
when I playback un-quantised from marbles it fine … but this means I cannot use ‘steps’.
I’m wondering whats the recommend way to get all my ducks in a row
Ive tried recalibrating my sequencer, but its still mis-aligned with Marbles.
( perhaps my multimeter is not accurate/calibrated well enough?)
buy a new Fluke multimeter (which will be factory calibrated) and then calibrate everything against that?
a bit concern this could turn into a game of chasing my tail, re-calibrating everything to match, or is this ‘normal’ in Eurorack.
note: spread = CCW , bias = 12 O’clock, so no offset present
EDIT: I guess one solution would be to reprogram all the scales in marbles, from my sequencer…
(should be easy enough to do, just let it repeatedly play a scale in the ‘ratios’ described in the marbles manual?!)
I’d personally do that yes. For best results I’d program a long musical phrase in the sequencer (make sure it contains all the notes, but some notes are used more often) then use that to create a scale in Marbles. If you only feed it a linear scale, you loose the ability to use the “scale carving” feature, which is one of the cool things about Marbles.
Can you check if Marbles’ miscalibration is on the input or output side?
Disable the external processing mode. Set STEPS fully CW, SPREAD fully CCW, and turn BIAS. It should generate perfect octaves. In particular, if BIAS is set to its minimum position (with the unipolar output range), the output should be 0.000V and should not cause any detuning when patched into a VCO. If all goes well at this step, then the miscalibration is on the input side, if not the miscalibration is on the output side.
Also, regarding your other question… I haven’t had the time to try on a real module, but I looked at the code and came up with the beginning of an explanation. In short, Marbles has three “output things”. When it’s set to external processing mode, with a common clock, only one “output thing” is active and records the notes. The two other “output things” are slaved to the first one and replay the sequence shifted in time, they don’t see the input signal. If you remove the common clock, these two other “output things” will no longer be slaved to the first one but they haven’t recorded any note because they were slaved, so they are stuck with whatever they were doing before you enabled external processing mode. I think I can come up with a software fix for that.
with 5v unipolar , it looks within a few cents of being perfect.
(at least first 4 octave, 5th is off… but perhaps this is hitting the limit of pitch input on plaits (around 6khz?) or something else)
there is about 15cents off, for BIAS = 0 and no jack… which seems pretty close to me.
so Marbles is good
I reckon its probably the sequencer, which in turns is likely due to my multimeter…
(as its both marbles and the sequencer ‘track’ plaits properly, perhaps its just a small offset?)
Im also starting to realise perhaps multimeters are not really accurate enough for calibration
Im wondering if perhaps the better approach for me is to calibrate the sequencer assuming plaits is calibrated - and then use a tuner, and tune the voltage to get the right pitch.
that’ll work for -3v to +5v … assuming plaits tracks over that voltage range nicely.
(I guess I can use the trick with BIAS you showed above with Marbles, to first calibrate Plaits)
thank you for the explanation, really interesting to hear about whats going on.
I’d incorrectly assumed that there was one buffer for ‘random’ values and they were distributed x1-x3,
but thinking about it, that wouldn’t work (in non-ext processing mode) , since when deja vi is enabled for X, you need the same values to go out on each X clock tick - so makes sense it’d store per output.
I guess the manuals description assumes that if you remove the patch cable from X clock, then the buffer for X1 , is copied to X2, X3?
yeah, Ive got a Squarp Pyramid, so my thought was I could save a song with all sorts of different scales programmed in it, then just replay different tracks to program each scale slot.
be a nice way to change things up a bit
is Marbles fixed to the 5 steps?, or does it just sort the notes into ‘buckets’ of pitches based on frequency of each pitch. e.g. could you have 7 steps, if you create more ‘buckets’.
(Im happy with 1,3,5 … so its just idle curiosity!)
I thought the “Bias” parameter was supposed to be able to accept 1v/o but the other night I plugged my Keystep into it and when I hit the keys absolutely nothing happened. An LFO into it worked fine though. Maybe I’m missing something?
Oh interesting. OK. So I’m wondering if I’ll be able to do something else I had in mind then. I’m about to buy a Digitone/Mutant Brain and I wanted to clock and send CV to the Marbles with it. My idea was to send clocks from the sequencer and different CV values per clock to the bias. I’m hoping this will be possible.
Anyone else using sequencers in interesting ways with Marbles?
For those who might be interested, i just investigated the alternate modes and tried to figure out what they do. I only listened to the t outputs.
Bias still sets the propability for a trigger on T1 or T3 to occur, but the decision is no longer made between those two. Instead a trigger can occur on both or on no channel according to your settings and the dice.
Polyrhythm machine. One channel gets multiplied, the other divided. Here is what ratios i heard going from 12 o’clock to CW (no guarantee).