Preemptive Step Sequencing?

Cwejman, Livewire, TipTop Audio, Doepfer etc... Get your euro on!

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
waltergoschen
Learning to Wiggle
Posts: 5
Joined: Fri Mar 20, 2020 7:55 am

Preemptive Step Sequencing?

Post by waltergoschen » Tue May 26, 2020 5:24 pm

Does anyone know why sequenced changes to certain parameters on the Basimilus are inaudible when triggered with just a single step, and seem to require the previous step to “prepare” it with a similar voltage first?

To be precise, I’m using Voltage Block to make a one-module drum kit out of the BIA. When I set up a step to have significantly less voltage than the previous one to (for example) trigger the noise on the attack parameter in the BIA, the step will end up sounding identical to the one before it in the sequence (despite how, if I hold down the step’s button, it’ll repeat the desired noise-infused sound). However, if I set the previous step to a similar voltage, the step I’m trying to affect will make the desired sound (though the sound of the previous one will remain unaffected).

So for some reason with the attack parameter, as well as the S/L/M parameter, I can’t adjust the sound of a step without adjusting the step before it. With other parameters, like pitch, I’m able to hear any changes to a single step the way I’d expect to, regardless of the settings of adjacent steps.

If anyone could help, I’d hugely appreciate it! Sorry if that was a prolix explanation.

User avatar
fruitsnake
Common Wiggler
Posts: 137
Joined: Sun Nov 18, 2018 5:56 pm
Location: Brooklyn

Re: Preemptive Step Sequencing?

Post by fruitsnake » Tue May 26, 2020 7:11 pm

I've never experienced anything like that with mine. Maybe the voltage from your sequencer just isn't dropping quite fast enough to line up with the trigger? Do you have another sequencer to try it out with and see if the same thing happens?

I usually sequence my BIA with my Mescaline sequencers.

User avatar
dragulasbruder
Wiggling with Experience
Posts: 252
Joined: Sun Oct 23, 2011 9:57 pm
Location: Chicago

Re: Preemptive Step Sequencing?

Post by dragulasbruder » Tue May 26, 2020 8:06 pm

Never used a BIA, so my opinion may be close to moot--but I have had issues like this arise when using virtual sample/holds on software modular systems (Clavia, VCV).

Are *all* of the controls on BIA actually updated in real time--or are parameters like SLM and noise amount actually immediately defined (and held) for a sound object (eg "drum strike on seq. step #1") at the time the voltage on the trigger input crosses a given frequency threshold?

My guess is this--all digital systems have latency. The latency on BIA for step-defined parameters (clocked by the trigger input) could possibly be shorter than the latency on the Voltage Block. We're talking on the order of microseconds here, but it matters. If BIA determines the value for S/L/M 0.1ms *before* Voltage Block has updated its output value, then the BIA's CV input is still seeing the voltage from the previous step--Voltage Block hasn't had time to update yet. On real-time-updated parameters like pitch, the BIA simply snaps up to the correct pitch 0.1ms late--basically imperceptible to us, but technically after the BIA has registered the strike for the given step.

*IF* this is actually what is wrong, then how do you solve this?? I have a number of options for you, but none of them are great, and all of them require another module yet. :(

1. Slew the trigger that clocks BIA, the faster the better. If you have any sort of DUSG clone or even a fast portamento module, this *might work*. Given the correct rise time, the BIA will still trigger, but the trigger from the sequencer won't cross the voltage threshold until after the Block's CV output has updated. Again, we're probably talking microseconds here--slew time will be extremely short. The advantage to this is that you can randomize or sequence slew time to get swing! If you don't have a slew you can try this with an envelope with a very short attack--only slightly longer than the near-instantaneous attack of the trigger.

2. Buy a gate delay, and make sure it can go short. I think Circuit Abbey makes a good one and it's also got a comparator--who doesn't need more comparators? Minimum delay on the BIA trigger should still sound in-time but get the correct sequence. Also you can sequence the delay time for swing.

3. Do you have a Disting, or similar digital swiss-army knife? Does it have a logic or comparator function (or even an envelope?) It's possible that the inherent latency of the software can delay the trigger *just* enough to arrive after the Voltage Block's CV output has updated to the intended value.

3. This next one is a method I use in-the-box all the time, but it might not work for you. I primarily use modular for music concrete type of shit and it's almost entirely in free time, latency is generally not a problem for me. If you have a logic inverter (sometimes called a NOT gate but a NOR gate will work for this too), there are two things you can try. First, try running the trigger that's being sent to the BIA through the inverter. The inverter will always output a high logic signal except for when the trigger is high--this means that the trigger will arrive late to the BIA by an amount of time equivalent to the length of the trigger. If you have a 1ms trigger, then great! If your module outputs 10ms triggers this could sound a little hairy if you're making 190bpm breakcore or something.

4. If that doesn't work out there's a far more extreme logic inverter solution that is also great--but it WILL NOT WORK if your trigger sequence for the BIA is coming from the Voltage Block. It will work but it will be a PITA if you use a clock that also has other functions (Pam's, etc), as the functions (LFO, divider, etc) will all be early with regard to the rest of your system. If you still want to do it, mult your master clock to the logic inverter. Use the un-inverted clock to clock the Voltage Block. Use the inverted clock to clock everything else in your system that used to receive the master clock. You've now functionally delayed your entire system to the Voltage Block by the pulsewidth of your master clock pulse--but since it's only a partial clock pulse behind the VB, "Step 1" still equals "Step 1"; you won't have to change your sequences. If your master clock is coming from a DAW full of MIDI-triggered samples, this could be a problem (but you could use track delays to re-synchronize your DAW with the sounds from the modular).

I might be wrong about the whole thing--might not be a latency issue at all! But that's my two cents. Prolix question, prolix solution: sorry if that was long or came off pedantic.
Last edited by dragulasbruder on Wed May 27, 2020 2:39 am, edited 2 times in total.
Now offering sliding-scale SYNTHESIS and AUDIO PRODUCTION LESSONS during the pandemic--please PM for details!

WTB Wish List: https://www.modulargrid.net/e/racks/view/1224624

MY MUSIC: https://www.soundcloud.com/wish-fulfillment

Aaronautical001
Wiggling with Experience
Posts: 285
Joined: Sat Jun 21, 2014 1:43 am
Location: UK

Re: Preemptive Step Sequencing?

Post by Aaronautical001 » Wed May 27, 2020 1:46 am

Worth double checking you haven’t got different divide settings on your VB that is controlling attack on the BIA? Or More generally making sure the VB is definitely doing what you think it is. I have both modules, but can’t claim to be an expert in this approach. But those original videos by Basek seemed to show the 2 can work well together.

Arneb
Ultra Wiggler
Posts: 788
Joined: Thu Jul 04, 2019 4:50 pm

Re: Preemptive Step Sequencing?

Post by Arneb » Wed May 27, 2020 1:59 am

waltergoschen wrote:
Tue May 26, 2020 5:24 pm
Does anyone know why sequenced changes to certain parameters on the Basimilus are inaudible when triggered with just a single step, and seem to require the previous step to “prepare” it with a similar voltage first?

To be precise, I’m using Voltage Block to make a one-module drum kit out of the BIA. When I set up a step to have significantly less voltage than the previous one to (for example) trigger the noise on the attack parameter in the BIA, the step will end up sounding identical to the one before it in the sequence (despite how, if I hold down the step’s button, it’ll repeat the desired noise-infused sound). However, if I set the previous step to a similar voltage, the step I’m trying to affect will make the desired sound (though the sound of the previous one will remain unaffected).

So for some reason with the attack parameter, as well as the S/L/M parameter, I can’t adjust the sound of a step without adjusting the step before it. With other parameters, like pitch, I’m able to hear any changes to a single step the way I’d expect to, regardless of the settings of adjacent steps.

If anyone could help, I’d hugely appreciate it! Sorry if that was a prolix explanation.
Sounds like a fairly common timing problem - like all electrical components the VB has some latency, it won't respond instantly to the step advance but will take a millisecond or so to change the output voltage. So, your BIA gets triggered before its attack parameter has actually changed. As for why this makes a difference for some parameters but not others, I guess that the BIA internally uses an S&H for the attack parameter but not the pitch one.

My usual recommendation to this is to just get the Ladik S-195 and run all your drum or envelope triggers, but not the VB clock, through it.

User avatar
Sandrine
Super Deluxe Wiggler
Posts: 2300
Joined: Tue Jun 09, 2015 10:28 pm
Location: BC Canada
Contact:

Re: Preemptive Step Sequencing?

Post by Sandrine » Thu May 28, 2020 10:12 am

I have a similar problem with the Beatstep Pro. The CV's are way behind the gates by a good 6mS per octave before they're stable. This introduces DAW audio-like latency when used with a gate delay module. It kind of ruined the whole thing when used with percussion, so what I did was advance the clock, or more accurately, delay the clock to the rest of the system and feed the original clock to the BSP. This worked!

waltergoschen
Learning to Wiggle
Posts: 5
Joined: Fri Mar 20, 2020 7:55 am

Re: Preemptive Step Sequencing?

Post by waltergoschen » Sun May 31, 2020 8:49 am

Thanks a lot for your help/advice!

I got a Pamela's New Workout a few days ago and I've been experimenting with micro-adjusting the timing and it does seem to fix this problem, though it seems weird that this is necessary, and it seems to potentially cause some other issues at the same time. I'm going to keep trying out some of these other suggestions to see what works best.

Post Reply

Return to “Eurorack Modules”