10-29-2019, 01:44 PM
(This post was last modified: 10-29-2019, 01:51 PM by Holon Mundi.)
(10-29-2019, 08:49 AM)XORadmin Wrote:(10-29-2019, 07:39 AM)Holon Mundi Wrote: I'm having trouble finding a good way to clock the Nerdseq in Bitwig using audio. My audio interface has no midi output and even if I could create some workaround by using a Keystep in between, it's not a good and lasting solution. I would love to see a setting for how Nerdseq interprets the incoming clock such as divisions and multiplications. Same goes for the clock and reset outputs. I have recorded the triggers from Nerdseq and time stretched it to fit and it's a pain to work with.
I don't really understand what you mean. Audio is not a usable solution for clocking. Or to be more detailed, it is a bad solution for clocking.
At least pure audio that you get out of your soundcard.
It is
1: Bipolar, while the clock inputs are digital inputs.
2: Audio outputs are mostly AC coupled, and so no steady signals can be archieved. All triggers of a several length will be filtered out and the signal will end up as short pulses, not even in the good voltage, with a falling curve. Not how clock signals should look like.
3: Noise in the audio signal can affect the quality.
I also don't know how multiplications or divisions should be involved here. Clock signals (where the NerdSEQ supports all standards for in and output) are used to synchronise multiple gear together. If you have a normal good Midi or 24ppqn or 1/16 clock source, the synchronising will be perfect. If not, then the source is the problem. And all timings and quality depends on may factors, once computers are involved. The factors are : The DAW you use, the operating system (and the current update level) you use, the Hardware you use (even the fastest computer can suck here), the audio/midi interface you use, all software that is running next to it and many more.
I dare oppose you since my experience is that audio is far more reliable using Bitwig and Ableton compared to midi. As I understand it the audio process has a higher priority since it needs to be real-time ti avoid clicks and it gets updated more often than midi in the computer. Correct me if you can. In my case I use a beefy MSI PC with Win 10 and a minimally populated test project in Bitwig 3.
1. Do you still have the option to program the Nerdseq to recognize positive voltage clocks? If so then please do that. It will make a huge difference for a lot of users.
2. I use a Motu 8M with DC coupled outputs. Most audio interfaces have DC-coupled headphone outputs so most nerds can use those without hassle.
3. Sure but isn't that a hypothetical problem? DAWs are known for extremely low noise floors and I clock nearly every other module in my system this way.
"If you have a normal good Midi or 24ppqn or 1/16 clock source, the synchronising will be perfect. If not, then the source is the problem."
I think your perspective might be a tad restrictive. If the device can't comply with the context it is made for (from a user perspective) the design is incomplete. Please read this in the right way, I'm still amazed by Nerdseq. Adding options that contribute to ease of use and flexibility is a very intelligent way to go since eurorack has so many possibilities and therefore has a greater number of outlier scenarios than a more homogeneous context (NES f ex, no modularity and perfect compatibility). We are a rowdy bunch with very different setups.
The usefulness of having multiplication options for clock in, out and reset seem self-explanatory to me. Using them you can alter the role that Nerdseq occupies in a system and lets you use it in more links of a clocking chain. It's much better to make a module that fixes the idiosyncrasies of other modules that require this-or-that clock input/output or has some fixed gate length or whatever.
(10-29-2019, 01:32 PM)Marizu Wrote: I regularly clock my modular using audio from my MPC live. I use one output for clock and another for reset.
Record a single click, normalise it so that the smaple is at the maximum amplitude and then make a short track that repeats and has the click on every step. If you want to go straight into the Nerdseq, then I think that you would need to have 24 clicks per quarter note.
Don't timestretch the sample, as that could introduce artefacts or reduce spikes at the leading edge of the pulse. I also have the MPC output set to maximum volume to make sure that it reaches/exceeds the voltage thresholds that the modular requres.
Thanks for the reply. I have done similar presets for other devices but it is messy and that could easily be avoided by adding a few options on the clock input. I did get artifacts at first but found settings that worked.
You are right about the volume in most cases. I recommend Motu Ultralight and 8M since they have 4.7V or 5V outputs that are DC-coupled. Another tip is Expert Sleepers ES-3 which goes to +-10V, is cheap 2nd hand but requires ADAT out from the interface.

