S&E
Full Member
Posts: 116
|
Post by S&E on Sept 14, 2021 3:42:02 GMT
Hello everyone. I am planning a 4-row standard AE set up. Planning to get the 2-tier starter, a second 2-tier case, and additional modules. I plan to use the 8-step sequencer, the 16-step sequencer, and 2 topograf drum sequencer. I need them to all be in sync with each other. What modules do I need to accomplish this? Also if possible I would like to have all the sequencers sync to midi (would be nice to know it is possible).
|
|
|
Post by admin on Sept 14, 2021 7:13:11 GMT
Hello everyone. I am planning a 4-row standard AE set up. Planning to get the 2-tier starter, a second 2-tier case, and additional modules. I plan to use the 8-step sequencer, the 16-step sequencer, and 2 topograf drum sequencer. I need them to all be in sync with each other. What modules do I need to accomplish this? Also if possible I would like to have all the sequencers sync to midi (would be nice to know it is possible). Hi, Welcome to the world of AE! You will get MIDI clock via the MASTER module. Usually this clock is too fast to be usable in a modular environment, that's where the BEATDIVIDER comes into play. It divides the MIDI clock signal into 1/2, 1/4, 1/8, 1/16 or 1/32 notes which you can then distribute to your sequencers. Personally I just use the multiples that are available on most modules to duplicate the clock signal, but some people found that through voltage drop there may be inconsistencies. That's where the BUFFERED MULTIPLE module comes into play - it can duplicate any signal without voltage drop. This was recently discussed here: forum.aemodular.com/thread/1558/sync-problems-multsMore info on all of these modules can be found on the Wiki: wiki.aemodular.com/pmwiki.php/AeManual/ModulesCheers, Carsten.
|
|
|
Post by funbun on Sept 14, 2021 9:11:21 GMT
I run three sequencers in my set up. I use either an LFO square/pulse wave or the clock from an rbss. Use either as a clock source assuming you aren't running some kind of external MIDI. To simplify here my sync process:
clock source > buffered multiple > standard multiple > sequencers, clock dividers, trigger processors or anything else that needs a clock signal.
|
|
|
Post by pt3r on Sept 14, 2021 11:24:25 GMT
That standard multiple in the chain can potentially cause sync issues if it sends out to multiple sequencers.
|
|
jan musi
New Member
My first rack just arrived! Woohoo!
Posts: 25
|
Post by jan musi on Sept 14, 2021 12:17:50 GMT
Yes to all of the above! It's also worth noting (especially in the context of a newcomer to AE or modular synthesis in general) that all sequencers have a CLOCK/CLK input so that they can be synchronized. Most modules (including most sequencers) do not have an internal clock of their own, so they are dependent on a clock signal to be patched into them via this input port. However, to make this a little more complicated, TOPOGRAF is one of the few modules that has an optional internal clock. Therefore, you can synchronize all of your sequencers by either: A) providing them all the same (copied, AKA multiple-d) input signal of any intensely pulsing nature, from anywhere (MASTER, BEAT DIVIDER, LFO, VCO, etc.), or... B) using a TOPOGRAF's internal clock as the "original" clock source and patching one of its output trigger signals (bass/snare/hat) to the other modules' CLOCK/CLK inputs. This would mean that other sequencers advance a step every time that the "original" TOPOGRAF generates a trigger on its TR 1/2/3 output signal. Hope this helps, and welcome to the beautifully reconfigurable world of AE!
|
|
|
Post by funbun on Sept 14, 2021 15:56:45 GMT
That standard multiple in the chain can potentially cause sync issues if it sends out to multiple sequencers. That's why you have the buffered multiple.
|
|
|
Post by pt3r on Sept 14, 2021 16:08:40 GMT
That standard multiple in the chain can potentially cause sync issues if it sends out to multiple sequencers. That's why you have the buffered multiple. that standard multiple after the buffered multiple can still cause sync issues if it goes out to multiple sequencers, cause those signals will not be exact copies.
|
|
|
Post by funbun on Sept 14, 2021 16:52:27 GMT
I understand, and you are correct. But, that's all I have to run my sequencers, MM divider, beat divider, TRIP, all from two outs from an rbss. Basically we're talking about building a complex sequencer out of smaller units like everything else in modular, and my buffered multi simply doesn't have enough outs. Given that I'm trying to do Steve Reich-like phasing, that's exactly what I want. Generally, I haven't had a problem with syncing, but that brings up another point:
We've both hit on exactly the same need for a master clock module. So far I've been using the rbss' internal clock as the master clock for my system via buffered multi. There just aren't enough outs either on the rbss or the buffered multi, so I use a standard multi until the signal get too weak to sync properly, then I run another line to the buffered if there is one available. It's always work around, and all work around have compromises.
For instance I still need a precision adder, but It'll be a while before I get one from Germany. If I want to play exactly the same melody in exactly the same time one two separate sequencers, I'll need precision adders and buffed multies. You often have to make do with what you have or half of what you need. You often come up with your best patches that way.
|
|
|
Post by solipsistnation on Sept 15, 2021 5:21:59 GMT
From my own experiments, sticking 1 unbuffered mult on the output of a buffered mult (like, if you have more than 8 things that need to receive buffered mult outputs so you unbuffered mult one of the outputs) will be generally good. Not all of the AE sequencers are quite as sensitive to lower clock pulses, so you could, for example, have your buffered mult set up with 1 clock heading to 4 buffered outputs, use 3 of the buffered outputs for the sensitive stuff, and then use the last buffered out to drive a mult and run your less sensitive stuff off of that. I've been doing something like this for a reset button push (+5V -> 3VCswitch, momentary switch outputs to -> buffered mult -> reset inputs and a mult driving a couple more resets) and if I use an unbuffered mult to a SEQ8, it sometimes does strange things, but if I use an unbuffered mult output for the reset on a SEQ16 it's fine. They seem to handle clock and reset slightly differently from one another.
You can get pretty complex with piles of sequencers-- one thing I also do is send the master clock to an MM-DIV and then use its various outputs to clock other things so they run at half-time or whatever. The MM-DIV outputs appear to be buffered so that seems to solve some of that. And then you can have a SEQ8 that gets one clock tick per 16 ticks to the SEQ16 and do more interesting timing things than just blatting out constant streams of 16th notes, although that's fun too. Or you can clock a SEQ8 from one channel of a TRIQ164, or from whatever else seems interesting.
|
|
|
Post by pt3r on Sept 15, 2021 6:11:56 GMT
Good point you mention there; not every module seem to require the same voltage to detect the high signal of a trigger. I have also been wondering what the minimal viable duration of a trigger must be for it to be detectable, the module specs don't seem to mention it anywhere though. For the moment I go for 30ms.
|
|
|
Post by keurslagerkurt on Sept 15, 2021 7:25:15 GMT
It's not only the trigger threshold that can vary slightly, it's also how much input current the trigger input draws (or in EE terms how much 'input impedance' the input has).
A 'perfect' module would not draw any current from the trigger, meaning it would only 'read' the voltage and not alter it. This would make it possible to use one unbuffered trigger for an endless amount of these 'perfect' modules. However, of course these modules don't exist, so there will always be a little current draw, depending on the design of the input stage of the module. This current draw will 'load' the trigger output from the other module, making the voltage drop. Enough of these voltage drops and your trigger will drop enough to be unreliable.
|
|
|
Post by keurslagerkurt on Sept 15, 2021 7:26:26 GMT
In case you have a breadboard of brAEdboard, you can easily make a buffer (one mcp602 = two buffers). There is a post in the brAEdboard guide by TTNM that elaborates on this!
|
|
S&E
Full Member
Posts: 116
|
Post by S&E on Sept 17, 2021 3:46:34 GMT
I had thought about the buffer module before, and it is nice to see confirmation here. BUT the uocoming heartbeat module does exactly what I wanted and also adds swing. So I will definitely get that when it ships, they knew they had a gap there.
Now I need something that can take audio trig (like from a live drummer) and generate gate and cv. Is there such module? If not it would be nice to know if one is planned. I can use the i/o and bring it from my euro but would be nice to have native.
Btw just took advantage of the 10% sale and placed my first order! Still need to get my 2-tier starter and some other modules from noisebug, but waiting for a 2-unit standard case to come in stock. Really excited to be getting into this.
Another question: do the patch cables come out easily or do they have some resistance? Trying to see if I move them somewhere in a carry bag whether the cables will stay put.
|
|
|
Post by admin on Sept 17, 2021 4:08:29 GMT
.. Now I need something that can take audio trig (like from a live drummer) and generate gate and cv. Is there such module? If not it would be nice to know if one is planned. I can use the i/o and bring it from my euro but would be nice to have native. .. Another question: do the patch cables come out easily or do they have some resistance? Trying to see if I move them somewhere in a carry bag whether the cables will stay put. For gate/CV from audio input, the closest I can think of is the SLEW/EDGE module which sends a trigger when the signal changes either up or down. The patch cables are not as tight as with Eurorack and don't have that "click", but they do have some resistance and don't fall out too easily.
|
|
S&E
Full Member
Posts: 116
|
Post by S&E on Sept 17, 2021 5:06:29 GMT
No, I don't think slew/edge module does trig to gate/cv adaption. It's ok, I have that capability in euro and with the 4 i/o module it should work. I have written AE directly to ask this question though.
|
|
|
Post by keurslagerkurt on Sept 17, 2021 6:53:09 GMT
No, I don't think slew/edge module does trig to gate/cv adaption. It's ok, I have that capability in euro and with the 4 i/o module it should work. I have written AE directly to ask this question though. An envelope follower has been requested a lot, and lately i have found myself sketching up some possible AE schematics for it. Will try and breadboard something up the coming months, but can't promise anything of course 😉
|
|
|
Post by pt3r on Sept 17, 2021 6:59:49 GMT
No, I don't think slew/edge module does trig to gate/cv adaption. It's ok, I have that capability in euro and with the 4 i/o module it should work. I have written AE directly to ask this question though. I'm building something that will hopefully do this, to accompany my AEclidian module. The firmware is ready. Come on Mouser get your stocks up to date.
|
|
|
Post by funbun on Sept 17, 2021 8:46:59 GMT
|
|
|
Post by keurslagerkurt on Sept 17, 2021 15:05:53 GMT
Now this is very interesting! Every day i believe a little bit more that one row of GRAINS might just be THE FUTURE
|
|