Trig Conditions and Song mode

Hello,

I must admit that although I love the Elektron workflow, I hate how inconsistent song mode is.

I am using rytm mkii as my main drum machine for a new project alongside A4. I also own a syntakt, and have owned octatrack and digitone.

I read the manual repeatedly and I still can’t make sense of the trig conditions’ behaviour in song mode.

I have entire songs sequenced into pattern chains in the song mode (mostly 4-pattern long chains). I am using the trig conditions to add variation and ghost notes (mostly for my snare hits). The conditions I use the most are 1:2, 2:2, 4:4. Here’s where it gets dicy:

I can’t make sense of how song mode counts the trig loops.

Does the count reset at every chain?
Does it continuously count throughout the entire song?
Does the count reset if a different pattern plays in the same chain?

For example, consider the following song:

01: A01 A02
02: A01 A01 A01 A01
03: A01 A02 A01 A03
04: A01 A01 A01 A04

If I have a trig conditions 1:2, 2:4, 4:4 in A01. When do they play? I have done repeated experiments and I can’t find a consistent answer for when the count resets.

It would make a lot of sense if the count resets every chain, and is maintained throughout the chain when a different pattern is inserted in between like in 03.

Can someone confirm if a clear logic is available or if this is a bug?

Thank you!

I always thought that it has to do with patterns. When you change patterns, you negate the first pattern’s trig condition if it has not yet played.

So, like if I put 2:2 on A01 and then change to A02 instead of letting A01 play through twice, the 2:2 will not play on A02 because it was exclusive to A01.

Basically, any trig condition which is dependent on the same pattern playing through more than once to achieve will be negated by changing to any other pattern.

Did you test it from that perspective already? I guess the rules could be different in song mode, but I don’t see why they would be.

Thank you for the reply.

I have been working on this project a long time and I have tested many combinations.

The main issue happens when:

01: A01 A01 A02 A01
02: A01 A01 A01 A01

My 2:4 condition will play twice in chain 02 (on the odd counts) because the count carried over from chain 01. It is extremely frustrating that the count doesn’t reset with a new chain.

Yeah, that’s very annoying for song mode to behave that way.

How different are A02 and A01? Is it possible to achieve A02 by trig locking and microtiming something with a trig condition on A01 to effectively recreate the variation that appears in A02?

If they’re drastically different from one another then this doesn’t always work, but on the digitakt or digitone I often use trig conditions to do this, like to place a hit on step 2 and microtime it backwards to right in front of step 1 then use like a 3:4 trig condition or a 4:4 trig condition to make the variation on step 2 replace the hit on step 1 when it gets to the appropriate count.

I have good luck with this but if the entire pattern structure changes, then this becomes a lot more difficult to implement.

Yes this brings me to the point that elektron song mode needs a major update. It is unbelievable that a device marketed at close to $2000 is expecting me to abuse “workarounds”

I can’t even program a fill into song mode. This is one of the reasons why I’m ending up with so many patterns.

Some will say oh you can insert a trigless trig to do a fill or “microtime” a step all the way to the side and use that, but i’m sorry i feel like I shouldn’t have to do any of this and I refuse to consider it an acceptable solution. Elektron needs to deliver an update to both their software and their user manual to make their song mode more useable.

I’m mostly frustrated because I am so many months into this project and I’m close to wrap up the EP yet making minor changes is proving to be such a hassle! I’m pushed to make the changes post recording in the daw and losing it for the live performance.

It’s a limitation and frustrating, for sure. I think that if it’s important to you then it’s definitely appropriate to submit a feature request.

I had no idea that’s an option. I will go look into it. Thank you!

Sure. There’s actually 2 ways to go about it. One is you can send an email to:

feature-request@elektron.se

with the model and the firmware version, any information regarding what you’re making a request towards, and then also a description of what you’re requesting them to add or change.

There is also a thread for feature requests and you can post it in there as well.

If it’s rytm mkii I’d use this thread and in this case it’s obvious what device you’re talking about so you don’t really have to specify:

I wouldn’t expect any immediate changes but it’s good to get the ball rolling by letting them know rather than hoping that they’ll make the change without anyone suggesting it.

Knowing these Quirks and working around them is the only Solution imho. I usually plan my Tracks ahead and place strategic Pattern Changes so that the Trig Conditions reset when I need them to.

A quick and dirty Solution is to insert a dummy Pattern with the minimum Length of 2 Trigs in a strategic Place and shorten the preceding Pattern accordingly. If done correctly nobody will notice.

3 Likes

Update:

I have found the main culprit in the sequencer logic.
None of the “workarounds” are working because:

The sequencer does not recognize the position of the pattern in the song. So if I make a change in the chorus, and I want to listen halfway in the song, the sequencer will reset all logic gates to 0 and start counting from there.

This means that the only way to hear the pattern play as intended would be to play the song from the beginning every time, or at least from the section of the song where a count restarts.

If you work in chains of 2 or 4 patterns (2 or 4 bar loops), which is the most common way of composing, then you must ensure the following:

1- play the chain from the begging every time you are listening for adjustments
2- make sure the chain does not start on the same pattern the previous chain ended with

This is a major workflow drawback and a source of confusion for people trying to use song mode. It is slowing me down significantly.

I have already submitted a feature request with Elektron, but let’s be honest— this is not a “feature” and not a “creativity inducing limitation”. Sadly, it is just a bug.

Yes, that’s why I suggested to strategically insert a dummy Pattern with the minimum Length of 2 Steps. That Way you can reset the Trig Conditions on your original Pattern.

E.g. if you have a Verse with 3x Pattern A1 (which are, say 64 Steps long), and also want the following Chorus to play Pattern A1, you shorten the last of those Verse Patterns to only 62 Steps and insert a new Song Row with e.g. Pattern A2 with a Length of 2 Steps and every Track muted that makes unwanted Noise. That Way, when your Chorus starts you trigger those 1st and X:X Trig Conditions correctly. If chopping off 2 Steps at the End is too much, you can also just cut off only 1 Step and play that Dummy Pattern at Double Time.

If I understand your Problem correctly, that is.

1 Like

Leto,

Take this song pattern for example:

01: A1 A2 A3 A4
02: A5 A5 A5 A5

Let’s say I programmed trig conditions in A5 for 1:4, 2:4, 3:4, 4:4 to get variation throughout the chain.

Now I made a change to the 3:4 condition in A5.

In order to hear the change, I must listen to A5 play 3 times., i.e. from the beginning of chain 02.

If I play the chain from the third pattern (to skip directly to the change) the sequencer will restart all trig counts.

If I have eight A5 patterns chained with trig conditions such as 7:8 and 64 steps at 90 bpm, I’m looking at a frustrating wait time just to hear my fill play, change it again, and play the whole thing again.