With my testing of the Meander pentatonic PES output compared to the Harmony II PES output, it appears that it gets the scale right. Using Grande Quant for visualization.
2 Likes
I’m assuming if the UI was set to some other scale (like not a C), that feeding in an unsupported scale would not change it?
That’s correct.
1 Like
By the way, I really like the new Harmony II panel port labels. Thank you.
1 Like
If you get a chance to try, I put the two UI modes into Phase Patterns, called “Expert mode”. It’s documented in the manual, too.
1 Like
Expert UI works great.
But:
- After a few RIB+, there’s no clock out. Seems to scale with Dur.
- Shift can go below zero. Not in the gui, but internally: With Shift = 0.00, press RIB- two times → Shift stays 0.00 in gui, but in order to start a shift greater than zero I have to press RIB+ three times.
- Initialize doesn’t reset shift. Maybe this is by design, but it’s counter intuitive to me.
@soundismovement thanks for this report!
It sounds like the first bug is easy to repro? If you have an exact repro scenario that would be great, but if it’s easy to find… What do you mean by “Seems to scale with Dur”?
It sounds like if I just limit the internal RIB to being non-negative that would fix the “shift below zero” issue? When that happens should the light go off and the whole cycle be terminated, or just continue the cycle and limit it? (If I’m going to address this, might as well do the best thing).
I’ll look at doing a custom initialize. That sounds like a good idea.
Thanks again!
I just logged issues for each of these, so I won’t forget:
Any chance for a flat light/ dark panel alt?
No plans to do that at the moment.
OK, I can duplicate the "no clock output’ bug. What’s happening is that you are asking it to delay the clock more than my delay can delay. Should I a) keep outputting clock, just limited to my max delay? b) let the session possibly glitch once, and allocate a larger delay, c?
Is it not possible rewrite it so that it internally wraps around a single delay? When Shift is at x.00, reset the delay internally?
I don’t understand.
Assuming a constant clock input, a shift of 1.00 is equal to a shift 0.00, 100.00 or -1000.00. The clock pulses align. So instead of delaying the input clock by Shift amount, you delay it by fmodf(Shift, 1.00f)
Alternatively, I think you could use a counter or timer instead of a buffer when dealing with binary clock signals. Detect the points where the signal goes high and low, and then replicate these on the clock output, shifted by a calculated amount of time later.
Ok, got it. Thanks.
I tried playing around with HarmonyII. It’s very fun in minor Pentatonic mode.
Harmony2 Test.vcv (5.9 KB) Sort of a boogie woogie piano arrangement.
edit: It looks like this when I patch it up, it only indicates 1 channel. I thought it was a bug, but it does switch right back to 4 channels if I change any setting, otherwise it sits at 1 until something happens.
I’ll take a look at that.
I’m having so much fun with it. Here’s another patch. This module is really like a secret sauce, it’s awesome!
Harmony2 Test2.vcv (7.1 KB)
edit: fixed some triggering issues, complex patch…
I’ll try that patch out later - I’ve got two racks, and my dev one doesn’t have many plugins in it.
I haven’t been able to get Phase Patterns to work as well as I would like, so I’m going to put that aside for now. Soon I will get a new Library build to put Harmony II into the library.
2 Likes