Start/Reset for ANLG output

[Simon]

IMPORTANT EDIT:

This feature is about to be implemented for FW 4.0 coming in January. So the last chance for feedback/comments regarding the implementation is now

Plan is to implement the following settings for An.1 and An.2:

• OFF -> output is never sending anything, set to 0V
• '1, '2, '3, '4, '6, '8, '12, '24 -> sends a 5ms 5V pulse every X ppq (current values)
• run (Gate) -> sends 0V when Nome is stopped, and constant 5V when playing (like a start/stop gate)
• StA (Start) -> sends a 5ms 5V pulse when the Nome plays, and another one when it stops, nothing on resync
• rES (Reset) -> sends a 5ms 5V pulse when the Nome plays or resyncs, nothing when it stops
• 2, 3, 4, 5, 6, ..., 99 -> sends a 5ms 5V pulse every X quarter note

When the "AnL" setting is set to On or PLA, it will not affect the outputs if they are set to run, StA, or rES.
The Reset pulse (rES), the Start pulse (StA) and the 2-99 quarternotes pulses will be sent about 0.5ms early.
The run gate will start about 1ms early. If you need the run gate to be even earlier, you can use the DIN Sync mode.


----------------------------------------


Original topic: viewtopic.php?t=66

This is about implementing more possibilities for the ANLG outputs.
Currently both An.1 and An.2 settings can be set to a pulse, ranging from 1 to 24 ppq.

Added to the list will be the following options:

• StA (Start) - sends a 0V -> 5V start/stop gate when the Midronome is playing
• rES (Reset) - sends a 5V Reset pulse when the Midronome is playing

Note that "the Midronome is playing" refers to when the device is sending MIDI Start on its MIDI Ports. And each clock on the ANLG port is linked to its corresponding MIDI Port (An.1 sends play/reset at the same time as MIDI Port 1, and An.2 at the same time as MIDI Port 2)


Feel free to comment here regarding:

• If you think some information is missing or unclear
• How you would like this feature to be implemented (buttons, settings, how to activate, how it is done, etc.)
• Why you think this feature is a good addition to the Midronome

[referenceaudio]

I'm really looking forward for this feature to be added!
A start/run signal is essential to perfectly sync up eurorack with daw and other midi devices. Can't press play on more than two devices as the same time

I think the description above is very clear and straight forward. A 5v gate is all most (if not all) eurorack modules with a dedicated run input needs.

[gtech]

For the reset feature, most important for me would be to set the sequence length, so I can reset my analog sequencers, clock dividers etc every 2 or 4 bars or some other number of beats.

It wouldn't necessarily need an additional setting. I could imagine it working if the number of pulses per quarter note could go to fractions, e.g. 1/2, 1/3, 1/4... 1/16... 1/32 etc.'

Thank you!

[Simon]

For the reset feature, most important for me would be to set the sequence length, so I can reset my analog sequencers, clock dividers etc every 2 or 4 bars or some other number of beats.

It wouldn't necessarily need an additional setting. I could imagine it working if the number of pulses per quarter note could go to fractions, e.g. 1/2, 1/3, 1/4... 1/16... 1/32 etc.'

Nice idea, I like it! And that's pretty easy to add as part of this feature.
Just so we're clear, say you set it to 1/8, that means every 8th quarter note? (so every 2 bars if the bar is in 4/4)

How would you suggest the display should show it? Right now we have the values 1, 2, 3, 4, 6, 8, 12, 24 which are all PPQ values.

I would suggest this list, for each output:

• /1, /2, ..., /24 -> sends a 5ms 5V pulse every X ppq (current values)
• StA (Start) -> sends a 0V->5V start/stop gate when the Midronome is playing/stopping
• rES (Reset) -> sends a 5ms 5V pulse when the Midronome plays, nothing when it stops
• 2, 3, 4, 5, 6, ..., 99 -> sends a 5ms 5V pulse every X quarter note

The / being quite weird looking on the display (some sort of: r' ).

Simon

[gtech]

Hi Simon,

That would be amazing if you could extend it to include all those options!

Since the "/" would look weird on the 7-segment display, a couple more options could be:

- d1, d2, d3... d24 (for "divide") -> sends a 5ms 5V pulse every X ppq (current values)
- 1S, 2S, 3S, 4S... 99S (for "steps") -> sends a 5ms 5V pulse every X quarter note/step

Alternatively, just one bar in front of the current values? '1, '2, '3... '24? plus plain numbers for the quarter notes

Thanks!

[Simon]

Thanks for the feedback, I like your suggestions

This is on the roadmap for Firmware 3.0 so you can expect it some time in 2024

Cheers
Simon

[amaeland]

Hi Simon, I have a related question. Would you consider adding a feature whereby the pulse of the analogue clock output can be adjusted?

To clarify, I have been having reliability issues with my Korg SQ-1 when syncing it via analogue to the Midronome. Every now and then, it would jump a few steps and get out of sync.

It occurred to me that it could be because the shape of the pulse was not to the SQ-1’s liking. So I tried routing the Midronome analogue out into my Doepfer clock divider, which I set to Trigger mode instead of Gate, and used one of the outputs from that to sync the SQ-1. Now it’s perfect! All I had to do was change the PPQ setting on the Midronome to compensate for using a divided timing.

Maybe it would be a good idea to have an alternative pulse type which is shorter, like a trigger pulse? Or could it be that the output from the Midronome is simply too loud or quiet for the SQ-1? I need to try running it through a VCA now that I think of it…if that works, then having a volume control for the pulse output would do the trick. Currently however, it seems I need to use additional hardware to get a sync signal that works well with the SQ-1.

[Simon]

I have been having reliability issues with my Korg SQ-1 when syncing it via analogue to the Midronome. Every now and then, it would jump a few steps and get out of sync.

I see - great info! All your ideas seem to be very on point

Would you mind creating a post about it in the “Midronome Bugs Report” forum and copy-pasting your message there? Then we can continue this discussion/investigation there, and even if the problem is not solvable with a Firmware update, we can write a workaround (for example using a volume knob) for other people who could have the same issue.

Cheers
Simon

[amaeland]

I have been having reliability issues with my Korg SQ-1 when syncing it via analogue to the Midronome. Every now and then, it would jump a few steps and get out of sync.

I see - great info! All your ideas seem to be very on point

Would you mind creating a post about it in the “Midronome Bugs Report” forum and copy-pasting your message there? Then we can continue this discussion/investigation there, and even if the problem is not solvable with a Firmware update, we can write a workaround (for example using a volume knob) for other people who could have the same issue.

Cheers
Simon

sorry for the late response, seems I am not getting email notifications from the forum.

Glad if it can help! Sure, I’ll make a separate thread.

[amaeland]

Also, some feedback that is more on topic:

The below quoted options are really nice, but might there be a need for a small delay between reset signals and clock? Certain sequencers may have a problem with receiving both signals at exactly the same time. I ran into this issue when clocking stuff from the Midronome via a clock divider in my modular. Since I did not have the split cable needed to use DIN sync, I thought I could send a reset signal to the clock divider by using Silent Way and my ES-3.

With the PDC in Logic being pretty good I was expecting the first pulse from the E-RM plugin to be accurately synced with the trigger generated by Silent Way. However, the clock divider did not reset as desired and a timing offset occurred. In hindsight it may have been because the Midronome analogue clock and Silent Way reset pulse came from different paths - the former signal being processed by the Midronome and passed on while the latter came via adat to the ES-3 before being converted to an analogue pulse.

In any case, I fixed the issue by setting a negative track delay on my Silent Way track. This way, the clock divider got reset a few milliseconds before receiving the clock itself.

Maybe this kind of problem will not happen when the reset signal is output from the Midronome, but a microscopic delay added to clock (not reset) outputs might be a good idea just to be sure? I don’t think it should be problematic to have the “2…99” output mode set in such a way that it is, say, 0.3 ms ahead of the regular clock output. You could still use that output mode for other periodic uses than clock resets, e.g. if you wanted to trigger an envelope every 8 bars or whatever. Would still be tight enough. Almost getting into Pamela’s Workout territory with that output mode. Awesome.

By the way, I suggested 0.3 ms because I believe there is a trigger delay setting on my Mutable Instruments Braids module that can be set between 0 and 300 microseconds. Too fast and it might not help with certain sequencers, dunno.

I would suggest this list, for each output:

• /1, /2, ..., /24 -> sends a 5ms 5V pulse every X ppq (current values)
• StA (Start) -> sends a 0V->5V start/stop gate when the Midronome is playing/stopping
• rES (Reset) -> sends a 5ms 5V pulse when the Midronome plays, nothing when it stops
• 2, 3, 4, 5, 6, ..., 99 -> sends a 5ms 5V pulse every X quarter note

[Simon]

sorry for the late response, seems I am not getting email notifications from the forum.

No worries at all, and if you want to receive notification emails you can turn them on here: ucp.php?i=ucp_notifications&mode=notification_options

Thanks for the long feedback about the reset pulse, it's a very very good point! In fact the DIN Sync mode already has so many very complicated mechanisms precisely because these vintage Roland machines need a "lot" of time (about 9ms according to the 808 tech manual) between a reset and the first clock. This makes things very difficult when people want their clock to start right away when they press play in the DAW for example

So yes I'll make sure to send the reset pulse 1-2ms early

[joncoe]

Adding my voice here as an upvote for this! I would be able to drop one of my modules that I am only using for the purpose of sending a reset into the system. Lovely!

[Simon]

Hi all

This is about to be implemented for FW 4.0 coming in January. So the last chance for feedback/comments regarding the implementation is now

Plan is to implement the following settings for An.1 and An.2:

• OFF -> output is never sending anything, set to 0V
• '1, '2, '3, '4, '6, '8, '12, '24-> sends a 5ms 5V pulse every X ppq (current values)
• run (Gate) -> sends 0V when Nome is stopped, and constant 5V when playing (like a start/stop gate)
• StA (Start) -> sends a 5ms 5V pulse when the Nome plays, and another one when it stops
• rES (Reset) -> sends a 5ms 5V pulse when the Nome plays, nothing when it stops
• 2, 3, 4, 5, 6, ..., 99 -> sends a 5ms 5V pulse every X quarter note

When the "AnL" setting is set to On or PLA, it will not affect the outputs if they are set to GAt, StA, or rES.
The Reset pulse (rES), the Start pulse (StA) and the 2-99 quarternotes pulses will be sent about 0.5ms early.

[barseglin]

Hi all

This is about to be implemented for FW 4.0 coming in January. So the last chance for feedback/comments regarding the implementation is now

Plan is to implement the following settings for An.1 and An.2:

• OFF -> output is never sending anything, set to 0V
• '1, '2, '3, '4, '6, '8, '12, '24-> sends a 5ms 5V pulse every X ppq (current values)
• GAt (Gate) -> sends 0V when Nome is stopped, and constant 5V when playing (like a start/stop gate)
• StA (Start) -> sends a 5ms 5V pulse when the Nome plays, and another one when it stops
• rES (Reset) -> sends a 5ms 5V pulse when the Nome plays, nothing when it stops
• 2, 3, 4, 5, 6, ..., 99 -> sends a 5ms 5V pulse every X quarter note

When the "AnL" setting is set to On or PLA, it will not affect the outputs if they are set to GAt, StA, or rES.
The Reset pulse (rES), the Start pulse (StA) and the 2-99 quarternotes pulses will be sent about 0.5ms early.


EDIT: one point raised by someone over email, GAt (Gate) is not very clear, maybe it should be called Run instead. Any thoughts?

Hi Simon!
I’m not too well-versed in all this, but for what it’s worth I think “Run” is the right word for this. I just got a Pam’s Pro Workout for the modular setup and trying to sync it with the Midronome is the goal for me.
Pam’s manual says that it expects 24ppqn; GREAT, Midronome can do that. But it’s also expecting some kind of running signal prior to the first beat for “pre-syncing”? I’m not too sure, but happy to talk things through if it would help.
It’s all written on Page 22 of this manual: https://busycircuits.com/docs/alm034-manual.pdf

[Simon]

Pam’s manual says that it expects 24ppqn; GREAT, Midronome can do that. But it’s also expecting some kind of running signal prior to the first beat for “pre-syncing”?

Yes I'm aware Actually you can already sync your Pam by setting the ANLG (An.L setting) to "din", which outputs a DIN Sync clock, which is precisely a 24ppq and a "run" (or gate") signal - try it

As for naming the setting "Run" instead of "Gate", well even the Pam's manual calls it "Gate" when it's an output, despite calling the input a "Run" input...

To make it clearer I can use both name Gate and Run in the manual, so that nobody is in doubt

[Simon]

I don’t think it should be problematic to have the “2…99” output mode set in such a way that it is, say, 0.3 ms ahead of the regular clock output.

Hi @amaeland, I'm currently implementing this feature, considering what you wrote

As written above, I can make the rES pulse sent 1ms early, while the StA pulse is sent exactly on time. I could also make sure the pulses for the 2-99 options are sent 1ms early, but don't you think this could be a problem that these pulses are never sent exactly on time with the clock?

I am quite against addng yet another setting to control whether the pulses are sent exactly on time with the clock or slightly early. I would prefer nailing the right timing for everybody.

Any thoughts?


PS: about the GAt signal, note that this will always start earlier than the clock. That's because it will follow the gate used in DIN Sync mode, which respects the DIN Sync specs set by Roland where the machines need at least a few ms to "prepare for play".

[amaeland]

I don’t think it should be problematic to have the “2…99” output mode set in such a way that it is, say, 0.3 ms ahead of the regular clock output.

Hi @amaeland, I'm currently implementing this feature, considering what you wrote

As written above, I can make the rES pulse sent 1ms early, while the StA pulse is sent exactly on time. I could also make sure the pulses for the 2-99 options are sent 1ms early, but don't you think this could be a problem that these pulses are never sent exactly on time with the clock?

I am quite against addng yet another setting to control whether the pulses are sent exactly on time with the clock or slightly early. I would prefer nailing the right timing for everybody.

Any thoughts?


PS: about the GAt signal, note that this will always start earlier than the clock. That's because it will follow the gate used in DIN Sync mode, which respects the DIN Sync specs set by Roland where the machines need at least a few ms to "prepare for play".

Having the Reset pulse 1 ms early sounds like a good idea, maybe it could even be possible to manually adjust the offset to cater for old sequencers that need an earlier «warning» than 1 ms? I have no idea how universally well 1 ms would work.

«run» is more logical than «GAt» btw imho, not that it matters too much as long as it’s explained in the marketing and manual

[Simon]

Ok, I'll follow PAM's naming convention, I think it's much easier when different manufacturers agree to a name. So run it is.

Regarding the timings, I'll have:

• rES and 2-99 options will send pulses about 0.5ms early
• The run gate will start about 1ms early