How to get Soft Takeover working?

[ChocolateHawkins]

I seem to be unable to get soft takeover to work with my hardware controllers (UC-16, and a Impulse 49 have the same issue).

I am able to set up the specific knob to control what I want, and soft takeover works for the first run through the dial. But every other time afterwards, it does not.

Am I missing something?

[trevlyns]

Have you checked out Kenny's video?...

Hope this helps!

[ChocolateHawkins]

I have. And several others. My soft takeover does not work as is shown in any videos. I am using an older theme, but that shouldn't impact it, should it?

[EvilDragon]

Theme has nothing to do with it, it's just a visual thing.

[ChocolateHawkins]

Thanks. I assumed as much. Is there anything I could be doing wrong? Both Absolute and Soft Takeover are selected. The first time I use the assigned knob, soft takeover works as it should. If I click off the track and back (and have moved the knob) soft takeover no longer works.

[nofish]

You do know that you need to "catch" (pass) Reaper's current parameter value with the knob for soft takeover to work ?

[ChocolateHawkins]

If by catch you mean pass through the current value, then yes. It's not a case of knob not working or me not understanding how soft takeover works. Soft takeover does not work. The issue I am having is that the parameters jump to meet the knob where it was last placed. If I move something to 127 then switch to another channel, move the knob to anything but 127 (let's say 1) and then go back to the original channel and move the knob, the linked parameter will jump to 1.

[nofish]

Originally Posted by ChocolateHawkins

If by catch you mean pass through the current value, then yes. It's not a case of knob not working or me not understanding how soft takeover works. Soft takeover does not work. The issue I am having is that the parameters jump to meet the knob where it was last placed. If I move something to 127 then switch to another channel, move the knob to anything but 127 (let's say 1) and then go back to the original channel and move the knob, the linked parameter will jump to 1.

Yes, that's what I meant. Sorry, just wanted to make sure it's not a matter of expectation how soft takeover works.
I'm not in front of Reaper currently, but I could check in some hours if I get the same behaviour with my controller and soft takeover and report back if noone chimes in till then.

[ChocolateHawkins]

Thank you.

[EpicSounds]

Not working for me either.

Switching tracks and coming back to the first plugin and the control jumps.

OSX. reaper v5.25. Novation Nocturn 49 controller.

[nofish]

I could now test and not working for me either, I also get a jump.
I'd think it's a bug.

[Colox]

I have tried soft takeover from an M-Audio Axiom, and my findings is that different plugins responds differently. On Metric Halo's channelstrip (and several other Metric Halo plugins) a plugin parameter which is controlled from a MIDI device just seems to hop back and forth in enormous increments/decrements. With some other plugins, I've tried Reaplugs and Nomad factory the most, it works pretty well.

But there's this quirky thing about it still:
When I use the "Learn" feature to connect the right parameter with the right control, the plugin parameters moves just fine, like expected. But later, like if I save the control configuration, unload and reload the same plugin, load the control configuration, the parameters on the plugin doesn't seem to respond at all.

I've found that I can make the parameters respond again though, by making a number of really hard twitchy turns on the MIDI controller. For example, if I'm using a slider on the Axiom, I grad the slider and jerk it fully up and down 2-4 times and then - like magic - the plugin parameter starts responding. Sometimes I have to jerk it around alot though.
But only the parameter which the slider is connected to starts working. If I have 8 sliders on the Axiom connected to 4 volume and 4 frequency parameters of an EQ plugin, then I have to do this with all 8 sliders to 'wake up' all parameters on the plugin to respond.

Once they all do respond, they keep responding like desirable. But only on that particular track. Move to a different track, and I have to do the jerking around there too, to wake the controls up. But once I've done this to all plugins that I need to respond, I can move from channel to channel and it works as intended.
Also, when I open an existing project again, I have to jerk around on every track to get everything working. Every time.

I don't have a secondary MIDI controller to compare with, so I'm not sure which of the plugins or the controller causes what, in all this. But it seems clear that different plugins responds differently.

I would of course appreciate if it wasn't like this in the first place

[ChocolateHawkins]

I've tried two controllers (UC-16 and the Impulse 49) and had the same results. Different plugins have no better or worse; it just doesn't work period. I will take this over to the bug reporting section.

[Justin]

Ah, when switching away, I can see how it'd be using the old settings. I think a simple timeout for crossing values would work, let me see if I can implement that.

[ChocolateHawkins]

Thanks Justin!

[Justin]

Head on over to the prerelease forum and check out 5.28pre4, it might help

[ChocolateHawkins]

Thanks again. I will check it out in a few hours when I'm at my machine!

[ChocolateHawkins]

Originally Posted by Justin

Head on over to the prerelease forum and check out 5.28pre4, it might help

This fixed it 100%. Thank you so much.

[ChocolateHawkins]

Sorry Justin. I still am getting jumps. It occurs when I've used the same knob on something else in a different bank (different CC) and then come back. Other than that it works flawlessly!

[Colox]

Originally Posted by ChocolateHawkins

I've tried two controllers (UC-16 and the Impulse 49) and had the same results. Different plugins have no better or worse; it just doesn't work period.

If we assume that most plugins responds the same, I was about to say that this could indicate differences between different MIDI controllers, the data they send.
However, I just realized that the data these controllers are all sending, should be very much alike. If I'm correctly informed, the MIDI protocol shouldn't be complicated enough to open up for such big variances in plugin response - I think.

Just a thought.

[ChocolateHawkins]

A CC is a CC, it shouldn't be anything different. Either way, Justin has it working in the latest pre-build. I had one hiccup with changing banks and using the same knob for different CCs. But I can't replicate it every time.

[DarkStar]

Hmm, I wonder what Justin changed, as Soft Takeover was and is working here (v5.27).

[Justin]

Originally Posted by ChocolateHawkins

Sorry Justin. I still am getting jumps. It occurs when I've used the same knob on something else in a different bank (different CC) and then come back. Other than that it works flawlessly!

REAPER doesn't know about the bank switches, so if you switch banks and turn the knob, the original CC stays locked to those parameters, which means that any movement will be considered valid.

Hmm, I wonder what Justin changed, as Soft Takeover was and is working here (v5.27).

It did work before, though there were a million corner cases when using soft takeover in various contexts (e.g. learned FX parameters but only affecting when track is selected, or using the "adjust selected track volume" action, etc). It would persist having the controller locked to the value even if the selection changed... anyway most of these cases are now handled in a better fashion.

[ChocolateHawkins]

Awesome. That all makes sense. The bank change thing doesn't really matter anyway as I'm getting a controller with more knobs. Each will have one function per channel strip. Thanks again.

[DarkStar]

^^^^
Fair enough - My testing was not poking around in the corners.

This came to mind:

A police officer sees a drunken man intently searching the ground near a lamppost and asks him what he is doing. The inebriate replies that he is looking for his car keys, and the officer helps for a few minutes without success then he asks whether the man is certain that he dropped the keys near the lamppost.

“No,” is the reply, “I lost the keys somewhere across the street.” “Then why look here?” asks the surprised and irritated officer. “The light is much better here,” the intoxicated man responds with aplomb.

[Justin]

Hmm perhaps an action for "reset all soft takeovers" would be helpful for if you have bank switched knobs...

Edit: the action actually exists, but does not work. Fixing!

[Colox]

Just reporting on what happened to me with 5.28pre.
I wrote above, that after setting the mapping up, I still needed to jerk the controller around a bit to 'wake up' the corresponding plugin parameter up. (Using M-Audio Axiom via USB).
With 5.28pre it is the same for me, except that I don't need to jerk the controller around as much anymore. At times, the control wakes up just by moving the controller.

Tried this with different plugin makes, including Reaplugs.

[ChocolateHawkins]

Originally Posted by Colox

Just reporting on what happened to me with 5.28pre.
I wrote above, that after setting the mapping up, I still needed to jerk the controller around a bit to 'wake up' the corresponding plugin parameter up. (Using M-Audio Axiom via USB).
With 5.28pre it is the same for me, except that I don't need to jerk the controller around as much anymore. At times, the control wakes up just by moving the controller.

Tried this with different plugin makes, including Reaplugs.

Even when you've passed through the existing parameters value? That is odd. Obviously the knob won't do anything until you've passed through the value, but having to shake it a few times is weird. I've tested again on both the UC-16 and my Impulse 49 and don't see that behaviour. I am getting the Akai MIDIMix in the mail tomorrow and will try it as well.

[Knob Twiddler]

Is it possible that it just not works with certain plugins as suggested above?
I tried it on this free one: http://socalabs.com/products/papu/
and the midilearn is not working anymore when soft-takeover is activated.

[braj]

I'm trying to use my BCR2000 to control Repro-5, I don't have it set for control messages in the Midi Hardware Settings page, nor set as a control surface. How do I get soft takeover working with it? I'm using Midi Learn from within Repro-5. BTW it sure would be nice to have midi feedback implemented.

[udaemon]

I have just got in contact with arturia because of a problem with the beatstep midi controller. I'm going to paste my initial question as well as the answer here - maybe someone can help me to understand how midi controlling works within reaper:

I first wrote to the arturia support:

I'm having a little problem with the beatstep. I use it mainly to control vst-instruments and effect-plugins in reaper. when I newly open a session and move any controller (the knobs) its assigned value in the daw immediately jumps to 50%. I have set up the assignments in absolute mode. Any other mode doesn't seem to work properly -- or at least after hours of trying to figure it out I still don't know how to set it up in the right manner.
An interesting observation is, I guess, that when I use the minilab in a literally similar setup it works fine right away.

josé from arturia replied as follows:

This is related to how the DAW or virtual instrument handle the matching of the parameter and the position of the knob or fader, This setting can be found on the either the DAW or the Virtual instrument, For example in Ableton Live we have the "Pick Up Mode", which let you configure how Live will react if there's a discrepancy between the knob position and the parameter position:
​​
Analog Lab 3 also has a similar option, is called Fader mode:
​​
This is not related to the Absolute or Relative options, this is a setting that needs to be configure in the DAW, so please read your DAW documentation to learn how to achieve this in your DAW.

Does anybody here know how reaper is behaving or adjusted concerning these modes?
I guess this is also related to what Colox reported earlier.

thanks!!