Control Surface Integration (CSI) Technical Feature Discussion

[Geoff Waddington]

Originally Posted by chinpou

Did I read on here once upon a time that there was an effort to have modifiers mirrored by qwerty keystrokes? I searched through the forum and couldn't find when it was discussed. Just wanna voice that I would greatly benefit from this!

I'm currently using a trial version of Bome MIDI Translator Pro to turn shift, cmd, opt, and ctrl into midi messages, and then SteerMouse to route four extra mouse buttons to shift, cmd, opt, and ctrl. It's working fine but I'd love to cut out as many middle men as I can ... and not have to pay for Bome hehe

We've had discussions about this.

On the surface (no pun intended ) it looks attractive, but the details, oh the details...

Questions like, should this be two way ?

So, if you hit Shift on the keyboard, does that cause CSI surfaces to be shifted too ?

If you want to modify a mouse action for editing only what hapens to the surface ?

If you modify the surface, what happens to the mouse, does your mouse editing suddenly change ?

What if you want shift on the surface but not Reaper or vice versa ?

Since the overall CSI design philosophy is to reflect the Reaper state, we decided not to do this.

Your approach seems to do a fine job, although perhaps a bit more costly

[jakeman19]

Originally Posted by Jur451c

What the title says! Indicating which tracks are being controlled as you bank up and down on the control surface would be a massive quality of life improvement, especially if you could have different colours for different surfaces.

Jur451c, read this thread and let me know if you have any issues. It took me a bit to get it to work. The only drawback is with VCA Spill(Show/hide children of selected tracks). It only works if you have all tracks showing. Been trying to figure out a workaround but haven't as of yet. Ophic's solution was what I ended up using.

Even in CSI v2.0 you still need to have the BankLeft/BankRight commands in the Track.zon file not in Buttons.zon file.
In the attachment, the gold faders are what it's controlling Reaper.

https://forum.cockos.com/showthread....24#post2534424

Jd

[strangy]

Finally made a surround panner for ReaSurroundPan, attached file.

There is two XY Pads and small FX menu.

I have to use latest version of TouchOSC because it supports separating OSC messages in XY Pad (legacy version only supports bundled). Also I removed feedback because of ReaSurroundPan for some reason receiving inverted messages (1...0 instead of 0...1), but sending back normal (0...1) so it make XY jumping to inverted position after touch.

Still very happy with the result, thank you, CSI is awesome!

[Puck]

Originally Posted by strangy

Finally made a surround panner for ReaSurroundPan, attached file.

There is two XY Pads and small FX menu.

I have to use latest version of TouchOSC because it supports separating OSC messages in XY Pad (legacy version only supports bundled). Also I removed feedback because of ReaSurroundPan for some reason receiving inverted messages (1...0 instead of 0...1), but sending back normal (0...1) so it make XY jumping to inverted position after touch.

Still very happy with the result, thank you, CSI is awesome!

Nice job!

Yeah I should have mentioned you needed the new TouchOSC.

You actually can just use a single XY control. All you need to do is add 2 OSC messages to one control.

In the two osc messages you need to change the trigger and argument sections.

For the x message, make sure the trigger is set to x and not y, and in the argument section just delete the “y” argument. Then do the same for the 2nd message (the Y one) where’d make the trigger Y and not x and deleted the x argument.

Try that first and if the feedback is still messed up, try adding “InvertFB+“ to your fx zone like this:

InvertFB+yourXControl     FXParam ….

To any control where the feedback is backwards.

[strangy]

Originally Posted by Puck

Nice job!

Try that first and if the feedback is still messed up, try adding “InvertFB+“ to your fx zone like this:

InvertFB+yourXControl     FXParam ….

To any control where the feedback is backwards.

Not working. I suppose it's only for buttons

[strangy]

Originally Posted by Puck

Nice job!

You actually can just use a single XY control. All you need to do is add 2 OSC messages to one control.

Yes, I did the same. Two XY pads is for L and R channels

[Puck]

Originally Posted by strangy

Yes, I did the same. Two XY pads is for L and R channels

Oh! I understand now. The image looked like it was one xy for x and one xy for y.

You are also probably correct on “InvertFB”

You can try inverting it somehow in TouchOSC. I have a couple ideas but rather than sending you down a rabbit hole I’ll try some stuff on my own and reply back if I have a solution. Not having feedback bothers me

(It only bothers me because CSI makes things so tightly integrated that it’s a shame when it’s not)

[MT4U]

Originally Posted by MixMonkey

You're welcome Make sure the toggle bullet is on.

Just as you said, Sire. Working like a charm now. Thanks again!

[MixMonkey]

Originally Posted by MT4U

Just as you said, Sire. Working like a charm now. Thanks again!

Excellent Thanks to @Navelpluisje for telling me about this. It's by far the easiest way of adding LED feedback to Reaper Actions that don't have it. Much easier than SWSToggles/ExclusiveToggles.

[Jur451c]

Originally Posted by jakeman19

Jur451c, read this thread and let me know if you have any issues. It took me a bit to get it to work. The only drawback is with VCA Spill(Show/hide children of selected tracks). It only works if you have all tracks showing. Been trying to figure out a workaround but haven't as of yet. Ophic's solution was what I ended up using.

Even in CSI v2.0 you still need to have the BankLeft/BankRight commands in the Track.zon file not in Buttons.zon file.
In the attachment, the gold faders are what it's controlling Reaper.

https://forum.cockos.com/showthread....24#post2534424

Jd

Thanks for this, I’ll see if I can get it going!

[strangy]

Originally Posted by Puck

Oh! I understand now. The image looked like it was one xy for x and one xy for y.

Yes, I made it late night, some confusion is leaked to GUI. Changed to ReaSurroundPan L and R.
I tried to replicate classic setup of surround panner in postproduction studio like this s6 Joystick Module

https://www.hhb.co.uk/prod/avid/pro-...ystick-module/

Originally Posted by Puck

You can try inverting it somehow in TouchOSC. I have a couple ideas but rather than sending you down a rabbit hole I’ll try some stuff on my own and reply back if I have a solution. Not having feedback bothers me

(It only bothers me because CSI makes things so tightly integrated that it’s a shame when it’s not)

Totally agree. I lost enthusiasm at the point where I understand that I should learn how to script in touchOSC to move forward. Maybe later.

[Geoff Waddington]

Time for the semi-annual donation drive.

Aside from a very few diehards, (many thanks to you folks), the donations from the community have been, of late, to put it bluntly, pathetic, quite disheartening actually

Please consider donating to the PayPal account -- waddingtongeoff@gmail.com, if you find CSI useful.

[Navelpluisje]

Is it possible to reload plugins without relaunching reaper? Everytime I compile the project I have to relaunch Reaper now. And because C++ is not my mother tonque I compile a lot of small changes.

[MT4U]

Originally Posted by Geoff Waddington

Time for the semi-annual donation drive.

Aside from a very few diehards, (many thanks to you folks), the donations from the community have been, of late, to put it bluntly, pathetic, quite disheartening actually

Please consider donating to the PayPal account -- waddingtongeoff@gmail.com, if you find CSI useful.

It is done, Geoff. Thanks for everything!

[Geoff Waddington]

Originally Posted by Navelpluisje

Is it possible to reload plugins without relaunching reaper? Everytime I compile the project I have to relaunch Reaper now. And because C++ is not my mother tonque I compile a lot of small changes.

There have been some reports of CSI not resetting, would love to get to the bottom of it.

You should be able to call a Reaper Action: "Reset all MIDI control surface devices"

Try putting a breakpoint here in control_surface_integrator_ui.cpp, should be around line 92:

    if(call == CSURF_EXT_RESET)
    {
       if(TheManager)
           TheManager->Init();
    }

Let me know what you find, as well as your OS version, Reaper version, etc.

[Geoff Waddington]

Originally Posted by MT4U

It is done, Geoff. Thanks for everything!

Thank you !

[Navelpluisje]

Originally Posted by Geoff Waddington

There have been some reports of CSI not resetting, would love to get to the bottom of it.

You should be able to call a Reaper Action: "Reset all MIDI control surface devices"

Try putting a breakpoint here in control_surface_integrator_ui.cpp, should be around line 92:

    if(call == CSURF_EXT_RESET)
    {
       if(TheManager)
           TheManager->Init();
    }

Let me know what you find, as well as your OS version, Reaper version, etc.

Thanx, Looks like I had the setup for debugging not sorrect. Now after a build it restarts reaper.
Tried

CSURF_EXT_RESET

a couple of times, and as always it works as intended.

Finally can debug as it should.

[Geoff Waddington]

New build is up.

CSI Exp.zip

Should cure the reset woes, I suspect this is the root cause of a bunch of other similar reports.

[Funkybot]

Just a head's up that I've been slowly and quietly and working on the FX pages on the new wiki.

The first page covers the how to of FX and Instrument Mapping :
https://github.com/GeoffAWaddington/...rument-Mapping

The second page covers all of the FX related mapping actions (I've never used FXParamRelative or FXGainReductionMeter so definitely need some Geoff input on those):
https://github.com/GeoffAWaddington/...apping-Actions

...there will be a third page that covers the different activation methods for FX. I've already removed some FX related pages and consolidated them into the pages above. I will continue to do that for the activation page.

Please everyone take a look and provide feedback. I've been leaning towards fewer, but longer pages so information isn't scattered about on 8 or 9 pages. But there's a lot more information on a single page and I know I can be verbose (Geoff, of course feel free to act as an editor). The content on the FX and Instrument mapping page is mostly new as well.

[Geoff Waddington]

Originally Posted by Funkybot

Just a head's up that I've been slowly and quietly and working on the FX pages on the new wiki.

The first page covers the how to of FX and Instrument Mapping :
https://github.com/GeoffAWaddington/...rument-Mapping

The second page covers all of the FX related mapping actions (I've never used FXParamRelative or FXGainReductionMeter so definitely need some Geoff input on those):
https://github.com/GeoffAWaddington/...apping-Actions

...there will be a third page that covers the different activation methods for FX. I've already removed some FX related pages and consolidated them into the pages above. I will continue to do that for the activation page.

Please everyone take a look and provide feedback. I've been leaning towards fewer, but longer pages so information isn't scattered about on 8 or 9 pages. But there's a lot more information on a single page and I know I can be verbose (Geoff, of course feel free to act as an editor). The content on the FX and Instrument mapping page is mostly new as well.

Thanks man, real nice work, as usual !

A few typos, etc., here and there, I might just go ahead and correct them, once I get the EuCon stuff a bit more complete -- it's taking more time than expected

As an overall, I notice there are references to previous versions here and there -- stuff like "this was introduced in 1.1".

Wonder if it would be cleaner to just leave those out -- taking the position that 2.0 is a clean slate, just wondering...

[Funkybot]

Originally Posted by Geoff Waddington

Thanks man, real nice work, as usual !

A few typos, etc., here and there, I might just go ahead and correct them, once I get the EuCon stuff a bit more complete -- it's taking more time than expected

As an overall, I notice there are references to previous versions here and there -- stuff like "this was introduced in 1.1".

Wonder if it would be cleaner to just leave those out -- taking the position that 2.0 is a clean slate, just wondering...

Agree on the clean slate. Will remove. Typos aren't uncommon but feel free to correct any if you see them. I tend to revise, revise, revise so might catch some on my own.

[Mage Prometheus]

Thank you, Funkybot. This is exactly what I need to study.

[Puck]

Here is what I believe to be the minimum needed to test multiple pages with OSC. Figured it should go in the Dev thread. Please check zones for syntax errors or something I maybe missing and also test this as well.

Here's what happens on my machine (Win 10, using local IP address and a touchscreen monitor)

1. CSI starts, Page1 loads and works just fine

2. Press button that does a

GoPage "Page2"

result - Page2 loads fine, all new text gets sent and displayed properly. At this point, I have feedback but I do not have control anymore. CSI doesn't display any information in the log when I push any buttons.

Restarting Reaper gets me back to step 1 again.

[MixMonkey]

As FFWD and REW now latch, could they send feedback? (so we can make the button light up when they're engaged)

[Geoff Waddington]

Originally Posted by MixMonkey

As FFWD and REW now latch, could they send feedback? (so we can make the button light up when they're engaged)

I tried that, but on the X-Touch no joy, does the MCU have lights on the FF and REW ?

[MixMonkey]

Originally Posted by Geoff Waddington

I tried that, but on the X-Touch no joy, does the MCU have lights on the FF and REW ?

I was thinking more of OSC

[Geoff Waddington]

Originally Posted by MixMonkey

I was thinking more of OSC

Ah, right

[MixMonkey]

Originally Posted by Geoff Waddington

Ah, right

+ someone's bound to come along with a surface that does have LEDs in the REW and FFWD buttons

[Geoff Waddington]

New build is up.

CSI Exp.zip

Added feedback to Rewind and Fast Forward.

[Funkybot]

Made some revisions to:

https://github.com/GeoffAWaddington/...rument-Mapping

...hopefully got all the typos and made a few things more clear. And yes, that page has gone through 30 revisions. It's madness how I work!

[MixMonkey]

Originally Posted by Geoff Waddington

New build is up.

CSI Exp.zip

Added feedback to Rewind and Fast Forward.

Working nicely, many thanks

[Geoff Waddington]

Originally Posted by Funkybot

Made some revisions to:

https://github.com/GeoffAWaddington/...rument-Mapping

...hopefully got all the typos and made a few things more clear. And yes, that page has gone through 30 revisions. It's madness how I work!

Nice !!

[MixMonkey]

Originally Posted by Puck

Here is what I believe to be the minimum needed to test multiple pages with OSC. Figured it should go in the Dev thread. Please check zones for syntax errors or something I maybe missing and also test this as well.

Here's what happens on my machine (Win 10, using local IP address and a touchscreen monitor)

1. CSI starts, Page1 loads and works just fine

2. Press button that does a

GoPage "Page2"

result - Page2 loads fine, all new text gets sent and displayed properly. At this point, I have feedback but I do not have control anymore. CSI doesn't display any information in the log when I push any buttons.

Restarting Reaper gets me back to step 1 again.

Exactly the same thing happens on my setup (macOS Monterey, local WiFi network and an iPad Air4).

I couldn't see anything wrong in the syntax, but I've never used Pages (!), so don't rely on that too much

[jakeman19]

Originally Posted by Geoff Waddington

I tried that, but on the X-Touch no joy, does the MCU have lights on the FF and REW ?

Geoff, yes the MCU has lights on each of the transport controls.

Jd

[Navelpluisje]

Originally Posted by Geoff Waddington

I tried that, but on the X-Touch no joy, does the MCU have lights on the FF and REW ?

The Faderport has

[Geoff Waddington]

Originally Posted by Puck

Here is what I believe to be the minimum needed to test multiple pages with OSC. Figured it should go in the Dev thread. Please check zones for syntax errors or something I maybe missing and also test this as well.

Here's what happens on my machine (Win 10, using local IP address and a touchscreen monitor)

1. CSI starts, Page1 loads and works just fine

2. Press button that does a

GoPage "Page2"

result - Page2 loads fine, all new text gets sent and displayed properly. At this point, I have feedback but I do not have control anymore. CSI doesn't display any information in the log when I push any buttons.

Restarting Reaper gets me back to step 1 again.

Had a look at your .zon and touch OSC files.

Was just about to ask for your CSI.ini file and then it hit me

Arrrghh !!

There is a fundamental design flaw in CSI.

Right now you add a Page, then add Surfaces.

That is only partially correct.

Surface .mst and Zone folders should be per Page, but I/O should be common

What really should happen is this:

Add surfaces, giving them names and I/O configurations -- ports, IP addresses, number of channels, etc.

(We could theoretically put channel start position in the Page-Surface definition, pretty esoteric to allow rearrangement of the surfaces per Page ? -- maybe not ? Seems more intuitive to have them in the Surface definition, dunno..)

Then you can define Pages, the available Surfaces should appear in dropdowns, just like .mst file and Zone folder choices.

All of this makes for a more complicated Config panel, but it is the right way to do it.

Wow, didn't see that coming, open to discussion as to how best to resolve this

[MixMonkey]

Originally Posted by Geoff Waddington

Wow, didn't see that coming, open to discussion as to how best to resolve this

Not sure I understand the problem here. Is this something that only affects OSC? Surely the hardware connections are the same on different pages, because you've set them the same?

[Puck]

Originally Posted by Geoff Waddington

Add surfaces, giving them names and I/O configurations -- ports, IP addresses, number of channels, etc.

So in my testing, I did keep this stuff the same. Or at least that was the intention.

However, the reason I stumbled upon this a few months back is:

I was trying to copy Eucon's Track list. So my intention was to flip to a second page with significantly more channels to have a bird's eye view of your tracks.

Here's my .ini contents just for the sake of being thorough:

Version 2.0
Page "Page1" FollowTCP NoSynchPages UseScrollLink 
OSCSurface "PucksTouchOSCSurfaceForReaper" 9062 9082 "OSCPageTest.ost" "OSCPageTest" 0 0  127.0.0.1 

Page "Page2" FollowTCP NoSynchPages NoScrollLink 
OSCSurface "OSCPageTestPage2" 9062 9082 "OSCPageTest.ost" "OSCPageTestpage2" 0 0  127.0.0.1

[Geoff Waddington]

Originally Posted by Puck

So in my testing, I did keep this stuff the same. Or at least that was the intention.

However, the reason I stumbled upon this a few months back is:

I was trying to copy Eucon's Track list. So my intention was to flip to a second page with significantly more channels to have a bird's eye view of your tracks.

Here's my .ini contents just for the sake of being thorough:

Version 2.0
Page "Page1" FollowTCP NoSynchPages UseScrollLink 
OSCSurface "PucksTouchOSCSurfaceForReaper" 9062 9082 "OSCPageTest.ost" "OSCPageTest" 0 0  127.0.0.1 

Page "Page2" FollowTCP NoSynchPages NoScrollLink 
OSCSurface "OSCPageTestPage2" 9062 9082 "OSCPageTest.ost" "OSCPageTestpage2" 0 0  127.0.0.1

Hmmm...

You have 0 channels on both surfaces...

[Puck]

Originally Posted by Geoff Waddington

Hmmm...

You have 0 channels on both surfaces...

I'll change that and try it.