Control Surface Integration (CSI) Technical Feature Discussion

[Geoff Waddington]

Originally Posted by Funkybot

One more small bug I mentioned in the somewhat recent past and just ran into again: the action "Reset all MIDI control surface devices" used to Reset CSI entirely. Now it disengages CSI, but CSI no longer re-engages. You actually have to go to Preferences -> Control/OSC/Web -> Control Surface Integrator and OK out of there to get it turn back on.

Yeah, that's a by-product of having CSI be more responsive when switching projects, we don't reset OSC and EuCon and our Midi devices.

The key is the name of the action "Reset all MIDI control surface devices". Reaper handles Midi for control surfaces in general, but in CSI we basically roll our own, handling the whole thing a bit differently along with OSC and EuCon.

Probably should note that in the wiki somewhere - under things that work differently in CSi compared to other control surface extensions

[Geoff Waddington]

Originally Posted by Cragster

with this new mst do i remove all the rotarypush widgets now that they are together with the rotarys ?

Up to you, you can keep both if you want to use the RotaryPush functionality in other zones.

[Geoff Waddington]

Originally Posted by Geoff Waddington

Cool, thanks.

I've made a EuCon page and linked it to all the actions in the EuCon Specific section of the action reference -- maybe just write a very short description of each action as it applies to EuCon, might want to include how you navigate Sends, remember the confusion from the other day, would be good to clarify how it works.

Just realized, I need to add you to collaborators, shoot me your handle, either here or by PM, I'll give you write access

[Funkybot]

Originally Posted by Geoff Waddington

Just realized, I need to add you to collaborators, shoot me your handle, either here or by PM, I'll give you write access

My Git handle is FunkybotsEvilTwin.

[Geoff Waddington]

Originally Posted by Funkybot

My Git handle is FunkybotsEvilTwin.

Cool, you should receive an invite.

[cjewellstudios]

Originally Posted by Funkybot

My Git handle is FunkybotsEvilTwin.

Best alternate handle ever. I need to change mine, I honestly never thought I'd be terribly active on the forums so regrettably I didn't put much thought into my name. I'm open to suggestions

Anyhow, as long as we are in the wiki, in the FB_MFT_RGB tab there is a table of colors. I don't know why but aqua doesn't actually work. I had to change it to 0 250 255 as opposed to 0 255 255. I remember there's something weird about the MFT RGB spec in that one number has to be zero and another number has to be 255 so I'm not sure if that's why or what.

[todoublez]

Originally Posted by Geoff Waddington

I have, use, and love my Console 1 -- it integrates very well into CSI.

If you don't load the Softube software, it sends and receives Midi just like most generic controllers.

I'm not sure about the Console 1 fader, did a slight bit of digging and can't find a definite answer.

If it doesn't transmit and receive Midi without the Console 1 software loaded, it may not work with CSI.

Also, as far as I'm concerned the lack of scribble strips (Track name displays) is a show stopper -- this is not a problem if you load the Softube software, they have an accompanying app that takes the place of scribble strip functionality.

It certainly looks like the same rugged construction that makes the Console 1 a great device, but can't give you a definitive answer on whether it sends/receives Midi without the software, and without said software, you have no visual indication of where you are.

Interested in seeing what others have to say about this.

Thx ! I'm glad that i havent buy it yet !
but I'll consider to get a console 1 instead
much appreciated yo info !

[Funkybot]

Originally Posted by Geoff Waddington

New EuCon only build is up.

Touch fix for PanWidth, PanL, PanR.

No zon changes right? I just gave this a go:

1. With PanL and PanR - no automation is being written here at all. Can another Eucon user confirm? I just double checked I'm on the 11/11 build and the Artist mix is showing PanL and PanR in the displays but no automation is written at all.

2. PanWidth works, but...

3. There's some unexpected behavior with TouchMode and Pan along with PanWidth, and maybe it's about my expectations but say I ramp a pan down from initial to minimum....then release. It jumps back to the initial value. Great!!! Now touch again....instead of starting at the initial value, now it jumps to minimum value from before. Great for creating tremolo-like panning effects, but probably not intended. Faders don't work this way anyway.

It's almost like it's storing 2 values: 1) what's the initial "start" value, 2) what's the last value I remember before being released...then it alternates between those two with each touch and release message. My expectation is it always picks up at the initial/start value on each new touch.

See the screen print below where you can see this behavior on the Pan and PanWidth lanes...

[Geoff Waddington]

Originally Posted by Funkybot

No zon changes right? I just gave this a go:

1. With PanL and PanR - no automation is being written here at all. Can another Eucon user confirm? I just double checked I'm on the 11/11 build and the Artist mix is showing PanL and PanR in the displays but no automation is written at all.

2. PanWidth works, but...

3. There's some unexpected behavior with TouchMode and Pan along with PanWidth, and maybe it's about my expectations but say I ramp a pan down from initial to minimum....then release. It jumps back to the initial value. Great!!! Now touch again....instead of starting at the initial value, now it jumps to minimum value from before. Great for creating tremolo-like panning effects, but probably not intended. Faders don't work this way anyway.

It's almost like it's storing 2 values: 1) what's the initial "start" value, 2) what's the last value I remember before being released...then it alternates between those two with each touch and release message. My expectation is it always picks up at the initial/start value on each new touch.

See the screen print below where you can see this behavior on the Pan and PanWidth lanes...

Hmmm...

That sure looks like either EuCon, CSI, or both, are earlier versions, are you sure both are the latest, and you are using the EuCon.zon from the build ?

[siniarch]

Originally Posted by Geoff Waddington

We are close to CSI version 1.0, I have been hesitant to post this, but I think it really is finally time.

First, a heartfelt thanks to all of the great contributors in this community, the four year journey has been remarkable and rewarding !!

Second, a heartfelt thanks to all those who have donated on the website, it is very much appreciated !

Which brings us to the third point...

There are two whose donations have been substantial and/or recurring, a special thanks to those two !

However, aside from those two, the sum total donations for 2020 are... wait for it... $310.

That's not a typo.

The only thing I can deduce from that is that CSI isn't really of that much use monetarily.

I'm considering taking down the donation website, as it costs money to keep alive, and so far this year, aside from the two mentioned, has not even earned its keep.

Just sayin'

Whoa, whoa, whoa. Let's not be so rash over here. I don't think there is any need to take this donation site down.

1st. You can have a free super easy to set up website on WIX.com. I'm assuming you could link a paypal pay button to it while still having a FREE plan from them? (yes it would have their logo, but who cares)

2nd. I do pay for hosting my site and in it I can host as many sub-sites as I want. So, one option would be that I could set you up with one here and it would cost you nothing. In the past I have done simple Wordpress websites with much success. But of course the issue is that at any point I could re-direct that sweet $310 paypal income to my bank account when you are not looking. hahahaha. So it might not be a good option.

3rd. I would be more than happy to pay for a website for you for the next 3 years and for the hosting of it so you can have your pay landing page and not have to pay for it. Of course it would be in a hosting service that is cheap like mine. hostmetro.com. I believe it comes out to about $2.99/month + the $15/year for the domain name.

4th. As I said before, I'll be donating a small portion of each sale of my controller to the CSI project because without it, I would have never even attempted to start this journey.

5th. I'll donate a little more shortly.

[uksnowy]

Originally Posted by siniarch

I don't have my studio up and running yet, but this is the CSI beta file that I used when I created that MST and Zon file. I'm guessing the crashing is coming from the new version and having to alter the files to conform with the latest updates to CSI.

http://www.siniarch.com/csiconfigfil...CSI%20beta.zip

I guess for now, it would be best to just install this previous version.

Best of luck.

Thank you I will give it a go..

[Geoff Waddington]

Originally Posted by siniarch

Whoa, whoa, whoa. Let's not be so rash over here. I don't think there is any need to take this donation site down.

1st. You can have a free super easy to set up website on WIX.com. I'm assuming you could link a paypal pay button to it while still having a FREE plan from them? (yes it would have their logo, but who cares)

2nd. I do pay for hosting my site and in it I can host as many sub-sites as I want. So, one option would be that I could set you up with one here and it would cost you nothing. In the past I have done simple Wordpress websites with much success. But of course the issue is that at any point I could re-direct that sweet $310 paypal income to my bank account when you are not looking. hahahaha. So it might not be a good option.

3rd. I would be more than happy to pay for a website for you for the next 3 years and for the hosting of it so you can have your pay landing page and not have to pay for it. Of course it would be in a hosting service that is cheap like mine. hostmetro.com. I believe it comes out to about $2.99/month + the $15/year for the domain name.

4th. As I said before, I'll be donating a small portion of each sale of my controller to the CSI project because without it, I would have never even attempted to start this journey.

5th. I'll donate a little more shortly.

Thanks for the kind thoughts, the reason i went with a paid site is the PayPal issue you mentioned, I couldn't find a way to get a free PayPal button on a free site.

That said I'm very happy to report things have changed in the last 24 hours, a bunch of the good folk here have stepped up to the plate and hit it out of the park, we're all good now, thanks everyone.

[Cragster]

Originally Posted by Geoff Waddington

Thanks for the kind thoughts, the reason i went with a paid site is the PayPal issue you mentioned, I couldn't find a way to get a free PayPal button on a free site.

That said I'm very happy to report things have changed in the last 24 hours, a bunch of the good folk here have stepped up to the plate and hit it out of the park, we're all good now, thanks everyone.

Thats great news !!!

[Geoff Waddington]

New build is up.

Hold is now working, and, due to the nature of the fix, acceleration/stepped should work again too (crosses fingers) -- let me know...

[Funkybot]

Hold and Stepped Parameters are both working here. Just reversed stepped parameters in the case of the MFTwister. But that's expected.

Will retest the Eucon Pans and dig more into that later. If there was an updated .zon file recently, I didn't install it yesterday, but I grabbed the one from today's build in case it was that.

[Geoff Waddington]

Originally Posted by Funkybot

Hold and Stepped Parameters are both working here. Just reversed stepped parameters in the case of the MFTwister. But that's expected.

Will retest the Eucon Pans and dig more into that later. If there was an updated .zon file recently, I didn't install it yesterday, but I grabbed the one from today's build in case it was that.

Fantastic news !

If you have an older version of CSI, EuCon, or both, you would get the behaviour you saw, hope that bug didn't come to life

[Geoff Waddington]

Ok, getting really close !

@Funkybot, is it possible, just as a test, to compensate for the MFT by rearranging on the action end ?

I'm talking about this:

Widget RotaryA1
	Encoder b0 00 7f [ < 3f 3e 3d 3c 3b 3a 39 38 36 33 2f > 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

And if the values on the action end are like this, only more of them:

[ (0.001,0.005,0.025,0.5,0.1) ]

change them to:

[ (0.1,0.5,0.025,0.005,0.001) ]

What happens ?

[Geoff Waddington]

Anyone else besides @MixMonkey having problems on MCU with:

Lower Displays

Faders jumping around incorrectly in Touch mode when mapped to Sends

[pwallin]

Congratulations for near finished V1.0 Geoff! Also thanks to others who have done a big job of helping us. I am in a same boat with some of others, by not yet donated for a simple reason...I barely have grasped it due it's complexity! Now if we could have a proper user manual with plenty of step by step examples starting from simplest systems to a more advanced multi-controller systems and some ready made confiqurations files I would even buy it!
Why not make it commercial? Reaper users are used to expecting free stuff, but if price tag would be fair enough it would not be an issue and would earn you much more than donations. While I have not donated it (yet) I have kept it's flag up in user groups for a some time now. Sometimes even in "dangerous" places full of "non-Reaper" people. I will continue to do that, because while I have a very little understanding of how to program my own stuff, CSI already has made my working so much smoother with my X-Touch. Personally I think CSI is a "must have" for anyone serious with their controller. Problem is that not everyone knows about it yet and I sure keep shouting about it in every conversations that has even remotely DAW controller specific.

[Funkybot]

Originally Posted by Geoff Waddington

Ok, getting really close !

@Funkybot, is it possible, just as a test, to compensate for the MFT by rearranging on the action end ?

I'm talking about this:

Widget RotaryA1
	Encoder b0 00 7f [ < 3f 3e 3d 3c 3b 3a 39 38 36 33 2f > 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

And if the values on the action end are like this, only more of them:

[ (0.001,0.005,0.025,0.5,0.1) ]

change them to:

[ (0.1,0.5,0.025,0.005,0.001) ]

What happens ?

I can make the changes in the .zon file easily enough and test in a bit, but just so I'm clear. Am I doing anything on the .mst?

Your .mst example is the same as my current-state .mst, so am I making changes there to reverse the values or keeping that the same?

For instance this (current state):

Widget RotaryA1
	Encoder b0 00 7f [ < 3f 3e 3d 3c 3b 3a 39 38 36 33 2f > 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

Then becomes this...

Widget RotaryA1
	Encoder b0 00 7f [ < 2f 33 36 38 39 3a 3b 3c 3d 3e 3f > 51 4d 4a 48 47 46 45 44 42 41 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

Or maybe this?

Widget RotaryA1
	Encoder b0 00 7f [ > 3f 3e 3d 3c 3b 3a 39 38 36 33 2f < 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

...or are we only talking changes to the .zon file.

...just want to make sure I'm testing this the right way before I start. Lest I go down a rabbit hole and start reporting problems that don't exist.

[Geoff Waddington]

Originally Posted by Funkybot

I can make the changes in the .zon file easily enough and test in a bit, but just so I'm clear. Am I doing anything on the .mst?

Your .mst example is the same as my current-state .mst, so am I making changes there to reverse the values or keeping that the same?

For instance this (current state):

Widget RotaryA1
	Encoder b0 00 7f [ < 3f 3e 3d 3c 3b 3a 39 38 36 33 2f > 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

Then becomes this...

Widget RotaryA1
	Encoder b0 00 7f [ < 2f 33 36 38 39 3a 3b 3c 3d 3e 3f > 51 4d 4a 48 47 46 45 44 42 41 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

Or maybe this?

Widget RotaryA1
	Encoder b0 00 7f [ > 3f 3e 3d 3c 3b 3a 39 38 36 33 2f < 41 42 43 44 45 46 47 48 4a 4d 51 ]
	FB_Fader7Bit b0 00 00
WidgetEnd

...or are we only talking changes to the .zon file.

...just want to make sure I'm testing this the right way before I start. Lest I go down a rabbit hole and start reporting problems that don't exist.

I would start by changing the zone file only -- reversing the value list.

[Funkybot]

It works for stepped parameters, but doesn't do anything to resolve continuous parameters. Those are still reversed.

[Funkybot]

The invert modifier doesn't work with encoders either best I can tell (if it's still around). I tried this with no luck:

Invert+RotaryB5	FXParam 1 "Input Gain"	[ (0.001,0.0015,0.002,0.0025,0.005,0.0075,0.0125,0.0215,0.033,0.05,0.075) ]

[Geoff Waddington]

Originally Posted by Funkybot

It works for stepped parameters, but doesn't do anything to resolve continuous parameters. Those are still reversed.

OK, do you have an example of a acceleration list for another normal encoder with the same number of steps as the MFT ?

If we had one of those I could build a special MFT widget with a lookup dictionary that translated the MFT numbers into the more familiar ones.

Originally Posted by Funkybot

The invert modifier doesn't work with encoders either best I can tell (if it's still around). I tried this with no luck:

Invert+RotaryB5	FXParam 1 "Input Gain"	[ (0.001,0.0015,0.002,0.0025,0.005,0.0075,0.0125,0.0215,0.033,0.05,0.075) ]

Invert is really only supposed to invert ones and zeros for switches -- like where Reaper bypass checked means you turn the surface light off -- that sort of thing.

[cjewellstudios]

Originally Posted by Geoff Waddington

OK, do you have an example of a acceleration list for another normal encoder with the same number of steps as the MFT ?

If we had one of those I could build a special MFT widget with a lookup dictionary that translated the MFT numbers into the more familiar ones.

I remember when we were introducing acceleration that the MFT had the most values by far. 11 values either way. I'm not saying its not out there, it certainly could be.

[Funkybot]

Originally Posted by cjewellstudios

I remember when we were introducing acceleration that the MFT had the most values by far. 11 values either way. I'm not saying its not out there, it certainly could be.

Yeah, the MFTwister has 11 values in either direction. By comparison X-Touch One encoders have 8 or 9 values in either direction. I don't think I've seen any other devices with 11. The challenge is that on the MCU/X-Touch, 41 means CCW turn, whereas on the MFTwister 41 means CW turn hence the inverted behavior. I don't think the MFTwister authors were thinking about MCU when they were coding up the firmware.

I have no idea how much complexity that one device adds to CSI, but there's quite a few of us with them, and it's a pretty popular little box. Would a "MFTEncoder" widget make it easier? Sort of like MFT_RGB processor? As an end-user that would be totally fine with me.

The values for the turns are below (1=slow, 12=fast).

CCW Turn 1 B0 00 3F
CCW Turn 2 B0 00 3E
CCW Turn 3 B0 00 3D
CCW Turn 4 B0 00 3C
CCW Turn 5 B0 00 3B
CCW Turn 6 B0 00 3A
CCW Turn 7 B0 00 39
CCW Turn 8 B0 00 38
CCW Turn 9 B0 00 36
CCW Turn 10 B0 00 33
CCW Turn 11 B0 00 2F
CW Turn 1 B0 00 41
CW Turn 2 B0 00 42
CW Turn 3 B0 00 43
CW Turn 4 B0 00 44
CW Turn 5 B0 00 45
CW Turn 6 B0 00 46
CW Turn 7 B0 00 47
CW Turn 8 B0 00 48
CW Turn 9 B0 00 4A
CW Turn 10 B0 00 4D
CW Turn 11 B0 00 51

[Geoff Waddington]

Ok, now we just need a normal set to map to — 11 values would be nice, but if we only have ten or nine or even 8 I think we can make it work, anyone got a set of values ?

[Funkybot]

Originally Posted by Geoff Waddington

Ok, now we just need a normal set to map to — 11 values would be nice, but if we only have ten or nine or even 8 I think we can make it work, anyone got a set of values ?

Here are the X-Touch One values...

CCW Turn 1 B0 10 41
CCW Turn 2 B0 10 42
CCW Turn 3 B0 10 43
CCW Turn 4 B0 10 44
CCW Turn 5 B0 10 45
CCW Turn 6 B0 10 46
CCW Turn 7 B0 10 47
CCW Turn 8 B0 10 48
CW Turn 1 B0 10 01
CW Turn 2 B0 10 02
CW Turn 3 B0 10 03
CW Turn 4 B0 10 04
CW Turn 5 B0 10 05
CW Turn 6 B0 10 06
CW Turn 7 B0 10 07
CW Turn 8 B0 10 08

[Geoff Waddington]

New build is up.

Originally Posted by Funkybot

Would a "MFTEncoder" widget make it easier?

Yup, "MFTEncoder" is exactly what I had in mind, and is in this build.

Just replace Encoder with MFTEncoder and leave everything else alone.

Give it a try and let me know...

[Funkybot]

Originally Posted by Geoff Waddington

New build is up.



Yup, "MFTEncoder" is exactly what I had in mind, and is in this build.

Just replace Encoder with MFTEncoder and leave everything else alone.

Give it a try and let me know...

Works like a charm!

[Geoff Waddington]

Originally Posted by Funkybot

Works like a charm!

Fantastic !

[Geoff Waddington]

Anyone else besides @MixMonkey having problems on MCU with:

Lower Displays

Faders jumping around incorrectly in Touch mode when mapped to Sends

Given that MFT Encoders are now working, I'm going to post CSI v0_90, on the way towards CSI v1_0

Please post any other outstanding bugs, we are very close

[Geoff Waddington]

New build is up.

CSI v0_90.zip

See link in sig below.

[cjewellstudios]

Originally Posted by Geoff Waddington

Anyone else besides @MixMonkey having problems on MCU with:

Lower Displays

Faders jumping around incorrectly in Touch mode when mapped to Sends

Given that MFT Encoders are now working, I'm going to post CSI v0_90, on the way towards CSI v1_0

Please post any other outstanding bugs, we are very close

I haven't confirmed the mftencoders myself but I am so happy to hear they are working!

As far as possible bugs:

When you close a mapped fx gui, all my encoders go blank.

Resetting all midi devices OR clicking on an fx that isn't mapped goes back to the home zone.

Clicking on a mapped fx returns to that fx zone.

Not the end of the world, and it sounds like you ran up against a wall trying to figure that out, so maybe this is more a feature than a bug

(Learned that from my Brother-in-law, bugs that you can't fix become features )

[Funkybot]

Here's a new issue:

Used to be able to cycle through stepped parameters using buttons like this...

Press+ButtonA3 FXParam 3 "TapeType"	[ 0.0 0.33 0.67 1.0 ]

Now, I think Press was removed as a modifier a while back and we decided to ignore release messages, but even this doesn't work as expected in the new builds...

ButtonA3 FXParam 3 "TapeType"	[ 0.0 0.33 0.67 1.0 ]

Using a Button widget with both Press and Release defined.

Widget ButtonA1
    Press b1 00 7f b1 00 00
    FB_MFT_RGB  b1 00 7f
WidgetEnd

What I'd like to have happen is each button press advances one step like in the May 5th and prior builds. And when we get to the end of the list it cycles, unless the first value is greater than the last value.

[Funkybot]

Originally Posted by cjewellstudios

I haven't confirmed the mftencoders myself but I am so happy to hear they are working!

Just a head's up: I ended up slowing down the acceleration values as a result of the new build. My old values were too fast. So when you do try it out: expect to make some adjustment to any encoder acceleration built into your .zon files.

Also, agree that it would be great if when CSI "lost an FX Focus" it just always returned to Home by default. That would pretty much solve the focus issues if that's possible.

[cjewellstudios]

Originally Posted by Funkybot

Just a head's up: I ended up slowing down the acceleration values as a result of the new build. My old values were too fast. So when you do try it out: expect to make some adjustment to any encoder acceleration built into your .zon files.

Also, agree that it would be great if when CSI "lost an FX Focus" it just always returned to Home by default. That would pretty much solve the focus issues if that's possible.

Agreed!

Thanks for the heads up

I noticed your acceleration differed from mine anyhow. We can compare after I test.

[Funkybot]

Eucon Pan Bugs revisited...

1. PanL and PanR do not write any automation at all. Doesn't matter what automation mode. The knobs move. But no automation actually gets written. To reproduce: set pan mode to Dual Pan, and try to write any automation of the Pan Left or Pan Right parameters.

2. PanL and PanR - first RotaryTouch centers the pans.

3. PanWidth - Reaper's PanWidth goes from -100% (negative stereo image) to 0 (mono) to 100% (stereo). Eucon PanWidth currently stops at mono and does not allow negative values.

...the Touch stuff with Pan and PanWidth is ok now so maybe one of my eucon.zon or one of my dll's was out of date last night.

[Funkybot]

EuCon Pan Bugs continued...

4. If you use your mouse to move PanL and PanR on Reaper, this is not reflected on the Artist Mix meters. They stay put. Now if you touch the Pan encoder on the Artist Mix, the pan knob in Reaper jumps immediately to the value on the Artist Mix display.

[Geoff Waddington]

Originally Posted by cjewellstudios

I haven't confirmed the mftencoders myself but I am so happy to hear they are working!

As far as possible bugs:

When you close a mapped fx gui, all my encoders go blank.

Resetting all midi devices OR clicking on an fx that isn't mapped goes back to the home zone.

Clicking on a mapped fx returns to that fx zone.

Not the end of the world, and it sounds like you ran up against a wall trying to figure that out, so maybe this is more a feature than a bug

(Learned that from my Brother-in-law, bugs that you can't fix become features )

Are you using FocusedFXNavigator ?

If so, there is an issue with Reaper not reporting FX losing focus, working to resolve.