Go Back   Cockos Incorporated Forums > REAPER Forums > MIDI Hardware, Control Surfaces, and OSC

Reply
 
Thread Tools Display Modes
Old 06-06-2018, 02:22 PM   #1241
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Freex View Post
I wonder how well it plays with it's brothers?
I suppose that the goal of Behringer is to have a collection of X-Touches work in MC mode work with (some) DAWs out of the box. No idea if/ow this can be accomplished.

I feel that it's very likely that by means of CSCERF any collection of X-Touch devices can be made work in an excellent user definable way with Reaper.

Geoff: what do you think about using the rather tiny (seemingly graphic enabled, or two lines à 6 characters) scribble strip of the "One" with CSURF ?

-Michael
mschnell is offline   Reply With Quote
Old 06-06-2018, 02:33 PM   #1242
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
I suppose that the goal of Behringer is to have a collection of X-Touches work in MC mode work with (some) DAWs out of the box. No idea if/ow this can be accomplished.

I feel that it's very likely that by means of CSCERF any collection of X-Touch devices can be made work in an excellent user definable way with Reaper.

Geoff: what do you think about using the rather tiny (seemingly graphic enabled, or two lines à 6 characters) scribble strip of the "One" with CSURF ?

-Michael
Don't know for sure, but if it's the typical Mackie style using sysex, no problem, we already do that.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-07-2018, 01:34 AM   #1243
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

FYI update

Not so much of a design change as a recognition of the way things are.

If you name your Widgets like this in your .rst file
Shift ---- ButtonWithRelease 90 46 00 7f
Option -- ButtonWithRelease 90 47 00 7f
Control - ButtonWithRelease 90 48 00 7f
Alt ------ ButtonWithRelease 90 49 00 7f

do not make an entry for these in your .axt files.
...
F8 UnpinSelectedTracks
Shift Shift
Option Option
Control Control
Alt Alt

Read TrackAutoMode 1
Write TrackAutoMode 3
...

If you want to use these Surface Widgets for something else, simply give them a different name:

OptionButton -- ButtonWithRelease 90 47 00 7f
AltButton ------ ButtonWithRelease 90 49 00 7f
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-07-2018, 05:40 AM   #1244
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Don't know for sure, but if it's the typical Mackie style using sysex, no problem, we already do that.
Picturing an affordable yet versatile setup using an X-Touch Compact and an X-Touch One supposedly would request some modifications to the typical MC handling.

E.g. there would be only a single scribble 2*6 strip.

Maybe it could be viable to have same show the track name when any of the faders is touched, ignoring that it's above the one fader of the "One".

Supposedly many other specific goodies might be worth considering.

-Michael
mschnell is offline   Reply With Quote
Old 06-08-2018, 02:03 AM   #1245
inertia
Human being with feelings
 
Join Date: Oct 2013
Posts: 800
Default

Quote:
Originally Posted by Geoff Waddington View Post
With the new build, you must also upgrade all of the .rst files in the CSI folder.
I completely reinstalled the build as well as replacing the CSI folder but it still crashes.
inertia is offline   Reply With Quote
Old 06-08-2018, 06:51 AM   #1246
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by inertia View Post
I completely reinstalled the build as well as replacing the CSI folder but it still crashes.
Hmmm...

Was it working before ?

What OS are you on ?

If Mac can you provide a crash dump ?
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-08-2018, 04:39 PM   #1247
MixMonkey
Human being with feelings
 
MixMonkey's Avatar
 
Join Date: Sep 2017
Location: London, England.
Posts: 4,870
Default

New build working here on macOS 10.3.5. Changed the MCU .axt file so that banking was 8 instead of 20 and reversed + and - so that the right bank button on the MCU makes the bank shift right instead of left.

Pinned the Master fader in Reaper to the MCU Master fader with F7 on the MCU. Master now ignored by banking.

Color working, made one green page and it banks across TCP and MCP, identifying which screen faders are currently being controlled. Not sure how this coloring idea is going to interact with all the other ways to color tracks though.
MixMonkey is online now   Reply With Quote
Old 06-08-2018, 05:00 PM   #1248
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by MixMonkey View Post
New build working here on macOS 10.3.5. Changed the MCU .axt file so that banking was 8 instead of 20 and reversed + and - so that the right bank button on the MCU makes the bank shift right instead of left.

Pinned the Master fader in Reaper to the MCU Master fader with F7 on the MCU. Master now ignored by banking.

Color working, made one green page and it banks across TCP and MCP, identifying which screen faders are currently being controlled. Not sure how this coloring idea is going to interact with all the other ways to color tracks though.
Thanks for testing !

Wow 10.3.5, do you really run Panther or is that a typo ?

As far as colour, yeah, it will not play nice with other colouring schemes right now, it was more a of a quick and dirty proof of concept navigation aid for those who don't have scribble strips -- e.g. X Touch Compact users.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-08-2018, 05:15 PM   #1249
MixMonkey
Human being with feelings
 
MixMonkey's Avatar
 
Join Date: Sep 2017
Location: London, England.
Posts: 4,870
Default

:-) typo 10.13.5 oops
MixMonkey is online now   Reply With Quote
Old 06-08-2018, 05:59 PM   #1250
MixMonkey
Human being with feelings
 
MixMonkey's Avatar
 
Join Date: Sep 2017
Location: London, England.
Posts: 4,870
Default

I just added an MCU XT by copying the MCU .rst and .axt files (found I had to add an MCU_XT folder in the axt folder) I just deleted everything apart from the 'channel' part of the MCU files and renamed them MCU_XT.

I changed the banking to 16 in the MCU.axt file and was good to go :-) I think I was expecting the MCU XT commands to be different from the MCU, but they're identical. The units are differentiated by their different MIDI connections. Is there anyway I can make the MCU XT the first 8 channels rather that the second 8? Not a big deal at this stage (I was thinking about swapping them around anyway)

I think it might be a good idea to have some 'end stops' so that you can't bank past the first or last tracks in a session. It can become very confusing, although the colors mitigate this to a degree. When you reach the first or last 'group' of 8 faders (or 16,20) you shouldn't be able to bank any further.

Also, since most of these surfaces have an MCU emulation mode, you might want to change the MCU.rst file in the download so that the bank is 8 and the direction is reversed. It will be a lot less confusing for people trying to get a foothold as most will have an 8 fader surface and expect BankRight to be +ve (same for ChannelRight). Or is that just me?
MixMonkey is online now   Reply With Quote
Old 06-08-2018, 06:44 PM   #1251
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by MixMonkey View Post
I just added an MCU XT by copying the MCU .rst and .axt files (found I had to add an MCU_XT folder in the axt folder) I just deleted everything apart from the 'channel' part of the MCU files and renamed them MCU_XT.

I changed the banking to 16 in the MCU.axt file and was good to go :-) I think I was expecting the MCU XT commands to be different from the MCU, but they're identical. The units are differentiated by their different MIDI connections. Is there anyway I can make the MCU XT the first 8 channels rather that the second 8? Not a big deal at this stage (I was thinking about swapping them around anyway)

I think it might be a good idea to have some 'end stops' so that you can't bank past the first or last tracks in a session. It can become very confusing, although the colors mitigate this to a degree. When you reach the first or last 'group' of 8 faders (or 16,20) you shouldn't be able to bank any further.

Also, since most of these surfaces have an MCU emulation mode, you might want to change the MCU.rst file in the download so that the bank is 8 and the direction is reversed. It will be a lot less confusing for people trying to get a foothold as most will have an 8 fader surface and expect BankRight to be +ve (same for ChannelRight). Or is that just me?
Great work on adding an extender folder, that’s exactly how it’s done.

As far as end stops, sounds like you have too many Virtual Sufaces defined, just remove any unnecessary ones .

To change the order, just add the Virtual Surfaces in the order you want them to appear on the page.

Good points on the defaults.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-09-2018, 04:29 AM   #1252
MixMonkey
Human being with feelings
 
MixMonkey's Avatar
 
Join Date: Sep 2017
Location: London, England.
Posts: 4,870
Default

Thanks Geoff- great work btw, this is really starting to take shape!

Any chance of some scribble strip development, I'm itching to see some track names on those LCDs :-)

Regarding the colors, it might be better to lighten the controlled bank, rather than color them- can you get at the blend mode at all? The way it is is fine for the moment though- easy to identify the controlled tracks.
MixMonkey is online now   Reply With Quote
Old 06-09-2018, 04:57 AM   #1253
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by MixMonkey View Post
Thanks Geoff- great work btw, this is really starting to take shape!

Any chance of some scribble strip development, I'm itching to see some track names on those LCDs :-)

Regarding the colors, it might be better to lighten the controlled bank, rather than color them- can you get at the blend mode at all? The way it is is fine for the moment though- easy to identify the controlled tracks.
What ? No scribble strips ?

They are working fine here with Avid Artists in Mackie mode.

Do you have Mackie MCU and Extender ?

As far as colours, early days, but you can set them by editing Page in CSI config.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-09-2018, 04:58 AM   #1254
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Change these 2 lines in MCU.rst

Stop Button 90 5d 00 7f
Play Button 90 5e 00 7f

to

Stop Button 9d 35 00 7f
Play Button 9d 36 00 7f


Voila, Stop and Play should work, if the data you previously posted for native mode doesn't contain a typo.

Try it and let me know how it turned out.
Woks partly

I first copied the appropriate "MCU" files and directory to be called "XTouchCompactNative", edited the XTouchCompactNative.rst file, and selected those in the config.

Then switched the surface to native mode and restarted Reaper.

Now I can push the "PLAY" and "STOP" buttons to start and stop playback in Reaper.

Moreover, when pushing the Reaper GUI buttons, the lights in the surface buttons react correctly.

But when using surface buttons, their lights always are put out when releasing them. This is because the XTouch always stops the button lights on release. So it is necessary for CSURF to send the "light on" command (identical to "button pressed" e.g. 9d 35 7f) after the "button released" (e.g. 9d 35 00) was received.

As I don't decently understand how the 3rd and 4th hex in "Play Button 9d 35 00 7f" is to be interpreted, I can't seem to improve this on my own.

BTW 1: When pushing "START" on the surface while playback is running, playback stops, but when pushing "START" on the Reaper GUI while playback is running, playback restarts at the cursor. Is that different behavior intended ?

BTW 2: There had been a discussion on using HEX for the Midi bytes in the configuration files. Now as I see the Midi loggin provided by CSURF, IMHO using hex is absolutely appropriate.

Thanks for working on CSURF !

-Michael
mschnell is offline   Reply With Quote
Old 06-09-2018, 05:49 AM   #1255
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
Woks partly

I first copied the appropriate "MCU" files and directory to be called "XTouchCompactNative", edited the XTouchCompactNative.rst file, and selected those in the config.

Then switched the surface to native mode and restarted Reaper.

Now I can push the "PLAY" and "STOP" buttons to start and stop playback in Reaper.

Moreover, when pushing the Reaper GUI buttons, the lights in the surface buttons react correctly.
Excellent !

Quote:
Originally Posted by mschnell View Post
But when using surface buttons, their lights always are put out when releasing them. This is because the XTouch always stops the button lights on release. So it is necessary for CSURF to send the "light on" command (identical to "button pressed" e.g. 9d 35 7f) after the "button released" (e.g. 9d 35 00) was received.

As I don't decently understand how the 3rd and 4th hex in "Play Button 9d 35 00 7f" is to be interpreted, I can't seem to improve this on my own.
Uggghhhh, one of my favourite control surface peeves.

IMHO the buttons have no business telling the lights what to do on any control surface -- the XTouch Compact is not alone, some other surfaces also have this behaviour !!

To me this is a design flaw, why bother having a feedback mechanism if the control surface short circuits and overrides this behaviour ?

I'll have to see if I can find a workaround.

Once again, Uggghhh.

I've never heard anywhere close to a convincing argument why this should be so.

Quote:
Originally Posted by mschnell View Post
BTW 1: When pushing "START" on the surface while playback is running, playback stops, but when pushing "START" on the Reaper GUI while playback is running, playback restarts at the cursor. Is that different behavior intended ?

Thanks for working on CSURF !

-Michael

Hmm... just tested that here and got the same results.

Coulda' swore Reaper used to do the same as the control surface now does, did Reaper change ?
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-09-2018, 09:11 AM   #1256
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
To me this is a design flaw, why bother having a feedback mechanism if the control surface short circuits and overrides this behavior ?

I'll have to see if I can find a workaround.

Once again, Uggghhh.

I've never heard anywhere close to a convincing argument why this should be so.
I totally agree. (The "silent" layer switching IMHO is a design glitch as well !! )

The behavior of any button can be configured to be "momentary" or "toggle". With "toggle" it will only send a message on release, and the light is set according to the then set state. I suppose this does not help a bit regarding CSURF.

I decently hope that you find a way to handle this. (Rather obviously CSURF should send the desired LED state any time it sees a button release message. )

As you seem to be determined to do so, I'm going to test faders and Rotaries tonight...

-Michael

Last edited by mschnell; 06-09-2018 at 09:33 AM.
mschnell is offline   Reply With Quote
Old 06-09-2018, 09:25 AM   #1257
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Coulda' swore Reaper used to do the same as the control surface now does, did Reaper change ?
In fact, IMHO, the action performed by CSURF (set cursor to current location and stop) is exactly what I would desire it to do !!!!

You could happily press "STOP" and then "START" to get the behavior Reaper provides when pressing "start" when running,

(Supposedly Reaper can be configured to show this behavior, as well.)

-Michael

Last edited by mschnell; 06-09-2018 at 09:31 AM.
mschnell is offline   Reply With Quote
Old 06-09-2018, 10:40 AM   #1258
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

BTW.

Now that X-Touch uses standard Midi messages and not the weired MC style messages, the Midi Channel is important.

Hence this should be considered in the rst file format.

To stay fully compatible and provide even more flexibility I suggest to allow for a command such as

hex x a

From that line on each character "x" (or maybe also "X") when appearing in one of the hex pairs, is replaced by a character "a".

By this you can easily handle switching the Midi Channel, and moreover you can do certain other nice tricks / optimizations with the hex notation.


Is there a way to do comments ? E.g. the C style line comment ("//") would be very welcome !




-Michael
mschnell is offline   Reply With Quote
Old 06-09-2018, 11:00 AM   #1259
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

The channel faders and the channel rotaries (assigned to balance) do work bidirectionally !!!!

Some weirdness:

To have the faders rotaries 1, 2, 3, ... assigned, I need to write the Midi message of a hypothetical fader/rotary one left of the left most one that really exists. Is this intended like that ? (What if the sequence of Midi messages starts with 0 ??? Happily with the X-Touch the faders start with 01 and the rotaries start with 0a, moreover this would be configurable.

When moving the fader on the surface the fader in the GUI reacts rather sluggishly . This is not the case in MC mode. I suppose this is due to the XTouch's behavior.

-Michael
mschnell is offline   Reply With Quote
Old 06-09-2018, 11:59 AM   #1260
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
BTW.

Now that X-Touch uses standard Midi messages and not the weired MC style messages, the Midi Channel is important.

Hence this should be considered in the rst file format.

To stay fully compatible and provide even more flexibility I suggest to allow for a command such as

hex x a

From that line on each character "x" (or maybe also "X") when appearing in one of the hex pairs, is replaced by a character "a".

By this you can easily handle switching the Midi Channel, and moreover you can do certain other nice tricks / optimizations with the hex notation.


Is there a way to do comments ? E.g. the C style line comment ("//") would be very welcome !




-Michael
Yup, you can even use / instead of //

CSI.ini itself can also be commented, but the comments will be blasted on the next write by CSI config panel in Reaper

As far as midi channels, I think you can do this already:

StopChannel9 Button 99 35 00 7f
StopChannelA Button 9a 35 00 7f
StopChannelB Button 9b 35 00 7f
StopChannelC Button 9c 35 00 7f

Then just use them normally in an .axt file like:

StopChannel9 SomeReaperAction
StopChannelA SomeOtherReaperAction
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-09-2018, 12:02 PM   #1261
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
The channel faders and the channel rotaries (assigned to balance) do work bidirectionally !!!!
Cool !!

Quote:
Originally Posted by mschnell View Post
Some weirdness:

To have the faders rotaries 1, 2, 3, ... assigned, I need to write the Midi message of a hypothetical fader/rotary one left of the left most one that really exists. Is this intended like that ? (What if the sequence of Midi messages starts with 0 ??? Happily with the X-Touch the faders start with 01 and the rotaries start with 0a, moreover this would be configurable.

When moving the fader on the surface the fader in the GUI reacts rather sluggishly . This is not the case in MC mode. I suppose this is due to the XTouch's behavior.

-Michael
Yes, I suspect as you do, this is all about the hardware/firmware in the surface itself.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-09-2018, 12:18 PM   #1262
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
As far as midi channels, I think you can do this already:
The idea was to switch (e.g.) the midi channel with a single line for the complete or part of the rst file.

-Michael
mschnell is offline   Reply With Quote
Old 06-09-2018, 12:21 PM   #1263
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Yes, I suspect as you do, this is all about the hardware/firmware in the surface itself.
That would be a mess, as I feel it's close to unusable (while the MC mode worked perfectly).

I'll do a test by having a jsfx create a CC ramp and route it back via loopmidi, hoping that CSURF is the culprit (when smooting the 7 bit faders) , as same might be updated accordingly .

-Michael
mschnell is offline   Reply With Quote
Old 06-10-2018, 02:15 AM   #1264
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Unfortunately I am not able to configure CSURF to use the Loop Midi device as an input and show the loop midi device output in the log. (I can receive the output of the loop midi device in a Reaper track just fine.)

When I do so I get in CSI.ini:

RealSurface XTouchCompactNative 9 1 4 4 XTouchCompactNative.rst

while with X-Touch as an input I have

RealSurface XTouchCompactNative 9 1 3 4 XTouchCompactNative.rst, which usually works (now, but only after restarting Reaper after changing the input device).

the "edit CSI.ini" trick does not seem to work, either.

Any help ? Otherwise I don't know how to test this issue.

BTW. Maybe this is yet another X-Touch design glitch: the device does feature a 5-pin Midi out socket, and maybe the USB output is limited to the speed of the serial MIDI. . Seemingly it's not possible to deactivate the serial Midi out. I don't know why in MC mode this does not seem to do any harm.

Edit:
I recorded the X-Touch output in a track and found that moving the fader as fast as I can results in a sequence of Midi events with a duration of 200 mSecs. I think this might be too long. But in MC mode I see a similar behavior.

-Michael

Last edited by mschnell; 06-10-2018 at 03:13 AM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 03:28 AM   #1265
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
Unfortunately I am not able to configure CSURF to use the Loop Midi device as an input and show the loop midi device output in the log. (I can receive the output of the loop midi device in a Reaper track just fine.)

When I do so I get in CSI.ini:

RealSurface XTouchCompactNative 9 1 4 4 XTouchCompactNative.rst

while with X-Touch as an input I have

RealSurface XTouchCompactNative 9 1 3 4 XTouchCompactNative.rst, which usually works (now, but only after restarting Reaper after changing the input device).

the "edit CSI.ini" trick does not seem to work, either.

Any help ? Otherwise I don't know how to test this issue.

BTW. Maybe this is yet another X-Touch design glitch: the device does feature a 5-pin Midi out socket, and maybe the USB output is limited to the speed of the serial MIDI. . Seemingly it's not possible to deactivate the serial Midi out. I don't know why in MC mode this does not seem to do any harm.

Edit:
I recorded the X-Touch output in a track and found that moving the fader as fast as I can results in a sequence of Midi events with a duration of 200 mSecs. I think this might be too long. But in MC mode I see a similar behavior.

-Michael
Could you please post the contents of your .rst file, just the section for faders.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-10-2018, 03:49 AM   #1266
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Could you please post the contents of your .rst file, just the section for faders.
Code:
Channel
ChannelFaderTouch ButtonWithRelease 90 68 00 7f
ChannelRotaryPush Button 90 20 00 7f
ChannelRotary Fader7Bit bd 09 00 7f
ChannelDisplayUpper DisplayUpper
ChannelDisplayLower DisplayLower
ChannelFader Fader7Bit bd 00 00 7f
ChannelRecordArm Button 90 00 00 7f
ChannelSolo Button 90 08 00 7f
ChannelMute Button 90 10 00 7f
ChannelSelect Button 90 18 00 7f
ChannelEnd
Up till now I only edited the "ChannelRotary" and the "ChannelFader" lines.

-Michael

Last edited by mschnell; 06-10-2018 at 04:20 AM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 03:52 AM   #1267
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Unfortunately after experimenting with the Loop Midi device, I can't get any messages from the surface in the log display any more (and of course no reaction either.) Sending messages to the X-Touch midi device still works fine.

Part of the old Midi-device assignment problem seems to be back....

Edit:

disabling the XTouch input device in Reaper's Midi Device configuration and restarting Reaper got got working again.

-Michael

Last edited by mschnell; 06-10-2018 at 04:20 AM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 04:31 AM   #1268
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Of course it was easy to get the channel "MUTE", "SOLO", "ARM", and "SELECT" buttons working, but of course the lights behave erroneously here,m as well.

-Michael
mschnell is offline   Reply With Quote
Old 06-10-2018, 04:38 AM   #1269
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Trying to assign the "ChannelRotaryPush", I found that with XTOUCH this is 9d 00 7f for the left most Rotary of the XTOUCH. As I need to set the second hex one lower than the value sent by the left mist channel control. What am I supposed to write here ?

ChannelRotaryPush does not have an effect right now, so I can't check trying e.g. 1f.

-Michael
mschnell is offline   Reply With Quote
Old 06-10-2018, 04:46 AM   #1270
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

What are we going to do with the eight non-channel rotaries, the X-Touch provides ?

The rst file just provides the jog-wheel for this.

in MC-Mode seemingly the X-Touch sends a series of button presses when turning the Rotaries 15 and 16 for "BANK" and "CHANNEL".

This does seem rather appropriate, but in native mode we will just see rotaries and CSURF will need to be able to handle these for "BANK" and "CHANNEL" left/right moves.

-Michael
mschnell is offline   Reply With Quote
Old 06-10-2018, 05:20 AM   #1271
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
Code:
Channel
ChannelFaderTouch ButtonWithRelease 90 68 00 7f
ChannelRotaryPush Button 90 20 00 7f
ChannelRotary Fader7Bit bd 09 00 7f
ChannelDisplayUpper DisplayUpper
ChannelDisplayLower DisplayLower
ChannelFader Fader7Bit bd 00 00 7f
ChannelRecordArm Button 90 00 00 7f
ChannelSolo Button 90 08 00 7f
ChannelMute Button 90 10 00 7f
ChannelSelect Button 90 18 00 7f
ChannelEnd
Up till now I only edited the "ChannelRotary" and the "ChannelFader" lines.

-Michael
At a quick glance looks good from here.

Seems like native mode is fighting you all the way
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-10-2018, 05:22 AM   #1272
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Seems like native mode is fighting you all the way
Obviously.

But I suppose CSI is intended to be able to support any kind of surface.

So I thing this is a viable effort.

-Michael

Last edited by mschnell; 06-10-2018 at 09:46 PM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 05:55 AM   #1273
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
But I suppose CSURF is intended to be able to support any kind of surface.
Well... yes, but if the hardware behaves well in one mode and poorly in another, it is not likely we can do much about that.

That doesn't mean we won't try, but just want to set expectations reasonably.


Quote:
Originally Posted by mschnell View Post
So I thing this is a viable effort.

-Michael
I'll help any way I can, just let me know if you find where things are slowing down.

Maybe there's a boot up mode that doesn't restrict the baud rate in native -- if indeed that is even the problem.

Good luck !
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-10-2018, 02:31 PM   #1274
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

I'd need to find another Midi via USB source creating a CC ramp and try to make CSI accept that device as a Midi input.

I'll keep up researching.

Do you intend to implement the "workaround" for the button lights (e.g. by sending out the appropriate state message with any button release - this is what my JSFX software for the X-Touch does) ?

(If not already provided) do you intend to support using a Rotary as a a sequence of button pushes (e.g. for "BANK" and "CHANEL" up/down, for devices with few buttons but many rotaries ?

-Michael

Last edited by mschnell; 06-10-2018 at 09:45 PM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 02:49 PM   #1275
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Later:

I got CSI accept the Loop Midi device, simply by disabling it as a Reaper Midi input device. Seems like this is the setting that should be used for any Midi device intended for CSURF !!!! (-> docs !!)

So I could send it a perfect Ramp. CSI works like charm !!!!

After a short while, I did find the really simple and unproblematic reason for the sluggishness: if I disable the CSURF Midi Log, all is fine !

Thanks for listening !
-Michael

Last edited by mschnell; 06-10-2018 at 09:45 PM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 04:38 PM   #1276
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
Do you intend to implement the "workaround" for the button lights (e.g. by sending out the appropriate state message with any button release - this is what my JSFX software for the X-Touch does) ?
Yes.

Quote:
Originally Posted by mschnell View Post
(If not already provided) do you intend to support using a Rotary as a a sequence of button pushes (e.g. for "BANK" and "CHANEL" up/down, for devices with few buttons but many rotaries ?

-Michael
Not quite sure, will have to think about that a bit.
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-10-2018, 04:42 PM   #1277
Geoff Waddington
Human being with feelings
 
Geoff Waddington's Avatar
 
Join Date: Mar 2009
Location: Dartmouth, Nova Scotia
Posts: 11,184
Default

Quote:
Originally Posted by mschnell View Post
Seems like this is the setting that should be used for any Midi device intended for CSURF !!!! (-> docs !!)
Except if you have a midi keyboard with integrated transport buttons, etc.

Quote:
Originally Posted by mschnell View Post
So I could send it a perfect Ramp. CSURF works like charm !!!!

After a short while, I did find the really simple and unproblematic reason for the sluggishness: if I disable the CSURF Midi Log, all is fine !

Thanks for listening !
-Michael
Great, makes sense that there should be no penalty for native, and I presume you mean CSI not CSURF
__________________
To install you need the CSI Software and Support Files
For installation instructions and documentation see the Wiki
Donate -- via PayPal to waddingtongeoff@gmail.com
Geoff Waddington is offline   Reply With Quote
Old 06-10-2018, 09:41 PM   #1278
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Except if you have a midi keyboard with integrated transport buttons, etc.
In such case it would actually be desirable to allow for the device being used as input as well for CSI as for tracks.

But in my latest test, this does not seem to be possible in a decent way. Multiple times I did not get messages seen by CSI until I deactivates the device in the Reaper Mid Device setup, and it did work reliably after that and restarting Reaper. (And seemingly Reaper automatically enables input from a newly found device in certain instances, making CSI non workable.) I don't think this is just me...

Quote:
Originally Posted by Geoff Waddington View Post
Great, makes sense that there should be no penalty for native, and I presume you mean CSI not CSURF
In a released version of CSI, the docs should state, that the log is very helpful when editing the rst files, but should be switched off in production mode, as it will impose severe latency when many messages are to be handled within a short period of time.

Sorry for me not correctly distinguishing between CSURF and CSI. The meaning of CSI of course is obvious, while CSRUF just lingers in the back of my mind .

-Michael

Last edited by mschnell; 06-10-2018 at 11:31 PM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 09:51 PM   #1279
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by Geoff Waddington View Post
Not quite sure, will have to think about that a bit.
It does seem to make sense to me.

I think it could be done by always sending a value of "64" to the rotary and when getting a lower / higher value from same, simulate a push of the [UP] / [DOWN] button and re-send "64".

-Michael

Last edited by mschnell; 06-10-2018 at 11:29 PM.
mschnell is offline   Reply With Quote
Old 06-10-2018, 11:34 PM   #1280
mschnell
Human being with feelings
 
mschnell's Avatar
 
Join Date: Jun 2013
Location: Krefeld, Germany
Posts: 14,687
Default

Quote:
Originally Posted by mschnell View Post
Trying to assign the "ChannelRotaryPush", I found that with XTOUCH this is 9d 00 7f for the left most Rotary of the XTOUCH. As I need to set the second hex one lower than the value sent by the left mist channel control. What am I supposed to write here ?

ChannelRotaryPush does not have an effect right now, so I can't check trying e.g. 1f.
Could you tell me how I should set the hex values for a channel control that start with number Zero for the left most channel ?

How can I see "ChannelRotaryPush" in action ?

BTW.: are there no surfaces that feature more than just a single Rotary per channel ?

Thanks,
-Michael
mschnell is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 01:48 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.