Dolby DSS input change triggering (server/HDMI)

[Ioannis Syrogiannis]

Dealing with users that are less and less aware of the individual servers and more and more working through a TMS, I was wondering if there is any way to (remotely) trigger a change on the input of a Cat.745. (Yes, still working.)
I do understand that having no use in a playlist, a cue is not there. Yet, are you aware of any way to trigger such a change, other than the server's user interface?

[Steve Guttag]

You can change via ASCII command by setting up an "Input" command. You can also set the unit to automatically, at the conclusion of a show switch to an HDMI input. It will switch back to "Server" at the start of the next scheduled show.

[Ioannis Syrogiannis]

Right.
I thought I could do that only through serial, but it seems that's not the case.
I can create a serial command for that, and the TMS will apparently use it as any other cue.
To be honest, I always considered the serial commands as a one-way road, even when I was using GPI. But that is not the case at all.

[Ioannis Syrogiannis]

Well, I did try, but it didn't bear any fruits. At least for the TMS I am using now (RosettaBridge).
While cues that are outbound (output) do appear, because they are cues that one may add to the playlists as well, the inbound (input) ones are not picked up by the system.
It would either have to be a GPI/O combination, like the one you suggested for another feature, Steve, than I don't recall right now, or adding another ser2net device in between the TMS (or whatever controlling device) and the serial port of the server.
(Edit: ...and use that separate device to translate the ASCII text in question.)

[Ekaterina Ochagova]

If it helps in any way, here's how we did it (the picture is attached).

[Leo Enticknap]

It can be done via RS232 input commands (which, from her diagram, appears to be how Ekaterina is sending "switch to HDMI" commands to the server), as so:

dss220.png

Version 4.9.something also introduced the capability for the DSS200/220 to handle IP automation commands. I can't remember if this is just outgoing commands, or whether input commands can be sent to the server that way, too. Sorry to be so vague, but it's been ages since I've had to do any setup/configuration work on a DSS: I'm now down to two service contract customers that still use them, only one of which is with cat745s. The IP API will be on dolbycustomer.com for download, or if you're not registered, I'm sure that Dolby would send it to you on request.

[Ioannis Syrogiannis]

You are right, Leo.
I have a hard time to go to the downloads of Dolby (I can log in just fine, but I get into an endless loop with "okta", is it only me?), but I dug up the release notes for 4.9.5.2 from my archives and on the 4.9 new features the 7.2 section refers to that.
The Sample Ethernet Configuration has examples for setting the address and port of the automation system, output cues and input cues.
I wonder if one could use more than one ethernet automation devices, but it doesn't matter much.

Here are examples:

<InputCueConfiguration>
<name>hdmi1</name>
<type>other</type>
<command>hdmi1</command>
</InputCueConfiguration>

<InputCueConfiguration>
<name>hdmi2</name>
<type>other</type>
<command>hdmi2</command>
</InputCueConfiguration>

The DSS import port is set to 9202

[Steve Guttag]

When one uses the Ethernet automation added in 4.9, you are limited to ONE and only ONE device. Furthermore, your logs will choke up with all of the Ethernet cue set ups.

[Ioannis Syrogiannis]

[...] Furthermore, your logs will choke up with all of the Ethernet cue set ups.

Therefore, would you advise against, Steve?

[Steve Guttag]

No, I've done it in several locations. One should just know its shortcomings.

[Ioannis Syrogiannis]

I tried to implement the Dolby Ethernet Automation way.
It would simplify things, and I am all for having things in a booth as simple as possible, for the next guy that will want to use or set it.
I stumbled upon one or two obstacles:
1) When I tried to add escape characters to the command string, "\0A", for instance, they seem to simply not being there at the .xml file afterwards. It doesn't seem to have a choice on the "device configuration" part for such a thing either (that is terminating strings with return, line feed, etc). Nor for setting UDP or TCP, but that is less of a problem.
2) For inbound communication, I used a doremi device to send text to IP and port 9202, according to the configuration file ("hdmi1"), but wasn't successful. I don't expect that the IP given on "device configuration" is binding for incoming as well, but could it be? Or, could it be that an escape character is necessary as well?
3) The script for purging a configuration refers to the .xml file one ingests, in order to import the configuration. I didn't find anything on how to remove the configuration from the system (undo "import..."). My way to remove my custom made cues was to import an .xml with no output cues.

[Steve Guttag]

To be clear, you need to create the XML file that you will use as an ETHERNET automation. I've never had the server create it. So, looking at one of mine, where I have a DSS200 talking to an eCNA automation via Ethernet, my xml file starts like this:

<?xml version="1.0" encoding="UTF-8"?>
<EthernetAutomationConfiguration
xmlns="http://www.dolby.com/dcinema/automation/v1/schemas/ethernet">
<!-- Socket address and port -->
<DeviceConfiguration>
<address>172.16.56.32</address>
<port>13001</port>
</DeviceConfiguration>
<!-- Output Cue Configuration -->
<OutputCueConfiguration>
<name>FORMAT_2D_133</name>
<type>other</type>
<command>CMD,EVT,0,MAC,46,N,??\0D\0A</command>
</OutputCueConfiguration>
<OutputCueConfiguration>
<name>START_3D_48_SCOPE</name>
<type>other</type>
<command>CMD,EVT,0,MAC,25,N,??\0D\0A</command>


The input section looks like this:

<!-- Input Cue Configuration -->
<InputCueConfiguration>
<name>stop</name>
<type>other</type>
<command>RPT,EVT,0,SENDF2,1127\0D\0A</command>
</InputCueConfiguration>
<InputCueConfiguration>
<name>reselect</name>
<type>other</type>
<command>RPT,EVT,0,SENDF4,1129\0D\0A</command>
</InputCueConfiguration>
<InputCueConfiguration>
<name>play</name>
<type>other</type>
<command>RPT,EVT,0,SENDF3,1128\0D\0A</command>
</InputCueConfiguration>
<InputCueConfiguration>
<name>pause</name>
<type>other</type>
<command>RPT,EVT,0,SENDF1,1126\0D\0A</command>
</InputCueConfiguration>
</EthernetAutomationConfiguration>

I got the schema from the release notes and then just made my inputs and outputs accordingly.

One then FTPs the file to the NORMAL FTP directory (do not put it in the automation directory...that is where the DSS will put it when it becomes active.

Open an SSH session and move to the scripts directory. Perform a:

./listAutomationConfigurations.sh localhost

This will give you the ordinal number of the various versions...if you only have one, then there will just be the one but it is the same script that works for serial as it does for Ethernet.

Once you have the ordinal number then it is a simple ./importAutomationConfiguration.sh localhost <ordinal number> MainStore

[Ioannis Syrogiannis]

Yes, I kind of went by the book.
I took the examples from Dolby's site. (The ones that can be downloaded where the DSS software is.)
I then made changes on names, types and device info. More or less, like that:

<?xml version="1.0" encoding="UTF-8"?>
-<EthernetAutomationConfiguration xmlns="http://www.dolby.com/dcinema/automation/v1/schemas/ethernet">
<!-- Socket address and port -->
-<DeviceConfiguration>
<address>172.18.20.91</address>
<port>1702</port>
</DeviceConfiguration>
<!-- Output Cue Configuration -->
-<OutputCueConfiguration>​
<name>E-House Up</name>
<type>light</type>
<command>csvS2HouseFull 1 \0A</command>
</OutputCueConfiguration>
-<OutputCueConfiguration>
​<name>E-Mute Mic1</name>
<type>light</type>
<command>csvS2MuteMic 1 \0A</command>
</OutputCueConfiguration>​
-<OutputCueConfiguration>
<name>E-UnMute Mic1</name>
<type>light</type>
<command>csvS2MuteMic 0 \0A</command>
</OutputCueConfiguration>​

For some reason, first time I imported (FTP and then running the script, not FTPing intothe automation... folder) I couldn't find the "\0A" escape characters on the (created in the automationConfig folder) file.
I did it again.
Yet, before I do it, I tried to use the purgeAutomationConfigurations.sh script.
It removed from the list the .xml files I FTPed to DSS, so nothing would show if I ran "listAutomationConfigurations.sh localhost" but not the configuration.
Even though those strings run well from different systems on the same cinema (Doremi/Alchemy), I wasn't lucky running them out of the Dolby.

On the input cues, it didn't cross my mind that I had to modify the ready example, so I left it as it was...

...
-<InputCueConfiguration>
<name>server</name>
<type>other</type>
<command>server</command>
</InputCueConfiguration>
-<InputCueConfiguration>
<name>hdmi1</name>
<type>other</type>
<command>hdmi1</command>
</InputCueConfiguration>
-<InputCueConfiguration>
...​

No matter how I tried to send that "hdmi1", using echo <string> | nc <IP> <port>, or having the doremi send to the server IP and 9202, or adding escape characters, or space and escape characters etc. etc. nothing seemed to move the server.

I haven't thought about what you show on your example, changing the command string on the configuration, that is. I thought that, given that they are inputs, they would probably be fine as they are. I will.
I will admit though, that it smells like I am blind to what I stare in the eyes.

[Ekaterina Ochagova]

The example from Dolby Release Notes works on Dolby DSS systems with the JNIOR automation controller (like it is in our case). I have never dealt with the doremi device and the nc utility you mentioned, so I really don't know how I can help you with your problems. I can just try to suggest that first you need to set your machine/device/utility (172.18.20.91) to server mode to listen to the given port (1702), then import the .xml into the DSS and wait for the connection to be established, after that send "hdmi1\r\n" string from 172.18.20.91.

[Ioannis Syrogiannis]

Thank you Ekaterina.
The server device in question, a Q-Sys core, already listens to that port and works well with other screen servers. It worth the while to check different escape characters, though.
I will definitely experiment on the input cues with different strings, but I hope that DSS will be able to "listen" to more than one clients/IPs via port 9202.
Once I get somewhere, I will make sure to drop a line here, in case someone benefits from it in the future.
(I don't know how Dolby DSS and future may come together, but none the less...)