mvrdevelopment / spec

DIN SPEC 15800 General Device Type Format (GDTF) and My Virtual Rig (MVR) File Format description DIN SPEC 15801
66 stars 15 forks source link

Add Protocol and Network information to Fixtures in MVR #94

Closed Bartel-C8 closed 1 year ago

Bartel-C8 commented 3 years ago

Is your feature request related to a problem? Please describe. New feature: It would be great if a MVR could contain more info about the used protocol and optional network information for used fixtures.

Describe the solution you'd like Add to a Fixture definition these (or more/less) tags:

Describe alternatives you've considered Custom user variables in MVR?

Additional context This way consoles/media-servers can automatically select/enable the right protocol (for each fixture) on import. Especially useful in the case of using Art-Net with unicast. The IP info is critical there...

Bartel-C8 commented 3 years ago

@moritzstaffel @klinzey : If I am allowed to "ping" you after the monthly meeting.

Roel Apers first posed this question in the monthly meeting of 26/04 and Moritz said it was "a good one". Happy to contribute/discuss this further.

moritzstaffel commented 3 years ago

This is part of the Proposal for Electrical and Data Components in GDTF.https://github.com/mvrdevelopment/gdtf-din-spec/pull/3 There was some presentation in the monthly meeting here. The work on MVR here is still due.

Bartel-C8 commented 3 years ago

Yes, I saw that part. But this configuration above is setup/project specific so, therefore should reside in MVR, no?

moritzstaffel commented 3 years ago

Yes, this is currently not done. But the option on how the device can connect comes from GDTF.

moritzstaffel commented 3 years ago

So when the device has two connection options, you can define two IP dresses and that part...

Bartel-C8 commented 2 years ago

@moritzstaffel : Any chance this will land in the next MVR version (soon)? Would be awesome! I saw this issue mentioned in https://github.com/mvrdevelopment/spec/pull/108 , but I see no reference fitting this issue (yet).

Thanks!

reneberhorst commented 1 year ago

https://github.com/mvrdevelopment/spec/blob/Network_protocols_proposal/mvr-spec.md addresses the suggestion. Hopefully everything is covered.

Bartel-C8 commented 1 year ago

@reneberhorst , thank you very much!

A thing that we would like to be seen added is to specify whether unicast or broadcast/multicast data is expected.

Suggestion: Add a unicast field to the (Art-Net / sACN / ...) protocols: https://github.com/mvrdevelopment/spec/blob/Network_protocols_proposal/mvr-spec.md#node-definition-protocols

unicast : true -> unicasted (both valid for e.g. Art-Net and sACN) unicast: false -> which means: broadcast for Art-Net and multicast for sACN

With this, a complete (basic) lighting setup can be fully made reproducible (import/export). E.g. a MVR could be used as a "project file" for a lighting console containing all needed/required settings.

Other observation : Table numbers are not unique anymore: Table 17 — Protocols Node Children VS Table 17 — Adresses Node Children (+typo in 'Adresses') Table 18 — Network Node Attributes VS Table 18 — Address Node Attributes (even same name)

Other question: How RDM will fit in this protocols list, as it mostly is done over DMX.

RDM over Art-Net is supported, via ArtRdm packets, so is this why it's added? Or RDMnet even? Should the RDM node explain which sub-protocol it uses to communicate?

moritzstaffel commented 1 year ago

@Bartel-C8 https://github.com/mvrdevelopment/spec/pull/162/files

We have added this via the transmission attribute. What do you thing?

Yes the RDM will be the RDM-Net, we need to change this.

Bartel-C8 commented 1 year ago

Thanks for the changes (especially the transmission field).

One thing from my previous comment (typo) that is still there: Table numbers are not unique anymore: Table 17 — Protocols Node Children VS Table 17 — Adresses Node Children (+typo in 'Adresses') Table 18 — Network Node Attributes VS Table 18 — Address Node Attributes (even same name)

Other question:

Is the description still valid with the changed type/default?:

Default Value when Optional     |   Description
NetworkInOut                    |   This is the interface name.Typically used "ethernet_x", "wireless_x", "loopback_x" (x starting at 1 and incrementing)

Other question I have now is the meaning of NetworkInOut_1. Is this the physical connector on the fixture? Where is this described, in the GDTF then?

Because, what would be the MVR for a fixture having 2 ethernet ports (internal unmanaged switch)? The fixture probably will have 1 IP address (of each type, IPv4 of IPv6), available on the 2 physical network ports.

But how should you describe then that both ports support a specific protocol? As only 0 or 1 times a protocol can be defined...

Actually more IP addresses should be no problem at all, on the same interface. This is especially the case on a IPv6 network where both a global (routed) and link-local (fe80::...) IPv6 address is available. But this is allowed, right? Or not?

        <Network geometry="NetworkInOut_1" IPv4="192.168.11.5" SubnetMask="255.255.0.0" />
        <Network geometry="NetworkInOut_1" IPv4="2.1.2.3" SubnetMask="255.0.0.0" />

Thanks

Bartel-C8 commented 1 year ago

I was AFK unfortunately for 15 mins on the call today, and just at that time I think this was discussed. I will watch back the recording. But may I ping you in the meantime about my question above? @moritzstaffel @reneberhorst

moritzstaffel commented 1 year ago

Sure any time. We also can have a call :)

Bartel-C8 commented 1 year ago

But maybe @reneberhorst can already answers some questions?

When is the DIN handover planned? Should we schedule a call this week?