search

AMWA-TV / nmos-testing

Testing tool for the AMWA NMOS Specifications
https://specs.amwa.tv/nmos-testing/
Apache License 2.0
42 stars 48 forks source link

IS-04 Node API test_24 and test_24_01 #517

Open firthmj opened 4 years ago

firthmj commented 4 years ago

Would it be possible to extend the result of this test to include the UUID of the first failing resource - it would make resolving issues a lot easier?

Also, is there a definition somewhere of what the grain_rate should be for each of resources it is being tested against - I can't find a definition of which flows count as "periodic", and what the grain_rate should be for anything other than video flows.

Also, none of the IS-04 examples for flows and sources at https://amwa-tv.github.io/nmos-discovery-registration/tags/v1.3/examples include a "grain_rate" section, again making working out what should be included more difficult.

garethsb commented 4 years ago

Would it be possible to extend the result of this test to include the UUID of the first failing resource - it would make resolving issues a lot easier?

Done - #519

I can't find a definition of which flows count as "periodic", and what the grain_rate should be for anything other than video flows.

Periodic Flows are any that have Grains regularly spaced in time. The test suite currently assumes that is all video, audio and mux Flows (see #358). Grain rate for video is specified in the schema as matching the frame rate (and that applies even for GOP-based compression scheme; one reference for that would be the NMOS Content Model - Grain definition). For audio it's less obvious because there are no boundaries in the RTP stream (my recommendation would be to just follow the video rate if for example the audio has come from the same SDI source). (For ANC (as opposed to plain) data, grain rate should probably be the same as the corresponding video rate.)

Also, none of the IS-04 examples for flows and sources

Agreed! The examples need brushing up and bringing up to date for v1.3 best practice. Miroslav made a start with https://github.com/AMWA-TV/nmos-discovery-registration/pull/130 but we need to finish that off.

firthmj commented 4 years ago

It was specifically the audio flows that I feel that need some definition somewhere of what the grain rate is, ideally with examples!

Is there somewhere on the NMOS specs pages that I should ask this? I don't think defining it really belongs in the testing tool repository.