significant failure of duration observation on chunked content tests on TVs #3 and #6 from December results

[jpiesing]

2 of the better TVs have significant failures of the duration observation in the chunked content test runs from December.

FAIL: Playback duration 56204.33ms does not match expected duration 30000.0ms +/- tolerance of 50ms. Detected duration is different by 26204.33ms. Allowed tolerance is 50ms and duration frame tolerance is 0. Starting missing frame number is 0. Ending missing frame number is 0.
FAIL: Playback duration 33883.42ms does not match expected duration 29960.0ms +/- tolerance of 50ms. Detected duration is different by 3923.42ms. Allowed tolerance is 50ms and duration frame tolerance is 0. Starting missing frame number is 1. Ending missing frame number is 0.
FAIL: Playback duration 51438.02ms does not match expected duration 30000.0ms +/- tolerance of 50ms. Detected duration is different by 21438.02ms. Allowed tolerance is 50ms and duration frame tolerance is 0. Starting missing frame number is 0. Ending missing frame number is 0.
FAIL: Playback duration 34946.37ms does not match expected duration 30000.0ms +/- tolerance of 50ms. Detected duration is different by 4946.37ms. Allowed tolerance is 50ms and duration frame tolerance is 0. Starting missing frame number is 0. Ending missing frame number is 0.

These look large enough that an inspection of the recordings is probably justified to confirm that the duration has been correctly calculated.

[yanj-github]

The longer duration offset was caused by multiple pauses during playback, where the status indicated 'waiting' when these pauses occurred.

[yanj-github]

This issue is also linked to the issue cta-wave/dpctf-tests#228
Once issue 228 is resolved this can be closed. No further changes required from observation framework.