timothy/ersatztv
1
0
Fork 0
Code Issues 58 Pull Requests 8 Actions Packages Projects Releases 9 Wiki Activity
3,386 Commits 70 Branches 158 Tags
f00dfde0c574b9dd6797d25dbcd706d89aecdf4e
Commit Graph
9 Commits
Author SHA1 Message Date
timothy 1359bb6135 fix(350): burst-read the first HLS segments so cold start isn't readrate-bound
Build ErsatzTV Image / CI image pin matches docker/ci (pull_request) Successful in 11s
Details
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 16s
Details
Build ErsatzTV Image / decisions.md append-only (pull_request) Successful in 15s
Details
Build ErsatzTV Image / Formatting (changed .cs conform to .editorconfig) (pull_request) Successful in 1m19s
Details
Build ErsatzTV Image / API docs in sync (OpenAPI + endpoint index) (pull_request) Successful in 1m20s
Details
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 5m52s
Details
Build ErsatzTV Image / Functional E2E (curl contracts) (pull_request) Successful in 15m14s
Details
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 19m12s
Details
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Details 
`-readrate 1.05` paces input reading at wall clock so a channel behaves like
live TV, but it applies from the very first read. With 4s HLS segments and the
segmenter waiting for the first one, the playlist could not appear sooner than
~4/1.05 = 3.8s, so every tune-in that did not win a work-ahead slot paid a
multi-second wait.

Add `-readrate_initial_burst` (FFmpeg >= 6.1) next to `-readrate` on the normal
playback path, gated on runtime capability detection via the existing
`FFmpegKnownOption`/`HasOption` machinery, whose option list had simply been
empty. Measured on real prod media: time-to-first-playlist 5369/5344ms ->
648/649ms.

Root cause detail: the cold-start bimodality earlier rounds could not explain
was never about the media. `HlsSessionWorker` grants an unthrottled start only
while `_workAheadCount < work_ahead_limit` (prod: 1), so concurrent tune-ins
fall back to the throttled path. Confirmed on prod with three concurrent tunes:
firstGop 866ms for the slot winner vs 3845ms and 6357ms. This also falsifies the
issue's ranked #1 driver — accurate-seek decode-discard measures 30-100ms on
real media, and probe caps 20-50ms; neither can account for seconds.

Still images are excluded: their video input is paced by the realtime filter and
takes no readrate, so a burst would only run a song's separate audio input ahead
of the video. Concat/WrapSegmenter keep the unburst single-arg constructor.

fixes #350
 2026-07-20 23:59:42 +02:00
Jason Dove 9bae8e73bf slightly increase throttled readrate in segmenter (#2534) 2025-10-14 14:53:56 -05:00
Jason Dove 65ff1f5502 improve live stream reliability (#2524) 2025-10-13 15:41:48 -05:00
Jason Dove 036b6e63c7 add new graphics engine (#2265) 
* spike new graphics engine

* fix remote watermarks; add graphics engine to vaapi

* add graphics engine to qsv
 2025-08-06 15:04:43 +00:00
Jason Dove 9587692486 optimize image playback (#1609) 2024-02-11 12:54:18 -06:00
Jason Dove 9471cb55dd upgrade from dotnet 7 to dotnet 8 (#1529) 
* upgrade sdk

* fix warnings in ersatztv.ffmpeg

* fix warnings in ersatztv.core

* fix warnings in ersatztv.infrastructure

* fix warnings in ersatztv.application

* disable analysis for migrations projects

* fix warnings in ersatztv.scanner

* fix warnings in ersatztv

* upgrade project framework

* update github actions and dockerfiles
 2023-12-30 13:29:57 -06:00
Jason Dove 14a88bd225 optimize ffmpeg capability cache (#1486) 
* minimize cached ffmpeg capabilities

* use set intersect

* try disabling work ahead on nvidia/windows
 2023-10-16 08:42:26 -05:00
Jason Dove 0550c60a78 allow older ffmpeg for testing (#1485) 
* allow older ffmpeg for testing

* use proper option name
 2023-10-14 21:13:18 -05:00
Jason Dove b16215fcd6 improve hls throttle (#1434) 
* throttle using ffmpeg option

* update ffmpeg version
 2023-09-14 19:28:15 -05:00