https://bugs.kde.org/show_bug.cgi?id=510445
--- Comment #6 from [email protected] --- I generated melt scripts (instead of rendering directly from the dialog) with vs. without TC overlay. They differ in the included target filename (intended); and the version WITH TC "2025*-overlayTC-1.mlt" has only one additional section almost at the end of file: <filter id="filter92"> <property name="argument">#timecode#</property> <property name="geometry">0%/0%:100%x100%:100%</property> <property name="family">Sans</property> <property name="size">48</property> <property name="weight">400</property> <property name="style">normal</property> <property name="fgcolour">#ffffff</property> <property name="bgcolour">#bb333333</property> <property name="olcolour">0x00000000</property> <property name="pad">0</property> <property name="halign">left</property> <property name="valign">top</property> <property name="outline">0</property> <property name="opacity">1.0</property> <property name="mlt_service">dynamictext</property> </filter> To me, this does NOT explain why the version without TC should not produce video output :-( But when I run the melt which was supplied with the kdenlive AppImage with 2025*-overlayNone-1.mlt produces a result file WITHOUT video; with 2025*-overlayTC-1.mlt produces a result file WITH video. Neither of these runs produces any error messages. Moreover, the normal log output is very similar, apart from some hex address values - and the output from running WITH the TC overlay has (roughly) these additional lines, almost at the end of the complete log: [in @ 0x7f16d4c1c480] filter context - w: 1920 h: 1080 fmt: 26 csp: bt709 range: tv, incoming frame - w: 1920 h: 1080 fmt: 26 csp: gbr range: tv pts_time: 4230.44 [fftdnoiz @ 0x7f16d6f0a400] Value 0.000000 for parameter 'amount' out of range [0.01 - 1] [in @ 0x7f16d6f09d80] Changing video frame properties on the fly is not supported by all filters. [in @ 0x7f16d6f09d80] filter context - w: 1920 h: 1080 fmt: 26 csp: bt709 range: tv, incoming frame - w: 1920 h: 1080 fmt: 26 csp: gbr range: tv pts_time: 4230.44 Current Position: 0 Fontconfig error: Cannot load default config file: No such file: (null) Current Position: 0 ... Current Position: 105884 [in @ 0x7f15b8034200] Changing video frame properties on the fly is not supported by all filters. [in @ 0x7f15b8034200] filter context - w: 1920 h: 1080 fmt: 26 csp: bt709 range: tv, incoming frame - w: 1920 h: 1080 fmt: 26 csp: gbr range: tv pts_time: 4235.4 [in @ 0x7f15b81ab380] Changing video frame properties on the fly is not supported by all filters. [in @ 0x7f15b81ab380] filter context - w: 1920 h: 1080 fmt: 26 csp: bt709 range: tv, incoming frame - w: 1920 h: 1080 fmt: 26 csp: gbr range: tv pts_time: 4235.4 So my current hypothesis is that - maybe - the first activation of the TC overlay (=filter) *might* have increased some total filter counter for the project - and maybe, when that same TC overlay was disabled later on, the "filter slot" it previously occupied would not be filled again - so some video processing queue might have remained open; or some "drop-all" default/dummy filter might slip into the place of the TC overlay -so that the video stream would be lost rather than written to the output file. I don't know if this hypothesis is valid or of any use to you... :-) -- You are receiving this mail because: You are watching all bug changes.
