x265 encoding error.

[murriato]

Wouldn't it be better and more appropriate for the ffmpeg command to be -colorspace bt2020nc/bt2020_ncl instead of -colorspace bt709?

Yes, if you upscale, it would be better to also use a color matrix conversion to bt2020.

Yes, I think so too. That's why I wanted to try not to let that happen.

And do you know why the ffmpeg command contains bt709? How could I solve it?

It's just that checking the debug logs the ffmpeg command always contains bt709. Even if Hybrid detects the input as 2020 then in the encoding debug log you see the ffmpeg commands with bt709. And even if I use a 2160p MKV input instead of an avs script, the same thing happens. I don't know what to do.

About the crash, the debug out complains about:

Error initializing the muxer for 3840:2064:0:48,scale=1920:1032: Invalid argument
Error opening output files: Invalid argument

for:

"C:\Program Files\Hybrid\64bit\ffmpeg.exe" -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "C:\Users\cebol\Desktop\ENCODES\MapNLQ.avs" -an -sn -vf format=yuv420p10le,crop= 3840:2064:0:48,scale=1920:1032 -color_primaries bt709 -color_trc bt709 -colorspace bt709 -color_range tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -sws_flags spline -f yuv4mpegpipe - | "C:\Program Files\Hybrid\64bit\x265.exe" --input - --fps 24000/1001 --output-depth 10 --y4m --profile main10 --level-idc 5.1 --qg-size 16 --tu-intra-depth 4 --tu-inter-depth 4 --limit-tu 4 --me star --subme 7 --limit-modes --max-merge 4 --no-early-skip --rskip 0 --no-open-gop --opt-ref-list-length-pps --min-keyint 23 --bframes 16 --weightb --ref 5 --rc-lookahead 60 --lookahead-slices 0 --crf 16.90 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --ipratio 1.30 --pbratio 1.20 --rd 4 --psy-rd 2.01 --rdoq-level 2 --psy-rdoq 2.01 --aq-mode 3 --no-cutree --vbv-maxrate 160000 --vbv-bufsize 160000 --hrd --vui-hrd-info --deblock=-3:-3 --no-sao --selective-sao 0 --no-mcstf --range limited --colorprim bt2020 --transfer smpte2084 --colormatrix bt2020nc --chromaloc 2 --hdr10 --hdr10-opt --sar 1:1 --output "C:\Users\cebol\Desktop\ENCODES\HYBRID\2024-09-17@14_43_08_5610_01.265"

=> seems like there is an unncessary space in the call generation.
I'll look at it after work today, should be easy to fix.

Couldn't reproduce the problem here that Hybrid adds a space after the 'crop='.
I adjusted the code a bit to:
a. add some additional debug output where I suspect this .
b. do an additional check where the crop part is computed.
=> please try the latest dev and if it still fails, share another debug output.

I have tried the new dev version and I still get the same problem.
I attach a debug output.


Thank you very much again!