OS/2, eCS & ArcaOS - Technical > Internet
Firefox 45.9.0 & Youtube
Holger Schuett:
Hi there,
I did some more testing and made an observation, but I am not quite sure if it is worth considering.As mentioned I have firefox 38 too and youtube works well. After Dimik had pointed out the possibility to look at the statistics I took a look at the statistics of the same video when opened with firefox 38 and firefox 45.9.
Looking at the codecs the videos in FF 38 always show something like VP8/Vorbis and in FF 45 VP9/Opus.
Is this important?
Regards
Holger
Dave Yeo:
--- Quote from: Holger Schuett on October 13, 2017, 07:04:40 pm ---Looking at the codecs the videos in FF 38 always show something like VP8/Vorbis and in FF 45 VP9/Opus.
Is this important?
--- End quote ---
It shouldn't be, assuming you have a powerful enough CPU. Of course that is assuming that our VP9 port works well.
Basically VP9 is more efficient at encoding/decoding a video, so higher (perhaps double) quality video at the same bitrate, or lower bitrate for the same quality.
If I get a chance, I'll try transcoding some videos here and test VP8 vs VP9 on local videos. I don't have the bandwidth to use Youtube.
Andy Willis:
I've been seeing this issue with Seamonkey.
The original video in the ticket says I can't play it due to geography.
The one Darius put in:
https://www.youtube.com/watch?v=MAigQuoCfD0&feature=youtu.be
does show the problem in seamonkey but oddly works in Firefox 45.9.0.
I just checked some others that I had seen were/are not working in Seamonkey and worked in Firefox too. I just checked and in both cases it is the VP9/opus.
Holger Schuett:
Hi Dave,
what would a powerful enough CPU be, I run a TP X61, so fairly old machine. Perhaps this is the problem?
Holger
Dariusz Piatkowski:
Hi Holger,
--- Quote from: Holger Schuett on October 14, 2017, 03:54:49 pm ---...what would a powerful enough CPU be, I run a TP X61, so fairly old machine. Perhaps this is the problem?...
--- End quote ---
Just my $0.02 in this, I do not think the starting of the playback should be impacted by the CPU power. If it starts correctly and then starts to miss (dropped frames, etc...) then maybe the CPU performance is the cause, but the playback should begin regerdless.
I suppose the other option is that due to MM instruction extensions for a particular range of CPUs the older stuff may not support the new instructions? But I would imagine something like this is really controlled at compile time where a target CPU/family is identified. I can not imagine we would be publishing code for the OS/2 platform that will correctly execute on just the latest set of CPU families.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version