ArrayArrayArrayArrayArrayArrayArrayArrayArray BrainModular BrainModular Users Forum 2018-11-27T16:51:13+02:00 https://brainmodular.org/forums/app.php/feed/topic/6322 2018-11-27T16:51:13+02:00 2018-11-27T16:51:13+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40452#p40452 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 27 Nov 2018, 15:51


]]>
2018-11-27T16:42:20+02:00 2018-11-27T16:42:20+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40451#p40451 <![CDATA[Video performance on usine hollyhock]]>
Video: MPEG4 Video (H264) 1024x576 30fps 12419kbps [V: DataHandler [eng] (h264 constrained baseline L3.2, yuv420p, 1024x576, 12419 kb/s)]
Audio: AAC 44100Hz stereo 150kbps [A: DataHandler [eng] (aac lc, 44100 Hz, stereo, 150 kb/s)]
and here's typical properties of footage that performs best at 1280x720
Video: JPEG 1280x720 30fps 60783kbps [V: Apple Alias Data Handler [eng] (mjpeg, yuvj422p, 1280x720, 60783 kb/s)]
Audio: PCM 44100Hz stereo 1411kbps [A: Apple Alias Data Handler [eng] (pcm_s16le, 44100 Hz, stereo, s16, 1411 kb/s)]
as you can see, one with no compression performs better even at higher resolution than one compressed at less resolution. both are mov files, that doesn't really matter, same with avi.

for middle ground, footage with following properties performed well even compressed (but not exactly HD resolution)
Video: MPEG4 Video (H264) 1024x576 24fps 7538kbps [V: DataHandler [eng] (h264 constrained baseline L3.2, yuv420p, 1024x576, 7538 kb/s)]
Audio: AAC 44100Hz stereo 152kbps [A: DataHandler [eng] (aac lc, 44100 Hz, stereo, 152 kb/s)]
as you can see it's mostly the same as the worst one, but bitrate is considerably lower and framerate is also less. so I guess if you're going to have compressed video, it's better if it's more compressed? I don't know if that's how it works.

another example, but with rocky performance (stuttering a bit from time to time dropping frames, but otherwise going smoothly)
Video: DivX 6 1280x720 25fps 2991kbps [V: mpeg4 simple profile, yuv420p, 1280x720, 2991 kb/s]
Audio: PCM 44100Hz stereo 1411kbps [A: pcm_s16le, 44100 Hz, 2 channels, s16, 1411 kb/s]

Statistics: Posted by x.iso — 27 Nov 2018, 15:42


]]>
2018-11-27T09:19:00+02:00 2018-11-27T09:19:00+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40447#p40447 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by joffo78 — 27 Nov 2018, 08:19


]]>
2018-11-27T02:51:55+02:00 2018-11-27T02:51:55+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40445#p40445 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 27 Nov 2018, 01:51


]]>
2018-11-26T18:18:06+02:00 2018-11-26T18:18:06+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40436#p40436 <![CDATA[Video performance on usine hollyhock]]> I manage To have only 19fps with only one flow in 1280/720

Statistics: Posted by joffo78 — 26 Nov 2018, 17:18


]]>
2018-11-20T15:16:56+02:00 2018-11-20T15:16:56+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40405#p40405 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 20 Nov 2018, 14:16


]]>
2018-11-20T15:12:55+02:00 2018-11-20T15:12:55+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40403#p40403 <![CDATA[Video performance on usine hollyhock]]>
"But I think Usine processes actual pixels from decoded video when you apply effects and output it futher, which affects CPU as well. at least I think that's how it works."
Hmm, I thought of that too.

Statistics: Posted by TimTight — 20 Nov 2018, 14:12


]]>
2018-11-15T21:26:44+02:00 2018-11-15T21:26:44+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40384#p40384 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 15 Nov 2018, 20:26


]]>
2018-11-15T20:23:47+02:00 2018-11-15T20:23:47+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40382#p40382 <![CDATA[Video performance on usine hollyhock]]> any idea to optimise a little ?
Thanks.
JF

Statistics: Posted by joffo78 — 15 Nov 2018, 19:23


]]>
BrainModular BrainModular Users Forum 2018-11-27T16:51:13+02:00 https://brainmodular.org/forums/app.php/feed/topic/6322 2018-11-27T16:51:13+02:00 2018-11-27T16:51:13+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40452#p40452 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 27 Nov 2018, 15:51


]]>
2018-11-27T16:42:20+02:00 2018-11-27T16:42:20+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40451#p40451 <![CDATA[Video performance on usine hollyhock]]>
Video: MPEG4 Video (H264) 1024x576 30fps 12419kbps [V: DataHandler [eng] (h264 constrained baseline L3.2, yuv420p, 1024x576, 12419 kb/s)]
Audio: AAC 44100Hz stereo 150kbps [A: DataHandler [eng] (aac lc, 44100 Hz, stereo, 150 kb/s)]
and here's typical properties of footage that performs best at 1280x720
Video: JPEG 1280x720 30fps 60783kbps [V: Apple Alias Data Handler [eng] (mjpeg, yuvj422p, 1280x720, 60783 kb/s)]
Audio: PCM 44100Hz stereo 1411kbps [A: Apple Alias Data Handler [eng] (pcm_s16le, 44100 Hz, stereo, s16, 1411 kb/s)]
as you can see, one with no compression performs better even at higher resolution than one compressed at less resolution. both are mov files, that doesn't really matter, same with avi.

for middle ground, footage with following properties performed well even compressed (but not exactly HD resolution)
Video: MPEG4 Video (H264) 1024x576 24fps 7538kbps [V: DataHandler [eng] (h264 constrained baseline L3.2, yuv420p, 1024x576, 7538 kb/s)]
Audio: AAC 44100Hz stereo 152kbps [A: DataHandler [eng] (aac lc, 44100 Hz, stereo, 152 kb/s)]
as you can see it's mostly the same as the worst one, but bitrate is considerably lower and framerate is also less. so I guess if you're going to have compressed video, it's better if it's more compressed? I don't know if that's how it works.

another example, but with rocky performance (stuttering a bit from time to time dropping frames, but otherwise going smoothly)
Video: DivX 6 1280x720 25fps 2991kbps [V: mpeg4 simple profile, yuv420p, 1280x720, 2991 kb/s]
Audio: PCM 44100Hz stereo 1411kbps [A: pcm_s16le, 44100 Hz, 2 channels, s16, 1411 kb/s]

Statistics: Posted by x.iso — 27 Nov 2018, 15:42


]]>
2018-11-27T09:19:00+02:00 2018-11-27T09:19:00+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40447#p40447 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by joffo78 — 27 Nov 2018, 08:19


]]>
2018-11-27T02:51:55+02:00 2018-11-27T02:51:55+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40445#p40445 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 27 Nov 2018, 01:51


]]>
2018-11-26T18:18:06+02:00 2018-11-26T18:18:06+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40436#p40436 <![CDATA[Video performance on usine hollyhock]]> I manage To have only 19fps with only one flow in 1280/720

Statistics: Posted by joffo78 — 26 Nov 2018, 17:18


]]>
2018-11-20T15:16:56+02:00 2018-11-20T15:16:56+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40405#p40405 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 20 Nov 2018, 14:16


]]>
2018-11-20T15:12:55+02:00 2018-11-20T15:12:55+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40403#p40403 <![CDATA[Video performance on usine hollyhock]]>
"But I think Usine processes actual pixels from decoded video when you apply effects and output it futher, which affects CPU as well. at least I think that's how it works."
Hmm, I thought of that too.

Statistics: Posted by TimTight — 20 Nov 2018, 14:12


]]>
2018-11-15T21:26:44+02:00 2018-11-15T21:26:44+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40384#p40384 <![CDATA[Video performance on usine hollyhock]]> Statistics: Posted by x.iso — 15 Nov 2018, 20:26


]]>
2018-11-15T20:23:47+02:00 2018-11-15T20:23:47+02:00 https://brainmodular.org/forums/viewtopic.php?t=6322&p=40382#p40382 <![CDATA[Video performance on usine hollyhock]]> any idea to optimise a little ?
Thanks.
JF

Statistics: Posted by joffo78 — 15 Nov 2018, 19:23


]]>