Send Motion-user mailing list submissions to
motion-user@lists.sourceforge.net
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/motion-user
or, via email, send a message with subject or body 'help' to
motion-user-requ...@lists.sourceforge.net
You can reach the person managing the list at
motion-user-ow...@lists.sourceforge.net
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Motion-user digest..."
Today's Topics:
1. Re: Raspi 8MP Cam Motion Detection PI2 vs PI3 (tosiara)
----------------------------------------------------------------------
Message: 1
Date: Thu, 28 Jun 2018 14:28:53 +0300
From: tosiara <tosi...@gmail.com>
To: Motion discussion list <motion-user@lists.sourceforge.net>
Subject: Re: [Motion-user] Raspi 8MP Cam Motion Detection PI2 vs PI3
Message-ID:
<cachtdwtv3lp83spx-xrdn8acj8ypky0vmuh6mxnvzy944b4...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
As I already wrote, there is a way to utilize all of the cores
I was able to get 1080p close to 30fps using ODROID C1 few years ago. It
was running close to 100% of cpu on all 4 cores and recording movies using
x264 encoder.
In real life I don't need 30 fps on my security system. So instead, I have
dual camera 720p running at 10 fps, and I still have half of my arm quad
core capacity to do some other stuff
If you don't want to experiment with params to find the most efficient
combination - just add more power. Like, buy Odroid C2 or similar quad core
and have fun
FWIW, motion 4.1 has h264 passthrough mode which people are using to save
even more CPU. But you will have to read the documentation about it
On Thu, Jun 28, 2018 at 1:51 PM, <free...@tango.lu> wrote:
> Hello,
>
> Actually, multiple people talk about the frame issues:
>
> https://github.com/ccrisan/motioneyeos/issues/227
>
> "@dmarkscouk unfortunately I'm afraid that's the best you can get with
> motionEyeOS. No hardware/accelerated encoding/decoding is performed by the
> motion daemon and therefore the performance is limited to what the CPU can
> offer.
>
> A slow SD card may impact the performance as well, while saving videos to
> a network share could make matters even worse.
>
> The reason why FNA is that fast is that it doesn't use the motion
> "backend" but rather makes use of the streamEye project to simply capture
> and stream the mjpeg frames without processing them in any way (hence the
> great performance).
>
> I know this is not ideal and is probably not what most of you were hoping
> for when decided to try motionEye(OS) but this is what we have."
>
>
> It was similar experience for me as well, when there is motion, the
> framerate drops so what is recorded at the end is more like a sequence of
> pictures instead of a "movie".
>
> I have now compiled the latest:
>
> motion Version 4.1.1+git20180626-43b9c2b, Copyright 2000-2017 Jeroen
> Vreeken/Folkert van Heusden/Kenneth Lavrsen/Motion-Project maintainers
>
>
> and will do some experiments but with the same config I can see already
> that it is only using 1 core again :( so I will not expect much improvement.
>
>
> On 2018-06-28 10:12, tosiara wrote:
>
>> Regarding HTOP observation. Motion is indeed mostly using one thread
>> for all the heavy execution, like frame encoding/decoding and motion
>> detection. On multi core platforms you can use netcam with v4l2://
>> protocol to offload input decoding - this will give you an extra
>> thread. And also extpipe for movie save - with H264 codec it can use
>> all the available cores.
>> There are also ARM NEON optimizations available that can save up to
>> 20% of CPU on the main motion thread, but it is unofficial and not
>> available in the main repo (created by github user "notorca"), you
>> would need to apply patch and recompile motion
>>
>> On Thu, Jun 28, 2018 at 10:52 AM, <free...@tango.lu> wrote:
>>
>> I had a Raspi 2 v1.2 runnin g since a year with an 8MP camera. The
>>>
>>> camera is inside looking into a narrow hallway.
>>> Unfortunately the board died and I had to swap it to a Raspi 3 (not
>>> the latest 3B).
>>>
>>> Anyone who test the 2 boards side by side with camera noticed a
>>> difference in the motion detection?
>>>
>>> One would assume that the 3 might be better detecting motion since
>>> it has a faster CPU but as I noticed in HTOP motion is usually using
>>> 1 core only.
>>>
>>> I have the stock motion in Raspbian Stretch:
>>>
>>> ii motion 4.0-1 armhf V4L capture program supporting motion
>>> detection
>>>
>>> My config:
>>>
>>> daemon on
>>> process_id_file /home/bell/motion.pid
>>> setup_mode off
>>> logfile /home/bell/motion.log
>>> log_level 5
>>> log_type all
>>> v4l2_palette 17
>>> input -1
>>> norm 0
>>> frequency 0
>>> rotate 0
>>> width 1280
>>> height 720
>>> framerate 30
>>> minimum_frame_time 0
>>> netcam_keepalive off
>>> netcam_tolerant_check off
>>> auto_brightness off
>>> brightness 0
>>> contrast 0
>>> saturation 0
>>> hue 0
>>> roundrobin_frames 1
>>> roundrobin_skip 1
>>> switchfilter off
>>> threshold 1500
>>> threshold_tune off
>>> noise_level 32
>>> noise_tune on
>>> despeckle_filter EedDl
>>> smart_mask_speed 0
>>> lightswitch 0
>>> minimum_motion_frames 1
>>> pre_capture 0
>>> post_capture 0
>>> event_gap 60
>>> max_movie_time 0
>>> emulate_motion off
>>> output_pictures off
>>> output_debug_pictures off
>>> quality 75
>>> picture_type jpeg
>>> ffmpeg_output_movies off
>>> ffmpeg_output_debug_movies off
>>> ffmpeg_timelapse 0
>>> ffmpeg_timelapse_mode daily
>>> ffmpeg_bps 500000
>>> ffmpeg_variable_bitrate 0
>>> ffmpeg_video_codec mpeg4
>>> use_extpipe off
>>> snapshot_interval 0
>>> locate_motion_mode off
>>> locate_motion_style box
>>> text_right %Y-%m-%d\n%T-%q
>>> text_changes off
>>> text_event %Y%m%d%H%M%S
>>> text_double on
>>> snapshot_filename %v-%Y%m%d%H%M%S-snapshot
>>> picture_filename %v-%Y%m%d%H%M%S-%q
>>> movie_filename %v-%Y%m%d%H%M%S
>>> timelapse_filename %Y%m%d-timelapse
>>> ipv6_enabled off
>>> stream_port 8081
>>> stream_quality 50
>>> stream_motion off
>>> stream_maxrate 100
>>> stream_localhost off
>>> stream_limit 0
>>> stream_auth_method 1
>>> stream_authentication admin:hold
>>> webcontrol_port 8082
>>> webcontrol_localhost on
>>> webcontrol_html_output on
>>> track_type 0
>>> track_auto off
>>> track_iomojo_id 0
>>> track_step_angle_x 10
>>> track_step_angle_y 10
>>> track_move_wait 10
>>> track_speed 255
>>> track_stepsize 40
>>> quiet on
>>>
>>> Sometimes it did not detect the motion if someone went by fast
>>> (although its rare). I hope that the upgrade will make things
>>> better. Any config tuning options are welcome.
>>>
>>>
>>> ------------------------------------------------------------
>> ------------------
>>
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>> _______________________________________________
>>> Motion-user mailing list
>>> Motion-user@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/motion-user [1]
>>> https://motion-project.github.io/ [2]
>>>
>>> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>>> [3]
>>>
>>
>>
>>
>> Links:
>> ------
>> [1] https://lists.sourceforge.net/lists/listinfo/motion-user
>> [2] https://motion-project.github.io/
>> [3] https://lists.sourceforge.net/lists/options/motion-user
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> Motion-user mailing list
>> Motion-user@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/motion-user
>> https://motion-project.github.io/
>>
>> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
------------------------------
Subject: Digest Footer
_______________________________________________
Motion-user mailing list
Motion-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/motion-user
------------------------------
End of Motion-user Digest, Vol 144, Issue 11
********************************************
