Send Motion-user mailing list submissions to
[email protected]
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
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Motion-user digest..."
Today's Topics:
1. Re: Motion 4.2.2 vs 4.1.1 CPU usage (tosiara)
2. Re: Motion 4.2.2 vs 4.1.1 CPU usage (Ruslan Matveev)
3. Re: Motion 4.2.2 vs 4.1.1 CPU usage (rmbusy)
----------------------------------------------------------------------
Message: 1
Date: Sat, 20 Jul 2019 20:27:49 +0300
From: tosiara <[email protected]>
To: Motion discussion list <[email protected]>
Subject: Re: [Motion-user] Motion 4.2.2 vs 4.1.1 CPU usage
Message-ID:
<cachtdwqxh4oxvywihgm7zigz7wyvts8scpof2copz4oo8uk...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Solution for most popular RSTP cameras is to use low-res secondary stream
for motion detection, and main high-res stream for pass through
On Sat, Jul 20, 2019 at 7:55 PM oleg.chekalin via Motion-user <
[email protected]> wrote:
> Hi All,
>
> Then if all is bad as is, could someone propose a solution, to avoid the
> high load impact to processor?
>
> I mean seam no way for capturing rtsp without coding/decoding date?
>
> Regards,
> Oleg
>
>
>
> ?????????? ? MI MAX 2
> tosiara <[email protected]> | ??: 20 ???. 2019 ?. 6:53 ?? | ?????????:
>
> I have tested RTSP H264 1280x960 20 fps running on ARM ODROID C1.
> Just running, without any recording it takes 130% CPU. 4.1.1, 4.2.2,
> Ubuntu 16.04 and 18.04 - the result was always the same. NEON optimizations
> reduced to 125%, not much. Pretty all the time motion was decoding H264.
> The decoding, as already mentioned, is very CPU intensive
>
> On Sat, Jul 20, 2019 at 5:36 PM Colin Law <[email protected]> wrote:
>
>> On Sat, 20 Jul 2019 at 15:24, John D. Gwinner <[email protected]> wrote:
>>
>>> Ok, that helps me with my project ? we need to get 15fps or 30fps; (it?s
>>> a medical project).
>>>
>>>
>>>
>>> If you?re getting 2-5 and that high a CPU rate, clearly we?ll never get
>>> to 15.
>>>
>>
>> It also depends dramatically on the format of the data. If it is mpjpeg
>> for example then motion need do very little to decode the image, if however
>> it is H264 then there is a vast amount of work required just extracting the
>> images from the stream. On my system running an H264 camera a lot more
>> processor time is spent decoding the stream than is spent in motion
>> detection.
>>
>> Colin
>>
>> _______________________________________________
>> Motion-user mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/motion-user
>> https://motion-project.github.io/
>>
>> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>
> _______________________________________________
> Motion-user mailing list
> [email protected]
> 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...
------------------------------
Message: 2
Date: Sat, 20 Jul 2019 20:30:35 +0300
From: Ruslan Matveev <[email protected]>
To: Motion discussion list <[email protected]>
Subject: Re: [Motion-user] Motion 4.2.2 vs 4.1.1 CPU usage
Message-ID:
<cabqvc7s75msjour6jr8wuhg6w+st7wx2jt1koxysmbkolby...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Wow! Thats what I've been saying to topic starter from very beginning ))
??, 20 ???. 2019 ?., 20:28 tosiara <[email protected]>:
> Solution for most popular RSTP cameras is to use low-res secondary stream
> for motion detection, and main high-res stream for pass through
>
> On Sat, Jul 20, 2019 at 7:55 PM oleg.chekalin via Motion-user <
> [email protected]> wrote:
>
>> Hi All,
>>
>> Then if all is bad as is, could someone propose a solution, to avoid the
>> high load impact to processor?
>>
>> I mean seam no way for capturing rtsp without coding/decoding date?
>>
>> Regards,
>> Oleg
>>
>>
>>
>> ?????????? ? MI MAX 2
>> tosiara <[email protected]> | ??: 20 ???. 2019 ?. 6:53 ?? | ?????????:
>>
>> I have tested RTSP H264 1280x960 20 fps running on ARM ODROID C1.
>> Just running, without any recording it takes 130% CPU. 4.1.1, 4.2.2,
>> Ubuntu 16.04 and 18.04 - the result was always the same. NEON optimizations
>> reduced to 125%, not much. Pretty all the time motion was decoding H264.
>> The decoding, as already mentioned, is very CPU intensive
>>
>> On Sat, Jul 20, 2019 at 5:36 PM Colin Law <[email protected]> wrote:
>>
>>> On Sat, 20 Jul 2019 at 15:24, John D. Gwinner <[email protected]> wrote:
>>>
>>>> Ok, that helps me with my project ? we need to get 15fps or 30fps;
>>>> (it?s a medical project).
>>>>
>>>>
>>>>
>>>> If you?re getting 2-5 and that high a CPU rate, clearly we?ll never get
>>>> to 15.
>>>>
>>>
>>> It also depends dramatically on the format of the data. If it is mpjpeg
>>> for example then motion need do very little to decode the image, if however
>>> it is H264 then there is a vast amount of work required just extracting the
>>> images from the stream. On my system running an H264 camera a lot more
>>> processor time is spent decoding the stream than is spent in motion
>>> detection.
>>>
>>> Colin
>>>
>>> _______________________________________________
>>> Motion-user mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/motion-user
>>> https://motion-project.github.io/
>>>
>>> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>>
>> _______________________________________________
>> Motion-user mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/motion-user
>> https://motion-project.github.io/
>>
>> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>
> _______________________________________________
> Motion-user mailing list
> [email protected]
> 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...
------------------------------
Message: 3
Date: Sat, 20 Jul 2019 11:11:12 -0700
From: rmbusy <[email protected]>
To: [email protected]
Subject: Re: [Motion-user] Motion 4.2.2 vs 4.1.1 CPU usage
Message-ID: <[email protected]>
Content-Type: text/plain; charset="utf-8"; Format="flowed"
I heard you in the beginning Rusian, but I really want to understand why
I'm seeing the increased CPU usage before I simply try reducing it by
other means.? Believe me, your suggestion is on my list.? :-)
--
Rob.
On 7/20/19 10:30 AM, Ruslan Matveev wrote:
> Wow! Thats what I've been saying to topic starter from very beginning ))
>
> ??, 20 ???. 2019 ?., 20:28 tosiara <[email protected]
> <mailto:[email protected]>>:
>
> Solution for most popular RSTP cameras is to use low-res secondary
> stream for motion detection, and main high-res stream for pass through
>
> On Sat, Jul 20, 2019 at 7:55 PM oleg.chekalin via Motion-user
> <[email protected]
> <mailto:[email protected]>> wrote:
>
> Hi All,
>
> Then if all is bad as is, could someone propose a solution, to
> avoid the high load impact to processor?
>
> I mean seam no way for capturing rtsp without coding/decoding
> date?
>
> Regards,
> Oleg
>
>
>
> ?????????? ? MI MAX 2
> tosiara <[email protected] <mailto:[email protected]>> | ??:
> 20 ???. 2019 ?. 6:53 ?? | ?????????:
>
> I have tested RTSP H264 1280x960 20 fps running on ARM
> ODROID C1.
> Just running, without any recording it takes 130% CPU.
> 4.1.1, 4.2.2, Ubuntu 16.04 and 18.04 - the result was
> always the same. NEON optimizations reduced to 125%, not
> much. Pretty all the time motion was decoding H264. The
> decoding, as already mentioned, is very CPU intensive
>
> On Sat, Jul 20, 2019 at 5:36 PM Colin Law
> <[email protected] <mailto:[email protected]>> wrote:
>
> On Sat, 20 Jul 2019 at 15:24, John D. Gwinner
> <[email protected] <mailto:[email protected]>> wrote:
>
> Ok, that helps me with my project ? we need to get
> 15fps or 30fps; (it?s a medical project).
>
> If you?re getting 2-5 and that high a CPU rate,
> clearly we?ll never get to 15.
>
>
> It also depends dramatically on the format of the
> data.? If it is mpjpeg for example then motion need do
> very little to decode the image, if however it is H264
> then there is a vast amount of work required just
> extracting the images from the stream.? On my system
> running an H264 camera a lot more processor time is
> spent decoding the stream than is spent in motion
> detection.
>
> Colin
> _______________________________________________
> Motion-user mailing list
> [email protected]
> <mailto:[email protected]>
> https://lists.sourceforge.net/lists/listinfo/motion-user
> https://motion-project.github.io/
>
> Unsubscribe:
> https://lists.sourceforge.net/lists/options/motion-user
>
> _______________________________________________
> Motion-user mailing list
> [email protected]
> <mailto:[email protected]>
> https://lists.sourceforge.net/lists/listinfo/motion-user
> https://motion-project.github.io/
>
> Unsubscribe:
> https://lists.sourceforge.net/lists/options/motion-user
>
> _______________________________________________
> Motion-user mailing list
> [email protected]
> <mailto:[email protected]>
> https://lists.sourceforge.net/lists/listinfo/motion-user
> https://motion-project.github.io/
>
> Unsubscribe: https://lists.sourceforge.net/lists/options/motion-user
>
>
>
> _______________________________________________
> Motion-user mailing list
> [email protected]
> 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...
------------------------------
------------------------------
Subject: Digest Footer
_______________________________________________
Motion-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/motion-user
------------------------------
End of Motion-user Digest, Vol 157, Issue 35
********************************************
