On Jul 22, 11:01 pm, Jerry Avins <j...@ieee.org> wrote:
>
> >> One thing to keep in mind is that allowed decoding latency, which
> >> the original poster didn't state. I wonder if the question really
> >> was,
> >> "If I were to re-encode the entire steam (over the entire week) and
> >> decode the entire stream all at once, what is the optimal bit rate?"
>
> > Latency in TV iOne can allow for it if the worst case is known. The
> > latency of my cable box is a few hundred ms.
>
> Text got lost:
> Latency in TV isn't much harder to deal with than latency in recordings.
> One can allow for it if the worst case is known. The latency of my cable
> box is a few hundred ms.
>
> Correction: I can't prove more than about 100 ms in the box itself.
>
> Jerry
What I referred to was latency in terms of encoding and decoding block
lengths, not in signal propagation ...
Julius
Reply by Vladimir Vassilevsky●July 23, 20082008-07-23
julius wrote:
> On Jul 22, 10:33 am, Jerry Avins <j...@ieee.org> wrote:
>
>>pawinll wrote:
>>
>>>Hi guys,
>>
>>> I have been recoding the bit-rate information of one video content
>>>for a week. This system used the variable bit-rate which varied as the
>>>complexity of video picture (eg. fast-moving picture requires higher
>>>bit-rate). I would like to determine the optimal bit-rate from this
>>>one-week information, do you have any efficient model beside using density
>>>estimation (plotting data histogram)?
>>
>>If the content is typical, just compute bits per week and divide to get
>>any interval you want (bits/day. bits/sec, etc.).
>>
>>Jerry
>
>
> One thing to keep in mind is that allowed decoding latency, which
> the original poster didn't state. I wonder if the question really
> was,
> "If I were to re-encode the entire steam (over the entire week) and
> decode the entire stream all at once, what is the optimal bit rate?"
There is one more thing to consider: the video compression schemes often
use a bit reservoir of several hundred kilobytes or so. This reservoir
is slowly filled during the normal operation and emptied in the cases of
the peak demand. The operation of this reservoir is more involved then a
simple buffer; it is controlled from the transmitter side.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
Reply by Jerry Avins●July 23, 20082008-07-23
Jerry Avins wrote:
> julius wrote:
>> On Jul 22, 10:33 am, Jerry Avins <j...@ieee.org> wrote:
>>> pawinll wrote:
>>>> Hi guys,
>>>> I have been recoding the bit-rate information of one video
>>>> content
>>>> for a week. This system used the variable bit-rate which varied as the
>>>> complexity of video picture (eg. fast-moving picture requires higher
>>>> bit-rate). I would like to determine the optimal bit-rate from this
>>>> one-week information, do you have any efficient model beside using
>>>> density
>>>> estimation (plotting data histogram)?
>>> If the content is typical, just compute bits per week and divide to get
>>> any interval you want (bits/day. bits/sec, etc.).
>>>
>>> Jerry
>>
>> One thing to keep in mind is that allowed decoding latency, which
>> the original poster didn't state. I wonder if the question really
>> was,
>> "If I were to re-encode the entire steam (over the entire week) and
>> decode the entire stream all at once, what is the optimal bit rate?"
>
> Latency in TV iOne can allow for it if the worst case is known. The
> latency of my cable box is a few hundred ms.
Text got lost:
Latency in TV isn't much harder to deal with than latency in recordings.
One can allow for it if the worst case is known. The latency of my cable
box is a few hundred ms.
Correction: I can't prove more than about 100 ms in the box itself.
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Jerry Avins●July 22, 20082008-07-22
julius wrote:
> On Jul 22, 10:33 am, Jerry Avins <j...@ieee.org> wrote:
>> pawinll wrote:
>>> Hi guys,
>>> I have been recoding the bit-rate information of one video content
>>> for a week. This system used the variable bit-rate which varied as the
>>> complexity of video picture (eg. fast-moving picture requires higher
>>> bit-rate). I would like to determine the optimal bit-rate from this
>>> one-week information, do you have any efficient model beside using density
>>> estimation (plotting data histogram)?
>> If the content is typical, just compute bits per week and divide to get
>> any interval you want (bits/day. bits/sec, etc.).
>>
>> Jerry
>
> One thing to keep in mind is that allowed decoding latency, which
> the original poster didn't state. I wonder if the question really
> was,
> "If I were to re-encode the entire steam (over the entire week) and
> decode the entire stream all at once, what is the optimal bit rate?"
Latency in TV iOne can allow for it if the worst case is known. The
latency of my cable box is a few hundred ms.
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by julius●July 22, 20082008-07-22
On Jul 22, 10:33 am, Jerry Avins <j...@ieee.org> wrote:
> pawinll wrote:
> > Hi guys,
>
> > I have been recoding the bit-rate information of one video content
> > for a week. This system used the variable bit-rate which varied as the
> > complexity of video picture (eg. fast-moving picture requires higher
> > bit-rate). I would like to determine the optimal bit-rate from this
> > one-week information, do you have any efficient model beside using density
> > estimation (plotting data histogram)?
>
> If the content is typical, just compute bits per week and divide to get
> any interval you want (bits/day. bits/sec, etc.).
>
> Jerry
One thing to keep in mind is that allowed decoding latency, which
the original poster didn't state. I wonder if the question really
was,
"If I were to re-encode the entire steam (over the entire week) and
decode the entire stream all at once, what is the optimal bit rate?"
Reply by Jerry Avins●July 22, 20082008-07-22
pawinll wrote:
> Hi guys,
>
> I have been recoding the bit-rate information of one video content
> for a week. This system used the variable bit-rate which varied as the
> complexity of video picture (eg. fast-moving picture requires higher
> bit-rate). I would like to determine the optimal bit-rate from this
> one-week information, do you have any efficient model beside using density
> estimation (plotting data histogram)?
If the content is typical, just compute bits per week and divide to get
any interval you want (bits/day. bits/sec, etc.).
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by pawinll●July 22, 20082008-07-22
Hi guys,
I have been recoding the bit-rate information of one video content
for a week. This system used the variable bit-rate which varied as the
complexity of video picture (eg. fast-moving picture requires higher
bit-rate). I would like to determine the optimal bit-rate from this
one-week information, do you have any efficient model beside using density
estimation (plotting data histogram)?