|
Home > Archive > Macromedia Flash Server > November 2005 > Re: RE: FlashComm Digest, Vol 11, Issue 55
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Re: RE: FlashComm Digest, Vol 11, Issue 55
|
|
| Naicu Octavian 2005-11-25, 7:45 am |
| I do not think you understand well what a buffer means.=20
=20
A buffer is a place to store data before it is displayed (in this case=
). Video data travels continuously from the server to the client. If tha=
t data travels fast enough (Broadband) it will go into the buffer and wa=
it there until it is displayed. If data is not travelling fast enough (D=
ialUp) the buffer is not filled fast enough and there is nothing to be d=
isplayed.
=20
=20
Fabio Sonnati <sonnati-6XAXFxi7MR1y6ZJYwPuT2FaTQe2KTcn/@public.gmane.org> wrote: It is possible, in n=
on realtime streaming, to change the buffer depth
on the fly to better exploit client bandwidth.
Ex. suppose your client has 350Kbit/s and you deliver to him
a 300Kbit/s stream. You can set the buffer time to 3 seconds,
when playing start, you are able to change the buffer deep
without forcing rebuffering. So, if after the playing is started,
you set buffer time to ie: 6 second, the player will try to fill
the buffer and with the bandwidth in excess (350-300) the
buffer will grow over the initial 3 seconds while playing.
This approach is usefull, because the inizial wait is 3 second,
but the buffer capacity is then greater and in the case of
a sudden bandwidth congestion there is a better capacity to react.
When the buffer empty, You must set buffer time to the
initial value of 3, and then again set it to 6 (or more) when
play begins.
This work only for non real-time stream.
FS
----- Original Message -----=20
From: "Graeme Bull"=20
To: "'FlashComm Mailing List'"=20
Sent: Friday, November 25, 2005 7:35 AM
Subject: RE: [FlashComm] RE: FlashComm Digest, Vol 11, Issue 55
> No, that's not quite possible in the way that you want to do it.
>
> You're going to have to calculate out the users bandwidth and then buff=
er
> accordingly. The only way to make it a better experience if it has to=20
> buffer
> after that point is to set a larger buffer. Then again, if they have to
> buffer, then set it bigger etc.
>
> The flash player should be buffering all the time, but if you set it to=
o=20
> low
> and it can't keep up, that's a network problem (bitrate is too high). =
Not=20
> a
> flash or fcs problem.
>
> -Graeme
>
> -----Original Message-----
> From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org] On Behalf Of Vivek
> Lakhanpal (Office)
> Sent: Friday, November 25, 2005 2:44 PM
> To: flashcomm-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> Subject: [FlashComm] RE: FlashComm Digest, Vol 11, Issue 55
>
> Hi everyone,
>
> Thanks for all those who have replied to my query. Let me list my scena=
rio
> more clearly for you all.
>
> 1. I am recording a live presentation in FCS application. It can be of=20
> hour
> length or a minute length.
> 2. After that I am relaying the same video with another application i.e=
..
> recorded video in my other application using FCS application.
> 3. Right now I am using FCS 1.6 i.e. patched version of FCS 1.5
>
>
> I have few doubts to all solutions provided in the list:
>
> 1. As per Naicu Octavian whenever we get "NetStream.Buffer.Empty" we
> increase the buffer length. But I think we are using some value to set
> buffer length and when it consumes the whole buffer it start filling it=
up
> again. So does it make any difference if we set that value to 10 or 20 =
or
> whatever until we get the whole video length?
>
> 2. As per HTH "we can reach metadata information. That way, you'll get =
the
> length of the stream. Just position the bufferLength on that value."
> What if the video length is of 1hr or more then that?
>
> 3. As per Sxren "Is it streaming or progressive .... progressive is a l=
ot
> better". As I made clear above it's the recorded video done through FCS
> application. Now I am not sure if we can do that i.e. change the type o=
f
> recording through FCS application.
>
> 4. As per Ed "FMS2 now provides a NetStream.Play.InsufficientBW" "But i=
t
> might provide you with some indications that you can use to adjust your
> buffer." First of all I was using FCS 1.6. Even if I upgrade to FMS2 I=20
> think
> that won't solve my problem.
>
> What i have asked is if I can relay the video and simultaneously fill t=
he
> buffer. So that it never get's emptied. Or if there is some nearby=20
> solution
> like through which I can give user a smooth running kind of experience.
>
> =3D-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =3D-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
>
> =3D-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =3D-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>=20
=3D-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=3D-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around=20
http://mail.yahoo.com=20
=3D-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=3D-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
|
|
|
|
|