Media Server :: Dynamic Buffering Based On Bandwidth And Datarate?
Jul 13, 2010
I'm trying to determine an initial value for NetStream.bufferTime based on the client's calculated bandwidth and the video's bitrate.As far as I can tell you can't do this, because you must have an initial buffer set BEFORE you call NetStream.play() (or use the default 2 seconds) and FMS does not call NetStream.onMetaData, where one would normally find the videodatarate, until after NetStream.play() is called.Even if you modify NetStream.bufferTime during the onMetaData call back it doesn't effect when the video starts playing or when the NetStatus event registers "NetStatus.Buffer.Full".
How do I find the Video Data Rate BEFORE I call NetStream.play?How do how do I reset the initial bufferTime before acctual play back begins and the NetStatus event fires with "NetStatus.Buffer.Full"
We are having a little problem implementing bandwidth detection with FMS, for some reason when enable it, bandwidth consumption spikes significantly I am wondering why? and if there a white paper somewhere that would talk about how properly implement this feature.
Is there anyway to utilise the bandwidth of the flash media server from the server side in such a way so that the streaming will happen from the flash media server to the user based on the user internet connection speed. So users with low bandwidth will receive less streaming data in their buffer and users with high bandwidth will receive more streaming data into their buffer from the server.
As stated in the title, I want to know how to determine if my server/host has native bandwidth detection. I have FMIS 3.5 installed on my Linux server. Is this a setting in FMIS or is this something I have to call my host to find out?
i am currently playing around with a flash media server trial in combination with the flowplayer.All in all i am testing the performance and general functionality for the company i work for.Anyways, i am trying to search a way to check the bandwidth of a client ( server sided ) and redirecting him to the right video stream.Sure this could be done by the backend flash application, but its out intention to perform this check server sided.
i setup the development versionof fms 3.5 server on windows 2003 server and it worked great in my tests.so we bought the server and are putting it in production now.the only problem is i never "tested" large FLV files.now if i put a large 50+megabytes flv file in the server it constantly buffers on the client.Is there a setting that defines the buffer size to increase the amount of buffer data that gets sent to the client?if i use a small flv file it works flawlessly but for large files it says "buffering" every 2-5seconds on the client.the client is just opening the test page that comes with the server the server has 3000mb/s to the internet and the client has a 20mb/s connection so i don't think bandwidth is an issue...
I'm trying to stream a webcam from 1 server to many client, what I really need to know is :1. How to adjust a bitrate video for bandwidth streaming, is it enough just using a Camera.setQuality() ?2. If u set Camera.setMode() at large resolution, for example at 1024x768 is it could make the bandwidth for client access increase too ?I mean is it affect the bandwidth for set the resolution at small (ex: 160x120) or set at large (ex: 1024x768) resolution.3. Same as number 2. just it was for video=new Video(w, h) . Is setting a video size affecting the bandwidth ?
Is it possible to limit globally the bandwidth usage of FMS 3.5 ? I don't want FMS to use all the bandwidth available via our ISP.I want my coworkers and other servers to be able to continue using Internet while sending/receiving data from/to FMS. I know it is possible to limit various bandwidths per Applications, but it seems there is no easy way - using the xml config files - to globally limit the server bandwidth.
I am trying to get the "bwcheck" application to work on FMS 4 on Amazon EC2, in order to detect my upload speed to the server and provide the best quality stream for my upload speed.
The tutorial I follow is this:
[URL]
Both are almost identical. However when I compile the .fla to .swf (having changed the URL for my server and application), and after I load the server-side code, I get the following error in my server-side log:
Method not found (recData)
I understand that apparently it is not finding the method, but I tried getting it off the function in the global scope, tried attaching it to the client, etc. but nothing worked. The original code of the server-side script is:
for ( i = 0; i < 1000; i++ ) { data += "S->C"; } Client.prototype.recData = function(data)
rtmfp downloaded from the node, and the upload bandwidth to another node, but the FMS bandwidth limitations Client.getBandwidthLimit () seems to only limit the bandwidth of the server to the client such as RTMP, ask how to limit the bandwidth between nodes, such as RTMFP.
I'm in the process of developing a realtime video chat application where multiple users can send video streams simultanuously. The number of users receiving the streams can be very big, e.g. 10 broadcasters and 500 receivers, each receiver should get all streams.
I use RTMFP connections to an FMS and streams are published in P2P multicast groups by passing the groupspec to the NetStream constructor. Currently I'm having problems with audio/video synchronization and video stream 'jumps' (not continuous). From what I read on other threads, this is related to the fact that there is not enough upstream bandwidth for sending the streams. So my questions are:
How to calculate the required upstream bandwidth on every peer for the given example of 10 broadcasters and 500 receivers (is it 10*bandwidth of one stream)?What settings (on NetStream, Camera, Microphone etc.) should be used for best results and how to adapt them based on the number of broadcasters?
I am running into some streaming bugs and am wondering if anyone else has come across them. Very grateful in advance for any help as I've been banging my head against this for a week now.
First of all, I'm finding that no matter what value I set for a NetStream's bufferTime property, the NetStream will not begin actually playing until the bufferLength property is over 2 seconds. This is not what Flash documentation says (that the default value of bufferLength is .1 seconds). Could this issue a function of the FMS configuration with my RTMP hosting account? Is it something I can resolve with my asc file?
Also: In my Actionscript 3 app, I need to pre-buffer a number of video streams so that any one of them can be triggered immediately later in the program. I do this by building an array of NetStreams (within a single NetConnection), then playing each of the streams I need to pre-buffer and immediately pausing each too. This is successfully causing them to buffer in the background, so that they're ready for immediate play when I need them. When I do this with just one NetStream, it works fine. But with more, the streams are behaving unpredictably. They are successfully buffering, but when I need to actually play them and I go to unpause them (Netstream.resume), the NetStreams are generally NOT playing back correctly. Their behavior is not consistent, and they mostly play with no video, or no audio OR video. The one exception is the FIRST NetStream that I started pre-buffering, which is the only one that's consistently playing back fine with its audio and video. Again, if I unpause any of the other NetStreams, they play back with just audio, or not at all.
When my streaming videos start they will buffer and begin to play as normal. However, as soon as the vid starts buffering again a problem arises. Basically what is happening is that Flash is buffering the video to a certain point, but instead of resuming playback from the point where the vid paused it now jumps instantly to the end of the newly buffered area without playing the vid portion in between. Which essentially means that it skips merrily through the video and never stops buffering. That said i'd prefer to get to the root of the prob, particularly if this is a codec/keyframe issue, since i could then re-export the vids appropriately.
I have confirmed this behavior in FMS 3.0.2 on Windows 2003, Vista, and Linux.
In prior versions of FMS if I had a publisher sending a live video feed to the server, then I could have subscribers connect to the server and set a bandwidth limit for them and the server would ensure that the subscribers only got as much bandwidth as I set for them. As of version 3, the server now ignores this limiting for live streams.
You can set a limit on the publisher which of course then limits the live stream that subscribers view, but that is not what I want. I want to have the publisher push a live stream at the quality I specify and then I need a way to give some subscribers a lower bandwidth version of that stream. This all worked fine in prior versions of FMS and I can't imagine this is intended behavior in version 3, but it seems like a major oversight.
The bug can be replicated by creating a simple flash app to stream live video to the server and then in a separate NetConnection, subscribe to that live stream and display it. Then on the server side, setup an application.onConnect to do a setBandwidthLimit on the subscriber and you can see that it is totally ignored. You get the full bandwidth stream even though the limit should be reduced for the subscriber.
We need to record and display daily/weekly/monthly statistics on vod application usage, we are working with multiple application and need statistics for each one of them individually parameters required are bandwidth, number of connection, view time ...
we are using RTMP with Flowplayer with dynamic bitrate switching and I'm a bit confused on the bitrate (kbps) that is called out in the code in order to configure the bandwidth check. I presume that this is the target bitrate if encoding using ABR or constant. How about for the rest of us that are using CRF? My understanding is that CRF is the way to go now for H264 (don't want to hijack the thread to a CRF vs ABR discussion). If we are using CRF, can we guess on the target bitrate that is called out in the bwcheck plugin?
We noticed that when we are streaming our own VOD file that we had copied to the server, which is an .MP4 file encoded at 1200 kbps using either HDS or RTMP (, which I suppose the server recodes on the fly ) we are getting the incoming stream at the average bitrate of 14 Mbit/s, and it looks like bursts of 25 Mbit every couple of seconds.
However if we try a sample file, such us sampl1_1500kbps.f4v we are getting an average of 1.2 Mbit/s with peaks of 1.5-2 Mbit. The tests are conducted in our company LAN. So what is the reason for this? And is there a way to control/restrict the output bandwidth for our own file without re-encoding? I'm not talking about several streams with different bitrate for the same source aka adaptive bitrate but rather about the single stream bw)
I am trying to get the "bwcheck" application to work on FMS 4 on Amazon EC2, in order to detect my upload speed to the server and provide the best quality stream for my upload speed.The tutorial I follow is this:{URL}Both are almost identical. However when I compile the .fla to .swf (having changed the URL for my server and application), and after I load the server-side code, I get the following error in my server-side log:Method not found (recData) understand that apparently it is not finding the method, but I tried getting it off the function in the global scope, tried attaching it to the client, etc. but nothing worked. The original code of the server-side script is:
I have downloaded and installed the Flash Media Server (FMS) trial version. Just purchased the software today and waiting for the activation serial number. I created a video gallery - [URL]. My goal is for these videos to stream just like vimeo. I am only using HD 1280 x 720 video. All videos are between 25 to 125 MB in size, originally .mov files converted to .flv using FFMPEG. Videos are between 30 seconds to 58 seconds in length.
I am using ColdFusion to upload and organize my videos and I have configured FMS to work with CF. The Issue is that these videos buffer the same way they did when I didn't have FMS installed. Press play, the video buffers, the video plays, then stops to buffer again. Stop and Buffer, Stop and Buffer. I can understand if there is buffering at the start of the video but not throughout. When I start the video over from the start, it has to buffer all over again!
I'm using the FLVPlayback to load streaming videos from a FMS. The bug that I'm running into is with the buffering state. Several times the video will enter a buffering state that never ends. The last video event that's fired in this scenario is the BUFFERING_STATE_ENTERED. But if I seek the video on the seekbar, it's playing away.
I followed the instructions located here:when i load the player and put the correct rtmp link, the video is stuck in bufferingI also tried the strobe media playback and getting the same issue.I also tried playing the sample.flv file located in the vod/media directory and getting the same problem
Does anyone know if Flash Media Server can be configured to reduce the image quality of a publish stream if available bandwidth goes down or there are delays in the connection?
When I (locally) play back videos recorded on poorer/intermittent connections, video frames are dropped, so the image freezes up.
I have made a sample application for publishing video through FMS using as3.In that I had used the same code as given in the sample provided by the adobe,
m_nc = new NetConnection("rtmpt://localhost:1935/Test"); private function onNetStatus(event:NetStatusEvent):void { switch (event.info.code) case "NetConnection.Connect.
