Geoff Shang said:> Hi: > > I don't think you'll be able to do this with EZStream, I don't think it has > the capability to send more than one stream instance, but I could be > wrong. > > Also, the reencode isn't going to match your standard input, as it's > looking for files ending in .mp3 which of course standard input won't do. > > Geoff.Thanks, I will keep that in mind. The worst-case scenario is 3 instances of ezstream, right ? Any suggestions for with what I could do this properly? I need to get an 128kb stream from a server, reendcode and rebroadcast it from a second one, at 3 different versions - 32/64/128. And I have a problem - no audio card on the second server - and I doubt I can get one added. :( Thank you, Dan
Dan V wrote:> I need to get an 128kb stream from a server, reendcode and rebroadcast it > from a second one, at 3 different versions - 32/64/128.Ok, well I'd set up a straight relay for the 128k version, then run two instances of streamTranscoder (from oddsock.org). This has the disadvantage of pulling the 128K stream 3 times, though if you got the transcoder to pull from your relay then those two would be local. Oddsock, have you had any thoughts about having multiple outputs from a single streamTranscoder instance? I'm currently runing two instances, both transcoding the same stream, and it occured to me that it should be at least theoretically possible to have one decode stage rather than having to have two. Geoff.
At 08:22 AM 1/19/2005, you wrote:>Oddsock, have you had any thoughts about having multiple outputs from a >single streamTranscoder instance? I'm currently runing two instances, >both transcoding the same stream, and it occured to me that it should be >at least theoretically possible to have one decode stage rather than >having to have two.it's certainly something that can be added, although right now it does not currently do this. I created ezstream as a "moderately-better" version of the example program that came with libshout. The intent was to make it a very simple and straightforward application as possible, not as feature-rich as other source clients...although you could argue that I violated that intent by adding the reencoding feature...either way, I'll look into it, but can't promise anything in the near term... oddsock