Any feature you'd like to see in Ultr@VNC? Just propose it here.


Postby cerb » 2004-09-19 17:18

this is a feature that would be very nice to test remotely connections and/or programs using sound. On the codec point of view this would be much easier than video, with many excellent and low-bandwidth sound codecs existing open source (GSM, CELP LTC etc).

Do you feel this useful function will ever be integrated in a VNC server? I think the only key issue to solve is the development of a direct recorder/virtual audio driver which allows sound output in the remote and/or recording to file. Once the recording is done coding and sending is trivial.

Does anything like this exist? If it does not, is there any third party software that allows some simple sound streaming from one PC to another. I do not require any accurate/nice reproduction, just something useful to understand what sounds are getting generated on the other side...

Postby UltraSam » 2004-09-19 21:59

Sound streaming is planned for Ultra V2 Beta.
Yes, there's already some GPL solution for this. I don't have the name in mind for now. (Rudi knows).

The Plan is to release V1 first.
We have to finish FileTransfer improvements and Multi-Monitor support.

V2 developpement should start by the end of 2004.
Admin & Developer
Admin & Developer
Posts: 466
Joined: 2004-04-26 20:55

Looking forward to it...

Postby ACN » 2004-10-07 15:39


I'll be looking forward to having sound going through my VNC connections as well.

Even if the only "serious" applications for this (as far as I can see?) would be getting the OS GUI "event" sounds (which might nevertheless be quite useful), and perhaps in some cases to provide voice communication (such as when doing support), I still think it would be a neat addition to VNC (even if only so I can see how smoothly I can get some game to run through a VNC connection on my GigaBit LAN ;)

Some points to consider when you start working on this (I hope all of these aren't blatantly obvious to all involved?):

- Of course some kind of compression of (digital) audio will be needed. I should think that the nice way to get this would simply be to support the existing codec scheme in windows, allowing the user (that is, the guy installing/configuring the server) to select among the installed codecs (which will allow users to use whatever current or future codec they wish). The client will of course need to have the same codec installed as well (just like you need the right codec to hear some sound file you download, but perhaps VNC could provide some option on the server to allow the client to download the codec from the server on the fly, if necessary? BTW. The same thing, downloading on the fly, could be considered for the DSM plugin as well, even for video/desktop-compresion codecs/algorithms, should such be implemented for use with VNC at some time.)

- Like I assume is the case with the current video (desktop) compression algorithm(s), great improvements can be had if the above (codec based) scheme is supplemented by less bandwidth demanding ways of transmitting "sounds". Thus you might consider (an option to) transmit only the "event" itself, instead of the actual audio produced, whenever some windows "event" sound is played (which would of course mean that the user would hear the sound he has configured for the even in question, but for such "events" that would probably be fine for most users). Likewise, it might be worth the while to send the actual MIDI (or whatever "synth data") stream, rather than having the server compress the played audio. This would also keep the quality of the music pristine, as it wouldn't need to be (lossy) compressed. (although there is a good chance that very few users will be using such a feature, which might make it impractical to waste development time on?)

- Like you have done with the "video-hook" driver, there might be advantages to doing something similar for audio? (ie. you might be able to obtain the "raw" audio-stream(s), before they are mixed and converted to analog. This might be especially useful, if some of the sources aren't supposed to be transmitted, such as the "synth" if doing the above "tweak", or perhaps the user only wants some sound sources to get through, such as the "mike", but not any MP3's that happen to be playing through the "wave" source etc.)

- Might some sound-cards (and/or their drivers) support some kind of hardware compression, that might be utilised to keep CPU loads low on the server? (I guess this would be adressed by supporting codecs, as such cards ought to come with codecs making use of any such features?)

Well, just my two cents as usual.


Posts: 7
Joined: 2004-10-07 14:02

Postby Rudi De Vos » 2004-10-07 17:00

It will more be something like ip-phone.

Vnc is mostly used for helpdesks, and some kind of phone
talk can be handy.

We already have trouble with debugging and maintaining the current parts.
Cisco, a well known player for ip phone, have an opensource project....integrating it is a lot faster and less bugs then writing
it self.
Rudi De Vos
Admin & Developer
Admin & Developer
Posts: 5984
Joined: 2004-04-23 10:21

Return to Feature requests

Who is online

Users browsing this forum: No registered users and 1 guest