Old 19th June 2005, 22:59   #1
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
Virtual audio NETWORK cable

I'd like to run my streaming source client on one computer and then do the encoding on a separate computer.

I can connect the sound card audio out on the source client computer to the sound card audio inp on the encoding computer. But, that requires converting from digital to analog, moving accross a cable and then converting from analog back to digital. The conversions and the cable both downgrade sound quality and introduce noise.

So, it seems to me that it would be possible to attach to the digital audio on one computer, transfer the data via tcp/ip to the other computer and then input it directly to the encoder.

Virtual audio cables are used within one computer to connect the output of one software program directly to the input of another program.

So, for programs that would work with a virtual audio cable, there's no reason why the output from a software progam on one computer couldn't go across the LAN to the input of a software program on another computer.

This is exactly what virtual audio cables do, except that there are two ethernet cards and tcp/ip between the input and output.

Does anyone know of software or a reliable technique to do this?

If my thinking on this is screwed up, please let me know.

This could also work on the input side of things. For example if you wanted to use a separate computer to run your call-in phone system and then use the Virtual audio network cables to connect the telephone audio to the studio console computer (the streaming source client computer).

Geoff
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 00:48   #2
djSpinnerCee
Forum King
 
djSpinnerCee's Avatar
 
Join Date: Aug 2004
Location: Hollis, Queens/The Bronx, NYC
Posts: 3,511
I'm not sure what you're asking exactly -- encoding (resampling/downsampling) is done by the WinAMP and the DSP -- this is the "source" that is sent to the DNAS server -- the server does not "touch" or modify the source in any way, it just distributes it to listeners.

So, you can create your source on PC A (DSP) and send it to PC B (DNAS) to be sent to listeners. You could make the DSP source the highest bitrate/samplrate possible so that there is no loss in the process and the resultant stream at the DNAS on PC B would still be 100% digital. This is how to get the source from one PC to another.

If you want to process the source more on PC B (ie to add voice, etc...), you could "listen" to the DNAS on the same PC -- This WinAMP/DSP in PC B could also send it's output to another "public" DNAS on either PC.

In effect, the WinAMP source program on PC A, is sent via TCP/IP to PC B, where it can be further processed... Using Soundcard output in teh DSPs would allow either WinAMP source to contain any or all of the audio inputs available to both PCs. There is a real time delay as source A gets to PC B, but as far as PC B is concerned, WinAMP B will hear source A in it's own time.

If the delay from using TCP/IP is unacceptable, a better idea would be to transfer the audio signal as an analog via Audio-Out...Line-In -- Either method has its drawbacks, but there is no perfect solution.

You didn't make clear what you really want to do, so something else to consider: My Winamp source program (playlist-type MP3s) is compiled from MP3s that exist on no less than 5 PCs on my LAN -- I use Windows Shares to make the MP3s of all my PCs available to one WinAMP running on one PC (the source) -- my LAN is certainly fast enough to load any kilo-bit MP3 over TCP/IP (I really use IPX, but it's the same thing -- over 100mbps Ethernet). So it depends on what you're trying to do.
djSpinnerCee is offline   Reply With Quote
Old 20th June 2005, 02:26   #3
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
I didn't know you could transfer the audio signal in real-time from one computer to the next via tcp/ip. That's exactly what I am talking about. So, if you can tell me how to do that, I would greatly appreciate it.

But, be sure that you understand what I am talking about.

I am NOT talking about writing the audio to disk and then having the other machine open the audio file stored on disk.

Nor, am I talking about encoding the audio on one computer and serving it so that the other computer can pick it up and decode it. (Which is what it sounds like you are talking about.)

If I were to encode on the studio (content creation) computer, then wouldn't need the second computer. I want to move the encoding to another computer on the LAN to reduce the load on the studio computer.

That's what I am doing now: My mixer, mike, cart machine, and telephone (for call-in) are all connected to the "studio" or "content creation" computer. Then the output from the sound card on the studio computer is connected to the input on the encoding computer. The main streaming server connects to the stream on my encoding computer and re-distributes it to the listeners. This is a pretty standard way of doing things.

What I am talking about is the functional equivalent of plugging the output of the sound card on computer A into the sound card input on computer B. BUT, doing it "virtually" - by replacing the sound cards with a "Virtual Audio Network cable".

Do you know what Virtual Audio Cables are? If you have program "A" that sends audio to the input of the sound card and then program "B" that uses the output from the soundcard as input, Virtual Audio Cables are drivers that move the signal from the output of one program to the input of another program, bypassing the sound card(s) altogether.

http://www.ntonyx.com/vac.htm

http://software.muzychenko.net/eng/

If you look at the above two links and then modify the concept so that the output is on one computer and the input is on another computer connected via some protocal, probably tcp/ip across the lan.

Geoff

Last edited by RadioLeft; 20th June 2005 at 02:41.
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 02:35   #4
likearock
Senior Member
 
Join Date: Jun 2005
Posts: 121
^ there is no need for that.

There are 2 possable scenarios that I can think of that you're trying to do here...

1. Use 1 computer to DJ, use another computer to "shoutcast" the stream to the public

2. Use 1 computer to DJ and "shoutcast" the stream to the public, and listen on another computer in your home at the same time.

Is either of those what you want to do? Either way, its really easy...but I don't think we want to waste time walking you through it if its not what you're looking for...

Hopefully you can be more specific?
likearock is offline   Reply With Quote
Old 20th June 2005, 03:01   #5
djSpinnerCee
Forum King
 
djSpinnerCee's Avatar
 
Join Date: Aug 2004
Location: Hollis, Queens/The Bronx, NYC
Posts: 3,511
Maybe you just don't know how a ShoutCAST works:

Believe it or not, if you run a shoutcast source, which is WinAMP+the DSP (set to soundcard iuput) and the DNAS server on PC A -- you can use any or all of PC A's soundcard "sources" as input -- the WinRecord mixer makes it so. The DSP can provide the full quality of any of your soundcard inputs.

On PC B, you can run WinAMP and "tune in" to the DNAS on PC A -- WinAMP's "output" on PC B is controlled by the WinOutput mixer (comonly known as Volume control) by the WAVE device of your soundcard -- This will be exactly the same digital source as PC A (but it will be behind time wise).

This is what you're talking about, and it takes a LAN, that is, two PCs connected to the same switch/hub, bith running TCP/IP. You may have found it under a million different names, but this is the essence of a ShoutCAST.

Read up on doing your own ShoutCAST -- download the DSP and DNAS, and try it out, you'll see that it's what you're looking for.

Just because just about everyone "talking" on these boards is doing the radio station thing, ShoutCAST can be and is used for many, many more things -- I've seen people use a WinAMP+DSP to bug offices, as an intercom, and even as a baby monitor (soundcard input and a MIC can go a long way), Keep in mind that WinAMP can use/mix/broadcast ALL soundcard inputs simultaneously, AND any disk based sources. When the DSP has a DNAS server to connect to, that source can reach anywhere in your home or on the internet.

And, it is really easy
djSpinnerCee is offline   Reply With Quote
Old 20th June 2005, 03:04   #6
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
Look:

Here's what I have been doing for five years:

I use one computer to "DJ".
I use another computer to Encode.

The output from the sound card on the DJ computer is plugged into the input on the sound card of the Encoding computer.

This works just fine. I've been doing it this way 24 hours per day for five years. But, it introduces noise and distortion from the processing in the sound cards and the cables. That's what I want to eliminate.

If you know what Virtual Audio Cables are, then you will understand what I am asking.

I want to eliminate the sound card on the DJ computer, the sound card on the encoding computer, and the cables that connect them and replace this equipment with software drivers. A software driver on the DJ computer would "talk" to a software driver on the encoding computer via the LAN. So the signal would go directly across the LAN instead of through soundcards and cables.

AND, this would be done without doing any conversions on the audio signal. It wouldn't be encoded. It wouldn't be saved to a disk for the other computer to read. The signal would go directly from the output of the DJ software on the DJ computer to the input of the encoder software on the encoding computer.

Virtual Audio Cables take the output of the DJ program and input it directly to the encoder program on the SAME computer.

I want to go from the output of the DJ program on the DJ computer to the input of the encoder on the encoding computer using Virtual Audio NETWORK Cables instead of the sound cards and physical cables.
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 03:18   #7
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
djSpinnerCee:

I keep asking the same question over and over again with different words and all I get back is people who want to explain streaming 101 to me.

Now, it appears that you understand what I have been asking.

If I understand you correctly, DJ or production software on computer A and the encoder on computer B can communicate across the network without encoding and decoding the signal, writing the audio to disk and reading it from the other computer, or by using cables that connect the sound card of one computer to the sound card of the other computer.

Is that what you are saying? If so, thank you.

I've been doing Windows Media under the configuration described in earlier posts and was not aware that the shoutcast components could connect directly over the LAN.

By the way, if you answered yes, then the way it works is by having drivers on the various machines that talk to each other directly - which is exactly the architecture I was describing.

Geoff
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 03:43   #8
djSpinnerCee
Forum King
 
djSpinnerCee's Avatar
 
Join Date: Aug 2004
Location: Hollis, Queens/The Bronx, NYC
Posts: 3,511
Help me out here: Details always help, dont be shy...

the DJ PC -- runs ? PC DJ or something like that?

the DJ PC is connected to the Encoder PC with the "Speaker" out on the DJ PC connected to the Encoder PC's Line-In, correct

The Encoder PC runs WinAMP/DSP and the DNAS -- and you want to avoid the analog connection between the two PCs?

Is there a reason why WinAMP and the DSP are not running on the DJ PC? -- WinAMP can encode the soundcard on the DJ PC.
djSpinnerCee is offline   Reply With Quote
Old 20th June 2005, 04:35   #9
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
First, I read the documentation.

The Shoutcast DSP plug-in for Winamp takes whatever input you have (microphone, pre-recorded music, station bumpers and ids, etc. mixes them down and encodes them to MP3.

The Shoutcast Server connects with the Winamp source, takes the already mp3 encoded audio and makes it available to listeners who connect to the server.

But, there are really three components:

1. Mixing down of mikes, music, commercial into the audio feed you want to broadcast.

2. Encoding of that feed.

3. Distribution of the feed to the listeners.

The Shoutcast DSP plug-in for Winamp combines functions 1 and 2.

What I am doing is:

Function 1 on computer A

(And recording the program in a no loss format Wave file. The wave file is for archive purposes only. For example, we have interviews with every Democratic Party candidate for President from the 2004 election - There were 11 of them at one point. As well as recordings of live events we have covered such as demonstrations against Bush when he went to Brussels Belgium and the voter protest marches in Washington, DC and San Francisco on March 19th 2001. In fact, our first day on the air was that March 19th, 2001. We broadcast live feeds from the events in DC and San Franscisco with an on-site correspondent in each city and three host/commentators in our Dallas studios.)

Which connects to computer B via sound cards and cables. Computer B provides Function 2.

Function 3 is provided by Warp Radio. Warp Radio connects as a listener to the stream on computer B, pulls the stream to their servers where listeners connect.

I already have function 1 and function 2 on two separate computers. (Remember that function 3 is provided by Warp Radio at the moment.)

I plan to keep it that way because we are adding shoutcast mp3 and icecast ogg in addition to Windows Media. The encoding machine can handle encoding all three formats and delivering them to the servers in the sky.

On computer A, we will still mix down the feed. But, we are replacing the wave file for archiving with flac encoded files since they take up half the disk space of a wave file.

Because of everything we have going at once: a host and two or three guests, some in the studio, some on the phone, station ids, bumpers, and some music, callin from the listening audience, the mix down computer has a lot going on.

We have been broadcasting an approximately 2 hour talk show each day, repeated until time for the next day's broadcast. - We haven't really followed a clock. Now, we are adding addtional programming and so we have to upgrade our software, use a clockwheel, and have our programs run on an exact schedule. That requires more resource intensive software. And, since we will be encoding in 4 formats (FLAC for archive on computer A, and ogg, wma, and mp3 on the compuer b) everything will run much more smoothly if we continue to use two computers.

If I can eliminate the sound cards and audio cables to connect the two computers, it will improve the sound quality and also make it easier to move the encoding computer into the control booth, where we will have our call-in system as well.

Hence, I am asking about "Virtual Audio NETWORK Cables"

Geoff
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 05:12   #10
djSpinnerCee
Forum King
 
djSpinnerCee's Avatar
 
Join Date: Aug 2004
Location: Hollis, Queens/The Bronx, NYC
Posts: 3,511
Ok

I still have questions:

To deal with your real issue -- if you "find" what you call Virtual audio cables, what they'll really be is an analog to digital coverter (an encoder) with a TCP/IP transport capability -- now here's your problem -- your source is analog right -- if you want to avoid the physical analog connection, you must run an encoder of some sort on the MixDown PC -- there's no way around it. TCP/IP only transfers digital bits. Does your MixDown application produce digital output? The Archiving process is also an encoding if the source is truly analog.

Some other things I don't "get" that may be in my way -- locations -- are the MixDown PC and the encoder PC in the same room? are they on the same LAN? and do they both run TCP/IP? Only ask this because there are (distance) limitations to a line-level analog connection -- noise does increase as you add distance. If the connection is less than 8 feet or so, using good cables, and there's still unnaceptable noise, I would look at the other hardware and other sources of the noise. Bringing the encoder PC in close proximity to the mixdown PC may solve all of your problems.

Like I said in the first post, the DSP that runs on the mixdown PC can encode to it's highest quality and send that to a DNAS on the encoder PC without any loss in quality and to a broadcast-bitrate DNAS that Warp Radio can "listen/connect" to. WinAMP runs pretty quietly, and shouldn't load the mixdown PC too much.

There is another concern I have that may really be a problem -- if you are recording the archive on the mixdown PC, the WinRecord mixer and thus the soundcard may not be available for any other application -- this includes the WinAMP/DSP, or any other encoder that may be part of the virtual network cables package -- something to think about in your design.
djSpinnerCee is offline   Reply With Quote
Old 20th June 2005, 07:22   #11
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
Here you go...

Yes. The machines will be on the same lan.

The mix down machine is in the broadcast studio.

The encoding computer will be moved to the control room, where the callin telephone system and the call screener work.

I won't use Winamp. I may use SAM3. But, through the evaluation period, we've lost ungodly amounts of time due to their poor support. So, that's iffy.

But, their system doesn't require sound cards at all (Will use one for monitoring - headphones for the host and guests.)

So I already know that the Virtual Audio Cables will work with SAM3 on a single machine. In that situation, there's no conversion. I'm thinking that all this would have to do is packetize, ship it across the network and unpack it. It's just bits. So, I don't see why all of the conversions you're talking about would be required.

The over head of what I am talking about would have to be less than what you are suggesting. And, I won't be running a streaming server locally at all.

If you come across the technology I'm looking for, I'd appreciate knowing about it. Once I get it worked out, I'll post the results here.

Geoff
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 08:25   #12
Wavestreaming
Major Dude
 
Wavestreaming's Avatar
 
Join Date: Apr 2004
Location: United Kingdom
Posts: 1,563
Send a message via AIM to Wavestreaming Send a message via Yahoo to Wavestreaming
I'm familiar with this setup. This is how most radio stations who stream online operate...

A studio computer with multiple soundcards with the playout software running, each soundcards output is fed to a different channel on the mixer.

One of the mixers outputs is then fed into a encoding PC which encodes and sends the feed to shoutcast/windows media server.

If I were you, i'd invest in some good quality sound cards and some good quality cables which run between them, job done.

I dont think you're gonna get very far with this virtual audio cable idea, it is to specific and highly unlikely shoutcast (or even windows) will support it without some very expensive hi-tech software. Then, I could be wrong
Wavestreaming is offline   Reply With Quote
Old 20th June 2005, 08:51   #13
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
Give me a break, please. I asked if anyone knew of such software and I have spent the last six hours having to justify my existence on this planet.
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 13:04   #14
dotme
Moderator
 
dotme's Avatar
 
Join Date: Feb 2005
Location: USA
Posts: 4,024
Quote:
Originally posted by RadioLeft
Give me a break, please. I asked if anyone knew of such software and I have spent the last six hours having to justify my existence on this planet.
Sorry - but was that necessary? I thought the answers provided were perfectly reasonable, and the people who answered certainly know their stuff. It was nice of them to offer suggestions. If you're not satisfied with the replies, there's always Google...
dotme is offline   Reply With Quote
Old 20th June 2005, 14:14   #15
elektrovert
Junior Member
 
Join Date: May 2005
Location: Dublin
Posts: 21
have you considered optical soundcard connections between your computers?
no loss of quality there.
elektrovert is offline   Reply With Quote
Old 20th June 2005, 14:18   #16
RadioLeft
Junior Member
 
Join Date: May 2005
Posts: 21
No, I haven't

Quote:
Originally posted by elektrovert
have you considered optical soundcard connections between your computers?
no loss of quality there.
Don't know what an optical sound card is. But it sounds expensive.
RadioLeft is offline   Reply With Quote
Old 20th June 2005, 15:11   #17
elektrovert
Junior Member
 
Join Date: May 2005
Location: Dublin
Posts: 21
Optical sound cards can be cheap!

http://www.pixmania.co.uk/uk/uk/7823...fichetechnique

for example.

These cards use light to transfer digital information across cables from one device to another (pc to mixer or pc to pc or whatever).

The best thing about it is it doesn't matter howmmany crossed cables or magnetic fields there are in the way, there's no interference coz you're using light to transfer the information.
I have 24 channels working across 3 cables that are less than 3/4 of a centimeter thick together!
it's great.
Just make sure the cards you get have optical inputs too.

Also you might want to consider s/pdif, which is digital transfer across phono cables.
try this:

http://www.zzounds.com/item--MDOAP2496

A classic card, dirt cheap for the standard of audio it produces.

I've been using one of these for about 5 years now and it ROCKS!

google Audiophile 2496 and check the results, or go onto somewhere like www.futuremusic.co.uk and search the forums for the same, this card is loved by one and all!!
elektrovert is offline   Reply With Quote
Old 21st June 2005, 00:36   #18
likearock
Senior Member
 
Join Date: Jun 2005
Posts: 121
Quote:
I use one computer to "DJ".
I use another computer to Encode.

The output from the sound card on the DJ computer is plugged into the input on the sound card of the Encoding computer.

This works just fine. I've been doing it this way 24 hours per day for five years. But, it introduces noise and distortion from the processing in the sound cards and the cables. That's what I want to eliminate.
I don't know, it seems so simple to me...

I have 5 stations running from 1 computer using 5 copies of OTSDJ. All stations are sourced from 5 seperate outboard soundcard "channels".

The noise introduction that you are talking about is from a soundcard mounted inside the computer (and possably crappy drivers). Over a period of time, this will happen even with the best onboard soundcards (especially if you are using RCA to MINI or any other type of RCA or 1/8" cable, they have NO sheild, or quality grounding...hence noise).

Get yourself a nice outboard multi-channel soundcard, and continue as you were.

Noise gone, no silly "virtual ...", just a dip into the pocket for a good quality piece of outboard equipment (that is, a soundcard that is not a card, but an outboard firewire unit or outboard encoder to PCI card like MOTU or something along those lines- that use Audio XLR and/or 1/4 T/R/S "tip ring sleave" cables...get one with as many stereo in/out pairs as you need).

Now you can actually encode your streams on the same computer you DJ on because the sound from your "soundcard" is processed nice and clean OUTSIDE the box leaving it with more power to do other things.

There is no other way if you want to do it right with the least amount of issues and maintanance.

Last edited by likearock; 21st June 2005 at 03:28.
likearock is offline   Reply With Quote
Old 5th May 2010, 14:38   #19
ejmarino
Junior Member
 
Join Date: May 2010
Posts: 1
NetJack

I think it's a little late to reply but, if someone like me is looking the same that this guy:

http://jackaudio.org/

http://netjack.sourceforge.net/

hope it helps
ejmarino is offline   Reply With Quote
Old 6th May 2010, 16:15   #20
DJ Matt
Senior Member
 
DJ Matt's Avatar
 
Join Date: May 2002
Location: Topeka, KS
Posts: 433
That network jack sounds hard, I've already attempted to find free or even cheap software to do this and you're better off buying soundcards with SPDIF I/O like an ECHO MIA or similar.

Matt...
Star104 Internet Radio
http://www.star104.net
Oldies104 Celebrating 14 years on the air! (Does anyone remember OldiesRadioNET.com?)
DJ Matt is offline   Reply With Quote
Old 7th May 2010, 07:40   #21
rockouthippie
Banned
 
rockouthippie's Avatar
 
Join Date: Jun 2004
Location: Oregon
Posts: 11,002
Why do you use one computer to encode and another to do your DJing? You could just stream the output of your sound mixer running multiple instances of Winamp. One to DJ, the other to encode the sound card mixer output. Allow multiple instances is in the options some place. In some Vista and later drivers recording from the sound output isn't supported so you could have a problem there. In that case, I'd just loop the line out back into the line in with a short 1/8th inch patch cord.

I don't see how you are going to send analog from one machine to another in any reasonable way other than a patch cable like you are using.
rockouthippie is offline   Reply With Quote
Reply
Go Back   Winamp & Shoutcast Forums > Shoutcast > Shoutcast Discussions

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump