yorkie98

Sometimes I don’t always read through a message properly, I totally missed that you were setting up a web radio station…
So, ignore pretty much everything I advised, and take the advice from AdamH.

The very best soundcard for BBP (nothing else comes close) is a Marian Trace Alpha. This however is not a multi I/O soundcard so you will need to either add a second normal soundcard for your editing/mixing or use the onboard one.
The machine spec you have quoted should easily cope with BBP and Stationplaylist in fact you’ll probably only use about 10% cpu on that.
A BBP machine does not actually need to be a server, its just recommended to build a server grade machine for the job so running Idrac is not needed unless you are running this for your own purposes.

Hope this helps.

Yorkie.

Jesse, Loving this preset so far, it’s powerful yet open, and respectful of the dynamics. It’s also really bright and crisp, kick drums keep their edge and don’t lose their attack.

The only track which IMO did not sound as good as I hoped was the Great Gig in the Sky, although this may be down to the source as my reference for this track is the 2003 remaster SACD.
It’s just the piano seems to develop a gritty, almost 8-bit sound during the quieter periods (this may be gating noise in the orignal master) but I certainly don’t hear this on my SACD version.
Aside from this observation (which may be down to the source anyway) the dynamics were handled brilliantly and the feel and emotion of the song shone through.

A great job so far, keep up the great work.

Yorkie.

[quote author=”JesseG”]AM does use pre-emphasis.

and the NTSC spec for AM allows for 10kHz of audio bandwidth. no reason to limit it to 5kHz unless you WANT to sound worse.

And no soundcard is ever "overkill" for anything, unless you’re talking about a 13 year old kid playing games or something. This is broadcast baby.[/quote]

True comment Jesse about the soundcard, I’m just such a skinflint at times, always looking for that economy/quality sweetspot. I’ll take my hat off to the first person who uses a Trace Alpha for AM..

My setup was for the European AM specification where I live in the UK, the audio bandwidth is 40hz -5khz.

So, Robert, set the HPF and LPF filters accordingly for your territory/license. As for the pre-emphasis, again, none used here in Eurozone AFAIK but if appropriate for your territory then of course set to required time constant.

All settings apply to the latest (.93) version..
To begin with, if you are using AM, you will need to de-select Stereo in the settings section. Secondly, you will not need pre-emphasis selected either (put down to 15us).
In the calibration screen, I would first of all apply the bandwidth to 5Khz, and the HPF to 30Hz, which it will be on anyway.
In the I/O config screen, Disable the RDS in if not already done and disable the main output (this will bring up the AM sub-menu when you go back into the calibration menu) and then set the L/R output as your main output card. This will then give you an AM compliant mono output. Go back to the calibration menu and adjust the AM asymmetry slider to the desired level (25% for 125% positive peaks). The difference will not be immediately obvious visually but if you (temporarily) put the final drive right to the top and then adjust the asymmetry slider up, you will see the bottom of the trace lifting upwards, this is your visual confirmation that BBP is now producing positive peaks.

The preset/audio settings you use are entirely down to your own taste and requirements, try a few and play with the sliders until you get a sound you like.
I’m not sure (maybe someone can help here) if the Tilt calibration is required for AM, I’m kinda guessing not but I could well be wrong here, CMIIW people..

Finally, unless you are using the E-Mu 1212 for a specific quality reason, this is major overkill for AM, the 192Khz operation is only required for FM stereo operation. ANY soundcard will be fine for AM, even an onboard card will give a more than good enough SNR for AM.

Hope this helps,

Yorkie.

Hi,
Ok for a start, you will not need the Behringer 1202 at all. Here is how you should configure your setup..

To begin with your computer upon which you have Breakaway installed, MUST be capable of 192Khz operation. The very best soundcard to use is a Marian Trace Alpha. If you can’t source one of these then I recommend an ESI Juli@ which are easy to source. For ease, I shall refer to the Juli@ card but if you get a Trace alpha then replace each reference as appropriate.

If your playout is on the same PC as you are running breakaway, you will need to set the Zara output to "Breakaway Pipeline 1".
If your playout is on a different computer and/or you have a mixer in your setup, the audio from this must be connected to the input of the Juli@ card.
This is where the magic begins.. The computer with breakaway running, not only processes your sound above and beyond industry standards, but it also acts as your stereo coder (and optionally RBDS/RDS coder if you plugin the Airomate software for €25).
You then connect the output of just one channel of the Juli@ card into your exciter (remember this is now a multiplexed signal). You will now be transmitting in stereo with fully processed audio.

You should think of the computer running Breakaway with the Juli@ (or Trace Alpha) card fitted as the equivalent of an Orban Optimod 8500 (only better). You connect L/R stereo audio into it and you get fully processed and multiplexed (transmitter ready) audio out. It almost sounds too good to be true I know but this is how advanced Breakaway is as a product.
It’s best to have the breakaway PC doing this job standalone but it should be ok to run a basic automation system such as Zara on the same machine if you wish or if you need to.

To answer your question "Is there a software based stereo generator that I can utilize?" the answer is YES, Breakaway has this built in. There are a handful of other programs which can do this also but Breakaway stands head and shoulders above all of them as the only professional quality processor AND stereo coder out there.
The most important thing to understand is that you must have a 192Khz output soundcard (most standard PC soundcards are 48Khz). Without this, none of this is possible.

I hope this helps.

Yorkie.

[quote author=”rebel”]re split processing on the AAC+ link, after looking at the issue further, I understand it may also be desirable to set the processing at the studio to include the AGC & MBL compression stages only (and perhaps light peak protection limiting, but no clipping), while the tx end will have the limiting & clipping sections. The theory is that if tonal rebalancing of the MBL section occurs only at the transmitter end, that it may uncover some of link noise previously masked by the link codec. In this case, it may only be possible to use BBP (at both ends of the link) if we have more advanced control, or at least the ability to disable some processing modules.[/quote]

@Rebel/Aaron… I in fact use BBP in a similar way, I have a main site, close to the studio which is fed by a 320Kbps unprocessed IP STL (running about -20dB below threshold to give ample headroom), this is then fed into BBP and into the transmitter. The L/R feed (via dsp) is fed into Edcast (a 3dB attenuator is applied in dsp before edcast as a public AAC+ stream is also created by edcast) and a private 160Kbps MP3 stream is sent to a streaming host. This stream contains the processing of the AGC and the MBL but the effect of the final clippers is somewhat lost due to encoding/decoding artifacts, meaning the potential for overshoots is re-introduced. This stream is then cascaded to remote transmitter sites where the stream audio is fed into BBP but the preset selected is "6dB protection clip". This effectively does not utilise the AGC or MBL but just feeds straight into the final clippers. The resulting MPX signal is then fed to the transmitters.

If you are using AAC+ for your link and this is ONLY for your link, not used as a webstream too, then I would tend to try to run this with no processing at all, you will get a better end result as a heavily processed signal is harder to encode into a limited kbps stream than unprocessed music as there is simply less detail and information there to squeeze in. Then let BBP do ALL the work at the TX end which is what it does best.

To my ears, all the transmitters sound tonally identical, there are no overshoots on any site and +/-75khz nailed up all the time. The only way anyone would notice these are IP fed is that the main site is approx 20 secs ahead of all the remote sites, but I’m hoping to address that sometime by introducing a delay to the mpx output to the main transmitter to bring them to within 1 sec of each other.