Forum Replies Created
-
Posts
-
Leif’s own description of the Reference Settings preset.
“This is the preset the Breakaway Core algorithm was created with. This preset is as transparent and neutral as they come — great on its own, or as a starting point for any custom sound.
Imagine that you’re designing an audio processor, either digital or analog. Before there is a user interface, you must still choose correct component values to make the circuit behave, and sound good. When that’s done, you decide which parameters should be user-adjustable and which should stay hard-coded. The user-adjustable ones must have a default value, which should ideally match what you came up with as sounding good while writing the algorithm. Those are the Reference Settings, and that is this preset.”
What preset are you using? Does the effect change with different presets? Try it with “Reference Settings”, as this is the least processed preset.
I believe that these are the default settings.
https://ceaforum.net/wp-content/uploads/2023/04/BAE-Pipeline.jpgIf you have the Breakaway Audio Control Panel program installed, you should just be able to modify the parameters to bring it back to its defaults.
BAE purposely only allows one line, as it was never designed to modify output to other audio programs. Many people have tried over the years to record the output, but I believe that the program is coded to prevent this. Maybe the re-write, when we see it, will relax this restriction.
Thank you for reporting. I will investigate further.
I usually leave the pipeline at the default 44100/48000, as this takes in any CDs or DVDs that I might play.Both of these products have a niche in the broadcast domain. Trying to compare their niche is possibly like comparing a Mazda 3 to a BMW or (insert your favourite luxury car here). Both will transport you from A to B in relative luxury, but there are definable differences in the two vehicles, and some of the differences may be significant to some drivers, but some may not.
Similarly, some broadcasters insist on the best of the best, and have the budgets to achieve that, whereas many have to produce the most competitive sound output they can achieve with far smaller budgets.
1. The core code and algorithms are almost identical, but there have been extensions to that core specific to Omnia 9, including the number of bands, undo etc. These extensions possibly could be incorporated in future releases of BA1, but at an increase in price.
2. Surprisingly, the soundcard doesn’t have to be particularly “high end”. It obviously needs to faithfully reproduce the source material without colouration. Remember that FM audio is “capped” at 50 Hz to 15 kHz to allow for the 4 kHz safe zone and pilot tome at 19 kHz, so having a card capable of playing sounds above that is a moot point.
3. I will need to speak to my engineer to provide the specifics of what additional hardware is required for MPX, as he is the “hands on” guy. I’ll try to provide specifics later.
4. There a many, many operational radio stations who have a Windows-based “black box” running BA1 at the core of their broadcasting chain. Some have gone the extra mile and built it into a 19 inch rack mount case, but many are running a NUC because it doesn’t need additional fans which are noisy, attract dust and wear out.
5. The demo version of Breakaway One is fully functional, so you can set it up to achieve everything you want for a parallel trial, bearing in mind that there will be a random voiceover inserted into the program material until the product is registered. Breakaway Marketing can help advise you on which modules you need, but I would definitely recommend the Remote Management module so that you can make tweaks from afar.
The problem is, the way that BAE is currently written, you must run the TEST option to firm up the buffer and sample rate options. The inputs and outputs are only scanned in the configuration process and only “live” options at that time are listed.
Breakaway Audio Enhancer is slated for a complete re-write as it was first coded when Windows XP was the OS of choice, so these types of suggestions will be considered for version 2.
I am seeking absolute clarification, but I believe that in FM mode, BAE limits the top end frequencies to less that 15kHz, to allow for the 4 kHz guard band either side of the 19 kHz pilot tone used to produce a stereo FM signal.
An FM stereo signal is broadcast as a mono signal 53 kHz wide, with the Right channel (23 – 53 kHz) “stacked” on top of the Left channel (50 Hz to 15 kHz). A pilot tone at 19 kHz separates the two channels, with a guard band of 4 kHz either side of that to allow for deviation. This means that the useable music frequencies cannot go above 15 kHz without encroaching on the guard band.
That looks OK, but it really depends on your particular hardware. If you find the jitter figure (when you turn on the “test” light) is higher than (say) 15%, try a different combination of buffers and buffer size until you get the lowest numbers. However, the larger the buffers, the more latency becomes noticeable, because it takes BAE some time to fill the buffers.
So, ultimately, it is a juggling act between minimal jitter and minimal latency. For instance, you might get lower jitter numbers with a buffer size of 882, but it may introduce a degree of latency. At the end of the day, if you can’t hear the dropped frames (jitter), it may not be a problem, but you should aim for less than 5% if you can achieve it.
If all of your material is at 44.1 why are you upscaling to 48K on output? There is no audible advantage, in fact there may be some audible artefacts as a result.
As the preceding article says, it’s all about getting the lowest possible jitter rate. You might find that 882/2 might yield a better result, but it might not.
I’m not aware of an import option (others correct me if I’m wrong), but, if you export a configuration zip file of your working version of BA1, you have a copy of all the settings of the working program. This is basically everything in the folder {Systemdrive}/users/{username}/AppData/Local/BreakawayOne.
If you do a fresh install, close BreakawayOne, then rename the BreakawayOne folder above to BreakawayOne_old. Now unzip the backup to the same folder name. Restart BreakawayOne and the setup should now be the same as the old.
I am unsure of the definitive answer to this without setting up some scenarios and viewing the results on an oscilloscope. However, I always align windows, Pipeline and BAE settings to the same value. I have heard some systems where music recorded at 48k played slower when down-converted to 44.1k
Having said that, I believe that the Windows settings are ignored if you select “KS”, because that is a lower level layer than the Windows options.
There is a neutral preset in ‘Reference Settings’. It just provides a constant output level, with very little colouration of the source material.
When using your USB device, you probably need to change the input of BAE to pipeline 1,and the output as the USB device.
Good research!
When you say that it uses twice the CPU and RAM, what does that take it to? My experience is that CPU and RAM usage that is around 50% and occasionally transients to 75% still leaves enough headroom provided that the PC is adequately cooled.As you probably know, the whole point of the LL path was intended for radio presenters to be able to hear their own foldback, without there being a delay. It is very confusing if you are hearing your voice 1 second delayed because of processing delays.
