NEW FM PROCESSOR

[The F Mister]

It does sound like the FM version will be the way to go... time to start saving up the pennies. Will the full-control version be less or more expensive that the FM version? And will it have presets?

The full-control version, if I make one, will have to be much more expensive. We've gotta remember where the competition is at. The least expensive competitor is probably DSP-X Mini, and that one is still over $2000. Subtracting the cost of a server grade rackmount PC and a capable sound card, there's still a great discrepancy in the cost. I wanted to make a version that assured that *anyone* could afford excellent processing if they wanted to -- and that version is Breakaway FM.

The full-control version (If I make one - it depends on a number of business-related things) would have presets, tcp/ip remote control, input fail-over, hardware watchdog support and any other feature needed to easily build a capable "Hardware" processor from a Server PC.

///Leif

That's a nice comparison but bare in mind that the mini or any other "hardware" processor out there run at much less latency than Breakaway FM. I wonder how much of the cleanliness remains when running at the same latency. I'm sure any other processor manufacturer could come up with algorithms that do the same as Breakaway FM considering that they wouldn't care about latency. But the past has proven that latency is a big issue to most. It's not for nothing they come up with low latency presets or low latency monitor paths.

This doesn't mean Breakaway FM isn't a great peace of programming. It just needs to be put in the right context when comparing it to the rest out there.

 

[konbaasiang]

F Mister, you have a good point!

However, for example in the states, with HD radio broadcasting, the analog path gets an 8 second delay. Processing latency is certainly not an issue in those cases. It's an issue in other cases, but it can be worked around, for example with a low-latency path for feeding the studio. With processing this inexpensive, it's even feasible to have a completely separate processor in the studio.

Certainly it was *possible* for someone else to come up with the same innovative processing algorithm. However -- they didn't.

The fact that they have separate low latency presets is an excellent way to prove my point. Certain hardware processors have low latency presets so that the people who need them can use them, but even when using the normal presets (which are presumably designed for those who want maximum quality and loudness and do not care about latency), they're *nowhere near* as clean at the same loudness level.

Listeners do not care about latency. Whether they care about good sound is another issue altogether (many certainly do not - at least not consciously), but one thing is for sure -- they want a loud and clear signal (distortion is usually attributed to bad reception) and lacking a point of reference, they don't know or care whether the audio is Live, or 200 milliseconds behind.

There will certainly be broadcasters who are unwilling or are unable to use a separate low latency monitoring chain, and need low latency on-air processing at any cost. My hat is off to them -- and they certainly have plenty of processors to choose from!

Breakaway FM, on the other hand, is for those broadcasters who want to make a big improvement in their stations sound quality, and are willing to sacrifice off-air monitoring to get it.

 

[GeneSavage]

The "select Breakaway as your soundcard" thing tripped me up, too.. :-[ It was when I installed the personal version and it TOLD me to select it (and I think popped up the soundcard screen) that I "got it" and was able to get the FM version to work.

It sounds like a full control version is going to be way, way WAY out of reach for me for some time... so as a compromise, will more presets be available to add over time and / or can we request preset characteristics for the FM version? I'm thinking things like, "similar to the "Oldies" preset but with a faster attack," not "gimme something 'punchier'" which is absolutely impossible to define and likely different for each person.

I was goofing around with the FM again last night and found myself drifting away from the Zenith preset to try some of the others... it looks like I missed that a couple of other presets gives you the max number of bands as well (Oldies as one example). My ears are beginning to tune to some of the gain riding, and I'm finding myself gravitating towards the reference preset, amazingly... not the reference heavy, just reference... reference.

I also spent a significant amount of time with everything cranked to 100 and was surprised that although it was significantly coloring some of the songs, #1 it wouldn't suck up what wasn't there and #2 it wasn't really fatiguing like your typical "nailed against the wall" processor. It's kind if intriguing to listen to songs you think you're familiar with through this setup.

I have to wonder what that kind of density could do for AM, music or talk?

It seems like everything buzzes pretty happily along at about 87 for each (level, power and speed); anything more & you're really starting to color the sound, less sounds more open but not as tight as it can be. The defaults were 50 on everything weren't they?

I still like cranking level and power all the way up and speed all the way down, but I went from Ministry's "Every Day Is Halloween" (fade) into Talk Talk's "It's My Life" and nearly got knocked out of my chair! I don't know if that's something about the texture of "Halloween" or if it's a perceptual thing with the slow attack time... but that particular transition would NOT work without cranking up the speed.

I still love this product; I hope my feedback is helping in some small way to make it even better!

...And talking about latency: try monitoring yourself on an internet stream! ha-ha-ha...

 

[konbaasiang]

NightAire, I guess you're running Vista . In Vista, Breakaway Personal tells you to do it, but in XP, it does it for you.

Breakaway Personal is a pure consumer product, so I had to make it as easy as possible. However, doing that also loses flexibility.

So, for Breakaway FM, maybe I need to provide some hints. I guess I was simply assuming that people would be familiar with the concept of virtual audio cables, and in hindsight, making that assumption was not very bright . I guess I just didn't think about that part at all. When you've done audio processing on PCs for as long as I have (12 years and counting), things just become second nature.

Thank you very much for your comments - I do appreciate them, and they are definitely helping make Breakaway FM an even better product!

The different number of bands in different presets is really pretty arbitrary. It's impossible to say that 7 bands is better than 6, or that 6 is better than 5. It's possible to make excellent sounding presets for any configuration -- they're just *different*. I based the Oldies preset on Helix, so that's where the 7 bands in that preset comes from . Helix was based on Zenith a long, long time ago, but Jesse has tweaked and perfected Helix for so long that they have nothing in common anymore, other than the number of bands. Internally, Helix uses Infinite:1 ratio, 7 bands, and 24dB of multiband range. Those are maximum values for all three parameters - you may have noticed that turning up Range or Power does nothing on Helix -- only turning downwards makes a difference. I was floored when I saw the parameters he was using for this preset, and I'm amazed at the sweet, stable sound he squeezed out of my core under those premises!

My new clipper would technically work just as well for AM as it does for FM and Streaming, but there's so much more going on with the sound on AM (such as low pass filters inside receivers, mandated 5 KHz low-pass for IBOC, MONO only etc) that there's not really any appeal for me to make an AM processor. It's not just as much fun - unlike FM, where you really CAN squeeze an earful of sound onto that carrier, crank up the volume, and just get lost in the music!

Defaults are indeed 50. The user interface controls add a relative adjustment to any factory preset (50 means no change), but each preset is different internally.

Lowering the speed can be addictive indeed! It injects an openness into the sound that you don't get when running things at the default speed. However, Speed 0 is indeed very slow -- it does take away a lot of re-equalizing power. It could work extremely well for a fine arts station though, in combination with well controlled source material (like you get if you feed your DJ *unprocessed audio*) and limited range.

Lowering the range would prevent the compressors from gaining and gaining and gaining during a fade-out, which means less of a surprise when the next tracks kick in. This could work well even with extremely low speeds.

I have indeed monitored myself on an internet stream! It's pretty cool if you use it to your advantage, because it enables you to hear what your last break sounded like!

Back in the day (2001-2002), I wrote an integrated radio studio in software OctiMax Studio -- way before SpacialAudio's SAM broadcaster. Unfortunately it never saw the light of day, but I did learn one very important lesson from the whole experience -- don't be an employee!

Anyway, this studio software was designed for live webcasting, and had an "off-air" bus where you could click a button to hear the vastly delayed web-return.

Listening to a 30 second web return is not without its pitfalls though -- you can imagine what happens when you hear the song fading, and realize you've in fact been broadcasting dead air for the last 30 seconds!!

So, to prevent this, I made it so that it automatically returned to live monitoring when there was less than 30 seconds left of the song in playing in one of the two air decks. Problem solved .

///Leif

 

[Sgeirk]

Raced it against an Omnia 6 today...the O6 sounds broader with it's stereo enhancement...but the difference really isn't that strikingly dramatic. I'm sure Breakaway would slay an 8200. What a clean AGC and multiband section.

 

[GeneSavage]

NightAire, I guess you're running Vista

Sadly, yes. Actually, once you set everything to Win 2000 settings & shut off the protection features, it's not a bad operating system. ;D

(Funny aside: the feature I can't live without now is the ability to have Vista automatically adjust column width in any window... why wasn't that in Windows 95? It removes a grain of sand from MY computing shoe, anyway.)

It's impossible to say that 7 bands is better than 6, or that 6 is better than 5.

Oh, I can say it's better! I may be wrong, but I can still say it... ;D You're right, of course... I've heard excellent tri-band units, and at the same time I once saw a 10-band that me salivating... then when I listened to it, it didn't sound right at ALL. It's not your bands, it's how you play them... or something.

With my admittedly odd mix of everything from acetate sources through the latest all-digital recordings, I'm all about re-eq-ing on the fly and creating approximately similar spectral densities from track to track without crushing the life out of it... I'm the nut who splits songs into 8 octaves in Adobe Audition, gently compresses each octave, then recombines the files. It can really bring some older tracks (older like pre-Elvis) to life if done properly!

Helix was based on Zenith a long, long time ago, but Jesse has tweaked and perfected Helix for so long that they have nothing in common anymore, other than the number of bands.

...Sounds like I need to give Helix another listen... thanks for the info!

My new clipper would technically work just as well for AM as it does for FM and Streaming, but there's so much more going on with the sound on AM (such as low pass filters inside receivers, mandated 5 KHz low-pass for IBOC, MONO only etc) that there's not really any appeal for me to make an AM processor.

Yeah, I wasn't thinking of a separate AM processor, just maybe feeding this into a final conditioner for AM; I just have to wonder what that density would do to improve AM coverage (since AM coverage is like 50% density). If somebody ever tries that, I'd love to hear the results!

I have indeed monitored myself on an internet stream! It's pretty cool if you use it to your advantage, because it enables you to hear what your last break sounded like!

loll... no joke! It's literally an instant aircheck session. THAT'S latency.

You answered a ton of questions here, some I hadn't even thought of (THANK YOU!), but I still wonder: will additional presets be added over time?

 

[Tom Wells]

I'll be testing it on my AM tonight.

OK, got it running on the laptop this afternoon, and it sure sounds like best possible FM I can imagine, saying this based on
fully well knowing my own redubb songs' audio. I will be trying it against the current hardware processsor this evening on my part 15 AM,
which I will record. I'll do string of cuts maybe 10 minutes A/B. I will be recording from AM radio detector directly to CD, then
I will upload the result at my podcast site, and post the address later these evening.
Current processor is an ART Pro VLA, used in two passes for one channel. Fist pass is limiting, fast, inf ratio, second pass is slow att/rel at 4:1.
Both passes set threshhold to -13 and share the work.
That said, I have always preferred the sound of fat , but 15 khz AM with no bottom clipped, either.
It will be interesting to see how I roll off what I hear as excess crispness for the my setup on AM, capable of 19khz.
I imagine the webcast version or personal version might work well for me if the CPU usage is less this FM version.
The program used every last cpu brain cell this laptop had, cept 5 or 6, sucked down a battery that would last 4 hours in regular
audio player mode, in 1 hour and 25 minutes.
The laptop was a little mini-heater for the car on the way home, like a wee little hand dryer on the seat next to me.

 

[konbaasiang]

Hi guys!

NightAire and Tom Wells, darn it -- here I thought this product was nearing completion, and you go and give me more work :-D.

Speaking of little mini-heaters, I guess you've lit another little fire inside me, and it isn't gonna go away unless I do something about it.

I'm getting excited about the challenge of AM!!!

So... Tell you what. If you can give me the information I need, I'll do it -- I'll put an AM-mode into Breakaway FM.

First of all, there's the assymetrical clipping. What is it, 99% negative and 130% positive?

Also, what's the NSRC pre-emphasis curve? Is it 75us, or something else?

Also, we'll need band limiting of course. That's easy enough (it's already doing 16 KHz band-limiting) so I can certainly add the option for lower cutoff frequencies. The options will have to be limited in a product of this measly price point -- how does 5.0, 7.5 and 10.0 KHz sound?

///Leif

 

[Tom Wells]

Even though I recorded to CD and only edited once, saved as 192 kbs, I can hear some degrading more than usual.
And there is some hum this time. The comparison is interesting.
It is classic grunge vs precision.
The gain boost is certainly effective, and I can really hear the multiband's advantage in the pipe organ sections.
The apparent loudness is very striking.
Viewing the wave form of the recorded audio, I see how it is well controlled to limit FM deviation.
I feel this limits AM "punch" and would hope that one version has more "room" to breathe for those 125% moments.
I notice it is almost so loud, there is no "room" left to hear the reverb "in".
I just now see your comments.. I love the 16 khz limit. It neatly matches the res of the 192 kb file.
I think it's +125% "Occaisionally" and whatever won't accidently become -100%, ever.
There's the AMAX standard, which someone else can link to. I let it fly to full 20khz, and adjust till it sounds good on every radio.
If you put in brickwalls, they need a "jump ramp" at the upper end of the curve to give good balance overall, or it muddies up right away on
modern limited bandwidth detectors, particularly in GM cars.
I don't think my AM server could handle this and Zara at the same time, but maybe an AM version would be less taxing on cpu.
I think I can make it punch some more. I'll be out of town a few days, but will play some more.
Breakaway really shines on "Islands in the Stream".

For those who don't know, this is airchecked off a Sony 1982 AM/FM table radio and it really is AM 1620.
Here is a link, I hope.

http://thomasjwells.podOmatic.com/entry/2008-10-15T21_59_52-07_00

 

[konbaasiang]

Hi Tom!

I'm listening to it right now -- very interesting! Breakaway FM does sound pretty good on AM, who would have thought!

16 KHz audio bandwidth is overkill though. Tuners don't pass anywhere near that high, and in your recording it appears to be gently rolling off above, plus the recording was made at 22 KHz sampling rate, so the hard limit for frequency response is 10 KHz.

[LOL, Goldfinger! That's been a while, awesome! I love that song (and movie).]

I just realized though, that it's impossible to do the -100 / +125 assymetrical clipping with a standard sound card.

The reason is very simple: Sound cards can't do assymetrical waveforms! Sound cards have capacitors in series with the output (to block DC), and will thus center the waveform whether we like it or not. For FM, I can successfully counteract this by doing subsonic boost, which is what the Tilt control does. For an assymetrical waveform, on the other hand, I would have to keep charging and charging that capacitor until it exploded. Of course, that won't happen -- I will run out of DAC headroom long before that.

So, -100/+125 is out, unless you want to heavily modify your sound card. Sound cards generally use single ended power supplies, so once you get rid of that capacitor you'll still have a 6 volt DC offset or so, which you'd have to deal with.

However, mono support (to save CPU), and adjustable low pass filtering, i can and will do!

///Leif

 

[konbaasiang]

Hm. It's a darn shame sound cards are 192 KHz, and not 1920 KHz.

If they were sampled at 1920 KHz instead, I could have done the AM modulation completely in software, and all you'd have to would be to boost the signal and feed your antenna! :-D

I wonder if there's an easy way in hardware to frequency-shift an AM signal, the way you can do with an FM signal (clip, and filter out the harmonics you DON'T want, if I understand correctly)..

Why is audio so much fun?? :-D

///Leif

 

[konbaasiang]

Done!

Breakaway FM version 0.90.31 beta.

New features:

Selectable Output Bandwidth 5 - 16 KHz
Selectable High Pass Filter 30 - 120 Hz
Mono/Stereo selectable.

Anybody got a stop-watch? ;D

http://bredband.leif.cx/browse/bafm

///Leif "Engineering-while-U-wait" Claesson

 

[Tom Wells]

I re-edited the file, made sure I saved as 192kbps, and uploaded the file again, so there is a new link.

http://thomasjwells.podOmatic.com/entry/2008-10-15T21_59_52-07_00

Then I listen and see that somehow, even though I saved as 192kbps, the player stream is only going to be 96k... blech.

I'll try a download and see if that's 192k.
I don't recall this ever happening before.
I can see the file is now only 21.6mb instead of 43mb, so it must be a 96k.
Guess I need to "go Pro" to get full fidelity. This is NOT making me happy.


The new buttons are good. It is good to be able to mono and cut the audio upper end, if only to hear the result.
I think I'm going to try the personal version next.

 

[konbaasiang]

It took me a while to understand what you meant, but I finally figured it out -- podomatic.com is transcoding your uploaded mp3 to lower quality!

That's pretty annoying indeed -- might i suggest using a format they know nothing about (such as ogg vorbis or flac), or maybe just putting the mp3 file in a zip file before uploading?


Cutting the audio upper end is actually a pretty important feature. Hearing the result is a nice side effect, but the real benefit is:

If you're running the audio processor at full bandwidth (16 KHz), and then low-pass filter after the processor (the way an AM transmitter and radio will), not only would you lose a lot of loudness (because of the energy filtered out) but peak modulation would *increase*, due to some harmonics that were previously keeping the peak level down, being filtered out. It's a double-edged sword, except that both edges are BAD.

On the other hand, if you let the audio processor do the filtering, no modulation will be wasted on frequencies that will be filtered out further down the chain, and absolute peak modulation control is maintained.

///Leif

 

[WNTIRadio]

I just realized though, that it's impossible to do the -100 / +125 assymetrical clipping with a standard sound card.

The reason is very simple: Sound cards can't do assymetrical waveforms! Sound cards have capacitors in series with the output (to block DC), and will thus center the waveform whether we like it or not. For FM, I can successfully counteract this by doing subsonic boost, which is what the Tilt control does. For an assymetrical waveform, on the other hand, I would have to keep charging and charging that capacitor until it exploded. Of course, that won't happen -- I will run out of DAC headroom long before that.

Ok, numbskull question here then... when I am playing back, say, a recording of my voice and I'm seeing it asymmetrical in the waveform on Adobe Audition, what I am hearing thru the monitors isn't actually the true recorded audio? Can it continuously re-center the DC offset with every sample? If the soundcard (I'm using an ASI 6122) can record asymmetry, why can't it play it back?

BTW, the Breakaway FM sounds great! I've tried the other software processors, such as Sound Solution (no way to get that to sound good) and the Sonos III (nice multiband section, just no real clipper) and was VERY impressed by Breakaway. I was skeptical upon seeing the first post, but it is tough to make this sound really bad and easy to make it sound good.

 

[konbaasiang]

Thanks, WNTIRadio!

I agree - the claim of "My processor sounds better than other processors" has been heard before, far too many times, and has usually been less than accurate, and with nothing to back it up!

However, most people who made that claim weren't crazy enough to actually provide a free software tool to simulate the whole transmission path, and provide real world, accurate samples of different audio processors to go with it .


Regarding voice assymetry, I believe that's a little different. The air pressure average does not increase or decrease as we're using our voicec. Analyzing an assymetric voice waveform usually finds an equal amount of positive and negative energy, it's just that one edge is narrower and peakier than the other, so it extends further.

The easiest way to show a waveform like this might be to open Adobe Audition, and create tone, selecting "Inv Sine" as the flavor. You will end up with a very obviously assymetrical signal, yet applying DC correction doesn't fix it. This type of signal can easily pass through a sound card. (Extra credit: Apply a +90 degree phase shift to this waveform with the Graphic Phase Shifter in Adobe Audition, and watch it turn completely symmetrical, looking like a smoothed sawtooth wave).


To be honest, I have not tried the assymetrical clipping, but the problem as i see it is that clipping the bottom edge more will likely not make those peaks "fatter", the way they would have to be to counter the top ones, which extend much further. Unless they balance out exactly, that capacitor at the output will re-center the waveform, and we'll be right back where we started -- a symmetrical signal -- except this time the contents will be unevenly clipped.

At least that's what I think, and it was reason enough for me not to spend the time trying (until I could get the hardware side worked out), but if you have any ideas, I'm listening .

It's not really a matter of not being able to reproduce it -- but rather, reproducing it accurately enough to serve the purpose, which is modulating an AM transmitter assymetrically.

Again, disclaimer -- I have very little experience in AM, other than the fact that the stereo and rds subcarriers on FM happen to be AM, so someone please feel free to step in and teach me how little I know!

Best regards,
///Leif

 

[cgould]

Another thing that the sound card capacitors will cause is the tilting of the clipped waveform of extremely low frequency content. The output would have to be flat to DC to keep from "pinching off" the carrier from time to time by tilted deep sub-sonic square waveforms. I would think your EQ controls would help, but it won't be good enough for prime time, though. Today's solid state AM Transmitters are capable of amazing specs, and to take full advantage, you need audio outputs flat to DC feeding them.

-C

 

[dspxscott]

Thanks, WNTIRadio!

I agree - the claim of "My processor sounds better than other processors" has been heard before, far too many times, and has usually been less than accurate, and with nothing to back it up!

However, most people who made that claim weren't crazy enough to actually provide a free software tool to simulate the whole transmission path, and provide real world, accurate samples of different audio processors to go with it .
Best regards,
///Leif

Hi Leif

You know I admire your programs but there are two key points that need to be made.

All of the recorded mpx clips from other processors have had the processing run through D/A converters (in the processors) and A/D converters on a PC. It is evident that the breakaway clip has a pure digital path from breakaway's processing to the flac file. This makes breakaway appear a magnitude cleaner.
The cleanliness that can be seen in the breakaway clips shows steep filters, whose delay leads me to my next point....

You have already admitted that breakaway has significant delay. Compromises have been made in all of the other processors in order to allow for live listening. Being able to analyse a significant portion of the waveform ahead of time allows any processor to precondition the control signals enough to make the audio shine.
In order to remain credible you need to be honest enough to state that breakaway can not be used in the same manner as the other processors it is being compared with, and I don’t mean in the small print.

It would be great to get a breakaway clip recorded after analogue conversion and even better if we could get a user contributable area of mpx files. I know the BW processors would look spectrally similar to breakaway if recorded through the D/A output with appropriate parameter settings and I would relish the chance to have several BW presets available for comparison.

If you need some bandwidth and storage for files give me a shout.

Let’s get the MPXtool community moving.

Keep up the good work and let’s as the man from Omnia says - 'raise the bar'.

 

[konbaasiang]

Hi Cornelius!

With a brick wall high pass filter on the input, and tilt correction on the output, the output is actually 100% accurate. It can't hold DC forever -- sooner or later it runs out of headroom as the tilt corrector keeps increasing the voltage to counteract the capacitor pulling it back to zero, but it doesn't need to, thanks to the high pass filter on the input. Square waves down to 30hz are *completely* flat on the scope.

Scott, you have very good points!

Breakaway FM is indeed an all-digital path. It actually makes *very* little difference. Even very cheap sound cards (Let's say Realtek HD, which is on board many motherboards today) are capable of a much cleaner output spectrum than the algorithms of most other processors. 99% of the the difference is in the algorithm, not the D/A->A/D path.

However, you're right. Fair is fair, so I will re-record the clip using an analog path, and consumer grade sound cards.

How about Realtek HD audio out (192 KHz) into EMU 0404 USB? The EMU has mediocre input frequency response, so I'll correct for Tilt, Arc and EQ with MpxTool, just like I did when recording the 8500, the Omnia 6 and the 8200, although I used a $1000+ LynxTwo card for those. It wouldn't make much sense to use a $1000 sound card with $199 software though (nobody would), so I'll use equipment in the right price range.

Regarding latency, I've never tried to hide it! When someone asks, I answer quite honestly that it's long -- too long for live monitoring, and that I recommend running a low latency version for feeding the studio.

I gave up on latency because, as has been proven time and time again, there's just no way to make it sound good enough while keeping latency low! Xylophones and saxophones aside, with todays pre-clipped CDs, a phase linear path is imperative to prevent further damage, and the 30 Hz phase linear high-pass filter alone already pushes latency over the edge.

You're absolutely right regarding a user upload area. I'm using the Joomla content management system, and I'll bet there's a way to allow user accounts to edit certain pages. That way, we can have a User page linked from the main torture test page, and I can add links to the main page upon request. I'll look at it as quickly as I can. Registrations will still have to be approved (to prevent spam / abuse) but I'll happily appoint other people to do it besides myself.

///Leif

 

[konbaasiang]

Oops. I'll blame the late hour -- I forgot that the torture test file for Breakaway FM is not actually MPX, but L/R, just like the 8200 (since neither contain a composite clipper). So, I won't run 192 KHz, but rather 48 KHz, and I can't use the EMU 0404 USB since my particular unit only has 1 working input channel (I fried the other along with a tuner when I plugged it into on MPX output and forgot to turn off phantom power, *ahem*).

So, I'll just use plain old motherboard line input and line output (different computers), with rat shack "interconnects". I trust this will be sufficiently low-end to prove my point.

In the name of full disclosure: The built-in stereo encoder in mpxtool (which kicks in when playing an L/R recording) has a brick wall phase linear low pass filter at 18.5 KHz to protect the pilot. However, if there is anything in the signal for this filter to actually filter out, that alone will create overshoots visible on the modulation monitor. This is part of the reason why it's there (to trap unfiltered clippers), and you can still look at the recordings themselves with a different analysis tool.

I may even add LPF on/off and a pilot level slider for the internal stereo coder, now that there's room in the gui for more controls.

///Leif