• Get involved.
    We want your input!
    Apply for Membership and join the conversations about everything related to broadcasting.

    After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
    If you use a disposable or false email address, your registration will be rejected.

    After your membership is approved, please take a minute to tell us a little bit about yourself.
    https://www.radiodiscussions.com/forums/introduce-yourself.1088/

    Thanks in advance and have fun!
    RadioDiscussions Administrators

NEW FM PROCESSOR

Sgeirk, with a heavily processed composite or L/R signal, to be honest, DAC quality is not much of an issue. Considering how much noise will be added to the signal even in favourable reception conditions, you could probably use a 10-bit DAC and not hear a difference in 99% of listener radios.

It comes down to flat frequency response, starting at DC and all the way up to 56000 Hz. Few sound cards have this - most roll off at the top and bottom - but you can compensate for that with the Tilt and PEQ controls in BaFM, provided you have an oscilloscope for calibration.

I have successfully used all of the following sound cards:

Lynx L22 PCI
LynxTWO PCI
Marian Trace 8 PCI
EMU 0404 USB
Realtek HD audio on-board
Sound Blaster Audigy 2 ZS (PCI)
Sound Blaster Audigy 2 ZS Notebook (PCMCIA)
M Audio Delta Audiophile 192 (PCI)
Edirol FA-101 (Firewire)

With calibration, all of these are basically dead-on accurate: No DAC-induced overshoots.

Dedicated computer is an excellent idea! Make it a server grade machine with a flash drive (no hard drive), and disconnect the ethernet cable once you're done. Those two actions take care of 99% of computer failures ;).

///Leif
 
Mr13013, thank you!

I can't wait to hear your findings :). I developed the basic clipper algorithm mostly by playing xylophone, saxophone, telephone through it and listening to the clipped result in Sennheiser HD650 cans. :)

Regarding presets:

I really want to be able to release a version at a price point that *anyone* interested could afford, even if they're a hobbyist on a shoestring budget. I refused to compromise audio quality or reliability -- so the only thing remaining was to limit preset editing. It's not that much of a limitation, really - you can get very far by adjusting the available controls, which actually each adjust several preset parameters on the fly.

The more I hear this question asked, the more I'm getting convinced that I need to make a full control version too, for the ones that really need it and can afford it (such as small to medium market FM stations). It will have to be much more expensive than the base version, but a brand new server grade pc, sound card and full-control version of Breakaway FM will still add ju to significantly less than the least expensive hardware processor available from competitors. :)

///Leif
 
I'm trying to get my mind wrapped around hooking up the output of a soundcard to the input of an exciter.

I'm guessing you would take the left and right balanced analog output and combine them into a bnc adaptor cable?

Depending on the soundcard, would the coaxial output feed a BNC cable? Sorry for all the questions, i tried e-mailing you and thanks for your help.
 
Hi Sgeirk!

Combining Left/Right isn't a good idea. Just use one of them, leave the other one hanging.

Whether to use balanced or not depends on the input to your exciter. Some exciters have both balanced and unbalanced input.

To go from Balanced XLR to Unbalanced BNC i've personally had good luck tying Cold and Ground together, and then using just Ground + Hot. I can't swear grounding cold is a good idea with every sound card (it's a short circuit), but it did the trick with LynxTWO.

It's really pretty straightforward - as long as the signal gets there with a good ground, there's not much that can go wrong.

However, don't forget to plug that BNC into your oscilloscope before you plug it into the exciter!

Tilt adjustment (to compensate for LF rolloff) and Parametric EQ (to compensate for HF rolloff) is absolutely imperative - you will lose several dB's of loudness otherwise for no advantage at all.

I didn't get your email yet. Please try to resend to [email protected], but make sure to remove both KFC's utensil and any luncheon meats first. :)

///Leif

///Leif
 
so what do you think Hot96? I notice the AGC really sucking up lower passages on some hissier, older titles, I've been able to get it to calm down somewhat. I'm still impressed. It's missing stereo enhancement. Perhaps that'll be in the more expensive, fully adjustable version?
 
i would think if using a card with digital in/out, no correction would be needed.Will try the digigram v222 using digital I/O .My exciter will take spdif,aes or toslink.
 
I should confess first that I am a webcaster, so my wants will be a little different from either an FM or a desktop user. This software can create crazy-clean density, but I'm not trying to squeeze an extra mile of coverage out of a transmitter.

Also, I've been using Sound Solution (a 5-band free DSP with enough controls to keep engineers up for YEARS at a time) for 3+ years with a present somebody else created because I was driving myself crazy setting and resetting and re-resetting the controls.

This preset has made me VERY happy ever since I put it in, and while I'm not the loudest thing on the web I'm pretty proud of our sound: clean, consistent, easy to listen to at soft or loud volumes even as I throw music from 6 decades at it.

It's not perfect, I've often wished for more multi-band dynamic eq, but I had never heard anything to compare.

With an earlier version of Breakaway I wasn't able to get it to install (Vista), but today I was finally able to install both the personal and the FM (.90.26 beta).

I'm sure it's more complicated than this, but it seems to me the presets basically determine how many bands of compression you have and the output eq. Being a multi-band freak (still waiting for that 1st 30-band compressor) ;) I went with the Zenith preset.

With the speed set at "0" the levels were still very consistent but the texture of your source passed through beautifully. Classical was smooth, pop was edgy... it only corrected for dynamic deficiencies & did some gain-riding to make it easier to listen in a noisy environment (for example, next to a computer).

For the most part, despite the slow attack & release, the density of songs stayed pretty much the same... an impressive feat for most processor, which usually require faster attack / release times to create a consistent-sounding average level.

Listening on the consumer version it was impressive how much the system could correct for eq issues. The only downside I heard was, when driven hard and with difficult source, there could be a split second of "was that distortion?" but it was too fast to even trust your own ears and rarely repeatable.

My conclusion was that #1 it handled dynamics a little differently from Sound Solution & that's something I'll have to listen to a while to decide if I like or not, #2, it was a HAIR dirtier than Sound Solution, and #3 it was better at on-the-fly re-eq-ing than Sound Solution. Over all, very close.

Then I switched to Breakaway FM.

Hey... this thing has a drive control! ;D

I backed the drive all the way off & the distortion (I thought) I'd heard before on the consumer version DISAPPEARED & hasn't been back.

From time to time I will think I'm hearing distortion in the bass, but I'll switch it off and / or switch back to Sound Solution and realize Breakway is just pulling deeper bass to the front, and if that deep bass is distorted, you're going to hear it. If I was willing to back off the power and / or range I'm sure I could minimize that.

I can't afford the FM right now, but I am really, REALLY looking forward to the webcaster's version & finding out what the price will be. Absolutely spectacular work... a friend was raving about this software this morning, and I think now I'm convinced... this may be the best processor EVER.
 
Interesting observations, NightAire!

I wonder if it's a matter of IM distortion. Sound Solution has no shortage of both THD distortion and severe aliasing (due to using non-oversampled clippers). However, it has extremely little IM distortion because of this.

Breakaway Personal uses its limiters very sparingly, but it's still look-ahead limiting, so if you're very sensitive to IM, that would explain what you heard.

Breakaway FM, on the other hand, uses oversampled advanced distortion cancelled clippers. Very little IM, and very little THD and Aliasing :).

The Webcaster version will have look-ahead limiters like Breakaway Personal, though. If you want to avoid IM, Breakaway FM is the right one to go with, even for webcasting.


By the way -- NEW VERSION OUT:

Version 0.90.28

New features: Winamp DSP Plug-in support, for using effect plug-ins and encoder plug-ins directly from Breakaway FM, without needing any other programs!

The feature has been in since 0.90.26, but it now has a brand new gui for loading the plugins, instead of having to edit ini-files.

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

///Leif
 
Installed software on dedicated server.works great with digigram v222 card.Really fun to play around with.Nice..
Some of the best code i've heard.Presets work great.congrats
 
Before I play with this, I've read a comment or two on www.stationplaylist.com's yahoo chat board that this program is causing 100% processor cycle usage on some fairly fast computers...can anyone comment on that before I play with something that sounds like "must try" software?
 
program will check your cpu for adequate power.i ran it on a HP vista (UGH) with no problems,but moved it over to a dedicated dell server i wasn't using.i'm running the full fm version,the home version is very light on cpu.hope this helps.
 
Hi Bob!

There was a bug in all versions up to 0.90.26 which could cause it to use 100% cpu while using the Test feature in the Configuration box.

It's fixed in 0.90.28.

Other than that, Oldiesstation is right - the installer will run a benchmark to check for adequate cpu power before modifying your system, so nothing lost by trying!

///Leif
 
BobOnTheJob said:
fairly fast computers

like what models of CPUs? A Pentium4 @ 3.4gHz is just NOT a fast computer anymore. Also - gHz means NOTHING anymore. ;)

An Intel Q6600 @ 2.4 gHz will walk all over a Quad P4 setup (at any speed), for instance... and it only costs $175 for OEM. Anyways, I just think that some people's ideas of "fairly fast" needs updating. ;) There's CPUs out there that walk all over the Q6600 now too, and there's mainboards supporting up to four of em!!! (yep, 16 cores)
 
I'm about to race it tomorrow. I already have it running on an Xfi Soundblaster on a P4 3.0gig w/1gig ram...very little running in background.

No issues so far...i'll see if it outputs to the exciter O.K....it already sounds fantastic running the line output. I'm extremely curious to scope it out then throw it into an exciter.
 
Leif,

Knowing me, I'm probably hyper-sensitive to IM... of course, then why don't I pick up the THD & aliasing in Sound Solution? Stupid, selective brain. :D

You can imagine where I have the Sound Solution clippers set (+6). I have tried a couple of times running without them at all but I really like having SOME kind of brick wall so I can set my levels and know I'm not going to overdrive my encoder.

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 so that when I decide I've been up too many nights fiddling with it and I just need to SET IT and get some sleep, I can set it and walk away? ;) (I think my only concern might be speeding up the attack times while leaving the release times alone, if it didn't mess up the sound.)

My friend & I were discussing Sound Solution this morning and he said it always sounded "plastic" to him, "not real," and Breakway sounds "real." I feel like Sound Solution can sound brittle and if pushed for heavy compression it sounds very pumpy, breathy, strained, just not pleasant at high densities.

Breakway sounds GREAT up to and including "brick wall" settings... as a matter of fact, for fun try setting everything to 100% and listen to something with lots of gaps like Queen's "We Will Rock You" or Gwen Stephani's 'Hollaback Girl"... despite the hysterical increase on the volume between the beats... the system STILL sounds like it isn't breaking a sweat, and somehow manages to not sound like it's pumping and breathing... I'm guessing it has to do with the look-ahead design... really impressive!

FWIW, I had zero problems running the FM on a 2.4 GHz quad-core with 2 GB of RAM using Vista. In fact, I left it running and surfed, ran Winamp, etc with no issues.

Can't wait to try out the new version!
 
NightAire,

Knowing me, I'm probably hyper-sensitive to IM... of course, then why don't I pick up the THD & aliasing in Sound Solution?

Good question. I can think of at least one very famous processing guru who has the same ailment, so you're in good company. :)

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.

the system STILL sounds like it isn't breaking a sweat, and somehow manages to not sound like it's pumping and breathing...

Thank you for noticing :). This is what sets it apart from any other processor I have ever heard. Some might call it FM -- F****** Magic! It isn't though. It's all science -- and there are good explanations for every single thing that goes on.

The ironic thing is how complicated the algorithms get, when you're aiming for "Natural" sound! It's highly counterintuitive. You'd think the shortest, simplest signal path would be the cleanest! With certain methods of measuring (such as Impulse Response) a simple clipper would measure the very cleanest. However, as wise men have said, Music is not about Sinewaves!

The CPU utilization is one indicator of the sheer amount of things going on under the hood in Breakaway FM.

///Leif
 
I'm feeling silly now. I tried earlier in the week, but could only get audio when I selected the soundcard as the input.
Then I heard feedback from/through the computer (laptop) microphone. I uninstalled, did a massive defrag sesssion, badly needed anyway,
then tried again tonight with lots more room to spare.
Still same result, sound card is a Soundmax in a HP/Compaq 8230.
No errors, just no audio despite each option of wave, DS, KS, making sure input device is selected.
Earlier in the week I was trying to feed audio from Zara automation, tonight I was trying using realaudio.
Never saw any line in signals or waveforms.
I might like to try the webcast/consumer version to run on my pt 15 AM server (different computer), but it may not have enough stamina.
I've been able to make computers work at most things, what I am doing wrong here?
I mean, I go all the way back to punch cards and 40k of hand-woven magnetic donut-n-wires CORE memory with
real write and sense lines...
So I hafta wonder what may be blocking the internal selection of the I/O from my wave output..
I hear the audio from real player, but see no levels in meters.
From Zara, I don't think I heard any output, even the "source" audio through Zara's output.
 
i had the same issue, within whatever playout program you're running, you should choose breakaway pipeline 1 2 or 3 as the output sound device...then within the breakaway i/o setup...choose whichever breakaway pipeline you selected in your playout software.

good luck.
 
Tom, Sgeirk's advice is spot on!

Breakaway Pipeline is essentially a virtual cable.

If you connect one end of Pipeline 1 to the input of Breakaway FM, you must also plug something into the other end of the cable, namely your player or automation system :).

If you want to feed ALL audio from any program in your computer to Breakaway FM, you can do it in this manner:

Control Panel -> Sounds and Audio Devices -> Audio -> Sound Playback Default Device: Breakaway Pipeline 1

If your computer supports Kernel Streaming (KS), I recommend using it, since it then bypasses most of the audio back-end in Windows, resulting in a much cleaner and trouble-free signal path.

///Leif
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom