Leif

Hi Brian!

No problem, thank you for the good questions 😉.

For timeframe, I’m not quite sure. The Winamp versions are ahead on the priority queue (although they’re not very difficult to do, I’ve done many Winamp plug-ins before). Vista programming, however, is completely new territory for me, and may take a while. I’m still debating whether I will actually be able to put up with Vista on my main work machine, or whether I should stick with XP on it.

So, I can’t promise a timeframe yet, because I simply don’t know myself. However, it shouldn’t take too many months.

I have no reason to believe that the native enhancement wouldn’t be available simultaneously. The native enhancements are in fact done as APOs, so basically Breakaway should just be added to the list. That’s how I hope it will work, anyway.

Breakaway will most likely be able to use any format you select.

The current version of Breakaway does not in fact bypass the chosen device. When installing, your currently selected sound card will be the default selected in the Breakaway wizard, and this is indeed the card that will be used for output.

Unless you want to play surround sound (more than 2 channels), there’s really no reason to be afraid of the current version in Vista – if it works reliably on your system, you can really just use it as is.

While the current version uses older Windows technologies (Kernel Streaming if available, WaveIn/WaveOut otherwise), there’s several pieces of code in place to ensure that it does run on Vista as well.

///Leif

Hi Steve!

Thanks for the info.

The Drive control in Volume Logic, actually switched presets. Much like the Less-More control in Orban’s professional broadcasting products, I designed 9 presets (Level 1 through 9) and the algorithm itself interpolated the in-between steps. The presets changed multiple core parameters at the same time, to process harder or lighter.

Breakaway on the other hand really only has one preset per preset. The three compression controls (range, power, speed) each adjust multiple core parameters, but they do so algorithmically – allowing near infinite variation. For example, if you want to retain more original long-term dynamics, but speed up the attack and release slightly, you can do that just as you have done (decrease Power, increase Speed).

I had a lot of fun writing the Bass Boost algorithm. Have you ever tried turning it all the way up? I dare say there’s no other Bass Boost algorithm quite like it 😉.

Multi-discount is not a bad idea. I’ll discuss it with my partner.

It would be very interesting showing it to Steve Jobs. Unfortunately from past experiences, the answer from a lot of people in that position, when you demo it, is: "We can do that ourselves, nothing to it". That is, until they try. Or even worse, they come up with something, and it "sounds fine" to them, and they ship. There’s SO many examples of this.

For example.. Almost every TV has something called "Smart Sound" which is supposed to equalize volume levels. Basically Breakaway’s job description. Does it work? Well, uh.. No.

I agree about the huge loudness differences. It’s nuts! If you look at most music mastered today (or the last 6-7 years) with an oscilloscope, you see how ruined the audio is from all the flat-tops. Pure distortion! Breakaway just happens to contain an accurate oscilloscope.. So if you didn’t already hear it, you get to see it 😉. (Gee, how did that get in there! What a coincidence!)

///Leif

Hi Brian!

1) Certainly! The Vista version, if all goes right, will implement its audio back-end as an APO. This way it will essentially get invited in to process the audio along the path to your sound card, instead of having to elbow itself in the way it does now.

The current version implements a virtual sound card (Breakaway Pipeline) and just has all windows apps use that, and then Breakaway is responsible for processing the audio and sending it to the real sound card. This works fine in XP (and it is the best and only way there), but in Vista, doing an APO is a much cleaner solution, and will work better.

2) I’m surprised nobody has asked this before!

Due to the Windows architecture (all versions), it’s impossible to have a Normal Priority thread, and a true Realtime Priority thread (priority 31), in the same process (instance).

So, in order to get the low latency audio engine stable enough, I had to very painstakingly write the code to support Breakaway running in two different processes, while communicating with one another. One of them is the audio engine, and the other is the user interface. You can easily find out which is which by maximizing the Breakaway window, and looking in Task Manager. The one that uses the most memory then, is the user interface.

If you don’t like having two processes, you can easily disable this feature. In Breakaway, go to the Settings, and uncheck "Audio Realtime Priority". That’s it.

When you uncheck Audio Realtime Priority, the audio engine will run as a Time Critical thread (priority 15), not true realtime, and thus the requirement for a second process goes away.

For reference, the Normal thread priority is 8.

Why should the audio engine run at such a high priority?
It comes down to this — if the system is completely clogged for just a moment, which thing gets to go first? Should it be the hard drive, the mouse pointer, or the audio?

I pick the Audio, because any tiny drop-out (even if it’s just a millisecond!) is very audible and distracting, but having the mouse pointer (extreme example) freeze for a millisecond is completely imperceptible.

///Leif

Hi Steve!

This is a good idea in theory. We don’t have any plans to do it yet, but it’s not impossible that it could happen at some point.

The difficult part is that having an extra device makes it much less portable. It will need to be its own little box, have its own volume control and user interface, have its own battery which must maintain a change, etc.

Getting it into the iPod itself (through Apple) would be a much cleaner solution, but I doubt that could happen – it’s a very tricky Business-Development issue, and Apple does not seem to be a company that likes to let any aspect of their product out of their grasp (as they would be if they licensed someone elses algorithm).

Of course, if Breakaway as it is gets popular enough, who knows – maybe they’d listen. We’ll just have to hope for the best and see what happens. If you want to help, tell everyone you know about Breakaway 😉.

Thanks for your comments! Yes, it was about time to release something again. I’ve been busy with audio processing all this time (since VL) but they’ve all been strictly Pro products. It’s been fun to do something for consumers again!

///Leif

Hi Benny!

I wish it was as simple as that.

However, describing Orban’s AGC only as Window-Gated neglects to mention that it’s actually two-band with variable coupling, both left and right, as well as switchable ratio. It also doesn’t say anything about attack/release characteristics, let alone mention that it’s impossible to make it fast enough and transparent enough at the same time – there are no parameter settings that make both overlap for all conceivable program material. Also, Orban certainly uses some kind of RMS averaging as well, or it would be much much worse than it is.

Describing Omnia’s AGC only as RMS-detected neglects to mention that it’s Wideband with no control signal filtering, making it attack bass way too much, and often making the audio much quieter once deep bass kicks in. Unfortunately, if this problem was rectified, it would cause problems all along down the chain, and manifest itself as extreme clipping distortion at the end when you’re playing bass-heavy material.

An AGC, while simple in concept, is very tricky to get right — especially with arbitrary program material, and a large gain budget (Reference Heavy at Range 50: 20dB AGC range, Infinite:1 ratio).

Precisely how Breakaway’s AGC works, is a well kept secret, other than what you can deduce from listening to it. Listening to it, my ears tell me it’s the best compromise so far. It adjusts quickly when it needs to, stands virtually still when it needs to, and is never caught by surprise – not by bass or anything else. It also doesn’t colour the sound. So, from this one can deduce that it’s certainly not peak-based, it’s certainly not 2-band, and it’s certainly not using the audio as control signal without any filtering.

(Why do I feel like Pinocchio all of a sudden? 😉 )

///Leif

Hi Bernard!

Thank you for the feedback.

Livelike’s WM8 problem isn’t related — that really was a WM8 issue, not a breakaway issue.

However, Breakaway Vista compatibility IS still a problem. I am working on a native vista version (using the new Vista APIs instead of the legacy interfaces which also work on Windows XP and Windows 2000) and when that is done, that will take care of this problem once and for all.

As it is, the audio engine runs with absolute maximum priority – it’s very strange how there are some machines it doesn’t run reliably on, yet it runs fine on others. (And on one of my XP machines, it runs fine with Tiny buffering!)

For now, the best suggestion I have is to open the settings window (the expert mode).

You could try switching the Input device interface to KS (kernel streaming). This should work fine even in Vista, and if the problem was related to Input, this should make it more robust.

If this also doesn’t help, try increasing the buffer size for the Output device. You can increase it well beyond the wizard’s "Huge" size by setting the parameters manually. You might try Buffer size 1024, Buffers 16, or maybe 2048/8, or maybe even 2048/16 (which will cause humongeous delay but good for testing purposes).

You can also try to turn the Audio Realtime Priority switch off. Theoretically, running the audio engine at maximum priority should yield the most robust audio processing, but this theory assumes that every other part of the audio engine also runs at this high priority. I’m not certain this is the case in Vista.

Also, since you’re a Winamp user – the Winamp Output plug-in version of Breakaway will be done before the Vista version, and will be included in the price with the System Wide version.

In fact, once you buy a Breakaway for Windows license, you get to use any or all of the upcoming standard user versions (Winamp Output, Vista Native) on one computer. Licensing-wise, we don’t differentiate between them. So, if you do get Breakaway to run acceptably with a longer buffer size, feel free to purchase it now – the Winamp version will be yours once it’s done, no extra charge. 🙂

The Winamp version will be as robust as Winamp all by itself, because it gets to process the audio *before* Winamp’s long output buffer — it doesn’t have to worry about passing audio through as quickly as possible the way the standalone version has to. Running it in Winamp is a cleaner solution overall – as long as you don’t mind that ONLY the Winamp audio is processed. This notably excludes YouTube videos and watching DVDs.

Please let me know if this helps at all.

///Leif