Why no cheap or free audio editors for Windows?

[headquest]

Okay, in that case Tracktion is not doing anything different after all. I believe that most hosts convert to 32-bit floatin point etc seamlessly/invisibly from the user's perspective (it's in the small print!)

Funny how much this stuff is just 'invisible' to me. .

Its invisible because Tracktion does its job well.

But if this is what is happening, then so are most of the hosts doing it well . All good stuff

However, in a 'proper' mastering situation the DAW used should be capable of editing an already mastered 16-bit file without having to change the bit-depth and re-dither. Tracktion can't do that because everything becomes 32-bit when imported.

Same in Audition, I have to admit...

Also, Tracktion's on-the-fly sample-rate conversion is probably not the very best available..

Can you say more about that?

Like Lunch Money, I agree that this has turned into an interesting and educational read!

One thing IS puzzling me though...

In Ableton Live 5 when I imported some OGG files LM sent me, they didn't convert to 32-bit .wav; they stayed as OGG files, running alongside .wav files in other tracks. The same happens with MP3 files in Live too. So conversion is not taking place... Live really does appear to be running a whole pile of files in different formats simultaneously

Guys, am I misunderstanding something there?

[headquest]

destructive editing is now in almost all hosts. Cubase/Nuendo, Pro tools, Sonar, Digital Performer... the only one I can think of is Logic, which is pretty sad.

Well, my other hosts are Ableton Live and Tracktion 2, neither of which do destructive editing. Pro Tools of course does... I didn't know those other ones did, so thanks for the information

I think that illustrates the point that the features one needs in an audio editor are partly linked to which host they are using. Live doesn't do surround or video for example, but I could do them in Audition (very well) if I needed to.

As a general rule I have chosen to use a range of "specialist" tools for different tasks rather than one "bog-standard comprehensive" that tries to be all things to all people. (Hence using Sibelius rather than choosing a sequencer that has crappy second-rate notation bolted on )

snapping to zero crossings is a feature of most host apps, but you need to read the manual usually to find out.

Again, not a feature of either of the hosts I use. See above!

[IIRs]

Also, Tracktion's on-the-fly sample-rate conversion is probably not the very best available..

Can you say more about that?

sample-rate conversion is a complicated process, and can use a lot of cpu if done at the highest possible quality.. I assume that Tracktion's algorithm was coded with one eye on the cpu %, but for mastering purposes I would always want to know that I am using the best sounding algo regardless of cpu consumption, as I will likely be processing it offline anyway.

[headquest]

Thanks IIRs - that makes good sense. Just out of interest, what do you personally think is the best sounding algo (regardless of cpu consumption)?

[IIRs]

In Ableton Live 5 when I imported some OGG files LM sent me, they didn't convert to 32-bit .wav; they stayed as OGG files, running alongside .wav files in other tracks. The same happens with MP3 files in Live too. So conversion is not taking place... Live really does appear to be running a whole pile of files in different formats simultaneously

Guys, am I misunderstanding something there?

Yes you misunderstand! the files are converted on-the-fly, ie: Live or Tracktion will either create temp files at the required samplerate / bit-depth, or convert the files in realtime on their way into the audio engine (using cpu cycles in the process)

Oggs & MP3's can't actually be directly played anyway, as the missing data needs to be recreated by the de-compression algo before it can be sent to an audio output..

[IIRs]

Thanks IIRs - that makes good sense. Just out of interest, what do you personally think is the best sounding algo (regardless of cpu consumption)?

I haven't done any critical listening to be honest, as I always stick to 44.1 or 88.2 myself.. usually I just convert in Soundforge because I only routinely do that for ringtones but I have a copy of R8Brain somewhere in case I ever need it (the specs looks good on paper iirc)

[headquest]

Aha! Thanks - an excellent explanation. Now makes perfect sense! And combining the points in both your posts, I assume too that when you are using OGG or MP3 the de-compression algo is tapping into some cpu as well.

Also, that explains Greg's point before about appearing to be using a range of different files simultaneously in T2.

Cheers!

[Lunch Money]

I'm quite certain that T isn't 'converting' when you import. I drop an OGG file into an edit, and it's there instantaneously. Doing a conversion would take at least some few seconds, probably more.

It must be "on the fly", but I'm not knowledgable enough to know how many steps of 'conversion' are going on before it hits the output. I know that it's instantaneous and I've never noticed (doesn't mean they're not there, just that I've not noticed!) any CPU spikes from using multiple filetypes.

Greg

[IIRs]

I'm quite certain that T isn't 'converting' when you import.

what do you think "preparing to play compressed or networked files" means? (Try opening a demo edit that you havent opened before, if you have one..)

<edit> what I mean is: in certain situations Tracktion is quite evidently creating temp files in a format that it can play directly..

[Lunch Money]

Different thing, though, IIRs. Drag an OGG file from your desktop into an edit. No "preparing to play compressed or networked files".

When it's preparing to play compressed or networked files, it's simply unpacking an archive. I'm not definitevely saying that it's NOT creating temp files EVER, but I haven't seen the evidence that it is, with the "preparing to play..." message being the only hint that it MAY happen sometimes (not evidence, per se since I don't see any temp files in the project directory and can quite safely wipe out my temp file directory). Dragging OGG from desktop certainly doesn't do this, though.

So basically, we both have educated guesses going on.

Greg

[IIRs]

So basically, we both have educated guesses going on.

Of course But the waveform display that appears on your imported ogg file must be created from a decoded WAV version, just as the audio that is sent to the outs must be: that single wavey line simply does not exist in the compressed version..

[Lunch Money]

Well, I've come full circle and now I believe you're right. Not being sarcastic, I actually DO think you're right after all. Mark this day in your calendar as the day Lunch Money graciously admitted (instead of grudgingly) that IIRs is right and that he's wrong.

OK, not TOTALLY graciously-- the reason for my change in heart is that I converted .OGG to .wav and it only took a few seconds.. like, 2 and a half.

I had it in my head that it would be a similar length of time as converting .wav to OGG, which takes considerably longer.

Given that short a conversion time, it seems quite acceptable and reasonable that it IS converting as it imports, even from drag'n'drop.

Greg

[IIRs]

[robenestobenz]

f**k me, someone admitted they were wrong on an internet messageboard. My deep respect. Flying pigs are tapping on my window, I must go now.

[BONES]

Normalising is pointless, or worse.

Normalising is essential to ensure the maximum use of available bit-depth. If you are not normalising your audio you have wasted bit-depth which equals lower quality.

There's not a thing in a "mastering" checklist of things to-do that an editor can do better, except perhaps export to a variety of file types, which to me is NOT part of the mastering process.

... except normalising.

There's not a thing in a "mastering" checklist of things to-do that an editor can do better, except perhaps export to a variety of file types, which to me is NOT part of the mastering process.

... except normalising.

To master in your audio editor would take the same amount of resources unless you have already rendered a stereo "mixdown" from your host.

Actually, because your editor is not real-time it will always use 100% of availble resources to do things as quickly as possible.

TBH I use CE2k far less now that I used to but I still think that there are too many things that it does better to ever consider being without. But what my host offers is the ability to do many things that may once have tempted me to spend big bucks on WaveLab or Audition to get through. Its just a matter of thinking things through a little more to come up with optimum workflows.