Sound design basics with AA 1.5...?

Hi guys,

I'm generally a composer for short and independent films, using Sibelius 6, and I've just started work on a friend's micro-short. For this he's lent me an old laptop that has AA 1.5 on it, and I thought I'd offer to do the sound design as well as the score as it's so short. I was wondering what technical things I should do with Audition right at the beginning before getting all creative? I know to set it to 48000, but have no other idea! Any help gratefully received.

Have no idea whether this will help.  AA, in general, is essentially an audio editor with multi-track record and mixing capability.  As such while it can generate sine waves & noise it lacks any tone generator/synth modules.  Additionally 1.5 does not support either VSTis or Dxis.

While it will load a MIDI file any MIDI capability can be described as rudimentary (which in building on its strengths is not necessarily to be construed as criticism).

That said with regard to recording, editing, mixing audio it is not merely powerful but flexible and intuitive.  AA1.5 does not support direct to disc recording so there is no need to assign project parameters in advance of opening a 'wav' (or other format) to edit.  A screen for the minimum necessary choices pops up when you hit 'record' (bit rate, sample freq.) for the first time in a project .  A project or 'session' does have to use files of the same 'type', bit rate sample freq., but the program will alter offending files on the fly with barely a hiccup.  Since everything is processed to a 'temp' file decisions about where files are stored, what format you want to store them in and at but bit rate sampl freq can be made as you save, after recording or processing.

AA 1.5 does (if I remember correctly) does support BWF 'wav's , so some time markers can be saved in a wav's header which can be useful when exporting a project to other apps.  The number of vid formats supported is limited but I'd typically generate a low-rez compressed AVI as a visual marker for cues.  It does have some SMPTE ability but have always found visual/audio sync'ng for short vids to be easier.  (and typically I don't use AA to assemble a complete, beginning to end, audio composite but work with clips, or chapters and use a video assembling app to put everything together.  But as there is no track limit (other then hardware limitations) there is no reason, particularly on a short film you can't do all the audio work in AA and simply save the mixed stereo file back to video master.  And I have worked on some music videos where the session had at least 96 audio files. 

But, generally speaking, I've found CEP/AA to be far more intuitively flexible then anything offered by Avid/PT (for example).  If you are unfamiliar with the program you probably want to do a little shake out prior to starting 48 hr to deadline caffeine session.  But lot of people here have used the program in all sorts ways so if you have specific questions as you start more targeted response should be forth coming.

good luck

Sounds like fun, Chris.  Just jump in and start playing.  Just bear in mind that if you want 5.1, the stereo mix you make in multitrack is only the starting grid, and you can open the encoder (Cntl +E) to access the surround controls.  Each track in Multitrack View becomes either a stereo or mono 'source' in the surround encoder (depending on how you assign it) -- be sure to use your ears because this can shift the level up to 6dB.

Otherwise, just play with it like Oretez was saying.  And come back with questions.  Good success to you!

Hi guys, many thanks for the help! AA certainly does seem intuitive and easy to use.

However, I seem to have hit one snag. When I exported a completed session as a test, it seemed to be much quieter than the raw audio. Did I miss something in the mixdown process or do I then need to master that file in a package like Cubase (which I have)?

Hmmm, that's usually only a complaint that we see with AA3, which handles volume differently than 1.5.  There's certainly nothing wrong with a mastering step being added, though I wouldn't call it either necessary or typical  Bottom line, if you have all faders at "0" and the master fader at "0," then the volume of tracks won't be changing at all.  So you've found it necessary to lower them somewhere, which is totally normal -- adding many tracks together raises the overall level of the combined sounds, as you'd expect.  Bear in mind that Dolby "Dialnorm" values are typically between -25 and -32 dB, which means that your average movie mix has a lot of headroom and dynamic range, unlike the senseless rape of dynamics done to music CDs and downloads. 

So when you're done, and you "print out" the mix to find that it never reaches "0", then either raise the master fader, master it in something else, or find your new favorite dynamics tool.  But if parts of your mix are plenty loud, while others are quiet, well, that should be just as you'd expect if it matches picture in mood and volume.

It helps to know details and process.  If you are mixing the audio in AA, 'printing' a master stereo track then it is relatively easy, within AA to open that track in edit view see what the amplitude to is, raise it, if necessary.  The track can be compressed (or limited) if 'peaks' on some transients are great enough that the 'main' part of the program is lower in amplitude then desired.  You might also simply be able to raise the level of the master fader.

(it is not particularly unusual for people to confuse 'limiting' the final stereo mix with 'mastering' . . . the file might, or might not, benefit from a 'mastering' stage but Cuebase will bring little (generally speaking) to table not already available in AA.  If you already know what you want to accomplish with a mastering stage, and know that a module in Cubase supports that more efficiently then AA then by all means use Cuebase.  But it sounds as if you are not quite sure what you want to accomplish with 'mastering'.  In that situation, in my experience, I've seldom found that tossing an 'app' included mastering 'pre-set' ever does much to improve the music.  Things like reverb, certain types of distortion & compression used to increase RMS loudness will frequently elicit positive initial responses, but for music meant to be listened to more then once they also, when over used or used unsubtly will increase the fatigue factor significantly.)

If you are saving the multitrack mix straight back to the vid file (no intermediate stereo file), if the issue is not simply a matter of needing to increase volume, you can still insert a compressor before the master out, bring transient peaks down and boost RMS amplitude.  If you are dealing with a final surround mix the details are more complicated but procedures are pretty much the same. 

In addition to the balance between peak transient & RMS levels there are other things that can effect perceived mixed volume.  Any mixer, soft or hard, can employ different 'panning laws' with regard to summing tracks panned hard left and right (and anywhere in-between).  (Not going to go into detail as this is another buzz word that opens another can of worms (hard limiting, mastering, multi-band compression are other topics that create endless debates).  Do not have AA 1.5 on this machine so can't check to see if it permitted options for how panned tracks were summed.  You'd think I'd remember, but don't.  But I would expect that it would.  Particularly when checking a 5.1 mix on studio stereo monitors how this pan summing works can make noticeable difference in perceived loudness of the material (in fact it can even if you're just mixing to stereo, but creating an effective 5.1 is a bit trickier).  Additionally creating a stereo mix can introduce phase issues among individual tracks that detracts from clarity, detail.  Can leave the mix sounding 'muddy' which in turn can effect perceived volume.  The fundamental frequencies for human speech (dialog) and conventional pop music all cluster in a relatively narrow band.  Piling all the fundamentals (and harmonics) on top of one another does not, typically, improve clarity and detail.  Even if you are not fighting phase cancellation issues (summed waves are in fact increasing amplitude) the fact that you've lost the detail of specific transients can make perceived levels seem lower.

If this all seems a bit complicated, it is.  But workable solutions can be achieved well in advance of mental 'clarity'.  In all likelihood you might want to look at 'summing laws' of the mixer and if variable see if one option works better then another. Then a little lite compression either on the master bus or if a file is printed prior to saving back to the vid, on the stereo file itself.

If you could post short samples of the conflicting audio that would also help with practical suggestions.