Very basic questions about midi to audio, and reverb

[thinkingMusic]

First, please forgive these really dumb questions (my excuse is that although I've been composing for years, I'm completely new to using hi-end samples):

I'm using VSL through GS2.5 and Cubase SX -- on one computer (P4, 2.8GHz, 2Gigs Ram, 2x200Gig HDs) -- and using (for now) only one In and Out on my sound card. Since neither GS nor my CPU/RAM will likely be able to handle a full orchestral ensemble all at once, I've been told that I will have to record sections to audio as I go along.

Is the standard approach (with my set-up) to record all the sample tracks that make up one instrument to one audio track?

If so, how should one go about handling reverb -- via each instrument's audio track? Or by then recording a number of these instrument audio tracks as an orchestral section, and then applying reverb to each section's audio track? Or by applying reverb to many sections, so that they receive the same reverb treatment?

Along similar lines, I notice that the "Release Control" in Performance Legato + Release Control samples seems to incorporate more natural reverb than is present in the non-RC samples. Wouldn't this necessitate recording each sample's track to audio, so that its own reverb characteristics can be taken into account, relative to the reverb added to other samples used for the same instrument?

Again, please forgive the vast expanses of ignorance in this post -- I'm new in these parts [[;)]]

Thanks, in advance, for any tips you'd care to share.

[david ender]

Hi Michael,


I'll try and give you some answers, but as always with these matters, there may be different opinions or approaches. btw, your questions aren't dumb, and most of us had to ask them some time ... [[;)]]

1. Yes, you would record all Midi/GS tracks for one instrument into a single audio track.

2. I normally route my audio tracks through a single group channel with a reverb insert (SIR for the cheapest solution) so every instrument gets the same homogeneous treatment - also, that's time- and RAM-saving. Sometimes I add some band saturation for a warmer and less pristine tone. That way, I can still add "special effects" to single audio channels. And if you want a single instrument to stand out distinctly, you can also route it to the master and give it its own reverb etc.

3. I can't answer your release control question now, but I think that the reverb applied to your audio tracks would smooth over the differences especially if there are other instruments, too.

I hope that helps!

Cheers,
David
VSL manuals

[thinkingMusic]

@david ender said:

1. Yes, you would record all Midi/GS tracks for one instrument into a single audio track.

2. I normally route my audio tracks through a single group channel with a reverb insert (SIR for the cheapest solution) so every instrument gets the same homogeneous treatment - also, that's time- and RAM-saving. Sometimes I add some band saturation for a warmer and less pristine tone. That way, I can still add "special effects" to single audio channels. And if you want a single instrument to stand out distinctly, you can also route it to the master and give it its own reverb etc. . .

I hope that helps!

Thank-you, David -- that certainly does! [:)] You've very clearly answered my question regarding recording midi to audio, and given me my homework on reverb. I'll look for some reading material on SIR, "band saturation" (EQ, I presume?) and group channels.

Thanks again!
- Michael

[Dietz]

I think "tape saturation" is the common wording. David thought of plugins like Steinberg's "Magneto", for example.

[thinkingMusic]

@Dietz said:

I think "tape saturation" is the common wording. David thought of plugins like Steinberg's "Magneto", for example.

Thanks, Dietz. . . as I begin to learn the lingo [[;)]]
- Michael

[david ender]

Sorry, my mistake - the "band" crept in from German ...

Cheers,
d.

[thinkingMusic]

@david ender said:

Sorry, my mistake - the "band" crept in from German ...d.

Actually, I'm totally impressed with everyone's English there! Did all of you spend time in anglophone countries, or is this level of multilingualism actually the norm in Austria?

- Michael

[david ender]

Hi Michael,

it's not really the norm here in Austria, but you're in luck with the VSL people [[;)]]

I myself was born in Australia (although I spent most of my life in Austria), and apart from being a translator, I have a special interest in languages and "English" things.

Cheers,
d.

[Jonathan Mitchell]

Michael, If your computer is beefy enough, you can get an smallish orchestra all together in a single MIDI rendering, rather than recording audio for each channel. I use a 3.4Ghz, 2Gb memory 250Gbx2 drive configuration with Sibelius 4 (or Cubasis) and GS3. I can load up about 120 GS samples (make sure you use the memory tweak trick - see the VSL GS forum). This means I can run 16 MIDI channels with a small orchestra format (Ft, Ob, Cl, Bsn, Tpt, Hn, Tbn, Timp, Perc, VlnI/II, Vla, Vc, Cb) together with a solo instrument if required.

I prefer to hear the whole thing together when I am polishing my own work. Since I always use a small orchestra format for my work, I can usually get away with it. However, you need to make sure that you are using every sample with no 'passengers' loaded that you do not use. My computer is at full stretch with this so I always record the audio through line-out via my amplifier into my old computer.

[PolarBear]

I once ran a full orchestra off my single PC (using Opus 1) with smaller specs than yours: AMD 2500+, 1.5GB RAM, two HD drives. Actually my latest composition didn't even fully need the hardware I got in this one PC. So my bet is you can do this, too! [;)] Although this was the MIDI/GS only version, no effects or mixing added at the composition stage. Just one temporary GS NFX reverb that was replaced later for the final mix you can hear at the link by 3 instances of SIR, quite some EQ and correct stereo positioning (composed everything dead-center). Which I wouldn't recommend because I also feel you'd get more out of a piece if you can hear the final rendition. There are more effective convoluters than SIR out there that do a lot better job and are suited for live use.

But you're right, once you have templates and want every step in the process of creating accessible at any time, things get a bit more power hungry. With deadlines around I guess I would alter my process, too.

Just my 2 cents,
PolarBear