EXS 24 Logic Sampler Users Group

Rapha\ufffdl TIDAS wrote:
> Can we enjoy songs with 24 exs24
> inserted, 48 audio tracks with vst plug ins inserted, plus emagic plug ins
> etc... Without stepping into TDM's world ?

I don't know - to me it seems that for truly complexed stuff it's still good
to have some hardware (or TDM for that matter) FX around, especially when it
comes to reverbs and compressors.
Right now I am using an Athlon 1gHz and while it's almost sufficient for
complete native mixes when dealing with audio only, I manage to max it out
easily when using VSTis (especially things such as the ES2). And of course
there's still no high quality native reverb for some tasks.
Personally I am thinking about either getting one of those DSP cards
(Powercore or UAD-1) or a new external unit for reverbs only. I seem to get
some reasonable results for small rooms and ambiences out of TrueVerb, RVerb
and FreeVerb2, but as soon as I want some more prominent reverb I wouldn't
know what to use but an external unit.

Anyways, here's some (rough) performance figures based on my system:
- 180 EXS voices without filter.
- 90 voices with filter.
- 250+ B4 voices (yeah, that thing is damn efficient)
- 27 audiotracks, 12 Fat EQs, 2 RenEQ 4band, 2 RenEQ 2band, 3 RenVox, 3
Compressors, 1 Endorphin (MB compressor), 2 RenVerbs, 1 StereoDelay, 1
TapeDelay, around 5-6 B4 voices, around 5-6 EVP88 voices, around 10
(unfiltered) EXS voices. That's what I used for a recent mix and Logic was
behaving pretty well. Of course all those weren't allways playing at one
time.

An even more rough figure from a friends system I recently worked with (an
Athlon 2000+):
- 350+ EXS voices without filter.
- around 220 EXS voices with filter (very rough estimation, had no time to
check more properly).
-  40+ audiotracks (loads of choire tracks and vocal snipplets), Fat or
Ren2band EQ on at least half of them, at least 10 RenVox, at least 10
compressors, *some* RenComps,  lots of delays (stereo and tape, at least
4-5), 3-4 RenVerbs, some Endorphins, at least 50 EXS voices (don't remember
which were filtered and which weren't), one ES2 (playing like around 2-3
voices of a 3 OSC, 2 filter patch), some EVP voices (a basic chord track)
and some other things (Ensemble, Phaser, Tremolo etc.) I don't exactly
remember... as said, that's just a very rough estimation, but it all went
fine and we didn't have to even think about inbetween bounces a single time,
actually we haven't even been watching the CPU meter, we were just working.
Oh well, that was all in 16bit.

My impression is that right now it should allready be possible to run a full
16bit mix natively in case you don't do 100% native synth stuff and in case
you have (or don't need) a better solution for compressors and reverbs.
I would say that in 1-3 years from now (of course you never know what sort
of performance boost those new 64bit CPUs and OS's will bring) you won't
need that much external stuff any longer - assuming that sound quality of
both VSTis and FX is increasing as well, but we're allready seeing that with
things such as Absynth, FM7 and the ES2. And of course assuming there's
further instruments finally coping up with Virusses, Nordleads and the likes
(if you need those at all, that is).

Regards,
Sascha