Opinions wanted from those using LASS with DP

[wolfetho]

I recently purchased LASS which I'm very happy with.
I was thinking of using each chair of LASS (A,B,C)
on the same MIDI channel, but a different MIDI track in DP.
Each chair is doubled with the other two chairs.
The reason that I'm thinking of using the same MIDI channels
is to cut down on the number of MIDI channels in Kontakt.
For example, I would record the Staccato articulations
from Violins I (A, B, C) all on MIDI channel 1 but using separate
MIDI TRACKS in DP.

The reason I want to record on different MIDI tracks in DP is to
be able to control each track independently with various CC controllers
for more realism.I could control the dynamics separately for each chair
using the mod wheel for each chair independently.

My question is- if I use this method (recording with same MIDI channel,
on different tracks), is there any disadvantage using this method, as opposed to using separate MIDI channels for each chair? I'm thinking that as long as I use different MIDI TRACKS, I can use the same MIDI channel,
and control each track independently.

Any opinions are appreciated.

Thanks,

Tom

[jroadrage]

If I understand you correctly, you would have several MIDI tracks in DP all assigned to the same output source which would correspond to several different LASS instruments in Kontakt that have the same input source?

Sounds messy to me. You'd basically be taking a nice symmetrical flow chart and tying a noose around the middle. You would have to keep track of A LOT of cc info to make sure that it didn't overlap and end up with some zipper effects or strange cc chasing anomalies. Why are you trying to conserve MIDI tracks in Kontakt?

[jroadrage]

Also, if you're really concerned about MIDI tracks you could always try using LASS in instrument banks in Kontakt. I have all of LASS loaded into my template and I only use 25.

[wolfetho]

John:

Thanks for your input. That's exactly why I wanted to ask
others in case there might be a reason that I hadn't thought of
for not using the same MIDI channels.

Let me explain this situation. I'm sequencing a fast repetitive
string part to create tension in a cue. I'm only using the spiccato
articulation, and doubling all three chairs (A, B, C) playing
each note of a suspended chord. So I have violins I (A, B, C) all playing
an E which is the top voice in the chord. Next, I have violins II (A, B, C)
all playing B which is the next lowest voice in the chord. I repeat this
for each note in the chord. Next, I have all of the violas playing the next noteand so on. So basically I have all three chairs of each instrument playing one note of the chord.

I really notice a difference when I can go in and control each chair independently using different velocities and releases for each section.
It sounds much more authentic to my ears.

How would you suggest that I use the banks in this scenario? Would I just load a multi of say violins I and do a multi record in DP on 3 separate MIDI channels (one for each chair in the section.)? This way I don't have to
make a new pass for each chair one at a time.

Thanks,

Tom

[jroadrage]

Here's my basic setup with a slave PC running VE PRO:

In DP I have 1 MIDI track for every size and section of the strings (Vlns I Full, C, A, B, FC - Vlns II Full, C, A B, FC, etc...) and these tracks output each to a bank in kontakt that holds all of the articulations for that section, along with the CPC script (which passes on cc info for each instrument in the bank when a program change message is sent). I have to use program changes instead of keyswitches (like the rest of my template) but it works pretty well.

It's nice because the setup is very manageable with only 2 kontakt instances and 25 MIDI tracks. Also, every instrument bank is basically identical and I edited the midnam file for Audio/MIDI setup so all the articulations are spelled right out in DP. Being able to control the cc info for every section independently, even if you're just redrawing copied cc1 curves, is really were LASS shines. If you group all the cc data down to one source then you're really missing out on that.

[wolfetho]

Thanks John-
Even though I record all three sections (A, B, C) at once
using multi-record in DP,they're each on a separate MIDI channel so I can go back and control CC data separately for each one. I'll make one pass
using the mod wheel for release times on section A. then I will do the same
for B, and then C.Then I will repeat the same process for velocity.
Then I will go back and repeat the same process using delay and humanization for each section. Is this what you're saying I should do?

Sorry for all of the questions. I'm very new to this whole way of
working.

Thanks,

Tom

[jroadrage]

It's always best to record as much independently as you can but I get pretty good results by:

1. Recording and or drawing the material on the C ensemble track until it sounds pretty good
2. Copy and pasting everything onto the A, B track (you can skip this if you use multirecord)
3. Manually drawing over the cc1 curves in the A and B tracks so that the shape is basically the same but there's some variation in values (+/- 15 or so usually works)
4. Randomizing the attack times and durations for the A and B ensembles (you can almost always get away with more than you think here, I've slid notes over almost an eight note for slower material, ignore how "off" it looks and just judge with your ears)
5. Varying the velocities usually around 20% or so
6. Varying legato time, the shorter the better usually, if you want either a gliss or a portamento try using it in only one ensemble, it can sound a lot more convincing that way
7. Listening back to each track individually and make sure that nothing jumps out as sounding fake.

Everything else like the tuning maps, anti-machine gun effect, release times, legato times and velocity tweaking I only dig into if I really need to, which is rare because the up front programming of the library was very carefully thought out.

It sounds like a lot but if you have some good editing chops in DP then it can go surprisingly quickly and is well worth the time.

[Gabe S.]

Hi.

Just so you know, the next update of LASS will include an "auto-arranger" script. So, they'll provide a multi that allows you to simply play full chords in and the script will decide how to to split up the chord to the different divisi patches in the multi. I know for sure it works with the legato patches which is cool because then you can have your legato transitions for each note in the chord while playing live. I'm not sure if it works with staccato but I imagine it could.

Cheers.
-gabe

[wolfetho]

Thanks guys for your comments.Its amazing how just
varying the velocity levels differently for each section makes
it sound so much better.

What I'm hearing if I don't vary the sections independently it doesn't sound
as smooth on playback. I guess that might be called smearing.There's a harsher
quality to the sound which doesn't sound as natural.Its almost as if the midrange is boosted slightly on the tracks that are controlled together
instead of independently.

Do you guys hear the same thing?

[jroadrage]

Hmm...you mean if you record/edit the A, B and C sections independently the playback sounds smoother than if they are all playing back identical MIDI data? I've never noticed this and I don't know what would cause something like that. Is it possible that there's some duplicate MIDI data that could be causing some phasing?

[wolfetho]

No, I mean if I edit each section differently, using slightly different
velocity levels (c111) it sounds much better than if I edit one section,
and simply copy and paste that into the other two sections,
so all three sections have identical velocity levels.

I've found the same is true for c1 and release times on shorter
articulations. This also sounds much better controlled
independently, than just duplicating the controller info
on the other two sections.

Tom

[Elektroakoustika]

Hey Tom,

Here's a few hints that I've learned through doing stuff like this.

First, I have the divisi sections in my template assigned to different MIDI tracks. When I'm recording a unison line, I use DP's Device Groups to make a MIDI track that records all the divisi sections at the same time into one MIDI track. I've found this saves a lot of time in copying parts to different tracks.

Now as far as smearing and other editing goes, I've set each divisi section's delay & humanization settings to make sure each section is different than the others. This works especially well for legato and if done well, gives the string sound a much more realistic sound. As far as dynamics go, I go into the dynamic curve and adjust the settings so again each of the divisi has a different way it changes dynamics according to how you use the modulation wheel.

Basically I went through all of LASS and asked myself "how can I set LASS up so it saves me a bunch of time?" I think these two things are really effective. Occasionally I'll record each section separately for unison lines, but thats a rare occasion on my recent discoveries.

Hope this helps.

-ea

[jroadrage]

When I'm recording a unison line, I use DP's Device Groups to make a MIDI track that records all the divisi sections at the same time into one MIDI track. I've found this saves a lot of time in copying parts to different tracks.

Do you use instrument banks with LASS? I could never get patch changes to transmit to anything in a MIDI device group. If I could find a way to get that to work it would be a huge timesaver, it would also be nice to have an A+B group for two-part divisi.

[wolfetho]

Using the MIDI Device Group in DP sounds like a great idea.
As I mentioned earlier, I use the multi-record and send
out to all three sections at once (A, B, C),
but I still have 3 separate MIDI tracks.I will have to
check that out.I'm sure AudioBro and NI will simplify and
refine things in LASS with each new update.

[Elektroakoustika]

Do you use instrument banks with LASS? I could never get patch changes to transmit to anything in a MIDI device group. If I could find a way to get that to work it would be a huge timesaver, it would also be nice to have an A+B group for two-part divisi.

I don't use instrument banks. I've thought about it but just haven't quite got into it yet. I currently just have all the different articulations set to different MIDI tracks. Which works fine since I'm usually mixing a lot of sounds together anyways. And sometimes I have half the section on one articulation and the other on another articulation.

Using the MIDI Device Group in DP sounds like a great idea.
As I mentioned earlier, I use the multi-record and send
out to all three sections at once (A, B, C),
but I still have 3 separate MIDI tracks.I will have to
check that out.I'm sure AudioBro and NI will simplify and
refine things in LASS with each new update.

I can't emphasize the delay and humanization settings enough. Those are incredible at making a wonderful sound come out. Really, just spend some time messing with them and you'll eventually get a really nice effect that sounds more realistic than any other string section.

cheers,
ea