Need help understanding Racks, Songs and States

Hello,

I have been using Cantabile Performer 3 for a few years now and although I use it live I have never gotten around to setting up Racks and such because I am still having difficulty understanding how to set it all up properly. So far I have been programming my songs this way: Create new song, add plugs I need for the song, create midi/audio routes, edit levels, tweak, etc. and save song. Then add to setlist. So when I load that song everything is mapped out the way I like. Every song has it’s own rack, so to speak. But of course, it’s unloading and reloading the plugs. I do get that…however, I learned how to program my songs really easily and quickly in this manner. And 95 percent of my songs load in less than a second. Only a handful of songs take a few seconds or longer because of samples and/or lots of plugs being loaded. It’s kind of a trade-off, in that when I first start Cantabile at the gig, it loads the last song very quickly and I’m ready to go immediately.

I assume if I WAS using preload and programming the ‘correct’ way it would take Cantabile considerable time to load everything initially at launch. So in this sense I really haven’t suffered, and in one way it might be better programming in this fashion, because I HAVE had crashes and minor instability issues in the past, but recovery was very quick because Cantabile was NOT having to load a bunch of stuff upon re-launch…I guess just a different way of doing things.

FWIW, a side note, I switched from Mac & Mainstage to Cantabile a few years ago when both of my Macbooks died and I had to start from scratch reprogramming in a big hurry. I remember at the time Mainstage taking like over 3 minutes to launch and the OS took over a minute to load (pre SS drives) and Mainstage would crash on occasion, but when it did, it took almost 5 min to reboot the OS and reload Mainstage ! In comparison, on my current system (Lenovo Thinkpad T450s, CORE i5, 250gig SSD, 8 gig RAM, WIN 10 Pro) it takes less than a min to boot and start Cantabile in this way. I imagine with a newer system it would be even faster.

Anyway, I would really like to grasp a good understanding of this rack/state/song thing.

What I’d like to have is maybe a single linked rack with ALL the plugs I will ever use in all of my songs. I want this rack to load when I start Cantabile and preload all songs in the setlist accessing that rack. I prefer to not have to use song states if I can help it. I just want to load a song and play. I’m trying to keep things simple as possible.

I have read the guide several times and still do not have a working understanding of these concepts. Any help is greatly appreciated.

Thanks !
Greg

Auch, that is a tough one. It seems to me we all do it slightly differently when setting up Cantabile, and we all twist the concepts to how we want them to work. But I like your approach about getting a proper grip of the concepts, and I have been struggling, too. I will happily offer my interpretation, I fully trust that more knowledgable and skilled users will correct and add information. One thing I notice is that you mention that you want to reuse one song having all the plugins, that is basically just removing one layer of abstraction from your system - but I just did the same, so I understand why you would want that. I will get back to that part, firstly just the generic stuff.

Song is the main entity or container, a song is normally considered exactly that, all the settings that you need to play a specific song. You organize your songs in setlists.

Song state or song part is used to organize your song in different parts, like verse and chorus, and then you can go through the song parts as the song progresses, to achieve different sounds and settings in each part.

Rack is a collection of sounds and settings, like a module - or a hardware rack. You have inputs and outputs, and you can send these around within the rack as you like. Normally you would either have a rack for sounds, e.g. piano or hammond, or effects, e.g. reverb. Personally I have racks for my main instruments (piano, e-piano, hohner pianos, hammond organ, transistor organ, various synths, saxophone (only effects), harmonica (only effects).

Rack state is used to prepare a specific ‘sound’, e.g. a honky tonk piano, a grand piano, a distorted rhodes, etc. It is kind of similar to have a preset in a plugin, but it allows you to add effects and other settings as well. In performer racks can be reused across different songs, so a linked rack is already loaded, but then it is also sharing settings across the songs which makes it easy to update all songs in one go, but also requires that all songs can use the same settings. An embedded rack is a single copy where you can make all the changes you like without affecting the other versions.

1 Like

And back to my setup with ‘one song to rule them all’. I play in a band where we play new songs all the time, and don’t spend much time refining the sounds or arrangements. As a consequence, I ended up using a fixed set of sounds, mostly piano and/or organ, in various combinations and keyboard splits.

In short, I have instrument racks for my main instruments, piano, e-piano (rhodes, wurlitzer), hohner, hammond organ, transistor organ, some synths, plus audio racks (effects) for my saxophone and harmonica. Then I decided to have a few basic settings, with solo or splits either 2/5, 3/4 or 5/2 octaves.

Based on the above I made a song with the routing to support solo/splits, and I made an AllInstruments rack, that contains the instrument racks (not the audio), and using a copy (actually symbolic links, but I think I will go away from that) I have a main rack and the secondary for splits.

I use LivePrompter to manage the song selection, and made an elaborate ‘protocol’ for selecting the sounds, using a banked program number to send all the information about split, sax/harmonica settings and selection of sounds for the two piano racks. I used reajs to decode the banked program number, it could probably be done in Cantabile, but it was easier for me to just program it.

So right now I have about 130 songs from this band, and only two of them could not fit into this schema. And I have assigned buttons on my keyboard to move between the sound selections, so when we - as usual - bring in two or three new songs for a rehearsal I just need to decide on the basics like what split and initial instruments I think will be suitable, and then I can change sounds on the fly. When I settle on something I just have to enter the correct banked program number in LivePrompter. The advantage is that new songs are very easy to set up (which I needed), the downside is that all our songs sounds alike, more or less.

There are so many ways to use racks and states. I personally would never load everything into one rack. If I did, my cpu load would no doubt peak, and yes I might crash as you have done. Just for example, I use a Piano rack. It has several different pianos, with different settings. In any particular song, I will select which piano I use, by using a state. Each state is determined by song, and I can also change pianos within the song with a midi signal sent by keyboard or footswitch. I think of a rack as a particular instrument sound module, although it can be imagined in other ways. If I use this rack in other songs, it will not be re-loaded, as it is already loaded into the system, and is immediately available. IMHO, putting all your eggs in one basket could really tax your system, unless you let the “All Inclusive Rack” only select certain instruments being used by state, within a song. No need to have everything “alive”, pushing the limits on your system. It is really important to run as lean as you can. It is no wonder you are crashing.
Also, you can search the forum for racks and states, and see how other people use them.

Regards

Corky

1 Like

Here is a good collection of threads on racks & states

1 Like

This is another example of why I’m confused. I thought the function of PRELOAD was to load all the plugs ever used in all my songs ‘up front’ ? What is the difference and benefit in creating and ‘pre-loading’ several different racks as opposed to just ONE rack if they are all going to be ‘pre-loaded’ anyway? I just don’t get it.

Is there a ‘proper way’ to do this ? Do I create a new song then create the rack, or is it the other way around ?

Maybe someone could post a more detailed explanation/example here?

Having read all that I found so far I simply cannot figure this out. I am fairly intelligent yet I feel stupid. The way I have done it so far makes programming my songs so very quick and easy. Only drawback is switching between songs is not as quick as they could be.

Again, any help is greatly appreciated.

Greg

Hi Greg

This is what I’ve observed in my own setup. Others may debunk this, but it is all about the song. Example…in song #1 I use a piano rack, and an organ rack, using 25% CPU. In song #2, I use an amp sim, compressor, reverb, and chorus, using 36% CPU. In song #3, I use a synth rack and a huge Kontakt sampled piano rack using 51% CPU.

If i run them seperate…no problem. But, if I put them all into one song, that adds up to 112% CPU. All the crackling, and dropouts would lead to a crash. Yes, I could put everything into one basket, and pick and choose, which would take quite an effort to deactivate the ones I wouldn’t use. When pre-loading, it’s all in memory, but I only use a small part of that memory when using songs to control what you use. Otherwise, everything is active and pushing CPU. I reuse several racks throughout the set list, so not as many things loaded in memory.

Again, others may debunk this, but that has been my experience since I started using Cantabile in 2016. I averaged 4 gigs a week until Covid, and I learned through experience what works well for me. I was in 6 bands at one time, and each one had a approx. a 120 song repertoire That’s roughly 720 songs that I had to keep up with. With my setup, I could load the night’s setlist for the respective band, and be on my way with no problems. I was always ready for the next song before a guitarist could reset his pedalboard.

BTW…no need to feel stupid. We were all in that position at one time or another, because there is quite a learning curve to this. I still feel that way when others come up with their elaborate bindings. Hope this helps you.

Cheers

Corky

P.S. I usually have a song at the end of each setlist that contains a rack with an organ, piano, synth, horns, in case of requests, or a guest is invited onstage. Those are ready in case I need it quick. They usually cover most anything thrown at me.

The beauty of Cantabile is making it your own. If this works for you, then go for it. I don’t know if there is a right way, or a better way to approach this. Just going by all the threads I’ve read on this site, everyone has their own way of setting things up.

I’ve done it both ways. If I have a song that requires something I don’t normally use, I would probably create a rack for that song. I have also created racks I know I am going to use. When IK Hammond was released, I created a rack for it, and created states within the rack…each state had a different preset, which I could select within the song.

Again, probably because you have so many active vsts. I keep things very lean, and don’t used many sampled instruments. I try to use as many modeled vsts as I can.

1 Like

There’s nothing “wrong” per se with loading one big rack in every song, but it’s a bit unflexible in putting together your songs.

I have dozens of racks that I use in my songs - mostly for different instruments, so I have an acoustic piano rack, a Wurly rack, a Rhodes rack, a string layer rack, a pad synth rack, a couple of solo synth racks, my favorite two virtual sax racks etc…

Each of these racks has different rack states - think of them as “presets” for a synth. So my acoustic piano rack may have different states like “Rock piano”, “Ballad piano”, “Solo Piano”, …; same with all other racks.

Now for any given song, I string together strictly the racks (i.e. “instruments”) I need in this specific song - for a plain piano song, I may simply need one acoustic piano rack, with possibly a string layer rack in addition. For a massive Pink Floyd arrangement, I may need a dozen of different instruments (piano, organ, pads, string machine, lead synth, sound effects from a sampler), so these songs will get complex.

The good thing is that since I have all my instruments wrapped in re-useable racks, the piano rack that is used both in the plain piano song AND the Pink Floyd arrangement will only be loaded once. And, since all my racks (“instruments”) are loaded at the beginning of the set list, loading any given song will be pretty quick.

This makes building your songs pretty easy and efficient - especially when you build your songs from a number of different “colors” - I have a couple of racks (acoustic piano, electric piano, organs) that get used in most of my songs, but a couple of others just get pulled in occasionally when I need that sound (e.g. virtual sax).

I’ll post pictures of some songs when I’m back at my music machine.

Hope things start making sense…

Cheers,

Torsten

:slight_smile: - similar approach here. I have four “rehearsal songs”:

  • Acoustic piano + options
  • Wurly + options
  • Rhodes + options
  • Two-manual hammond with four different setups on buttons (of course Corky’s favorite hammond sound is on button 1 :wink: )

The first three all have different options for the upper keyboard - two different flavors of hammond leads, plus brass and a universal polysynth. I can select the options quickly via buttons on my keyboard.

These four songs always live at the top of my Cantabile setlist where I can reach them with two clicks, so I have a quick standard setup away that I can survive with…

1 Like

Well, of course it is! Where else could it possibly go? :wink:
Thanks for the compliment!

I have to admit, I’ve messed with it a bit - I’ve mixed it with a classic 888 000 000, blending between them with my organizer script. So with modwheel at 0, I have classic rock, and MW at full is all Corky.

This is my ultimate combination - I have the “bare bones” classic hammond lead (with percussion and vibrato on controllers for variety), and with a touch of the mod wheel, I can add texture and overtones to taste. I do have direct drawbar access on my keyboard, but I rarely use it - blending between settings with the modwheel is usually enough and far more predictable :wink:

1 Like

Actually, the drawbar settings were meant to adjusted per style and taste. I was going for that real Hammond feel on the swell pedal. I like how you setup the mod wheel for blending drawbars. I’ve been experimenting with it. You know how I love my E-MU controller, but it has no sliders for drawbar use. I use your "organ"izer script on it now, so I am digging my controller again. :wink:
I am moving this discussion to the organ page since it’s OT.

1 Like

Here as promised some pictures of typical song setups with my racks.

First: a two-manual hammond song:

Both keyboards (upper and lower) playing the same rack, containing the IK Hammond. Add a bit of delay and reverb from a separate “effects” rack, and feed it all into my master rack (contains a bit of routing and my master limiters). That’s it.

Now here’s one of the above-mentioned “Rehearsal” songs:

The main keyboard plays an acoustic piano, optionally mixing a string and a Hammond layer to it. The upper keyboard plays the IK organ (red), the alto sax (orange) or the solo synth (yellow), the routing controlled by song state.

Add a bit of effects, volume control (_MainVolume, _SoloVolume) via MIDI controllers and a master rack, and you’re done…

Here’s a more complex one: my setup for our version of “Shine On You Crazy Diamond” - interpreted very loosely :wink:


A mix of piano, DX7 rhodes and pads on my lower keyboard, strings, organ, synth solo and tenor sax on the upper keyboard. All in varying configurations, with volumes and routing controlled by song states

So you can see that I compose my song configurations of different instruments in racks, combined with effects and volume control racks. It’s a modular approach - like Lego for grown-ups :wink:

My rack states are a mix of “generic” presets (like “Saw Strings” or “TE Live Piano”), which get used in a number of songs, and some sounds dialled in specifically for a song (“Shine On Drone”, “Shine On Intro”. As an example, here is the state list of my Hive_Solo rack:


You see that the rack contains only Hive as the main instrument, plus a lightweight EQ that I use for patch-specific EQing - it helps to cut off too much bass or boost the brilliance a bit.

The patches are a mix of generic (“Moog Lead”) and song-specific (“Radio Gaga Bass”) - I’ll have to clean up the list at some time :wink:

Hope this helps!

Cheers,

Torsten

2 Likes

Mmh, which controller are you talking about? I had (and still own) E-MU hardware. I’m not aware of controllers by E-MU (it’s not a very popular brand here). I have a VOCE drawbar controller, actually I had two of them, one is gone, broken, stolen, maybe sold, who know?
Do you remember the VOCE clones? What a crap!

1 Like

Moved discussion to Organ tips page since this is off topic to thread.

I never get around to ebaying those :face_with_raised_eyebrow:

1 Like

I have mine as well. I now feel as if I wasn’t the only sucker. :wink:

2 Likes

Had the DMI-64 (horrific) and later the V3 (better not so much).
Guess there were lots of suckers, hungry of the Hammond sound.
I achieved best results in emulation, in '90s, using samplers and outstanding libraries (E-MU, Northstar, Best Service, etc.). The real trouble was the Leslie slow/fast transition.

image

image

I also used these until I found “Charlie”, which was a precursor to VB3. The E-MU module was actually pretty good, but was limited to what you got.