I tried today’s version hoping I could finally start using it on stage, but a few issues still prevent me from moving forward.
Quite often some notes on the first beat of a Part won’t play, as @Giuseppe_Loffredo mentioned.
In the same scope, exiting a loop is always buggy, as if VL was “surprised” and not ready to play the rest of the score, panicking while looking for the next page in his binder! Should it be a buffer problem?
Sysex are now working almost fine, except when they’re fired as chased events. As an example, I have a Song where the Start Marker is set at 3.1. The configuration Sysex is at 1.0 and is sent as soon as I activate the Song, but its data are unrecognizable by the sound module.
The last issue is less critical. In Actions and Shortcuts I linked Program Changes to Song Select and it is working really fine, but if I send any new PC while the Song is playing, it starts again from the beginning just after the end of the current bar.
I am still 100% Midi and don’t use Audio Tracks at the moment, so my VSTLive based setup hasn’t hit the road yet…
@musicullum , I will send you a simple Project with a single Song containing only Midi Tracks playing Halion Sonic, so you will (not) hear the missing notes!
May I end with a request? It will be extremely useful to have the left view of the selected Song (and just this one) automatically expanded so we can see all of its Parts and watch the progress of each one.
we can reproduce one case (thanks for the project Jihem)
What is your FlexLoop setting, maybe “Part End”, that could be problematic because it raises a contradiction. Will check but would help to know.
3. will check
4. sure, probably FlexLoop is active and then it is supposed to do that probably.
I am not very familiar with the Flexloop settings yet. I will study them tonight. Until now, I set a length for each Loop and didn’t pay attention to other parameters…
Thank you Felician, I have an acoustic piano gig tonight but I will probably spend the end of the night testing all the Flexloop options, as soon as I’m back home!
I wrote “Loop” but I meant Flexloop, whose settings are still a little blurry in my mind… But I should probably get the logic or Parts first, and this is far from obvious. Let’s take this example:
(No Default Layers checked in Preferences)
New Project
Create a Song Track
Add a MIDI Track (no need for data)
Split the Song Track at 3.1.1 using the Scissors tool
→ Now we have:
- Part 1 (D) with Trigger 0003.1.1
- Part 1 without Trigger
- The first 2 bars or the Song are out of any Part, is this normal?
- If I select Part 1 and try to set its Trigger at 1.1.1, I have this error message; Part Triggers cannot be set to top of the Song
- If I put the playhead at 2.1.1 and click on the Trigger of the same Part 1, another error message tells me that Trigger time cannot be earlier than previous Part Trigger
- So I try to drag the Part 1 in first position, above Part 1 (D) → CRASH
- The only thing I can do to fill that void is using the double arrow to move the left border of Part 1 (D) to position 1.1.1, then do a new split at 3.1.1
Split the Song Track at 5.1.1, a new Part 1 (D)(D) is created, its trigger is at 0005.1.1
Click twice on Stop to set the playhead at 0
Open the Layers Tab and create a Layer
Open its VSTi Editor
Play the Song, though it is empty, anyway…
At the end of the Part 1 (D), the Editor is closed and there’s no Layer visible
So my questions are:
If the following Parts don’t have a Layer, why is the first Part’s Layer still sounding?
Why is it sounding during the bars BEFORE the first Part?
Why can’t I edit this Layer’s sounds while I still can hear them during the rest of the Song?
The real is question is probably: Am I completely out?
Indeed, Flexloop is a so cool feature that it’s always active in my projects. That said, I still don’t understand what’s happening.
In actions, Program Changes on channel 1 are bound to Song Select, with no Offset.
I send PC #12 → Song #12 is playing. OK
While it’s playing, I send PC #7 → Song #12 plays till the end of the current bar, then starts again from the beginning. I expected no action, or maybe activation of Song #7, but not the same Song to start from 0.
While from that point all below is true, I don’t 100% sure you’re handling this the 100% right method (maybe SPLIT shouldn’t be allowed for SngTrack? @musicullum ?)
But for now @Jihem : Forget split tool for song track and leave song track as an indicator or your PARTS. So create PARTS as you do w.o. touching SngTrack. Things will work then correctly I think.
It’s the other way around here - Part 1, then Part 1 (D) -, but working with the next to come version.
If however I moved the (D) part to the top, I’d end in a situation you described, this will be fixed too. Who would do that
That was fixed already too. All move/duplicate etc Part actions would crash except for regular cut/copy/paste (use as workaround until you have the next version).
You probably have no default Layers in preferences activated, but in any case: when the Part at 5.1.1 is selected, the editor of the previous Part is closed as its Instrument is no longer active, that is intented. You can activate the upper right item in the editor to have it re-open when its Part becomes activated again.
All in all, thanks for your input, and pls check if all issues werer resolved with the next version (tomorrow, we hope), as it is an important one.
That’s what I have understood.
I’m used to play my Midi tracks on a hardware sound module (Yamaha Motif Rack ES) and I’m now expecting to replace it with a single multi-channel instance of Halion, supposing that it will be less power consuming than an instance per Track, am I right?
For that purpose, I have no other choice than using Midi Tracks and not Instruments Tracks in Cubase, and then opening an instance of Halion as a global VST. Unfortunately, at this point, when I export as VL Project, I lose the sounds and settings of my Halion rack.
After importing the project into VL, I have to create a Layer with Halion, and then set the output of each Midi Track - one by one, a multiple selection would be a real time savior - towards that Layer. The next step is to play the Song and search for the most appropriate sounds and settings for every channel in my unique Halion Layer. If I already divided the Song in Parts to take advantage of Flexloop, editing Halion is only possible when playing the first Part, while the VSTi is still sounding during the rest of the Song. This is a problem to me.
I’m wondering if I should duplicate the first Part, and have a new instance of Halion with every new Part created, but I think it will unnecessarily multiply the load for the PC.
A solution I thought of was obviously to load Halion on a Global Layer, but I don’t know of a fast and simple way to init its instance at the beginning of each Song.
What I probably miss is a sort of “semi-global” Layer at a Song level. All the instruments loaded on this Layer would be available and editable all along the Song, whatever Part is selected or playing.
Or, once again, maybe I’m taking the whole workflow the bad way…
… presumably you would save Halion settings as a preset and then call that up with MIDI.
I’ve been thinking about the way we use VL for a couple of days and I wonder about the paradigm and the views we have within VL when live.
From a keyboard players point of view I’m used to having a setup of my keyboards that itself doesn’t vary - a couple of hardware synths plus some VSTs. I wonder if the ability to replicate that in VL would be of benefit so that the project contains a global collection of layers and stacks - i.e. a virtual rack - that can be saved as a preset and called up by a song. (Essentially I do this already but each layer is embedded in the song and each layers VST already has its own preset.) I understand the logic of having song level layers because it makes a song mostly self contained - but only to a degree if global layers and stacks are being used. I’m sure a lot of us use basically the same set up but tweak it for each song so maybe it makes sense to have all the settings in one place rather than have to duplicate it in each song.
By extension one or more of these virtual racks could contain processing for guitars or vocals.
Does this make sense? Maybe I’m overthinking or just on the wrong track.
Maybe I’m wrong and would love to hear @musicullum (than I will know that for god)
I’m not sure if loading multiple instances of a VST instrument is exactly multiplying the reserved resources. For example I’m using Spitfire BBCSSO a lot for symphonic sounds and while one instance allows to select one instrument only, loading all instances for a symphonic “besetzung”, it does nothing with my resources… I mean it does… because loading such a large amount of samples… but the “empthy” instances aren’t multiplying they “needs”.
Long story short, I’m not sure if using multiple instrument instances vs. one instance with multiple sounds will result mentionable difference. If that above is true, probably would solve your need Jihem
That entirely depends on how a plugin handles its own resources.
Samplers with huge libraries often feature internal global managment of samples, so when you create a new instance, it does not necessarily load the same samples multiple times.
When VST Live creates a new instance of a Layer, it uses the default Layer which creates the programmed Instrument and loads the stored state of the plugin when it was saved, then passes it to the plugin (when saving, the inverse process takes place). This content is a black box just as much as the plugin itself is; it is again entirely up to the plugin what that means in terms of resources of any kind.
When an Instrument is created, VST Live instanciates the plugin by opening it, but no state is assigned because we have no idea what that is. We load the plugins state then, and it will be saved along with the object it was saved with (Layer, Part, Song, or project).
Then I will be limited to 128 presets for the whole project, right?
Just to make sure, when a Song has a first Part from 1.1.1 to 3.1.1 and a second Part for the rest of the tune, do I have to copy the Part 1 VSTi Layer to Part 2 to have access to the instrument editor while listening any location of that Song? Right now, I can hear the VSTi all Song long but I’m unable to tweak the sounds when the playhead is outside Part 1.
