Can you share a screenshot or copy of the webconf homepage?
AFAIK, all the options i referred are available in both, Oram & Vangelis versions.
1 Like
You are right. I never saw it. Funny to still discover new things after a year 
4 Likes
Ok, so I found the bold press options for loading Z3S – cool Thank you very much for pointing that out!!!
But there still is something I don’t understand:
– Load Replace ChainS (plural : ) replaces ALL chains.
– Load Merge Chains adds the chain to the existing ones. Close, but not there yet
– Load loads Everything
I see why you’d make a Difference between loading everything (including Sequences) and loading almost everything (not including Sequences), but then, “Load replace chains” will overwrite everything else you had set up in the project. It may replace the selected chain and keep sequences for that chain, I suppose, but if there’s more than one chain in your project, this option does not make too much sense, because it’s overwriting all the other chains. Or am I, again, missing something fundamental?
…or is it another bold press on the “load Replace Chains” Option?
I don’t understand. What do you want to do? I thought you wanted to load chains separately, so we explained this can be done with load-replace… but suddenly you want a different thing?
Please, clarify your use-case .
Regards
use case:
I am just missing a Save/Load Level between Processor Preset-Level and “complete State of the entire machine including all other chains too“-level.
1 Like
In your video, you selected “Load Replace Chains” which as you have already noted (plural) replaces all existing chains with the new snapshot. If you select “Load Merge Chains” then the existing state remains intact and the chains within the loading snapshot are added (merged) which sounds like what you are requesting. There is not a “Replace Chain” but you can do this in 2 actions by deleting the chain the merging chains.
As previously requested - explain what version of firmware you are running and what exactly you want to do… and RTFM.
[Edit] Please be aware that loading snapshots is a slow action. It is not intended for live use. It should ideally be done at the start of a session / performance. ZS3 is the in-performance mechanism for modifying the state quickly.
1 Like
Yes. That merges it. The new chain gets added to the right in the mixer.
But I tried to replace that one chain.
I plan to prepare a handful of complete Instrument chains with their custom FX chains and everything, and then use the “LOAD REPLACE CHAIN(s)” to load an entire Instrument chain when I need it. I know, I could merge and then go back in, delete or mute the chain I wanted replaced, and use the new, merged one. This might look like a tiny inconvenience, but in a dark, hazy and chaotic club situation, where I need all of my brain’s bandwidth on the modular synth, every additional button press or menu dive is a new source of failure
So, sorry for being a bit picky.
I know, you can build entire multitimbral setups and recall them, but my case is a bit special in the sense that when I start a track on stage, I don’t know yet which instruments I will need. I have no “Songs” in the classical sense. It’s all on-the-spot Improvised Jams. Where sometimes I want two channels of Rhodes and a Polysynth, then I’ll need to add a drum sampler, than swap it for a DX7 or something for the next tune. I am looking for a way to quickly build, change and evolve the Chains setup in the moment, from a list of prepared chains. If that’s not possible, what other options do I have in the Zynthian ecosystem to achieve such a thing?
You may want to consider it this way:
If part way through your set you decide you need a DX7 sound instead of a MiniMoog sound, you probably wouldn’t run off to the van, grab the DX7, run back to stage, derig the MiniMoog and rig the DX7.
Similarly with zynthian, we advise against changing snapshot mid-performance. You are asking for something that we purposefully do not implement or support.
Snapshots are for whole performances. You should consider what instruments you may want to use during your performance and load them.
1 Like
I see… Well, that does kind of make sense
In that scenario, I’d probably have both the Minimoog and the DX7 ready to go. Sitting on top of a Rhodes and another Hammond B3 Rig to the side. To expand your analogy, my next question would be: Can the stage handle the weight of a DX7 AND the Minimoog and the Rhodes and the Hammond at the same time? → Do muted Chains still use up CPU?
…if not, there might be a “Mute-state Management”-route I could take then
I have a rig with six implementations of fluid synth which are selected by zs3’s and loaded from one snapshot. The levels are modified to enable the solo instruments or combinations, almost like drawbars on an organ …
Snapshots should be considered like loading samples in from floppy disks as we used to do back in the day.
There is of course resource constraints and this is something that each user needs to consider for their configuration. Carrying on the analogy, if you run out of DX7 voices then you may need to buy another DX7.
There is currently no way to disable resource usage for individual engines or for whole chains. There is a ticket tracking a request for this functionality and I am looking at that but it is not a high priority.
Most synth engines ramp their resource usage based on actual usage, i.e. the more notes played, the higher the resource usage. There are a few engines that have constant (full) resource usage. setBfree is an example. Audio processors tend to be a mix of these, using resources and ramping them further based on the amount of processing required.
Hi @nemoy. I am afraid my friend that, on such demanding premises of procedural beats without a laid out arrangement plan, spurted on-the-fly by exoteric eurorack modules with CV-Midi converted for the external wilderness, the Zynthian might not be for you, since it is neither Ableton nor MaxMsp, and does not aspire at being that.
Going from your words, I suggest you consider building yourself a further custom and headless Pi5 Zynthian, besides your official kit. That would easily give you 8+8 midi channels/chains without audible xruns, even at high sample rates and short sample buffers. You could then switch live, with four incremental +/- pedals, between 4 Midi channels with fixed presets for each of the four Midi sources that you are planning to use. Old style, everything set, patches always available and very reliable, no program changes, no glitches, no compromise on release trails.
Or… You could tap into the Euclidean powers of Pure Data (available as a special processor/generator), and make a virtual eurorack of the Zynth itself!
Just ideas.
Cheers.
2 Likes
I mean, @nemoy’s use case is not that eccentric. Let’s remind ourselves what the four chains were described as:
1x Pianoteq for rhodes, wurli & Pianos
1x Hammond Organ
1x Polysynth
1x Sample Loop player for little percussion loops. (havent gotten this far yet)
Neither of these has complex interactions with other chains, nor does it require on-the-fly switching between different processors.
If a ZS3 could be configured (say, after initially saving it) to only apply to certain chains in the Zynthian state, then this would solve his issue.
But I wonder if ZS3 is actually the right tool in this case. It seems to me that presets would do everything that @nemoy needs here, and perhaps it is just a question of how quickly the relevant ones are accessible.
Have you looked at preset favorites? Any preset can be marked as favorite, including user presets that you yourself have created. These then show up in a dedicated list of favorites, which is accessible in the “Banks” view of the preset selector (press the “Preset” button again when already looking at a preset list).
The question then would be whether selecting preset favorites is a fast enough workflow for your use case. Selecting a favorite requires three steps:
- Selecting the relevant chain, by encoder wheel or tapping.
- Bold-press the “Preset” button to get into presets. If the “Favorites” bank was the last one you were looking at in the preset selector view, then bold-pressing the “Preset” button from the mixer view will take you directly to your list of favorites.
- Select (probably tapping is fastest) your preferred preset.
- Press “Back” again because loading that preset will open the parameters view, as opposed to taking you back to the mixer directly.
This is four interactions, whereas a nice list of ZS3 presets could be one instead if it did what you’re looking for.
It would be a nice touch for Zynthian to optimize this workflow, e.g.:
- To allow the the “Favorites” bank to always come up in performances when “Preset” is bold-pressed, as opposed to requiring fiddly navigation through the preset bank tree beforehand.
- To allow loading a preset without immediately proceeding to the parameters view.
- To allow navigating between chains with Left/Right buttons when the preset favorites list is showing.
- Or instead, as mentioned earlier, to allow customizing a ZS3 such that it will ignore certain chains or settings when loading.
I think there are plenty of ways that Zynthian could improve on the workflow for this use case without requiring a whole new UX paradigm, and without asking for simply more machines with more horsepower.
4 Likes
I think that the presented performance scenario, according to our new user’s own words, aims at a sort of improvisational approach. While in the beginning he seemed kind of resolved to target fairly standard requirements,
later he maybe started envisaging a more intuitive, and free, on-the-fly management of chain presets.
I was pointing out that a preset change map requires planning, and a clear design of the music’s development, being hardly compatible with an approach of extemporary orchestration. Both paradigms are legitimate and viable on their own, but written arrangement and unscripted improvisation are somehow in contrast.
This is not to say that they cannot be mingled, and there is plenty of excellent Jazz and Fusion repertoire to demonstrate the validity of this route, but having a Midi messages framework in place, for timbres which sometimes will be chosen on-the-spot, is technically tricky, unless the musician simply deploys a set of physical controls, each sending a selectable program change, which is intrinsically feasible irrespective of the Zynthian technological environment.
Probably, like @jpetso suggests, it’s just a matter of devising a handy and fast mechanism for live preset recall, avoiding program changes and ZS3 altogether, but this is something synthetists have been doing on stage, with various degree of freedom and success, since the early times of Tangerine Dream, Schulze and Jarre.
Regards 
1 Like