fernandraynaud

FWIW, it's been my observation that keeping one's older setup intact is a smart way to go. For one thing, you wouldn't get much cash for it, if you wanted to sell it. For another, the older machines have been open to upgrades. In my case it's a Cheese Grater dual Xeon 5,1 MacPro that I got for $300, and that's been very inexpensively upgraded over the years to two 3.47 Ghz 6 core CPUs, 64GB RAM, a Metal GPU and ample storage. It's running Mojave and Logic 10.5.1, without heroic workarounds, just firmware updates. It runs all the older plugins, including 32 bit ones. Logic 10.5.1 has most of the important new features, like live loops. It's all configured, it ran great, and it still runs great. Huge surprise: with 12 cores and 24 threads at 3.47 GHz, it's not much slower than my M1 MAX Macbook Pro. On some Logic operations it seems to work faster. With two big 4K displays cross wired I can run them together, instantly flipping one or both screens between them, and passing data over Gigabit Ethernet. Best of two worlds. The only downside is the > 300 watt (ouch!) power draw when everything is running, so it's great to be able to do a lot of work on the M1 alone. But I'm really glad I never broke down the Cheese Grater. Having both systems available is a sure way to keep keeping on.

Not sure why, but an edited files doesn't always get accepted. This needs more testing.

Michael, in West Coast US suport, threw out the depressing dead-end thought that "it may not be possible in Logic". Oh, Michael, that's not likely! We are already half way home. The regions coming from "the little hand" are correctly labeled and they play the right Instruments in the Orchestrator patch, IF we match up the MIDI channels. Dragging the regions one by one into empty track space, we DO get independent tracks. BUT then an Opus Multi gets in the way with the mystery BYPASS button. AAAGH!

Orchestrator is, thank Heaven, not going to write any Superhero 9th symphony for you. But as a little assistant it could be helpful, if only there was a fast way to get from A to MIDI to independent Regions/tracks. This here tries to illustrate a working method, starting with two OPUS Orchestrator instances. OP1 has a single stereo out, and OP2 is multitimbral, with a full 16 stereo outs, both with the same Orchestrator preset loaded. OP1 is where you develop your "prompts", by analogy with ChatGPT. You then export the Orchestrator's MIDI output, using the "little hand", to "fan out" into the multitimbral OP2. At this point OP1 and OP2 should SOUND the same, although the OP1 MIDI data is all merged, and the OP2 regions share too many settings to be useful. They are like tracks created "with duplicate settings", or a summing stack, and we know what a pain those are. Your mission, should you decide to accept it, is to easily break the OP2 stacked tracks into independent Logic tracks, each separately automatable, playing the correct instrument that Orchestrator assigned. It's perhaps feasible, and less memory-intensive, to keep them in one single OPUS instance, but without sharing settings? Who knows. Come join the developers in Germany, the Support staff in California, and everybody else, in beating our heads against this wall.

My method so far is to create a single output instance of Opus, load Orchestrator and input my "prompt" chords there. Once it's usable I add a 16 stereo outs multi-timbral instance of Opus that I load WITH THE SAME ORCHESTRATOR PRESET. Play the base track into it. That makes the "little hand" in the lower right hand corner show a white dot and record it. After stopping the transport, the "little hand "goes back to waiting. Drag the "little hand" to the multi-output Opus instance. That tends to assign the instrument tracks correctly. Now the fun part. How to pull these out into individual tracks with the right names, distinct settings and volumes. One at a time works, leaving some cleanup to do. Strange that I can't find a "break track-stack-like multi into separate tracks" function.

Orchestrator started life as some sort of hallucination for the naive kids who think it will write symphonic blockbuster music for them while they TikTok. Just fork over the cash. But it's a great tool if used on sections of compositions instead. They are over half-way to usable with the MIDI export "little hand". They just need to fix bugs and define a quick workflow to create individual tracks in Logic. Of course, the joy of finding tracks playing with default Logic instruments is at least better than playing GM patches, LOL. If you drag the export into a 16 stereo outputs multi-timbral instance of Opus with the same orchestrator preset, it plays ok, leaving the single Multi to untangle into separate tracks. Right? Anyone know how?

I've found EW support via chat very helpful. Michael knows the program. Now that the "little hand" export can be dropped into a multi-timbral instance of Opus, it works. Each track has an individual MIDI channel that controls the right instrument, e.g. 3CL for 3 clarinets. But all the tracks share settings, including volumę automation. Not sure how to separate out the tracks without some settings, like MIDI channel, getting shifted. Still not a HUGE amount of work. I keep asking EW for a RECIPE !!!

Is anybody finding a way to quickly map out an Orchestrator "little hand" MIDI export into individual Logic tracks? It works fine onto a multi-timbral instance of Opus, but then all the tracks share settings, including Volume automation.

I just spoke to EW. The developers are incorporating it into the next OPUS update. A beta should be available in a week or so. Not sure when it will reach mortals. For now, all you need to do is use a plain text editor and paste this into the .plist Articulation file. The SWITCHES tab then works and you can add your desired trigger Keyswitches. The questions are whether it matters where you add it, and how it acts when you re-export an Articulation Set .plist. <key>Switches</key> <array> <dict> </dict> </array>

OK, I found a solution, or at least a workaround. The trick is to insert an empty SWITCHES section into the Articulation Set .plist XML file that Opus exports, and that is missing any SWITCHES sections. This is what made adding Keyswitches or otherwise editing the SWITCHES tab in the GUI impossible. After a placeholder SWITCHES section is added, Logic's GUI Articulation Set editor works normally in all 3 tabs. I'll provide instructions after we complete some more testing.

Unfortunately, the Articulation Sets that Opus exports are incomplete. and break how they normally work in Logic. You normally have 3 tabs in the Articulation Set editor. Data flows from left to right. The first SWITCHES tab defines triggers that you can apply while playing live, or recording a track. These are consumed by the next tab, labeled ARTICULATIONS, so they do not litter your piano roll or score. Then each articulation has a matching line in OUTPUTS, that defines what actually gets sent to your instrument, such as keyswitch notes, or CCAC codes, or program changes, etc. The complete 3 tab set is extremely flexible. Some Instruments have dozens of articulations. I use a secondary keyboard or an Akai APC Mini or a Novation Launchpad, to send keyswitch note-ons that I have defined in the SWITCHES tab, while I play with the right hand. I spoke to East West support today and they confirmed that Opus exports Articulation Sets with an empty SWITCHES tab. Worse yet, I can't enter ANYTHING in this window. The + sign doesn't work to add triggers. Yes, I can modify Articulations after recording, in the editors, but I prefer being able to (also) change articulations while recording the way a player does, and it's broken. Does anyone know a workaround? East West says they will "look into it", but that it's working "as intended" !?

Thank you so much!

In the context of Prem Nath's needs, of course you're right. I was wondering if it wasn't being extrapolated off the context .

With all due respect, honestly, David, are you telling people that they can base hardware purchase decisions on your running Logic Pro 10.6x "with no problem at all" on a "MacBook Air 1.3 GHz i5 — MacOS X 11.0.1 — 4 GB RAM"? Is that what people should expect? That may be true in some situations, but overall I must be missing something, because when I'm using Logic with wavetable synths and sample libraries with multiple instruments, each of them with multiple articulations, the session is often using well over 32 GB and it would be impossible with 4GB of RAM, unless you're constantly swapping to disk, or bouncing every track as you go. Neither of those scenarios make for a nice workflow. And we've all seen a moderately complex project easily overload a 4 core i7 running at 3 to 4 Ghz, or a dual Xeon 12 core 3.47 Ghz system. Members here have often despaired of this situation, and begged for solutions. Has Logic 10.6 on Big Sur somehow advanced so far as to completely change all that?

Those of us running on Classic Mac Pro towers only need to wait for the update fashion to loop around, like skinny ties, back to Intel chips, or at least to "revolutionary" PowerPC chips Oh, and yes you CAN shoehorn Catalina into a Classic Mac Pro. But this hard sell update pressure is getting annoying. .