David Baay

I realized after posting that this explains why you don't use the audio metronome.

Yes, I've seen a lot of requests for this over the years. Kind of funny that it got implemented as an incidental benefit of the Export function. I hadn't really thought about it until the OP brought it up again.

Logically, PDC can only work by adding delay. So, conversely, overriding it can only work by subtracting delay. It's not possible to subtract delay from a real-time input, so PDC and PDC override have to work at the output. But the issue with existing vs. real-time MIDI could potentially be handled better by delaying the output of existing MIDI to the synth rather than delaying the audio output from the synth. 'Aways Echo Current' facilitates the composing process by allowing a user to switch instruments simply by changing track focus. In conjunction with 'Allow MIDI Recording without an Armed Track', the creative process can proceed very quickly. A major downside of this feature, however, is that a MIDI track that initially has no input assigned will default to All Inputs - Omni when it gets auto-echoed, which can cause a lot of unexpected problems if you don't realize it. Personally, I just don't work fast enough for it to be of much value to me, and I like to have control over what's being echoed at any given moment, regardless of where the focus is.

That's gotta be it. I overlooked the key word 'files' in the title.

Is it a VST3? I've had trouble with AAS Chromaphone VST3 hanging Cakewalk, though others don't seem to have a problem with it, and other VST3s load fine for me. For Chromophone I have to use the VST2 version.

That's odd. Is this affecting both projects and project templates, or just a particular project? Most of those settings are different than what I typically use, especially the MIDI metronome. It shouldn't matter but it's a bit unusual, so that would be the first thing I'd want to check. Any particular reason you're not using the audio metronome - don't like the sounds on offer?

Coincidentally, a youtube video was just linked in another thread that shows pretty much exactly this process - just doing it in a multitrack project. And he sets every measure from start to finish while I usually start with 'rough set' as described, and dial it in only where really needed. But setting every measure like this is valid if you know the tempo is that variable, and are more concerned with super-tight timing than audio artifacts. I swear: A. That's not my video, and B. I did not see it before posting. ;^)

If you just want to get the task done: - Bounce the Master bus to a 'Master Bounce' track that outputs directly to Main Outs - Solo that track along with the one you want to record on. - Enable PDC Override and record away,

I've ony listened briefly on crummy laptop speakers, but I'm guessing this suggestion would apply broadly; Bring up the level of the instrument(s) defining the harmonic structure. The chord changes are cool but not as audible as they should be everywhere. I think this is partly what makes it sound 'Asian' - the melody is too dominant. A switch to - or addition of - a lead instrument with more sustain on the choruses could also help that. Nice work otherwise. I like it.

I took the above to mean he wasn't auto-echoing the offending track, but it's not a 100% clear.

I've posted about this many time in the past, and usually just refer to those older posts, but decided to start fresh. As usual, this assumes some knowledge of functions, menus and keyboard shortcuts: The general approach I would recommend, whether you use Melodyne, Audiosnap or Set Measure/Beat At Now - or maybe even Fit to Improv with a MIDI guide track - is this: - Get the timeline aligned to the MP3 in a new, separate project, starting at 136bpm. - Enable Audiosnap in Autostretch mode (or Melodyne if you're using that), and delete all the tempo changes. - Reset the initial tempo to 136bpm if necessary, and bounce the audio to render the stretching . - Import the rendered .WAV into your other project. To get the most precise results with the greatest flexibiity in choosing how much you mangle the audio with stretching, I recommend setting the timeline manually using Set Measure/Beat At Now. - Start by importing the MP3 to a new 136bpm project, and deleting the dead air that MP3 encoding adds (and/or your collaborator left in) at the start - Drag the clip to align the first downbeat to the nearest bar line (i.e. 1:01:000 if it starts on a downbeat or 2:01:000 if there are pickup beats). - If isn't starting at 1:01, snap the Now time to 2:01 and use Set Measure/Beat At Now (Shift+M) to pin down that point as the reference for subesequent timeline 'Sets'. - Start play back and stop the transport at the downbeat transient that should be hitting the 9:01 barline, and SM/BAN that point to Measure 9, Beat 1 (or Measure 10 if the first downbeat is at 2:01). - Go through the rest of the clip using a combination of playback (spacebar), previous/next measure and tab to transient keyboard shortcuts to put the Now time on downbeat transients at 8- or 16-measure intervals, and 'Set' those beats out the the last downbeat or the point where a decelerando starts if there is one. (Note: If the rhythm doesn't have strong transients on downbeats, you can set a different beat or even a fractional beat somewhere else in the bar). once you get the hang of using the shortcuts, this goes pretty quickly. - Enable the playback metronome, and listen through the clip, setting additional points as needed to tighten up the timeline to the audio. Understnad that SM/BAN is entering tempo changes to make timeline fit the audio; there's no audio stretching going on at this point. - If you find at some point that you've made things worse, it's best to Undo your way back to the last 'Set' before the mistake was made. With a contiguous clip that starts at 1:01:000, you can delete tempos, and not mess things up, but when doing this with MIDI or multiple audio clips that don;ta ll start at time zero, it's imperitive that you use Undo. And it's best to do progressive Save As with a new name at regular intervals so you can easily back up to a 'good' state if needed later on. - When it sounds good throughout, do a final Save As, go to clip properties in the Inspector, enable Audiosnap in Autostretch mode, choose Online and Offline stretching algorithms (Elastique Pro may or may not sound the best), and delete all your carefully set tempos (eek!), leaving only the initial tempo. If that tempo got changed by the set at 9:01 or an earlier one (likely), change it back to 136.00. - Have another listen to see how the online stretching sounds, and if it's all good, Bounce to Clip (or Track if you prefer to preserve the online stretching for comparison), and drag or export/import that clip into your working project. - Have a beer, go for a walk or take a nap. One final thought: The issue I have withe Melodyne, and the reason I prefer Set Measure/Beat At Now is that Melodyne tends to produce a hyper-smooth tempo map with interpolated tempos where no beats even exist, and when you 'flatten' the tempo by deleting them, there's so much streching going on everywhere that significant distortion results. If a recording has relatively tight timing to begin with, it's best not to set it at a resolution finer than a measure, and preferably less if it's not absolutely needed to eliminate obvious drift. I've recently experimented with a procedure for using Melodyne to extract tempos, and then thin them down to beats or measures, but it's a bit fiddly, and ultimately not really an improvement over Set Measure/Beat At Now, either in speed or quality of the end result.

Yes, as I mentioned earlier, Cakewalk can't differentially delay existing content, and not delay new content generated by the same synth output. When Input Echo is enabled, the assumption is that you plan to rehearse or record new content so it overrides the delay compensation. That's unavoidably going to cause existing content to play early. I thought the OP made it clear he's experiencing this issue without having Input Echo enabled on the problem track(s), automatically or otherwise.

I just finished playing around with this for an hour, and couldn't reproduce a consistent problem. A couple of times I had all synths go out of sync with the metronome and recorded audio when I was making too many changes at once (e.g. adding tracks, toggling input echo, record arming and PDC override), and starting recording without running playback to reinitialize the PDC amount. But I never saw just one synth get out of sync. I does appear I was wrong earlier when I was thinking the automatic input echo built into synth audio tracks would cause them to be subject to PDC override. I believe that was a problem back when synth recording was first introduced, but now override is only triggered by enabling MIDI echo to a soft synth, and that all worked as expected as well, except that synth recording isn't compensated for PDC like audio recording. I dunno... must be project/plugin/platform/procedure-specific.

It's about input echo being enabled on audio tracks. And synth tracks and the synth side of a Simple Instrument track are audio tracks, I now recall having a discussion with the Noel Borthwick about this years ago, and the issue is that you can't separate the audio stream generated by existing MIDI from a stream generated by realtime MIDI. Synth tracks have to honor PDC Override so that you can perform and record in realtime with no latency, and the side-effect of that is that existing MIDI will play back out of sync when recording with PDC Override enabled. The real solution to all of this is just to keep Mastering plugins out of your project until you're done tracking.

Ah, but it is live. Ijust realized that, though it's not visible, synth audio tracks are invisibly input monitored to echo the synth output from the track's input to the output. So it's working as designed, and you'll just need to freeze the existing synth tracks at least temporarily.

Yes, that's expected. It's not possible to compensate existing content, and not compensate live input on the same track. Hard to say what might be going on without knowing where in the routing the PDC is being introduced, and how all the tracks relate to it. I know Guitar Rig, itself, can induce PDC with some patches, and you can't override PDC on tracks that are in the delayed path. What plugin is inducing the PDC, and where is it?

To clarify further, Plugin Delay Compensation is active on all tracks that aren;t affected by the PDC-inducing plugin, and the button of the same name is actually 'PDC Override',. Enabling it overrides PDC on input-monitored tracks. But it sounds like the OP understands this. The question is what are the MIDI tracks driving... soft synths, input monitored hardware synths or direct-monitored hardware synths. Live soft synth audio gets buffer up in advance, and behaves just like rendered/recorded audio tracks. But MIDI to hardware ports isn't compensated for PDC so audio from hardware synths can get out of sync one way or the other, depending on how they're monitored, and whether you're playing back or recording.

Pause enables 'CPU Conservation Mode' that limits UI updates one per second.

Looked at it briefly last night. Didn't see anything particularly unusual about the set, except for the mono inputs to Aux tracks getting set to stereo as you mentioned. It could be related to how the template is structured when all the components are already in the project on the first save. Possibly your progressive build-up bypassed a bug. That'll be difficult to track down. Maybe best to let the Baker look one of the broken templates alongside the working one.

In Views > Event List, hitting the Insert key on your PC keyboard makes a copy of the event at the current Now time. You can then double-click the Kind field, select Control, OK, and then set the controller number and value. More on the above: https://www.cakewalk.com/Documentation?product=Cakewalk&language=3&help=EditingMIDI.45.html Overview of other options: https://www.cakewalk.com/Documentation?product=Cakewalk&language=3&help=EditingMIDI.44.html Details on drawing controllers in the PRV Controller Pane: https://www.cakewalk.com/Documentation?product=Cakewalk&language=3&help=EditingMIDI.20.html In a lot of ways, the easiest is to use a CC7 automation envelope In a MIDI track, but that will require some general knowlege of working with envelopes as well as how to add one for CC7 so you'll want to begin at the beginning: https://www.cakewalk.com/Documentation?product=Cakewalk&language=3&help=Automation.11.html

I haven't used it as much as I should, but I've generally recorded the Jamstix MIDI output early in the game and switched it to MIDI Playback mode. Then I can re-arrange and edit the MIDI to get the best fit to the music I've already composed, and maybe move some parts to another VST if I'm not getting the sound I want for a particular kit piece from Jamstix. In that case, I typically take the quick and dirty path of moving the MIDI for that part to the relevant note number on another synth track . The more 'elegant' way would be to use a Drum Map on the MIDI output to send that part to the desired note number(s) on the desired synth, but it takes more time to set that up, and since It's usually a one-time thing for a particular project, I don't bother.

Track templates are saved as .cwx files in this default location: C:\Cakewalk Content\Cakewalk Core\Track Templates

If you save a copy of the track template without the plugins loaded, does that cause trouble by itself when re-loaded? If so, maybe share a copy of that 'empty' template for others to try and see if it's reproducible on other platforms.

For one thing because MIDI CC messages are, in fact, discrete events, not a continuous curve as implied by an automation line. When using automation, you have little or no control over the density of controller messages being sent (not sure whether or how Studio One addresses this). Not such a big deal when working with soft synths, but when working with a multi-timbral hardware modules on a single MIDI port, it can be very important to minimize the event density by using no more than the necessary number of control changes to get the desired change in sound. That's not to say a new approach isn't possible, just that there are reasons for the way it's currently implemented.

Not possible with a recorded click. You'ld just have to start your project two measures in. But when using the live metronome, there's a count-in option in metronome preferences.