David Baay

I was searching for one of the longer threads about continuous track scrolling and found this post that actually complained about track-by-track scrolling before continuous (a.k.a. fractional) scrolling was introduced in X3b... to each his own. http://forum.cakewalk.com/Smooth-Scrolling-of-Tracks-m1615686.aspx And here's the big one complaining about it. Post #13 by Seth Kellogg explains why it was implemented: http://forum.cakewalk.com/Im-Disapointed-in-X3-m2918795.aspx#291886 This may not be the definitive thread, but here's one discussion about the loss of 'rebuild layers'. More than one poster pointed out that the result of rebuild layers was a little unpredictable and the last post points out the the Re-use Existing Lanes recording option largely made rebuilding unnecessary. I prefer to have control over what ends up where, and generally have it set to Create New Lane. http://forum.cakewalk.com/X2-Take-lanes-rebuild-layers-m2686731.aspx

All of this - especially lanes vs. layers - was discussed ad nauseum on the old forum as the changes were introduced. Editing a lot of audio takes is not part of my workflow so I can't say definitively that layers weren't better in some respects, but as with so many software evolutions it's likely you just need to embrace the new ways of doing things and ask specific questions about how best to accomplish tasks that seem to have gotten more difficult/awkward.

As a piano/keyboard player, I have the same envious thought every time I see a musician sitting under a tree in the park playing a stringed box full of air.

Goto measure 121, Shift+M, enter measure 143, beat 1, Enter. No fuss, no muss, no fixing the meter, tweaking mistaken detection results or removing superfluous tempo changes due to the piece 'breathing' a little within the context of a fixed tempo.

I should add that this thread may seem a little confusing because the first post is not not asking the same question as the title. The answer to the title is: MIDI already follows project tempo changes by default. What is really being asked is how to 'Set Project From MIDI Clip' which is what I posted. Once you've done that, the MIDI will follow any tempo changes you make from there.

First of all, don't have any tempo changes in the project before you start recording, and don't have CW syncing to tempo from some external MIDI clock. A MIDI clip won't really have a tempo of its own unless you take it out of the project as a MIDI file or turn it into a groove clip. It has 'inherent' musical tempo based on the absolute time that goes by between quarter-note events at the current project tempo, but that doesn't have to match the project tempo. If you change the project tempo by any direct method other than Shift+M mentioned above, the MIDI will follow that change proportionally because MIDI event timestamps are based on where they fall in the timeline at the time they're recorded, not on absolute time.

If you know the target tempo: - Snap the Now Time to measure = Current Tempo +1 - Shift+M to open Set Measure/Beat At Now - Enter measure = Target Tempo + 1, beat =1 and OK. If you don't know the target tempo: - Slide everything to align the first downbeat at 1:01:000, or 2:01:000 if you need a pick-up measure. - Snap the Now time to that measure, and Shift+M to set the matching measure and beat. - Listen to playback without the metronome running, and count out 8 or 16 measures. - Snap the Now time to the downbeat that should be 9:01 or 17:01, and 'Set' that. - if the original was recorded to a click, and the tempo that CW sets is close to a whole number, you can undo the Set, and do the first procedure for setting a known tempo. - Or you can just keep using SM/BAN to set additional point to lock the timeline to the project as needed.

Shift+M is Set Measure/Beat At Now. It allows you to tell Cakewalk where measures/beats are in your MIDI, and automatically calculates and inserts tempo changes to align the timeline to the MIDI, adjusting start times and durations of MIDI events to preserve the absolute playback timing of the performance. I live by it.

I'm not 100% sure of your routing since the screenshot of the inspector for track 13 is not showing I/O assignments. But I suspect the above is part of the problem. With Input Echo enabled on both tracks and inputs set to respond to the same channel from the controller, both instruments will respond to any input. If you want only the focused track to be heard, you should disable the forced Input Echo on both, and enable 'Always Echo Current MIDI Track' in Preferences > MIDI Playback and Recording.

The way I posted is still valid. Process > Length > 50%. Or drag-stretch to 50%.

Not sure I completely understand the setup. Are you crossfading two different synths/patches using MIDI volume or two audio tracks with volume controlled by MIDI? In the case of synths, the result is going to depend on the response of each synth to MIDI volume, and there's no standard/convention on this in the MIDI spec. And in the case of both audio and synth tracks, the result is going to depend on how you set the MIDI endpoints of the group, the volume of each track at the high endpoint, and how much correlation there is between the signals. With the right tweaks to the grouping end points and max track volumes, you should be able to get pretty close in either case, but it's going to take some fiddling; there's no simple recipe you can apply that's going to work for all sound sources.

Simple procedure for re-setting the project to a a specific whole-number overall tempo: Snap Now Time to Measure = Current Tempo + 1 and 'Set' Measure = Target Tempo +1, Beat 1. I don't usually have a need to achieve a specific tempo. I usually just want to align the timeline to a live performance without altering it, and the tempo is whatever i played. But once the timeline is aligned, the tempos can be altered as needed. I most often just smooth out any excessive/unintended variation, but one could just as easily change the average tempo of the whole piece or sections of it which seems to be your goal. If you're interested, I'd be willing to have a go at 'Setting' the piece you're working with. I might be able to give you some pointers on how to get what you want without too much brain strain.

Missed your response until now. As you have apparently figured out, with SM/BAN you tell CW what measure beat an event should fall on, and it calculates and adjusts the tempo from the previous 'Set' point (or time zero if nothing has yet been set) to make that happen. It's the keyboard equivalent of dragging the bars/beats to align with a note/transient in a track. In the case of MIDI, CW recalculates note start times and durations to preserve the absolute playback timing at the new tempo. And it sets a matching tempo node at the now time to act as an 'anchor' for the next 'Set'.

You have MIDI being passed Thru a soft synth or by some hardware path back to the track input when it's accepting everything and new Note Ons are cutting off sounding notes (some synths would use a new 'voice' to play the duplicate note but some cannot), You might need to disable MIDI Out from the soft synth and/or check for some hardware MIDI loopback since the input was set to "All External Inputs". In any case it's always a best practice to set the track input specifically to the port and channel that you want to record/echo to avoid problems like this.

So is Cakewalk set to WDM driver mode? Normally I would recommend ASIO mode for a MOTU interface, but ASIO does not natively support driver sharing and the Cakewalk audio engine would need to be suspended to hear another application. If you've recently changed to ASIO driver mode, that would explain sudden appearance of a problem.

I'm down with these two, and have specifically requested the first one in the past. I also brought this up in a discussion about editing tempo envelopes recently that instantaneous changes should always be done with jump segments for simplicity and consistency. Mark's video showing four nodes being used for two changes is not a desirable default behavior in my view. My OCD self says. "Yeah, that would be nice", but my rational self says, "Don't bother, few people can reliably discern anything less than a 1dB change anyway"

Possibly the option to have CW release the driver when not in focus has become unchecked....? Preferences > Audio Playback and Recording > Suspend Audio Engine When Cakewalk is Not in Focus FWIW, I don't let Windows use my MOTU drivers. I leave onboard sound enabled for use by Windows and generic multimedia apps and route the soundcard audio to my studio monitors using an external mixer. I have found this to be the least problematic setup over the years.

Change the track input to the specific port and channel your keyboard is on.

I'm afraid you got the terminology backwards here. 'Focus' is indicated by the highlighted track name.

No, there's no mechanism available to 'nudge' a loop. The closest you will come is using keyboard shortcuts to set From and Thru selection points, and then using Shift+L to set the loop to the selection.

I misunderstood the problem. I thought it was that the volume widget wasn't staying at a new position, not that it wasn't having any affect. Glad you got it figured.

Sounds like the mixer is echoing its output back to its input which is more commonly a default setup with onboard soundcards but can happen with any interface that has internal mixing/routing capabilities. In any case, routing of audio in the Mackie would be the first place to look.

What's the timing reference and routing of MIDI and audio? Are you saying the sound from an external hardware synth is sounding ahead of the Cakewalk's audio metronome when enabled on playback with the synth output monitored through Cakewalk? And is the MIDI quantized to the grid in Cakewalk? If you're direct-monitoring the synth and your audio buffer is high, you might conceivably hear the synth a bit ahead of Cakewalk's audio metronome and/or audio tracks playing back, but usually MIDI transmission and synth response delays would mostly cancel out any output latency, and the typical 2-3ms discrepancies would be pretty hard to hear in any case. Cakewalk has a Timing Offset setting to sync MIDI-driven hardware synths with audio, but usually it's used to do the opposite - delay audio to sync with the delayed synth response. In any case, you'll want to understand exactly where the sync error is being introduced before you go tweaking anything. Timing Offset can have undesirable effects in other contexts, so I'd start be making sure it's 0ms. Preferences > Audio > Sync and Caching > Timing Offset (msec) If we're talking about soft synths the above is all irrelevant.

Step 1; Verify there's signal at the output of the mixer by connecting it to a standalone recording/monitoring/metering device.

Checked this out and found that once an automation envelope exists, it's not possible to alter the level with a snapshot at a position before the first node in the envelope; the new node takes on the value of that first node, regardless of what you set. This looks like a bug. But creating a snapshot of the level at time zero before any automation envelope exists worked as expected.