brundlefly

The numeric peak resets to -Inf every time you start playback and holds the highest peak encountered during playback up to the Now time. Playback has not yet reached the higher peak in the waveform.

This issue is almost certainly plugin-related, and eliminating UAD from an offending project would be a good place to start troubleshooting. Cakewalk has had various issues with UAD over the years, at least some of which had to be fixed by UAD as I recall. Again, I would be more inclined to attribute any variability in the output file to FX plugins that have random elements, intentional or otherwise. Standard operating procedure for troubleshooting an issue like this is to start stripping the project down until the problem goes away. A good way to do this with a big project is to take a binary approach: Disable half the buses/tracks/plugins and check the export to determine which half has the issue, then disable half of that half, etc. One way or another this is most likely to be a project-specific issue that can only be properly diagnosed by getting hands on with the project and all the plugins (and the UAD hardware if the problem can't be reproduced without it).

Wild guess: If you're depending on Always Echo Current MIDI Track in preferences to echo live MIDI input on the focused track and the focus is on the audio track, the live MIDI track will be silent. If that's the issue, you can either move the focus to the MID/Instrument track by clicking on the track name, or manually enable the Input Echo button on the MIDI/Instrument track so it doesn't matter which track has focus. https://legacy.cakewalk.com/Documentation?product=Cakewalk&language=3&help=Playback.35.html#1067525

It's working here to "lasso" the notes of several measures on several tracks in the PRV by right-click and drag, and then copy-pasting later in the same tracks. I recommend pre-selecting the tracks in the PRV's track pane by swiping their numbers from top to bottom (or Ctrl+clicking from bottom to top), leaving the lowest-numbered one focused so the paste will end up in the corresponding tracks. And then right-click to lasso the range. Also if you want to cut-paste, you'll need to make sure Auto-Lock is disabled in the the PRV track pane in order to be able to cut from the non-focused tracks.

Personally I'm happy to see Cakewalk get "monetized" again with Next and Sonar. For one thing it creates a revenue stream to sell to a future owner along with the intellectual property should the current owner ever decide to get out of the game for any reason . And while the "stigma" of using a free DAW instead of a "professional" one never bothered me, it will help to restore Cakewalk's reputation as a developer to be reckoned with by all the major players, including plugin and control surface vendors, which will be good for all of us. The king is dead. Long live the king.!

Cakewalk has always supported .01 bpm precision and I don't recall ever hearing of an issue getting that. Where are you entering the tempo?

The best tool and best results depend on the use case. Audiosnap works well if you have audio at a fixed tempo already matching the project tempo, and just want to change that overall tempo. It does take some processing power, depending on which algorithm is used, but I've never found that I have to raise my ASIO buffer by more than 2X at the most to accomodate it when applied to a good number of tracks. If you're talking about aligning the project tempo to an existing track that wasn't recorded to a click, Set Measure/Beat At Now or Melodyne can both do a good job. SM/BAN takes a little more time but gives you a lot more control over how strictly you lock in the tempo and is better for audio that is a mix of fixed and rubato tempos or changing meters. But both of those are offline processes and won't add any real-time load once completed so I don't think that's what the OP is about. Once the project tempo is aligned to the audio by any method, you can "flatten" or reduce the variability of the tempo changes using either Audiosnap or Melodyne to render the resultant audio stretching, but it requires the most advanced (and expensive) version of Melodyne to do this across multiple tracks simultaneously in real time without rendering each track to file on at a time.

Yes, it's about the interaction of CW with the sample clock and control of the output port. Plus there's apparently some crosstalk between the audio and video hardware that should be eliminated by moving the audio chores to a separate piece of hardware. You can get an audio interface that doesn't include MIDI DIN ports, but they are a good thing to have in case you eventually acquire an older piece of gear that doesn't have USB MIDI. EDIT: Also, both the D/A converters and analog sections of a dedicated interface will almost certainly sound better and have a lower noise floor.

This is a little weird; the 256-sample difference on the input buffer should only amount to a ~5ms difference in the required offset.

Are you sure you're not listening to the onboard wavetable synth? Those have notoriously bad latency that would be consistent with a 46ms offset. Or possibly the synth patch you're using on the Roland has a very slow attack...? In that case you should move the MIDI ahead of the grid or use the Time+ control in the track to compensate rather than using a system-wide offset. Other than that, I can't really imagine another cause and would expect the MOTU/Roland combo to work better than that.

These are all symptoms of inferior audio drivers and firmware in cheap onboard sound chips. I have some of the same issues (and others) on my laptop when using Realtek/WASAPI output. You will need to get a standalone interface with native ASIO drivers from a respected manufacturer to fully eliminate these issues. Buy used if money is an obstacle.

I have mostly avoided them over the years because I didn''t like their licensing and maintenance model. I had a huge multi-hour battle getting one suddenly and inexplicibly failing plugin re-activated a few months back and decided then that I will no longer buy anything new or renew anything from them after this year. This decision was only reinforced by the recent anouncement that no new licences will be perpetual.

Have not noticed this but have long wished that all plugins and Prochannel level and peak indicators would reset just as track meters do when playback is restarted. I have a shortcut bound to reset track meters at any time, but it has no effect on Prochannel modules.

I resisted the temptation to load up on plugins for decades, relying mostly on Sonar-bundled content plus hardware synths and FX modules for 30 years! And I have a lot of more-or-less finished compositions (if not fully produced recordings) to show for it because I spent more time just playing and developing musical ideas. A few years ago I caught the plugin bug (I blame this forum, actually!) and loaded up on all kinds of instruments and FX, many of which remain virtually untouched because I still tend to just sit down and play/record solo piano which requires very little in the way of processing and mixing. But I'm not as productive as I used to be because I still spend too much time keeping plugins, plugin managers and licenses updated and messing around searching for the perfect piano or accompaniment sounds or just familiarizing myself with some plugin so I won't get bogged down trying to figure it out when I really need it... which might be never! I haven't bought anything new for quite a while now and think I'll be able to hold the line because maintaining what I have is already pretty overwhelming. But I do enjoy just learning how things work and marveling at the heights that DSP and physical modeling technology have reached. But AI is just around the corner, and I'm sure there will be related toys to tempt me. I'm especially looking forward to the version of CW that will be able to analyze all my projects and create a finished song "in the style of David Baay" from musical keywords I enter. ;^)

I tried using dxishell for some DX plugins that didn't get ported to 64-bit, but it tended to be problematic so I abandoned them. For Sonitus and other plugins that did get ported there was no need.

In the amateur world: - Solo piano compositions based on I, IV, V triads played (or worse yet, step-sequenced) on a freebie VST. - The endless hours people spend trying to "humanize" sequenced parts that could be so much better spent learning to play a MIDI keyboard/controller in real time. In the professional world: - AUTOTUNE!!!

We never got VST versions, just 64-bit versions of the DX plugins.

The need for such a large offset is quite unexpected in any case. Even if your interface has quite high hidden latency on top of the 512-sample buffer, you should not be seeing more than 14-15ms of audio input latency plus maybe 3-4 milliseconds of MIDI transmission and response delay. Ideally you would get the buffer down to 128 samples or less and not need more than 6-8ms offset. But 46ms doesn't make sense in any scenario.

This setting shifts the MIDI grid against the audio clock (including the audio metronome) for both playback and recording and will affect where MIDI is laid down on the grid when recording. This setting was originally intended to compensate for MIDI transmission and response delays of hardware synths to sync them with playback of recorded audio - delays that are typically on the order of a few milliseconds. Higher values will cause problems in other contexts.

Clip/lane muted?

This is quite possible if using different online (real-time) and offline (fast-bounce) stretching algrorithms. When using the same algorithm for both, the difference will usully be more subtle. The Groove-Clip algorithm does not typically yield the best result overall for offline rendering, but it generally works well for online rendering and it's quality issues tend more toward distortion than "hiccups".

That's the way I interpreted it as well. Process > Length at 200% should do the trick.

I have HPF enabled on a lot of projects and have never seen/heard that. Can't think offhand what action might cause it either.

All the time: https://en.wikipedia.org/wiki/Street_light_interference_phenomenon

You mean with just a tempo change in the tempo map? is it just one discrete change or a series of changes over a short period (which some people expect to be "smoother" but isn't usually musically necessary and can actually cause problems as Jonesey noted). What kind of noise? Is this just involving MIDI or also stretching-enabled audio clips? Some plugins (mostly delays) can behave badly with tempo changes; otherwise should not be a problem in general.