bitflipper

Plenty of bruises on my forehead from beating it against the proverbial wall. Not from anything music- or recording-related, though. Those hard-earned badges of honor came from trying to understand people, not technology.

I had a similar puzzle once. It turned out there was a duplicate clip buried behind the "real" one. That's why you need to examine the project - it could turn out to be something totally unexpected.

To be fair, you have to first be familiar with the concept of assigning different MIDI channels to key splits before you can even begin to look it up in the manual. In any endeavor, knowing what can be done is 90% of the process; figuring out how to do it is the easy part.

Try inserting the synth with just "MIDI Source" and "First Synth Audio Output" checked. See if that works. Dim Pro isn't actually a multi-timbral synth, but does allow layering. Are you trying to direct each layer to a separate track? I don't know if that's even possible, but it would be the only reason I can think of for using multiple mono audio outputs with this instrument.

What you'll need to do is go into the Launchkey's settings and find where it lets you assign different MIDI channels to the lower and upper portions. Any MIDI controller that supports splits should have that capability. You know what they say: when all else fails, read the manual.

That does indeed look like it could be a bug. Unfortunately, I was unable to duplicate it. I followed your (excellent!) video as closely as possible, first overlaying two clips and then muting one of them via the clip context menu. Then I bounced to a new track. The muted clip was not included in the bounce, having been ignored completely, as expected. Does it act the same way if you bounce to clips rather than to a new track? By that I mean select all clips, right-click on a clip and choose "Bounce to Clip(s)". Technically, it's the same as bouncing to another track except the new clip stays in the original track.

It's probably not a plugin, either. Easy way to check: create a new (empty) project and see if the noise is still there.

You can also CTL-drag plugins from one track's fx bin to another track to duplicate them.

^^^ This. Not an EW Play user here, either, but I assume it has a visual representation of the mod wheel in the UI, and that it is animated to reflect the value of either CC1 or CC11. If it's not moving, that would be a clue that the instrument isn't seeing mod wheel events. There may also be an expression slider that determines how the instrument responds to CC11. When it's all the way down, CC11 would be ignored. I don't know if this particular instrument offers that feature, but it is a common one with orchestral libraries. I'd guess it's probably not a MIDI channel problem, since the track does play, which indicates the notes are on the right channel. If you're putting CCs in manually via the PRV, they should automatically go on the same MIDI channel.

That pop is unavoidable when the audio's numerical value at the split point is anything other than zero, because otherwise you're telling Cakewalk to go from X to nothing in an instant. Speakers can't go from X to zero instantly, but they'll try to comply and move as fast as they can. That results in the pop you hear. It's the sound of your speakers being kneecapped. Ouch. The term "zero crossing" just means a point in the waveform where the sample value happens to be zero. It's called a "crossing" because unless the clip is silent that zero value occurs as the waveform is crossing from a positive to a negative value, or vice versa. Open Preferences (press "P") and scroll down to Customization -> Snap to Grid. There is a checkbox labeled "Snap to Nearest Audio Zero Crossings". Even if you don't have grid snap enabled generally, with this option selected you'll always select a zero crossing when splitting a clip. No more pops.

Could be a lot of things, unfortunately. Electromagnetic interference (EMI) from a nearby appliance, for example. We'll need more clues in order to help. One thing it's definitely not is a Cakewalk bug.

^^^ Well, there ya go. gustabo saves the day! Sounds like Play's version of Kontakt's DSD preload buffer adjustment. It determines how hard your disk drive has to work to stream samples.

I have experienced similar problems in the past with various instruments. Not with EW Play, as that hasn't been part of my toolkit for a very long time. But it's not necessarily specific to any one instrument. It's a speed issue with streaming buffers and/or disk I/O (e.g. if you're short on RAM). Unfortunately it's probably not something a DAW can address. If Play has an option similar to Kontakt's preload buffer size setting, that might help (depends on whether the issue is slow disk I/O or too-little RAM). Easiest solution: turn off 64-bit processing. Trust me, you will not be able to hear any difference. But you'll be using half the memory. There are some instruments that just don't like fast bounce, but having to wait on a real-time bounce is annoying. Fortunately, in your case 32-bit processing apparently gives your system enough of a breather to mitigate the symptoms, so that's a better solution.

"I'll read the manual. Maybe she can make me a sandwich." Glad to see the return of humor to the genre, which has always suffered from taking itself too seriously.

Another tip: when you are hand-planting drum hits in the PRV, turn grid snap off. Some of them will be noticeably "off", and they can be nudged - using your ears, not the grid - to correct them. One more: if you must quantize something, make it the kick. You can't advance the snare unless it has something to advance relative to. A steady kick sets up the listener's expectations when anticipating the next beat, so that when the snare or ride cymbal doesn't exactly coincide with the kick it'll have the desired effect. And yes, as Cactus notes, you can "play" drums with nothing more than a drumstick and a microphone. Or even just your hands on the desk.

For your existing drum track, try this. In the PRV, click on the left pane to select every snare hit, and then hit Delete. Now record a new snare track and manually play along on your keyboard controller. Do it a few times and you'll gradually refine your timing to not only fit the song but to enhance it. Don't have a keyboard? Pick up something like this for fifty bucks. What I don't recommend is using your computer's keyboard as demonstrated in the video above, because timing isn't the only factor in humanization. Velocity is just as important. If deleting all your snare hits is too scary, mute them instead. That way, they'll still show up in the PRV as a guide but won't play.

The only thing guaranteed to make your drums sound just as fake as quantization is "humanization". Don't do it. Someday, some clever programmer will invent a smart humanization tool, but it doesn't exist at the moment. Inserting random timing swings does not make the rhythm more human. Yes, human drummers are naturally inconsistent compared to a PLL oscillator. But the variations are not random. A good drummer advances or retards hits in order to add excitement or to relax the groove. When hits don't land on the grid with microsecond resolution, it's for a musical purpose. A random number generator cannot do that. If you want more natural-sounding drums, and learning to play drums or enlisting a real drummer are not options, you can still enter MIDI data by hand. You don't have to be a keyboard player, either. Anybody can tap their fingers. Watch this for some inspiration (note: the presenter is not a keyboard player). Here's something similar but using Cakewalk and SI-Drums:

You're in luck. MMultiAnalyzer for 15 EUR at VSTBuzz. Time to check under the couch cushions for change!

Yeh, you're good. Noel was able to come up with a clever fix, and it works. And no, I don't have a magic formula to squeeze 16 instruments into one instance of Omnisphere. It's 8 tracks times 2 patches each = 16 patches.

Maybe you're thinking of the sidechain kludge that lets you show use multiple inputs of SPAN to display two tracks. It's a clunky kluge, though, and more trouble than it's worth, IMO. There are a number of products out there that do a much better job of this, and while not free they are fairly inexpensive. The best of those would be MMultiAnalyzer from Meldaproduction. It's $72 but periodically goes on sale for half that. Then there's the multi-channel version of SPAN, called SPAN Plus, for $45 and does 95% of what MMultiAnalyzer does. Both are very easy to use. Bluecat Audio's FreqAnalyst Multi is similar but adds a couple unique features for $99, and has been on sale in the past but less often than Melda's.

You're right, that really is worth a look. It would certainly be a time-saver for anyone who regularly mixes other peoples' tracks.

Looks like nobody around here uses that controller. That doesn't mean it won't work, though, and I wouldn't advise fixing the problem by buying another controller. Certainly not another from the same manufacturer, since it'll likely be functionally identical to the one you've already got. I assume you're trying to play a soft synth. By default, the output of a soft synth is muted except during playback. To hear it in real time, you have to explicitly tell Cakewalk to let the synth send audio to your speakers. That's controlled by the Input Echo button, as described here. Have a read through that article and see if any lightbulbs go off. If that doesn't help, or isn't the information you're looking for, come back and we can give you some troubleshooting tips.

I wondered about this, too. Omnisphere only supports eight voices, so more than eight outputs would seem superfluous. It has spurred much speculation among the Spectrasonics user community. The best hypothesis I've heard is that the 9th output allows you to route the aux bus to its own output (e.g. a common reverb bus), something Omnisphere had been unable to do if you've maxed out the multi with a full complement of 8 voices. This would allow you to tweak the reverb amount even after you've frozen the plugin. That makes sense, although I'd have never thought of it myself because I almost always do my mixing within multitimbral instruments - whether Omnisphere, Kontakt, Superior Drummer or the TTS-1 - and then route the multi to a single stereo pair of outputs. Treating a multi as a single instrument simplifies mixing. You'd have to be a seriously heavy-duty Omnisphere user to actually need that 9th output. However, such power users do exist, and they are important to Spectrasonics because they tend to make soundtracks for high-profile television series such as BBC nature documentaries. That's good PR. My theory is that one of those VIPs asked for the feature and Spectrasonics granted their request, unaware that it might cause problems for the thousands of users who aren't famous composers.

Nor mine. But I don't have that particular library. In the upper-right of the UI there are three dots. Click on those to expand a few settings, one of which is maximum polyphony. It should be 256 by default. Also check the preload and stream buffer sizes; not sure but I suspect these may be set automatically based on the amount of RAM you have.

I'm pretty sure all of the LABS instruments are polyphonic. Which one are we talking about?