It's compensated for by the fact that the whole sound will be delayed by the slowest process that causes the most latency. So in other words, the audio is output after the final processing is complete, and therefore all your audio will be in sync.

SonicExplorer
bitflipper, do you mean it is only a potential concern for some types of EQ plugs which do NOT report PDC?  Or does your statement apply regardless of whether or not PDC is reported? 

 
Sorry, reading back my explanation I see that it must have been confusing, because I conflated two different (but related) things: latency and phase cancellation. I should have just ended it with the PDC part, but sometimes my fingers take off on their own (as my bandmates can attest when I'm soloing onstage).
 
Here's how those two things are related...incorrect latency compensation can cause phase issues; plugins that cause phase shifts can also cause phase issues. I connected those seemingly unrelated dots because they both involve time delays, and both result in similarly unpleasant-sounding effects. 
 
To clarify, EQs are no more prone to misreporting latency than any other plugin. However, it is in their nature to cause phase shifts. Those phase shifts are normally not noticeable - except when run in parallel. That's the tie-in.
 

I think I figured out the problem I was experiencing....   I'm using Superior Drummer 2 and it has a feature called X-Drums.  It allows extra kit pieces to be layered on top of each other.  For example two snares, or two kicks.  BUT....for some reason the pairs don't always fire perfectly in concert.  It varies based on the kit piece, and possibly also due to the humanize function I suspect.  I think maybe the samples themselves which are not consistent between kits in terms of where the transients align.  One snare sample may not align with another snare sample just by virtue of the actual recorded samples.  Best I can tell (speculate) these are the factors involved.  Sort of defeats the value of the concept of X-Drums though....

That's a common challenge when layering sounds. Phase relationships between the layers can both increase and decrease tonal complexity. When it's a bunch of human voices or violins, it usually comes out sounding pretty good. When it's layered kick or bass samples, it's a toss-up. The difference often comes down to the frequency ranges and the inherent complexity of each layer. Higher frequencies and more complex sounds (like voices) tend to blend better.
 
Fact is, you cannot get two drum samples to ever be perfectly in phase with one another. Not unless they're literally identical, which would indeed defeat the purpose of layering. That's because you can only ever match phase for one specific frequency and its harmonics. The more inharmonic content there is (e.g. drums), the more phase differences will exist.
 
That said, it's still worth fiddling with phase to see what effect it has. It's why you use two microphones on a guitar speaker and move them around until you find the tone you like. It's why you nudge drum or bass samples and listen for when it starts sounding fatter. Superior Drummer does let you offset the X-Drum sample for that reason. In the kit definition screen, enable the envelope section. There is a knob labeled "offset". The knob's overly touchy though, so you may want to type in offset values since you'll likely settle on a value of just a few milliseconds.

bitflipper
Fact is, you cannot get two drum samples to ever be perfectly in phase with one another. Not unless they're literally identical, which would indeed defeat the purpose of layering. That's because you can only ever match phase for one specific frequency and its harmonics. The more inharmonic content there is (e.g. drums), the more phase differences will exist.

Indeed.
 
Myself, I don't like to layer multiple drum sounds, certainly not in their raw form. It seems a bit too unpredictable to me.
 
What I do like to do, is to decide what I feel the core individual drum sound is lacking, and find a way to add that.
 
Say I've got a kick drum which I generally like, but is lacking a bit of low end thump. I might layer in another kick, but that second kick would be low-passed as low as maybe 70hz. So I'd just be using the low end of it, and getting rid of a lot of phase issues.
 
Or I might not even use another kick. One thing I like is to get a low sine wave, and sidechain gate that off the kick. Tune the sine wave to a frequency that suits the song, either root or fifth, somewhere between 50-100hz. Adds plenty of weight to weedy kick sounds. But we're leaving all the highly inharmonic transient work to the original drum sound.

What about the mix knob on the 64 eq that cakewalk added a couple of years ago?

Don't know about the CW EQ, but every EQ I own that has a "mix" control isn't really a mix control like you'd see on compressors, reverbs or delays.
 
It can't be, because you cannot run an equalizer in parallel (or the equivalent, mixing wet and dry signals within the plugin) without risking weird phase cancellations. I have several EQs from Meldaproduction, for example, that feature a "mix" control, but it's actually a global gain adjustment that applies to all bands.

Wow, good discussion, thanks for the ideas guys.  And I will definitely look into that feature in Superior 2 that apparently lets you adjust sample alignment.  
 
Tonight I spent quite a bit of time messing with the parallel buses I have set up.  I don't understand why, but for some reason it seems I'm getting phase issues or something similar.  The minute I enable a Sonitus EQ, or a compressor, or whatever on a bus, it seems like I instantly lose a little something, even if the plug isn't doing anything.  I found by disabling the entire FX bin on a parallel bus it actually sounded more pure.  But even then, the sound altered slightly depending on how high I raised the fader - and in a way that goes beyond just pure volume increase.   I hadn't done a lot of parallel busing in a DAW before so some of these side effects are new to me.  Even when using a paralel reverb bus (for snare as an example) it seems to behave strangely as I raise the fader.  It alters the underlying tone of the snare, not just the reverb.  I'm so accustomed to working in analog studios many years ago that some of the concepts I had used aren't exactly behaving the same in a DAW. So I'm having a lot of pain trying to figure out what's going on in certain cases, and how to control things.   Ugh.   I hate computers....

Sonic

Sounds stupidly basic, but we often forget that when you do any kind of parallel processing you are summing two signals. You are going to get both destructive and constructive interference, meaning some frequencies are attenuated while others are boosted. Sometimes, the results are unexpectedly good, sometimes they're unexpectedly bad, and sometimes just plain unexpected. It's what makes experimentation fun!
 
This phenomenon bites me more often than I like to admit. Why is the snare suddenly too loud? Oh, yeh, I just ran a send from it to the reverb bus. Duh. Everything you run to the reverb bus gets louder.
 
However, running identical signals to two busses should only make it louder. If it also affects tone, then there's something else going on. It could be a plugin that's not being properly latency-compensated, but more likely one of the two signal paths has some kind of filtering happening.
 
 

∞