Smart Quantize

I find if the drums are programmed REALLY well, you can keep them perfectly in time, and then play your other instruments on top of the drums and just pull the time around with them.

Most important part of programming drums IMO is varying the velocity of the Hi Hat hits, and getting ghost notes up in the snare drum. Unless its the most basic rock beat, most drummers have some sauce happening with their snare stick between the main backbeat hits. Its essential to get those in there and balanced properly for programming to sound anything like an actual drummer.

John said,"EZ Drummer's feature of Tap to Find is really great...probably makes this a lot easier now".

Oh boy, do I feel that. I've sent emails to Toontracks asking if they'll ever include that feature in Superior Drummer as an update and got stonewalled.

I find it odd that EZ Drummer has it, and yet they haven't updated SD with it. That definitely makes Superior somewhat Inferior, no? It would save me a lot of time and hassle. I don't feel like buying EZ Drummer just for that feature, but I'd be glad to pay a reasonable update charge.

Today I discovered some issues I've never really noticed. First, I was lowering velocity of a Superior Drummer snare hit in Logic, and when I reached a certain point, the sample changed tone, I just couldn't get it low enough with the same tone. 

Often, I use the stereo drums option instead of having individual tracks for drums, it's easier and saves DSP power, (my computer's long in the tooth). Superior's mixer works pretty well. So, I tried the Waves CLA 76 "PA" preset on my stereo drum tracks which is a very good sounding Blue Stripe emulation. I was exactly what I needed to keep the snare and kick strong, but not volume choppy. Unfortunately, it brought up the high hat along with it, so there was no workaround I'd used before. I guess I could copy the high hat to a new track?

Now I know why better engineers than me here use all the drum tracks, arrgghh..

So, can one adjust the velocity of drum beats in the piano roll section of Logic for volume, and still use a compressor? Print individual tracks as audio, then compress?

Just noticed popmann's post saying,"And yes--I always find it's better to recut everything TO the sequenced drums--after you print them off as audio like you'd have tracking a real kit. leaving it MIDI until mixdown is a bad idea on a number of levels.

Popmann, can you please elaborate , perhaps I've been doing something wrong leaving them midi.

Sure, MIDI sequencing doesn't work like an audio clock. It works as an overlaid reference to the clock at BEST....if you get the drums where they feel good, recording them as audio ensures that never changes. Leaving them MIDI both leave open the ability to accidentally alter it--either by altering the midi itself or the sampler's response curve/samples and even now it's built in randomization....or by the application doing miscalculation of the latency compensation that is altered literally every time you insert a plug in on any channel that changes those requirements. Not to mention that MIDI "tracks" don't actually transfer from DAW to DAW well and certainly not accurately enough for drum's timing subtleties. So, long term archive, you HAVE to print audio--so, it's not an extra step--it's just ensuring you do the step early so the rest of your tracks always have a foundation that never shifts under them.

Anyway--the idea is simply that MIDI is used to build a part you can't play. Once it sounds like you want it to--you need to render it so that it's made real and move on.

That's the dirty lowdown. 

You can experiment with nudging elements of midi grooves, but the key is to get one or two bars to feel the way you want and build upon that.

I remember reading an article in the day about the Kool And The Gang production team, and how they spent time getting the one bar groove right before building the song and arrangement.

If you can play drums with your fingers but not a kit, pad controllers are useful. Midi scaling is something to consider also. If too many of your midi velocities are over 100, try scaling the data back, when the data sent to your kit is softer, it usually starts to sound more realistic, since most drummers don't play as loud as they can all the time.

It really comes down to the # of samples used.  I know you use Superior, but if Logic's drums are similar, I'm gonna guess that they use a different sample set (assuming there's a little round-robin going on) at each level.  For instance - from MIDI velocity 50-65 a certain set of samples is triggered, so while the difference between 55 and 60 may negligible (but for volume), there's a very noticeable tonal difference between 50 and 49.   I have always wondered exactly how they build these libraries, I'll bet some people here know.. but I speculate that you want at least 4 or 5 different samples for each level such that when you've got a repeated note, it sounds more human.  With 127 different volumes that's 500+ samples that have to be loaded just for one drum.  On the other hand if you create samples for a volume range of 10 or so, now you're down to having 50 or 60 samples at the ready, and I'm guessing some of these companies use quite a bit less than that.  On the other hand, with processing speeds and memory getting increasingly powerful, maybe I'm wrong.

When they say a drum has 16 layers, they mean that the 127 possible velocities are divided into 16 groups. so, 0-7, 8-15, 16-23, etc.. One sample per velocity layer. You can do some fancy cross-fading to make it seem like there are more layers, but that's what they're doing so they don't have to manage 127 individual samples per drum, without considering "round-robin" sampling.

Select all notes in the piano roll editor - then on the left there's a velocity slider - I think it takes them down percentage-wise, but I'm not sure.

Scaling a selection of midi velocities should be possible in every DAW.

If you use logic x, go read the manual chapter on auto save backups

Select Midi Region -> Right Click -> Bounce In Place

Right, that's what I mean....they made it easy--only doing that throws away the MIDI.

In Cubase, You're going to "export audio mixdown"....and the source can be any mixer channel(s)....you check all the enabled outputs of the VI (or group channels you've assigned them to for submixing).....check "import into pool and project"....and when it's done, you have however many new audio tracks on top of the MIDI and the VI is still active in the rack. You simply click the power button, deactivating the VI (but leaving all the settings stored) and I mute the MIDI tracks--but, end of the day you've deactivated what they're pointing to, so it's not technically going to make any sound if you leave them "playing"....I just prefer to cleanly mute them and toss them into a folder to get them out of my view.

Thanks Chuck, my bad, I didn't bounce to a new track like I should have. I do some things I shouldn't because my computer's so anemic, I try to save dsp anywhere I can. As soon as the summer passes and I'm working again, a new computer's in the top five of my to do list.