Too much hit variation in sampled drumkits?

[Gwilym]

Okay, this is something that's been quietly driving me mad for almost 15 years now, and I've never managed to find a useful search term to get answers about it. Hopefully someone here can give me a bit of perspective.

In short: is this a normal amount of snare variation? Because to my ears, there are two distinct snare recordings in this, and it is utterly random which one plays on a given beat: https://drive.google.com/file/d/1qUM...ew?usp=sharing

Those are both great-sounding snare hits! But it's so inconsistent about which one I get that it almost feels like the sort of thing a real-life drummer would get fired for. Is it? Am I just being persnickety?

Here's the same beat on a different kit. Same thing: https://drive.google.com/file/d/1xvk...ew?usp=sharing

This happens with every single virtual drumkit I own. Granted, most of these are from the same Kontakt library: EastWest's Colossus. The same behaviour also happens in Goliath, the Play engine version of that library.

(Play actually has a CC command for 'reset round robin,' supposedly to avoid unwanted variation, but it doesn't seem to make a difference here, perhaps because it's a randomised thing rather than a sequential one.)

So the obvious solution could be not to use Colossus/Goliath drums and instead get something better. But what makes me hesitant here is that I can use a kit from the Kontakt Factory Library, and it does the same thing: https://drive.google.com/file/d/1MpZ...ew?usp=sharing

So is this normal? Is this actually desirable? If so, is this the sort of variation that I should be using a compressor to tame? If so if so, do I apply that compression to just the snare? Or the kit as a whole?

Obviously, this is less noticeable in a full song where it's not just the drums on their own - but it is still noticeable, and I don't think it's *only* because I'm listening for it.

And just in case there's something about my drum programming that's directly causing this, I've attached a screenshot of that beat. It's super simple and extremely on-the-grid, but I get the same snare variation thing even when I'm playing live on a MIDI keyboard, so I don't think it's a quirk of this particular bit of sequencing (even so: please still feel free to give pointers about the sequencing if anything jumps out - as short and basic as the clip is, it's pretty indicative of what my drum programming tends to look like, so anything that improves this would improve everything else).

Thanks. And I'm sorry for the blind URLs; I didn't realise they'd turn out that way. I assure you they're all 20-second MP3s. If this is against forum rules I can remove them and find another less obfuscating place to upload them.

Oh and if you can't even hear what I'm talking about, please say so! That'd be useful information too.

[valy]

Doesn't it have to do with velocity layers? Like velocity 30-59 = one sample, 60-80 is the next one, etc.

Most drum software lets you choose which layers trigger which sample, or at least turn off all but one sample if you want the same snare hit every time (which I would honestly prefer to your examples).

Or, you can figure out the velocity thresholds and change the velocity of the undesired sample to something that will trigger the other one.

[Gwilym]

I probably should've said; those hits are all at 100 velocity, and I can't see anything that might be altering them.

And yeah, in Kontakt I have the ability to dig into the mapping and tweak things, but it's so complex that it's really beyond my abilities.

Probably the fastest way to absolutely guarantee the same snare each time would be to load it into a sampler, and have the MIDI trigger that instead of the snare in the kit. I'm just hoping there might be something I'm overlooking that would be more workflow-friendly than doing this.

[paul_c]

Its probably better than the variation over whether the drummer will actually turn up for the recording session that day; or whether he'll bring his girlfriend in, to hang around the studio and generally criticise things, or smoke in the control room, or have an argument on the phone in a quiet section, etc.

AIUI its deliberate and I've never really worried about controlling which one of the snare samples sounds at any given hit, once its in the mix. There's bigger issues to be concerned with.

[planetnine]

Originally Posted by Gwilym

I probably should've said; those hits are all at 100 velocity, and I can't see anything that might be altering them.

And yeah, in Kontakt I have the ability to dig into the mapping and tweak things, but it's so complex that it's really beyond my abilities.

Probably the fastest way to absolutely guarantee the same snare each time would be to load it into a sampler, and have the MIDI trigger that instead of the snare in the kit. I'm just hoping there might be something I'm overlooking that would be more workflow-friendly than doing this.

But drum hits do naturally vary in sound. Drum and other instruments are designed to not play the same sample twice in a row (round robin) or they sound unrealistic, artificial.

Kontact experts should be able to tell you how to defeat the variation if you don’t want it, but as a drummer I’ll tell you that’s how real snares sound...


>

[Crashwaggon]

Originally Posted by planetnine

But drum hits do naturally vary in sound. Drum and other instruments are designed to not play the same sample twice in a row (round robin) or they sound unrealistic, artificial.

Kontact experts should be able to tell you how to defeat the variation if you don’t want it, but as a drummer I’ll tell you that’s how real snares sound...


>

Exactly.

Fwiw the first audio file the OP posted is fine. Nothing wrong with it, and would work well in a mix. Those little differences between is what makes it work.
Perfect is boring.

A drummer just doesn't hit the snare at the exact same spot and same velocity each time. That's damn near impossible. I've tried.

[valy]

Yeah, but some of those samples don't even sound like they are from the same drum (especially the second link). At least to my ears.

The first one wasn't bad.

[Gwilym]

Thanks everyone Yeah, the second and third clips do just sound dreadful in general, but they were specifically chosen for that reason - I loaded different kits until the snare variation thing really stood out. So they're close to the worst-case scenario as far as this behaviour is concerned.

The first clip is the kit I was actually using in the track, and is more indicative of how things usually turn out. So if that sounds fine to people's ears, that's actually good news and I'll try to just let it be. I feel a bit silly for how hard I've fixated on this in recent years (haha and for my "you'd fire the drummer" comment; indirectly besmirching an entire profession), but I did make this thread hoping the response might be "no, it's fine; you're being silly" so yeah - good news.

Plus if I ever decide in a particular piece that no, I absolutely cannot tolerate any snare variety at all, then I'll use a sampler to trigger a single snare wav, or even just drop the wav onto the track. I just won't try to make it part of my regular workflow.