Watto's upright piano sampled

Sampler and Sampling discussion (techniques, tips and tricks, etc.)
RELATED
PRODUCTS

Post

G'day guys,

I've been fiddling with just two octaves so far - what do you think?

http://simonwatson.com.au/kvr/piano_demo.mp3

It has some really light compression and reverb on it. This is just me mucking about, trying to play within two octaves :hihi:

There are 3 velocity layers (don't know if I have the patience for more), and I don't think I have my curve set up perfectly yet, because that demo isn't really hitting the hard notes at all, I don't think. It's pretty bright, but better bright than dull, no?

I'm basically doing this so I can play my own piano on my gigs, but I'm thinking about sharing it with KVR - dunno if I'd charge a small fee or not. Any thoughts appreciated.

:)
I've joined Lurkers Anonymous.

Post

Sounds great to me--I can recognize the piano tone. :)

Have I already asked you what you use to mic the piano?

Post

I'm always open for new pianos! As for cost, as long as it's affordable(not talking what it's worth, just what I can afford!), then I'm up for it as well!

I love the demo. A lot actually. The character of the piano is very nice and intimate. Right up my alley!

So, when can we take a shot at this bad boy?

Brent
My host is better than your host

Post

Cheers, Greg :)

Yeah, I use two AT 3031's in between the wall and the back of the upright, about halfway up, facing like this:

__________ (piano)

. . . \ / . . . (mics)
__________ (wall)

Very unorthodox, but I do that so I can sing while playing without too much bleed :)
I've joined Lurkers Anonymous.

Post

Cheers, Brent :)

Dunno - if I have time, I'd like to finish it before Thursday night (we've got an Ecliptic gig), so we'll see :)
I've joined Lurkers Anonymous.

Post

Thursday is good! Let me know if you need any help mapping or testing!

Brent
My host is better than your host

Post

Well, I'm encouraged to continue, so we'll see how we go :)
I've joined Lurkers Anonymous.

Post

hey that sounds pretty good!

Post

Thanks, bluedad! :D
I've joined Lurkers Anonymous.

Post

did you get the 7 bass samples i posted in that thread over there??

@watto n bluedad
Last edited by dover666 on Tue Nov 27, 2007 12:27 pm, edited 1 time in total.

Post

Nup - I'm forcing myself to play bass whenever I need it :)

Thanks anyway, mate :D
I've joined Lurkers Anonymous.

Post

i just want to know if they're good!! :cry:

Post

Oh, right :hihi:

I'll have a look and comment in the other thread :)
I've joined Lurkers Anonymous.

Post

Good to see someone including the wall when recording an upright. Seriously--a lot of upright samples seem to be done with the piano in the middle of a room.

One suggestion: record the hard strikes much louder, as loud as you can, by:

1. Hitting the keys as hard as you can.
2. Setting the gain so that these hard strikes come to the max level they can without clipping. The result should be at a level that makes the raw sample too loud for you at full volume to safely listen to with headphones or close-by speakers. In other words, in the pain dominion. Not joking here: go for the loudest possible sound without clipping. This will mean you absolutely must make several takes for each sample, since you can't monitor the sound live: you must instead record it and then lower the levels before playback. (Unless you place speakers\monitors a few yards from where you are sitting.) Be very careful here not to hurt your hearing.
3. Then adjust the lower level strikes to that loudness. Let the user set the curves and\or use filters. If you instead go for an overall smoothness of velocity shifts and try to create a good p or pp layer, you will be forced to hold back on the hardest strikes. (Sorry if I'm off base, here, but in the demo mp3, the hard strikes really don't sound hard at all.) If you want good soft sounds, a three layer instrument is going to have to use filters, etc to create smooth transitions. Much better to have good hard layers and use filters for soft sounds than to have a good soft layer and no presence at high velocities because they were recorded too low. You can still create a warm sound from these bright, loud samples.

In other words, don't normalize the hardest samples. Unless you find that you CAN normalize them--raise their volume still louder than you already have by recording them this way. But record them at their highest possible level and then create the other samples so that they descend in volume from that highest level. This is the way that Yamaha and Kurzweil record samples. (And Ensoniq used this method, too, when they were still around.) Gives you the most harmonic content, so the user can then assign flters, velocity curves, etc, but still have the harmonics there to come through when wanted. (If you have a pro hardware instrument from one of these companies and want to test this, set the volume to half-way up and play a note hard. (But be aware that some of them reduce the juice to the headphone jack, so you may have to test through the outs instead.) Then slowly raise the volume without any velocity curve or filter. You'll find that you can't play it at maximum value without causing physical pain. The expectation is that you'll send the outs to a board, where the volume can be reduced and\or sent over the heads of the audience. If the volume was reduced inside the keyboard, the mixing board might have to instead raise the volume to match that of other instruments, like electric guitars, and thus might distort the sound.

Post

Hey mate,

Some interesting thoughts, but I'm a piano player, and the loudness you're talking about is almost impossible to acheive when actually playing :shrug:

I started whacking each key as hard as I could, and it just didn't seem natural at all to me - when you have all five fingers spread over the keys, you can't hit it anywhere near as hard as that, and when I DID try the super-hard one, it sounded more artificial to me. This is purely meant for playing pop music, and even though I'm famous for slamming the crap out of my piano, I still couldn't hit a chord as hard as I could hit a single note with a few tries.

I said in my first post that the velocity curve is a bit off, because there ARE harder strikes than you're hearing there, but I'm not sure I want to whack the crap out of the piano, or do ridiculous pianissimo either - it's simply not how the piano is played, IMO. Above all, I wanted this to be pretty simple, so there are only 3 velocity layers - if you had a super-hard strike and a super-soft strike, there's a lot of territory for the one layer to cover in between.

That said, I appreciate you taking the time to post, and I'll read your post again when I wake up properly :hihi:
I've joined Lurkers Anonymous.

Post Reply

Return to “Samplers, Sampling & Sample Libraries”