Your Sampling strategies for keyboard instruments?

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

Post

niktu wrote:I would luv to hear some other perspectives on RRs and the every-note-sampled approach for keyboard/synth instruments.
How long is a piece of string?

It all depends on the instrument. If someone was to commission me to sample a monster string quartet that'll be featured in isolation in movie soundtracks (and charging a wodge of money for it), I'd do it chromatically and in great detail. On the other hand, if it's just a string pad that's likely to be buried in a pop song mix, I might do every minor third. For some of the hardware manufacturers I worked with, done stuff less than that.

And for some noises, I might do one sample covering the entire keyboard range.... or maybe every 5th, maybe smoothing the transitions between the samples with a hint of positional crossfade.

It all depends on what it is and the application it will be used in. And the price it will be sold at - something for £300 requires more detail, perhaps, than something for a fiver!

It is kind of amusing that some of the revered samples people want today were a few kilobytes in length, 8- or 12-bit with a 32kHz sample rate (if that) and spanned the entire keyboard range. The aforementioned M1 had 2Mb (or was it 4Mb?) of ROM yet people still love those sounds.

And I am always tickled when people sample things like the old LinnDrum at 96kHz, 24-bit, with round robin and multiple velocity zones - the thing had a dodgy 10kHz bandwidth, 8-bit, simply re-triggered the same sample and had a dynamic range of 2 ... on or off! The whole lot was crammed into about 100kb of ROM on separate chips ... and then you see a 96/24, 4Gb library of it released! Makes me giggle :)

Cheers,


Stephen

Post

hollowsun wrote:And I am always tickled when people sample things like the old LinnDrum at 96kHz, 24-bit, with round robin and multiple velocity zones - the thing had a dodgy 10kHz bandwidth, 8-bit, simply re-triggered the same sample and had a dynamic range of 2 ... on or off! The whole lot was crammed into about 100kb of ROM on separate chips ... and then you see a 96/24, 4Gb library of it released! Makes me giggle :)
It's probably just marketing strategy, those kind of libraries doesn't make any sense to me too!

Post

hollowsun wrote:
It all depends on the instrument. If someone was to commission me to sample a monster string quartet that'll be featured in isolation in movie soundtracks (and charging a wodge of money for it), I'd do it chromatically and in great detail. On the other hand, if it's just a string pad that's likely to be buried in a pop song mix, I might do every minor third. For some of the hardware manufacturers I worked with, done stuff less than that.


And I am always tickled when people sample things like the old LinnDrum at 96kHz, 24-bit, with round robin and multiple velocity zones - the thing had a dodgy 10kHz bandwidth, 8-bit, simply re-triggered the same sample and had a dynamic range of 2 ... on or off! The whole lot was crammed into about 100kb of ROM on separate chips ... and then you see a 96/24, 4Gb library of it released! Makes me giggle :)

Cheers,


Stephen
yes I suppose it depends on the target how to go about the sampling...price and usage...

I have to say Im not sure about the idea of making a 96k library (or even 24 bit)...but I do think there is a lot to be said for round robin...because I find it quite audible when the same sample is repeated over and over - and it always sounds nicer to me when you hear a real linn drum recorded vs a single hat, kick and snare sequenced in a DAW...I think the ear picks up the repetition...

In terms of the super small ROM machines...its funny - but I guess there is circuitary and digital programming that comes together to make a nice noise - even if it is not realistic. I think in a DAW if you try to put those ROM samples and play them back with a perfect neutral sampler and pristine converter - it doesnt sound too great. An example of this was a sample bank from an unnammed developer of the D50...they used partials as the basis in kontakt. It sounded harsh. Compared to similar patches of a D50 playing back the partials in its presets...which sounded warm and musical as the D50 is...

Unfortunately I find most soft samplers to be best suited at playing back recorded material and get results from having to record in tons of information as opposed to taking a small piece of data and shaping it into a musical instrument....the softsamplers not so great as creative soundshaping samplers like old hardware samplers or romplers...that is the bit that makes me laugh - the fact that an akai s1000 can sound more musical than a softsampler in 2013....
niktu wrote:I'm doing 3 round-robins (though I sample 5+ for problem takes & random artefacts), 6 octaves (or there abouts), minor third intervals and aiming my peak levels at around -3dBFS. No normalising, no stretching and generally avoiding post-processing as much as possible.

I'm keeping the minor pitch variations (nuance?) between notes on my DCO synths and going to try Expert Sleepers Silent Way when I have another crack at my VCO mono synths. It apparently has some clever 'listening' trick to send out the correct pitch. Another KVR poster suggested it in another thread and have since looked into it and it sounds promising. Looking forward to trying it out.

mkdr - Indeed.
yes silent way tunes the synth according to the input during set-up...

thats an interesting strategy 3 RR with minor thirds...will try that and see if I can hear the difference...

Post Reply

Return to “Samplers, Sampling & Sample Libraries”