Question for sound bank designers

[Tugdual]

How do you, sound bank designers, approach the overall design? Do you envisage the sound bank as a whole with a project in mind or is it a random compilation of sounds you did before? Do you envisage to have that many pads, that many leads, that many basses, and adjust spectrum so that they mix together well?
Also something which I think often miss on bank description is the expected number of voices. I think that if 2 stacked voices are used to build attack and then release separately, we should expect 16 voices providing that only 8 notes polyphony remain.

[Razmo]

In my case, it was simply the need for Ambient/Fantasy types of sounds, which there are never that many off, in the factory sounds of a machine... I simply needed them to make my own music... so you could say that my approach was predetermined, when it came to the genre, but the actual sounds themselves was a result of experimenting with the synth engine... i never decided in advance how many of each category I wanted, though I understood the need to have at least a good mix... It probably depend on what you're aiming for... a specialized bank with roughly the same type of sounds, or a mixed back of tricks.

So with that as a guideline, I simply just messed around with the synth engine, until something came up that I found nice... it was only in the end phase of the bank design, that I went for a specific type of sound, simply to try and even out the different types a bit.

Personally I'm not very good at planning a lot of stuff ahead of time... what ticks my clock is the fun of sound design, and I need to feel an urge to try something specific if I am to get a good result out of it, or I will simply cut corners because i get bored with it... telling myslef that "Now I need to make 128 sounds with only PADS" would be certain kill... it will be fun for the first 16 sounds, then I'll be dying to create something else

I see my soundbank as a work of art, rather than a product made to please anyone... that's the sheer truth actually... it was made for me, not any of you, and the only reason I decided to sell it, was because so many asked for it actually.

[RobustAmerican]

How do you, sound bank designers, approach the overall design? Do you envisage the sound bank as a whole with a project in mind or is it a random compilation of sounds you did before?

Hello Tugdual,

I don't really think there is a right or wrong way to go about it. If there was we probably wouldn't have so many great and varied options from so many sound designers. I love variety.

Do you envisage to have that many pads, that many leads, that many basses, and adjust spectrum so that they mix together well?

No, I simply make patches until I have way too many. Sometimes I will shuffle the final lineup so it is more balanced. During the whole process I am constantly producing little song/demos with the patches. (Can't help myself) This lets me hear how things are going to work in a musical context. Often, this results in more changes to the patches, sometimes not. This also gives me a firm grasp on what types of patches are still needed for the final lineup. Having said that, I like the idea of being able to produce a complete song with just the patches in the bank. Drums and all.

Also something which I think often miss on bank description is the expected number of voices. I think that if 2 stacked voices are used to build attack and then release separately, we should expect 16 voices providing that only 8 notes polyphony remain.

I don't worry about the voice count when making the patches. However, I do include the voice mode in my descriptions. (Single Layer, Dual Layer/Stacked and Split) I have no idea if the end user has the 8 voice or 16 voice version of the Rev2. I just supply the patches. It is up to the user to manage how the available voices are to be used.