Check this out...multisampling every note is really tedious, and sampling every open string then transposing the root note to cover a range of notes sounds sketchy because the formant doesn't shift (the difference between splits doesn't bother me so much, because different notes do sound different on bass).
So I tried sampling every open string, then copying samples to fill in the notes between samples. For example, I sampled E and A, and copied the E to what would be become the F, F#, G, and G# samples.
Now, here's the cool part: I used the transpose function for audio to shift the pitches. So for the F, I transposed a copied E up one semitone. For F#, I transposed another copied E up two semitones...etc. There's a Bass Solo transposition algorithm (how
extremely convenient!!), and it really does the job.
As to sound quality, it doesn't sound at all like standard sample transposition because the transposition algorithm seems to preserve the formant. In some ways, I prefer it to the sample-for-each-string approach because there's more consistency (it's really hard to sample each note with the same attack, tone, level, etc.) yet there's a timbral change when changing notes among strings.
This technique really speeds up the sampling process, to say the least.