While all that you said about strings is true, it's not necessarily true that the above is a waste of time. I've found string samples that literally were much more delayed on certain notes than on others. That's the fault of the person recording the samples. After 10,000 takes, working late on an empty stomach, he accepted some samples he should have done over. Or the person chopping them up and loading them into their proprietary software perhaps allowed a few milliseconds before the first transient. Whatever happened, the result was that some notes were slow-starters.David Polich wrote:As a LONGtime programmer/sound designer who has worked for Yamaha, Korg, Roland, Dave Smith, Native, and Steinberg, I can tell you you are wasting your time looking for clues in the attack portions of raw samples.
I see none of that in VSL's Dimension Strings, which is really an amazing string library. But in old libraries from Garritan, MOTU, Kontakt, and such, there were occasional inconsistencies. And some libraries whose names I forget just had a whole different feel in the attacks, as if someone had a completely different philosophy about how string samples should work.
Shooshie