A decade ago, true legato sampling was considered to be a specialty feature reserved mostly for premium, flagship products by larger developers. However nowadays, true legato is more or less a minimum requirement anytime a developer releases a product based on an acoustic instrument of voice that will be used in a melodic musical context (especially bowed strings).
While there are myriad approaches to true legato sampling, many of them closely guarded in-house secrets, there are a few concepts that I can share that will be more or less unchangeable regardless of the specific method you choose for your product(s).
True legato sampling involves recording the individual intervals in between notes in up and down directions usually up to an octave. Let’s use the note C3 as an example, a “round” of legatos intervals for this would involve capturing the following.
We will talk about a couple of specific ways of capturing these intervals below but the basic concept above remains the same.
The most time consuming way to record legato transitions is to record each interval transition leading into a full length sustain of 6-8 seconds (or however long you want your sustains to be).
Those are the basics of true legato sampling, as you are probably starting to see there are a million ways you can approach recording true legato intervals. How you approach true legato in your products is going to come down to the budget and time you have available, not to mention the skill of your scripter.
There is no right or wrong way to approach things. It helps to do your research, experiment and don’t be afraid of making mistakes. Many developers have changed their approach to true legato sampling over years of experimenting with different approaches.