It is a learning curve to get where you want. But using GP blows away hardware alternatives in terms if quality, flexibility, etc.
One aspect of all this that has taken me a bit to learn is how being aware of ram usage. I just dug in originally creating new rackspaces freely with different sample libraries.
If you just have a single “Gig file” for everything (without using “Predictive Loading”) you basically can access all your rackspaces freely and with no loading time. The downside is every new rackspace (especially if they are sample oriented, rather than soft synths or physically modeled) increases the ram you use on start up.
I have 32 GB of ram in my “old laptop”. So far, I am not having a problem with lots of rackspace and sample libraries, but I think I “may” be getting close (in terms of ram usage) and loading up a gig file (I think) takes about 6 mins.
So, I splurged on a new laptop with 64 GB of ram (expandable to 128). This is crazy for most people, but I wanted to “future proof” my system.
But, really like the comfort of having two duplicate laptops, so I am trying to think a bit about ram usage now that I understand things a bit better. So, I may shift to more physically modeled sounds, reuse rackspaces more, use variations with bypassing plugins more, etc.
So, let me at least try to respond to a couple questions:
Q:But how many songs in a rackspace?
A song (setlist Mode) will usually require one rackspace and variation. If you reuse variations of the same rackspace for different songs, this will conserve ram
How about version control and backups? Just save lots of versions?
A: I basically use the same large Gig file for everything. This can (and does) use a lot of ram. But, I have access to every rackpace immediately. I now end up saving this Gig file on a Google Drive and on my new duplicate laptop. But, if I want to do something that I am unsure about, create a new Gig file to “experiment”.
Don’t know if any of this was helpful to you, but…