I haven't tested it, but I think I read somewhere that the SD card reader is connected to the CPU via a relatively slow internal USB bus, which strangles data throughput so much that it doesn't matter what you use. The internal SSD SATA connector suffers from the same.
There's also something wrong with how the Force's own software handles opening files. If your project or keygroup program or drum program uses a lot of separate files, it will take many times longer to load than on an MPC. But if you combine all the WAV files into one big "chain" sample file, so that all samples share the same file, it will speed up loading by an order of magnitude. So it's not so much about how much DATA or disk block accesses there are, it's how many FILES it uses. Sounds incredibly dumb but that's the way it is with the Force. The MPC will also benefit from sample chaining, but not as much as the Force.
My
own experience with sample chaining
A certain Rhodes sample set I converted from Kontakt and reduced to Akai limits in Ableton, 165 megabytes, 418 wavs, takes over 60 seconds to load on my Force 3.2.3. But when all the samples are chained in one WAV, it loads in 4 seconds
So it looks like, in theory it could read at least 40 megabytes per second from the SSD (
at least) , but it needs over 130 milliseconds
per file. What the actual fsck can it possibly do for 0.13 seconds, for a single file??? That would mean an incredible processing rate of
eight files per second. I'm sure this would also get Roger Linn's approval, well programmed, boys, good job.
All in all, I'd say, it is what it is, and there's nothing you can do to make it any significantly better. The biggest slowdown is the software, which you can't change.
If you want to test different file systems, it should be fairly trivial and straight-forward to do. Buy several identical memory cards, format them to different file systems, copy all the same files to each card, and measure the loading times.