i cant directly anwer your question… (ive used Final Cut at university in the past on macs, and back then storage space was always an issue - we had to render small clips around 5 mins at a time and dump to zip disks LOL)
but, i use the 4-slot Drobo (v1), for video editing with windows movie make (smaller scale im sure) but its always been fine working off the drobo for source, and destination material. (only my cpu made it slow here and there but that was due to an older machine, half a gig of ram, and lots of fancy effects i was using)
at the high level, if i had a DroboElite and certain software etc i’d try the following…
(in between all tests, its good to let the drobo take some rest to get itself back in synch/auto defragged etc etc, so as not to skew the tests)
id probably say, have the main software installed on a defragged main system drive (if it doesnt take up too much space with base install and plugins etc)
have your source material on one of the volumes, and your final outputs on another volume.
(only if you already have 2 volumes - if you just have the one its ok to try).
then see how a benchmark file performs, using scratch discs on your main System drive,
and then scratch disks on your droboElite. (try and have a meaty benchmark, eg something that will take say 5 mins of rendering time and something meaty like say 1gb of output minimum.
if its windows you are using, check your swapfile size (and if you have more ram and space, you could set it to a higher min and max value)
a fair test might be to set up the preps, and then
reboot, and try test one. (repeat to see)
reboot and try test 2 (repeat to see)
when you get a rough idea, you can swap your benchmark for something more similar to what you think you might be doing… eg (if you are simply compiling small clips together, into a large high-bit-rate format, then main limiter will be the throughput rate of the drives. and if you are going to be adding lots of fancy effects, then main limiter will probably be more to do with ram and cpu calculations, rather than raw drobo data throughput - if you see what i mean?
hope this helps a bit, and if you do any tests, would be cool to know how things went
(ages ago, lightwave by newtek had a raytracing juggling man as a benchmark on sgi machines i remember my friend buying one - it was the same render routine used by all platforms and people to see how many hours it took to render on their various systems setups etc :)[hr]
btw when i mention defrags, i mean the auto defrags built into to drobo,
for you to defrag your system (non-drobo) drive.
eg, DONT defrag the drobo… (you might cause problems etc - let it auto defrag itself etc)