Thanks DG and William.
More freaked out than ever. Please understand that this is not a gripe @ VSL... clearly -every- lib maker is heading in the same direction requiring HUMUNGOUS systems. To a non-orchestral mockup user a 24GB system or dual processor system -would- be considered -big-... even these days. OTOH, to my friends who do video editing? That's -nothing-.
But the overall msg I get is that if you want to use the entire Dimensions Strings, count on RAM/CPU to be like loading 24 VI Pro instruments. And that -is- HUGE, to me. If one extrapolates that out to an entire 'Dimensions Orchestra' of brass and winds, even a single tonic chord of say a Mozart Symphony, will require enough RAM/CPU for 40+ VIs. Holy mackerel.
So... if one goes down the Dimensions path, you really -are- heading into a new 'dimension'. Instead of loading one VI with 16 violins, you're loading 16 VIs... each with separate guys.
Now, one may say that this is no big deal because one can 'mix n match' libs and 'freeze' sections, but then, I submit that it creates yet another level of 'planning' which separates one from the score.
IOW: In a perfect world, Dimensions is great because with auto-divisi and the separate 'players' you approach the ability to have a 1-1 correspondence between a score you could hand to real players... and a great mock up. That is -very- appealing. But the cost is a water-cooled network of PCs.
Or, in a world where the end product -is- the mock-up, I guess it's a moot point because you can freeze and mix n match. IOW: you can plan your product based on your resources.
Sadly, in -my- low-rent world, where I have to blend live audio, with MIDI orchestra -and- be able to add or edit tracks at any stage of production... right up to delivery. And then give real players parts. So for -me- having one stave that says '1st Violins' is MUCH easier to deal with than, say, having 5 MIDI tracks covering '1st Violins' that alternate between Dimensions or Chamber Strings or Appasionata depending on the passage. I never wanted to do that. I just want ONE string lib that does it all. Maybe D/S is a step in that ultimate direction.
Again, I'm not -complaining-... the technology is where it is and VSL is at the cutting edge. One can't complain because time machines haven't yet been invented. But still... this is far from a panacea and I think all potential users should think HARD about what D/S means. To -me- it's a clear signal from VSL that this is the direction they'll be going in from now on... and you better have the proper kit or don't bother. Fair enough, video editors deal with the same thing: If you don't have THE biggest/fastest machine on the block, don't whine about how your video won't render.
Also, I think this is -good- for the state of the art. It's seems the only way to compete with the 'orchestral construction kits' I loathe so much. Love the approach.
But still... I think it's fair to say that if I get D/S I'll also be saving for at least -2- new machines.
Finally, all ranting aside: I am very curious as to how users will decide to use D/S. IOW: will it find -occasional- use in complex passages, while Orchestral Strings and C/S will do the majority of the heavy lifting? Or will it become -the- primary lib and other libs do 'fills'... say only where there are very static backgrounds? See where I'm going? If these things are so resource intense, how -do- they fit into day to day scoring if one has to -plan- where they are used?
---JC
