galaxy II download vs boxed edition

[Oxytoxine]

hello fellow pianists (and hopefully uli, the developer of the galaxy piano);

this is my first post on this forum, so hello to you all :)

i'm strongly interested in galaxy II but unsure whether to choose the boxed 24bit or the download 16 bit versions of all the instruments. i'm aware of the technical differences between the two: 24bit vs 16bit, half tones vs whole tones sampled, 50s vs 30s sample lenght, the different modes of resonance samples etc., but i do not know which of these parameters really do make an audible difference and which are rather "cosmetics". my experience up to now with piano sample libs is rather limited (i currently use truepianos and PMI bösendorfer 290), therefore i would be very thankful for some help.

to cut my questions short:

a) is there an audible difference between the two versions? should the download version suit the needs of solo jazz/classic/pop recordings?
b) how big is the system performance difference caused by the larger sample set size and the DFD disc streaming mode?

longer form:

a) what is the "real-world" difference between the 16 and 24bit samples? have the samples been re-recorded at 16bits or where they dithered? will i have a problem with dither noise summing up when playing with high polyphony? or in the end when rendering projects due to "double dithering"? and does the absence of chromatic sampling result really in an noticeable difference?
compared to the bösendorfer, the galaxy vienna grand seems to be duller, but i don't know whether this is just because it's another instrument differently recorded or whether this is actually the 16bit "lofi-effect", but i like the overall playability/balance/feeling/more velocity layers etc. of the vienna much more.

b) the reason why i don't simply buy the boxed 24bit version (making the above mentioned troubles obsolete ;-) is that there seem to be performance problems with the full version (about 9gb/instrument) that do not occur with the download version (about 2 gb/instrument). is this mainly due to the simple sample size difference or rather the DFD mode?
i have a C2D E6750 2.66 ghz 2GB ram computer, but using the PMI bösendorfer 290 in kontakt 2 causes some CPU overload on my system at high polyphoonies if using resonance scripts, so i'm a bit afraid). is there hope the "big" galaxy II would be satisifed with this resources?
(the download version does run smoothly without any problems and rarely exceeds 10% CPU even at really high polyphony, also disk streaming seems not to be a problem at all).

sorry for the many questions and the unstructured way of presenting them; i'm fairly new to the technical aspects of vstis and any comments/tips/experiences would be highly appreciated!

thank you very much and stay tuned

oxy

ps: i also posted a similar thread in the actual galaxy II forum, but it does not seem to be very frequented..:-(

[Per Lichtman]

Since no one else had chimed in, I thought I'd present my limited thoughts on the subject.

- The PMI "modeled sustain" or "modeled resonance" patches are some of the most intense ones I have played in a Kontakt library. None of the other patches in for instance their Emperor library come anywhere close to the CPU usage of the modeled convolution sustain one with polyphony being reduced to less than a quarter of the other patches on some systems and CPU spiking to up to 5 times the usage. This is why I tend to use the other patches in the library.

- The Galaxy II full version lets you turn all sorts of aspects of the sound on or off, including most use of IR, and is thus very tweakable to different systems. You can try it for yourself at their site http://www.try-sound.com/. My experience was that in normal usage it was significantly less CPU intense than the PMI patch in question.

- Chromatic sampling can make a significant difference, especially in the higher register.

[Per Lichtman]

Also, on a more specific note of CPU usage, here is a quote from Uli during the development of the library.


Hi everybody,

my name is Uli and I’m one of the producers of Galaxy II. I hope I can clarify a few things, which were coming up in this forum so far.
...

- There are no system requirements on the website simply because we are building this site bit by bit in the moment. There is so much to do. The support area including the FAQs ist also still empty. This will change next week. One of the testers tried it on a Mac G4 Powerbook and ranked GII more CPU efficient than Akoustik Piano, which basically has the same engine (Kontakt 2).


EDIT: Although I just realized many subsequent posts contradicted Uli's statement. Still, I'd expect performance to be consistent with what I described.

[Oxytoxine]

Per;

thank you for this very concrete, informative and helpful post!:)

i'm happy to hear that this resonance script patch seems to be outstanding demanding and that this is not the case for all sampled pianos in kontakt.
now i can really be looking forward to the 7cg and white grand i ordered, after having played the pmi bösendorfer i thought that if already this "small" (ehm..) piano causes such problems i would never be able to run really huge libraries.
i've in the meanwhile also got the galaxy II bösendorfer DE and really like it, and it's good to know that the "full" version will be even nicer!

thanks again and stay tuned

oxy

[Per Lichtman]

Thanks Oxy, you're very gracious. :)

If and when you get the boxed version, make sure you have the 1.1 patch update installed as that seems to have been important for performance. Feel free to post some your further experiences to help others thinking about the DE vs. boxed edition as I only gave impressions rather than concrete numbers and would have loved to have helped more.