Vienna Symphonic Library Forum
Forum Statistics

196,946 users have contributed to 43,043 threads and 258,499 posts.

In the past 24 hours, we have 2 new thread(s), 15 new post(s) and 54 new user(s).

  • VI, ProToolsHD, FXPansion wrapper - WinXP report

    I just spent a few hours with a new install of Brass I, II and Special on our Windows ProTools HD machine. I'm running it with the latest version of the FXpansion VST-RTAS wrapper. I have the rest of the Symphonic Cube installed on a separate Mac G5 running Logic Pro.

    Basically, the VI plugin does seem to work without major issue on our system. I've run 8 instances simultaneoulsy so far, with a mix of instruments and matrices, and the hit on CPU overhead is barely noticeable (system is a Digidesign certified Dell Precision 690 - Quadcore 3GHz Xeons with 4GB Ram). The issues I've come across are generally fairly minor, and I feel could probably be very easily addressed by the VSL team exploring the wrapped version of the VI plugin a little further, and perhaps refining it a little - if necessary as a dedicated plugin version designed specifically to be wrapped.

    1. There are some GUI issues - when the plugin is first instantiated, the main information windows to the left and right of the GUI appear as large grey blocks. Only when you press the buttons to the top right of the GUI are the proper windows restored.

    2. Pretty much ALL the Standard Edition Perf-Legato-Speed matrices do not work correctly. The perf-leg-slow articulation does not sound in any of the standard versions of this matrix for any instrument (pretty sure I tried them all). Obviously this is a fairly big deal, but I suspect it's something fairly minor in the matrix programming, which could easily be addressed.

    3. Strangely, the perf-legato-speed matrices for Piccolo Trumpet and Vienna Horns work fine. Looking at them, I can only assume that this is because neither of these matrices use a perf-leg-slow articulation in the first cell. Instead they use a perf-legato-no_vib-sus articulation in it's place.

    4. Even more strangely, all the L2 (Extended Edition) perf-legato-speed matrices work fine, even though they use a perf-leg-slow articulation in cell 1. I could be missing something, but if these work fine, I see no reason why the Standard Edition versions of the matrices should fail to work correctly.

    5. In a lot of matrices, I am not hearing release samples if relatively short notes are played. e.g. TTB Perf-Repetition-Combi - alot of the articulations, if I play short notes which don't play to the end of the sample, I get a rather unnatural cutoff of the note, with no obvious release samples. I don't know if this is a limitation of the programming, a bug/oversight, or an issue specific to the wrapped version fo the plugin, but it does handicap the authenticity of the sound somewhat.

    In short, a few small issues, which may be critical under some circumstances (like if you don't own the Extended Versions), but my impression is that they could easily be sorted out with a bit of coding/testing time. More impotantly, it seems that this means of running the VI plugin in ProTools has huge potential, at least on a powerful machine, and I can see myself getting more instances than I will ever need on this system alone.

    Finally, I should add that I am more impressed by both the sound of the library and the scripting of the plugin, than I have pretty much ever been with a library or plugin. It is a quite remarkable achievement, and I can't wait to start writing in earnest with this awesome library.

    Jules Bromley

  • thank you jules for such a detailed report, although in a first run i can only comment on the display issue: this seems to be rather an issue with video driver or the wrapper (not drawing the window at initial start), sounds like a timing issue.

    basically this is another report that PT|HD on XP with a RTAS-VST wrapper works with usable performance but i have to admit we don't have such a system here and i'm not sure if we can find some in a reasonable time and distance. however on the other points you will receive a reply from team members who know more about it.

    regarding your other thread and AU validation - have you already tried to validate VI explicitely from the respective logic menue? i've heard that sometimes the *automatic validation* fails ...
    christian

    and remember: only a CRAY can run an endless loop in just three seconds.
  • Unfortunately it looks like my initial optimism was a bit premature. Although I have been able to work with up to four instances of the Vienna Instrument plugin, additional instances in a proper working session start to introduce serious instability to the system, resulting in Access Violation errors, freezes, crashes and even bluescreens.

    I've no idea what the problem is. The host CPU meter is very low (around 15%), PCI load is also low, and the system is only showing around 1GB of memory in use by ProTools, leaving 3GB of real memory available.

    This is a real blow, because four mono-timbral instances of VI doesn't really get you very far in an arrangement. I'm going to have to think carefully about how on earth we can run our vew Vienna Symphonic Cube (in part at least) on our Windows ProTools machine.

    If I have any additional revelations, I'll report further. For now it's back to the drawing board.

    Jules