Vienna Symphonic Library Forum
Forum Statistics

201,041 users have contributed to 43,227 threads and 259,193 posts.

In the past 24 hours, we have 4 new thread(s), 25 new post(s) and 51 new user(s).

  • jamio, the streaming engine of the vienna instruments is finely tuned - any user settings can only worsen the performance (read: changing of any buffer sizes would interfere with other optimizations)
    basically you're limited to the data throughput in randomn access mode which your harddrive(s) is able to deliver.
    christian

    and remember: only a CRAY can run an endless loop in just three seconds.
  • That's the kind of info I was curious about. Thanks for explaining that cm.

    On a related note (I think it's related), I noticed that in Logic, I really need to leave my audio buffer at 256, or even 512, when using the Vienna VI's. Anything lower causes clicks and pops. No other sample libraries/engines I own, require that high of a buffer setting. That's really why I was asking, in case I needed to adjust something.

    It's certainly not the end of the world, as the nature of the sounds themselves (strings in my case), seem to still be fairly playable, even with the added latency. And I can always nudge the MIDI region forward as well, if needed.

    Still, it would be nice to know why that's happening. I'm using Logic 7.2.1, OSX 10.4.6. My Vienna Library resides on an external FW800 drive.

  • i wonder if the samples are streamed, why loading time of a patch is quite similar to loading it into halion's ram...
    opening a song with a lot of vi instances is pretty hell
    shouldn't the vi's be loading the patches signifficantly faster?

  • last edited
    last edited

    @Sangit said:

    i wonder if the samples are streamed, why loading time of a patch is quite similar to loading it into halion's ram...
    opening a song with a lot of vi instances is pretty hell
    shouldn't the vi's be loading the patches signifficantly faster?

    Loading what into Halion's RAM? You can't load VI into anything but the VI player.

    DG

  • I believe Sangit was referring to loading any large sample set into a sampler, where the whole sound is loaded into RAM.

    I too, was skeptical of the load times of the Vienna VI's at first, but loading some of the performances pre-sets... I would imagine that even loading just the first second or two of all those different articulations would still require a lot going into RAM. That would make sense to me at least...

  • sangit, i don't know any HALion patch which is loading as many samples as a ViennaInstruments patch .... the usual (no - an unusual tiny) portion of the sample gets loaded into RAM - if a sample is triggered, the rest gets streamed from disk (as bytes flow to your soundcard's buffer on the front they get filled up on the back from disk)

    dg, i might be wrong, but isn't there an option you can check in HALion to play everything from RAM?
    christian

    and remember: only a CRAY can run an endless loop in just three seconds.
  • last edited
    last edited

    @Jaimo said:

    I believe Sangit was referring to loading any large sample set into a sampler, where the whole sound is loaded into RAM.

    Ah, I see. It is very easy to check; just look at Task Manager and see how fast it all is loading.

    DG

  • i am comparing loading time of an epic horn performance legato patch (930 samples) which i use with halion3 to a performance patch of e.g. the vi woodwinds. loading time is nearly the same. indeed halion is a bit faster.

  • last edited
    last edited

    @Sangit said:

    i am comparing loading time of an epic horn performance legato patch (930 samples) which i use with halion to a performance patch of e.g. the vi woodwinds. loading time is nearly the same. indeed halion is a bit faster.

    OK, well you need to compare like with like. The other thing is that the first patch takes longer due to the dongle issue, so it needs to be a subsequent patch.

    I only have GS3 to compare with, but because the sample loading is so efficient the actual patch takes a much shorter time than in GS3, although the actual loading into RAM is not much different (as far as I can tell).

    DG

  • yes actually i was thinking about the dongle issue to be a reason.
    pity copy protection goes at cost of performance... thanks software pirates!

  • last edited
    last edited

    @Sangit said:

    yes actually i was thinking about the dongle issue to be a reason.
    pity copy protection goes at cost of performance... thanks software pirates!

    Yes it is a pity, but it is only the first patch that is slowed down considerably by this.

    DG

  • ok - then second pity is i load mostly one patch per instance. so its always the first LOL [:)]))

  • last edited
    last edited

    @Sangit said:

    yes actually i was thinking about the dongle issue to be a reason.
    pity copy protection goes at cost of performance... thanks software pirates!

    Yes it is a pity, but it is only the first patch that is slowed down considerably by this.

    DG

    Hey, that's good to know.

    So you would suggest, when first opening a Vienna VI, maybe load a small patch first, to "wake the communication with the dongle up", so to speak. Then, after that, load the larger presets?

  • last edited
    last edited

    @Sangit said:

    yes actually i was thinking about the dongle issue to be a reason.
    pity copy protection goes at cost of performance... thanks software pirates!

    Yes it is a pity, but it is only the first patch that is slowed down considerably by this.

    DG

    Hey, that's good to know.

    So you would suggest, when first opening a Vienna VI, maybe load a small patch first, to "wake the communication with the dongle up", so to speak. Then, after that, load the larger presets?
    You don't even need to load anything (AFAIK); opening Vienna VI does the wake-up call on its own.

    I load a blank template for all instruments with all my controllers set already. Then all I have to do is load matrices until I have what I need.

    DG

  • One question regarding the DFD:

    Is it possible to turn it off for some samples ?
    Because for short samples like Pizzicato or Spicato you save like 5MB (or even less ... at least in Kontakt for some patches) by using DFD instead of completely into RAM loaded samples. So imo it's not worth to stress the harddisc for these 5MB RAM.

    So is there an option to totally turn off the streaming for some samples or maybe is it even already, that the system reconizes, that only a few MB would be saved and don't uses DFD ?

  • last edited
    last edited

    @LazyPeon said:

    One question regarding the DFD:

    Is it possible to turn it off for some samples ?
    Because for short samples like Pizzicato or Spicato you save like 5MB (or even less ... at least in Kontakt for some patches) by using DFD instead of completely into RAM loaded samples. So imo it's not worth to stress the harddisc for these 5MB RAM.

    So is there an option to totally turn off the streaming for some samples or maybe is it even already, that the system reconizes, that only a few MB would be saved and don't uses DFD ?

    No.

    DG

  • Hmm, why not ? Would be a really important function in my opinion. Especially in near feature, when it's possible to use 8GB, 16GB and more.
    I mean, you would be limited by the speed of your harddrive while you still had gigs of RAM free for the usage with samples.

    Maybe there's a chance to see such a feature in an update ?