Vienna Symphonic Library Forum
Forum Statistics

200,722 users have contributed to 43,209 threads and 259,123 posts.

In the past 24 hours, we have 3 new thread(s), 8 new post(s) and 61 new user(s).

  • Real world stress tests...pops/clicks...buffers...bidule??

    Am i missing something...

    according to the VI stress tests ( http://community.vsl.co.at/viewtopic.php?t=7977 ) you can run 11 instances of VI with just 12 secs of Latency and the CPU at 85% on an iMac G5.


    I recently set up a slave 1.25GHz G4 powermac with 2GB RAM running orchestral strings I. I tested it running in Cubase and plogue bidule.

    Im loading two patches...staccato Vls and Sus Violins. (Im nowhere near the RAM limit)

    I can play slow single notes fine, but any fast runs or multi voice chords cause pops and clicks (lots of them).

    These pops and clicks usually start at about 30-40% CPU.

    So,

    is it at all practical to be running VI with the CPU at 85% as indicated in the stress tests? or am i doing something wrong to be having pops and clicks at 30% CPU?




    Next topic...
    I mentioned I tested VI in Cubase and plogue bidule. I actually had more success (sort of) using Cubase. In Cubase i was able to load up serveral Violin and Viola articulations in 2 instances and have them play back with little or no pops and clicks.....

    The catch?

    I had to set the buffer setting so high that my latency was 210ms!

    In Bidule it seems that no matter how high I set the Buffer, the pops and clicks dont go away and the latency doesnt noticeably increase. Any ideas why? Also does "FFT" do anything when using bidule?

    Thanks in advance for any help...

    i fell backwards into a job setting up a VI rig for a music production house i work for. I've been trying to get this thing going for a few weeks. We have extra G4s lying around so we thought we could use them as slaves....but if they cant even play back a couple of patches in one instance of VI...i think we'll have to shell out for new minis (and Vienna should probably change their "system requirements" [*-)] )

  • I have same problem ccuts, in Bidule but using MacMini Intel. 5 instances and I start getting pops and clicks, but CPU is far from being maxed out. Setting buffer does change things a little, but pops and clicks remain no mater what.

  • thanks Vagn Luv for your reply. are you able to operate fine with 4 instances or less of VI? what percentage of CPU are you showing?

    So does anyone know why there would be all of these pop and clicks with out maxing the CPU?

    Chris

  • I also had problems with Bidule. Bidule taxes the CPU much harder than other hosts. On the PC, if you run it ASIO mode it was hitting 95% on the PC CPU where other hosts were hitting 30%. If you run on one of the other modes (PC) it's much better but the trade off is unexceptable latency, somewhere up in the 1200ms range.

    I think your Mac problem is related. I didn't get good results running it on my Macbook Pro either. The Rax preforms much better. Overall, Bidule doesn't seem to be very efficent for big sampler VI. I don't think they really designed it for this kind of use. Too bad, as the flexiblity would probably offer a lot for VI control.

    After experimenting with it on 4 PC's and one Mac, I gave up on Bidule. I'm now using V-stack on the PC (and currently trying out Forte) and Rax on the Mac instead.

  • thanks tripit...

    i had read so much goodstuff about Bidule on these forums, I thought it seemed like the only way to go. Rax and V-stack are no longer available...

    chris

  • Yeah, it's a bummer that the two that seem to work best have both been dropped.

    Forte has an GUI/graphic issue with GVI, which I also use, but hopefully they will get that resolved. Otherwise, it might work out well for PC.

    I have also done some limited testing using VSI with Live as a host on mac. It's probably not the most elegant solution, but it might be a way one could go.