Vienna Symphonic Library Forum
Forum Statistics

194,249 users have contributed to 42,914 threads and 257,941 posts.

In the past 24 hours, we have 2 new thread(s), 15 new post(s) and 97 new user(s).

  • Kage49, are you sure that Cubase and MIR don't try to access the same audio port? You have to tell both applications that they _must not_ touch the audio ports used by the other program. In Cubase, it's done in the VST-connections (F4 by default); in MIR it's under Audio Settings / Audio Routing.

    Apart from that: What are the system-wide latency settings of the RME? And how many buffers does MIR use additionally ("Buffer Multiplier" ... it's under the Preferences / Advanced)?

    Kind regards,


    /Dietz - Vienna Symphonic Library
  • Hi Dietz, thanks for responding.

    MIR is using ADAT3+4 for output, which is routed through the RayDAT internal mixer to ADAT3+4 input.

    Cubase has ADAT3+4 input as the "input" in VST connections, and ADAT1+2 as output. The buffer on the RME is 256. The multiplier on MIR right now is x16. (i kept increasing it because I also suspected that the pops were due to not having enough latency for MIR to handle all the instruments).

    Also, before the RayDAT I had 2 M-Audio cards and I made MIR use one and Cubase the other. The output from MIR into Cubase was using spdif cable. I had the exact same problem, this is the main reason I bought an RME just to be sure :(

    It might be worth noting that there is a guaranteed pop and/or total silence for half a second when I maximize or minimize a process with a priority set to "high" like Cubase... is this due to me having lowered the MIR process to "normal"?


  • We discussed your problem today, but we really can't see a good reason for the behavior you describe. One possible culprit is the video card, though: Did you ever try to bring down the "Venue Details" in MIR's Preferences to a value like 0.5 or 0.25 - just for testing? The Venues won't look as beautiful as before, but it will take off some strain from the graphics engine.

    Apart from that, the next MIR update to Vers. 1.2 (which is just around the corner) could solve the issue, as it introduces a different handling of process priority.

    Kind regards,


    /Dietz - Vienna Symphonic Library
  • I've tested with two cards already, no luck.

    I'll try the following, in order, to troubleshoot:

    1- Lower venue details, try.

    2- Wait for MIR update. try.

    3- Update to Cubase 5.5, try.

    4- Downgrade to Vista 64. try.

    If all of the above doesn't work, I might explore changing my mobo or RAM chips.. if not I'm just screwed [:(]


  • I just re-read your messages and discovered the following sentence:

    "I tried changing from a radeon HD 4850 x2 to an Nvidia GTX 460.."

    ... did you ever have the chance to try one of the more recent ATI cards? In our tests they have proven to be the best choice for MIR's needs, most of the time.


    /Dietz - Vienna Symphonic Library
  • I have another PC with a radeon 58xx card (not sure which), I'll find some time to try this too...

    But I just tried something: I lowered Cubase's priority to Normal and set Mir's output directly to the monitors... I did not hear any pops/clicks :/

    Of course I cannot record MIR's output because Cubase jitters like crazy, due to  the low priority. So is it possible Cubase is hogging the CPU resources because it's a higher priority than MIR? (Cubase > High, MIR > Normal)

    I have to ask: Anyone tried Windows 7 64x + Cubase 5.1 + MIR on one PC? 

    hmmm....


  •  Three other things to test:

    1. Try the Affinity trick with Cubase
    2. Raise the Cubase buffer to maximum and then lower it until the pops start.
    3. Make sure the video window is not open, even if hidden.

    DG


  • At the risk of sounding like the biggest idiot in this forum... Cubase needed more latency, that's all it was :(

    It never crossed my mind! why would Cubase need more than 256 buffer size? It's just sending out midi data and receiving one bus input... I've run Cubase with 80 VI's as vst instruments, each with multiple effect inserts from Vienna Suite running at 256 buffer size and not one pop! But for one input bus I need to increase buffer size? Ugh.. 

    Anyways, all's well that ends well I guess. I tested recording Battle of the Heroes multiple times and no pops/clicks so far. Thanks Dietz and DG and sorry to waste your time. I appreciate the help :)


  • Glad to hear it's all sorted out. One more thing you can try is to record directly from the MIR outputs. This will mean that Cubase doesn't need to have an input bus at all. Does that mean Cubase can run at a lower latency?

    DG


  • Great news, Kage49! May I change the thread's subject to "Solved" then? :-)

    BTW - DG's suggestions refers to RME's DigiCheck. This has a "Global Record" utility which allows for direct-to-disk recording from any of their audio interface's audio paths. You can get it from [URL=http://www.rme-audio.de/en_downloads.php?page=content/downloads/en_downloads_driver&subpage=content/downloads/en_downloads_digicheck]here[/URL].

    Kind regards,


    /Dietz - Vienna Symphonic Library
  • Yes you can change the title Dietz. :) Thanks DG and Dietz, I'll try the digicheck thing later today just for fun.