@Simon Ravn said:

[quote=christofori][quote=Simon Ravn]I don't use any firewire harddrives. I use two WD 500 GB RE2 SATA drives internally for sample playback.

Simon, are you using OS X.49? I also have a new mac pro dual 2.6, with 9GB Ram & a bunch of 500GB SATA2 drive, while suffering an assortment of dropouts, no matter how many instruments are loaded.

I'm going to do an experiment- create a new boot drive (OS X.4[H]on my 10k rpm WD Raptor, install nothing but DP & VI, use a 500GB internal SATA2 for samples, & see what happens. I'll report my results to the forum.

Mark
Mark, I have X.4.9 but do not use DP. I use LP 7.2.3. I predict your installation will not work if the way VSL VI handles its heavy processing load conflicts with the other hard drives competing for ram. I need to look at Herb's examples of the Mac setups for stress testing if they had more than 2 internal hard drives. Unfortunately, I thought only two Mac setups were mentioned and one was an iMac.
I have a question. When you have multiple internal SATA drives on one computer, does the operating system switch its attention from one drive to another at times feeding the RAM so that one drive cannot monopolize the system and this is causing sound drops and clicks and pops? In other words, could the problem be similar to the firewire clogging problem I experienced? If you had one internal SATA drive versus multiple ones would that make the problem go away? Does the fact I have the new VSL VI on an external USB or firewire while I have the Pro edition VSL, Colossus, and EW Symphonic Orchestra Platinum on an internal SATA identify the location of the conflict since my system is working? Would the problem others are having relate to multiple internal harddrives not counting the system hard drive? Would the VSL VI libraries on an isolated firewire bus and not on the internal hard drives solve the problem? An external firewire is not that expensive. This is an important question to me as I contemplate a new system. All I am saying is that I had the problem too prior to eliminating sharing of firewire resources with anything but VSL VI. Forgive me if my questions seem foolish as I am trying to learn to avoid these unpleasant experiences. I had wanted to upgrade to a Mac 3.0 ghz quadcore pro and put all my libraries on one computer with separate hard drives for each section and max out the ram at 16 gb. But it seems that may not work based on what I am reading. I.E. Not all ram is accessible, multiple hardrives may be a probelm, etc. Thanks. [*-)]

This is from Herb's stress testing. Also, most of the Windows systems tested used firewire. The sound management is MADI on one and built in on the other. My guess is the problem may go away with limiting VSL VI to an isolated firewire bus until this gets sorted out. But then, I could be wrong.

First Vienna Instruments Stress Test results on Mac OS X

So here they are, the first official test results with the new Vienna Instruments as Audio Units on Mac.
As on the PC platform, we have tested the Vienna Instruments on various computers and doublechecked on similar machines. All computers are fully equipped DAW´s that are used in everyday studio work, no additional tuning.

More tests are in the works, with different processors and hosts. Test results for the Quad will be released as well, we expect more Vienna Instruments to be loaded with maximum amount of RAM.

The limit of RAM usage is about 2.8 GB for the Vienna Instruments in a computer equipped with 4 GB RAM, that translates to 47000 samples.

The testing routine: Play a (really stressy!) stress test song with different loading options (more instruments, less RAM, and the other way around). All instruments are playing all the time, using all kinds of matrices and patches, and of course including additional MIDI data like ModWheel and different assigned sliders.

You always see the maximum Vienna Instruments possible in the given setup without any sonic trouble.

TEST COMPUTER 1

Processor: G5, 2 x 2.5 GHz,
Ram: 4 GB
OS 10.4.2
Data storage: Firewire 800
Host application: Logic 7.1.1
Soundcard: RME HDSP MADI with 648 MADI Interface

Latency at 512 Samples (12ms) / 2.5 GB Ram usage
Maximum Vienna Instrument instances: 24 (CPU 90%)
Maximum Samples loaded: 45000
Maximum polyphony: 300 stereo voices / (CPU 70%)

Latency at 256 Samples (6 ms) / 1.74 GB Ram usage
Maximum Vienna Instrument instances: 23 (CPU 70%)
Maximum polyphony: 200 stereo voices / (CPU 75%)

TEST COMPUTER 2

Processor: iMac G5 single, 1.8 GHz,
Ram: 2 GB
OS 10.4.5
Data storage: Firewire 400
Host application: Logic 7.1
Soundcard: Built in Audio

Latency at 512 Samples (12ms) / 1.08 GB Ram usage
Maximum Vienna Instrument instances: 11 (CPU 85%)
Maximum Samples loaded: 17400
Maximum polyphony: 250 stereo voices / (CPU 80%)

Latency at 256 Samples (6 ms) / 1 GB Ram usage
Maximum Vienna Instrument instances: 9 (CPU 85%)
Maximum polyphony: 160 stereo voices / (CPU 80%)

Last edited by herb on Mon Mar 13, 2006 4:55 pm; edited 2 times in total

@Another User said:

When will there be diplomacy and professionalism with regards to these matters? Will it ever change at VSL?

since i'm not working for a govement my need for diplomacy is fortunately limited but i'm allowing myself the exaggerated opinion to be sufficiently professional.
if you read through my posts, what might take you a while because this one is # 5.000, you should find i'm rather rarely getting emotional, although occasionally i'm not afraid to dispute with our users (note: i'm not wording it customers) intending to track something down to a rational cause.
christian

Still waiting for that PM or e-mail from you Christian...you know, the one where you're concerned enough to want to find out more about my issues so that together we can work on solving the problem?

midphase, you have already written to support@vsl.co.at providing as many facts about your machine and setup as possible?
such a request should also include RAM and device configuration, installed software and steps already taken trying to solve the issue. sometimes, if not often apperently unrelated details are leading us to a solution.
christian

@Another User said:

in the USA it's considered bad business and even opens you up to litigation to bring emotions to a corporation's public relations

It's not impossible that I'd make it through the first morning at that corporation, but no way would I last an entire day.

In Britain emotion has been banned altogether. If you want to smile or weep you have to go to Italy.

hahaha

midphase, thanks for sending some details to our support. the email also came to my eyes and a few things come into my mind:
4,5 GB RAM, this seem to be mixed RAM sticks - are they of the same timing?
digi core driver 7.1, is there a reason for not using their latest - usually newer drivers are performing better
digi002 + KORE 2, not sure now which interface is used how for output and what exactly does the 44.1 to 48 kHz conversion (logic?)
do you remember if the update from 10.4.8 to 10.4.9 went without noticeable oddities and have you applied the pro application update?
i didn't notice any other sample libraries listed, only plugins, so to what compares SE

i have the strong feeling you should temporarily remove some components to get a better testable situation, i'm not sure we can even duplicate your setup in detail - please keep our mac support updated.
christian

Hello Christian,

Thanks for bringing this back to a normal....civilized discussion.


You ask:

4,5 GB RAM, this seem to be mixed RAM sticks - are they of the same timing?

They are all PC3200U, I believe that their timings match....I bought them at different times (but always in matched pairs)...can you offer some advice as to how I may go about checking for timing differences (ie. by looking at the chips for some label marks)?

digi core driver 7.1, is there a reason for not using their latest - usually newer drivers are performing better

I'm also running Pro Tools 7.1 (have not updated to 7.3 yet) and Digidesign recommends keeping the same audio driver as the version of Pro Tools that you're using. According to their literature, there are no substantial differences between the AudioCore 7.1 and 7.3 so I just use what they recommend.

digi002 + KORE 2, not sure now which interface is used how for output and what exactly does the 44.1 to 48 kHz conversion (logic?)

Digi002 is the only interface that is used by all of my programs (Kore is only used in standalone more). My DigiCore driver is set at 48khz and my Logic Pro sessiono is set up at 16bit 48khz also.

do you remember if the update from 10.4.8 to 10.4.9 went without noticeable oddities and have you applied the pro application update?

Yes, no oddities. Also keep in mind that I did a fresh reformat and reinstall last week-end on a brand new drive (I was getting the clicks on my old drive too so I don't think the new one is at fault), so the problems have remained even on the newly re-installed OS.


i didn't notice any other sample libraries listed, only plugins, so to what compares SE

Several of my libraries have been converted from Giga to EXS24 or K2. A good comparison might be something like ArtVista VGP which is a pretty big grand piano instrument, chromatically sampled with several velocity layers and release samples. I think that it's fairly safe to say that it compares well to something like the VSL SE staccato flute? I want to remind you that 1 single instance of VSL VI causes clicks and pops even though my performance meters are not even close to maxing out. Could this really be a 48khz vs. 44.1khz issue?

Hi,

Just read this thread with interest. I'm a newcomber running VSL with Mac OSX 10.8 on a MacBookPro, 2 gig RAM, Logic Pro 7.3 and a 002 Digidesign interface. I've got similar problems to the ones mentioned here, but they're not the dealbreaker. My trouble is playback. When I get the CoreAudio Overload message, I'll hit continue and half the time I lose playback at this point, have to shut everything down, and maybe will be able to continue with a reboot. Very often playback remains disabled for a day or two. All in all, my ratio of recording to troubleshooting is about 50/50.
I'm taking the system in to have it checked as soon as I can. Am told 2 Gigs should be enough RAM to carry me for a while, but it's not happening. Question now is... should I go to 3 RAM, or just throw in the towel and wait till I can get a machine or update, or something that will make my system run.
I too would love to finally establish whether I'm doing something wrong, or OSX10.8/Logic/VSL is a bad match. From this thread, I sense an elephant in the living room...
Thanks. I'll be watching this thread to see what the conclusions are.

tom, how many GHz has the CPU and is it a Core Duo or a Core 2 duo and have you installed the 7.3 standalone drivers?
anything else running when you receive the overload message?
the digi is connected via firewire, right? something else on the FW-bus?
christian

I kinda suspect that nowadays 2gigs just doesn't cut it anymore for serious audio production. And I also think that the speed of the processor plays a big role in things as apps like VSL push the limits of technology.

Here's an update from my tests:

It appears as if VSL VI runs fine by itself, but when I throw in some healthy dose of Kontakt 2, then the clicks and pops appear. I'm actually not even sure anymore who is doing the clicking and popping, wheather it's VSL VI or Kontakt 2 when VSL VI is present (soloing the tracks makes the clicks and pops go away so that doesn't help determine who is at fault).

Either way, there seems to be some sort of conflict between VSL VI and Kontakt 2 as they both vie for the hard drive on which both sets of samples are on. I tried to move the VSL SE files to my system drive and point the VSL Directory utility to it, but this actually made the problem worse.

Switching from 48khz to 44.1 doesn't really affect things a whole lot, and neither does increasing the audio buffer.

The CPU doesn't appear to affect things either, in the sense that the clicks and pops simply appear when both Kontakt 2 and VSL VI are present in the same Logic sequence, regardless of how small the instruments loaded are or if it's just one instance of each.

I suspect that on faster machines, the adverse effects might be minimized to the point to where they are not perceptible....could this be because of faster busses, or a different SATA chip?

Either way, I seem to have isolated the culprit to some conflict between VSL VI and Kontakt 2

Christofori, sorry for the long delay in responding to your email.
I've done some different tests (with VI SE) with different drive configurations, and I can say that there is no real difference in performance.
Again, my computer is a Mac Pro with 9GB RAM.

I tried putting the VI samples on a singe 500GB SATA2 drive, a Raid of 2 of those drives, and a 10,000 rpm Western Digital SATA drive. All performed the same. I thought that VI might perform better on the 10,000 rpm drive, since my *gasp* EWQLSO performs better on my PC with that drive.

I think the bottom line is that many of us are loading up our Macs with a bunch of Ram, and trying to coax our machines into accessing more that the OS or our host sequencers will allow in a stable manner. I've tried the various methods of loading samples in different programs, inside and outside the sequencer, but this tends to make things less stable. I think perhaps Digital Performer is problematic in this type of situation as well.

The most dissapointing thing is that I have to run all my VI SE instruments on a single computer, and yet I can only use a fraction of them at one time. Ideally I would be able to create a sketch orchestra (very basic instruments), but I'm ending up relying on my QL instruments, since they seem to be more stable running in Kontakt.

Mark


Mark[/quote]
Mark, I have X.4.9 but do not use DP. I use LP 7.2.3. I predict your installation will not work if the way VSL VI handles its heavy processing load conflicts with the other hard drives competing for ram. I need to look at Herb's examples of the Mac setups for stress testing if they had more than 2 internal hard drives. Unfortunately, I thought only two Mac setups were mentioned and one was an iMac.
I have a question. When you have multiple internal SATA drives on one computer, does the operating system switch its attention from one drive to another at times feeding the RAM so that one drive cannot monopolize the system and this is causing sound drops and clicks and pops? In other words, could the problem be similar to the firewire clogging problem I experienced? If you had one internal SATA drive versus multiple ones would that make the problem go away? Does the fact I have the new VSL VI on an external USB or firewire while I have the Pro edition VSL, Colossus, and EW Symphonic Orchestra Platinum on an internal SATA identify the location of the conflict since my system is working? Would the problem others are having relate to multiple internal harddrives not counting the system hard drive? Would the VSL VI libraries on an isolated firewire bus and not on the internal hard drives solve the problem? An external firewire is not that expensive. This is an important question to me as I contemplate a new system. All I am saying is that I had the problem too prior to eliminating sharing of firewire resources with anything but VSL VI. Forgive me if my questions seem foolish as I am trying to learn to avoid these unpleasant experiences. I had wanted to upgrade to a Mac 3.0 ghz quadcore pro and put all my libraries on one computer with separate hard drives for each section and max out the ram at 16 gb. But it seems that may not work based on what I am reading. I.E. Not all ram is accessible, multiple hardrives may be a probelm, etc. Thanks. [*-)][/quote]

Here are my experiences relevant to this discussion. I tend to instantiate multiple VI instruments in Logic, usually with a single patch loaded in each one. I consistently get the click/pop problems - that is, they don't go away after some playback activity, as they have for others who have posted here.

The pops/clicks arise pretty consistently when I instantiate more than 3 or 4 VI plugins, though not if I load a single plugin and stuff it 3 or 4 patches. I've come to accept that if I want more than 3 VI instances, I have to bounce my tracks. (Bouncing makes the artifacts disappear.)

My disk activity and CPU activity meters in Logic indicate fairly low levels of activity, so it doesn't seem like hard drive speed is the issue. I have my Appassionata sounds on a separate fast drive. (In one case I tried moving some of the sounds to yet another drive, to no effect.)

Running a Mac G4 1.24 Dual under OSX 10.4.9, 1.75 gigs RAM, Logic 7.1.1, and exclusively using Appassionata Strings right now.

Hi Michael,

Please note that our system requirements of an Apple G4 1 GHz with 1 GB RAM represent the absolute minimum system performance to run one single instance of the Vienna Instruments on its own with no other plugins involved!

This is why, considering your current setup, 3 or 4 instances of the Vienna Instruments might very well tax your system beyond its limits.

Best, Marnix

Just wanted to add my name to the list of users having this problem. I get dropouts in the sound at pretty regular intervals. They don't go away no matter how long I use the instruments for or how many times I play the same note. They occur even if I only have one instance running only one patch. I'm using the built-in audio rather than an external soundcard and I've tried putting the VI data on an external FW800 drive and the internal one - it makes no difference in either case. Does anyone have any suggestions?

Martin

mosso, i'm confused now ... or i missed something .. gigastudio orchstra on a mac?
christian

Sorry I guess my profile should be clearer... [:O]ops:

I have a slave PC (Windows XP SP2) that I'm using to run GS, GVI and VI. I was hoping to keep the PC for the Giga libraries only and use the MacBook Pro for VI. Problems with VI occur only on the Mac.

Thanks in advance for any help!

Martin

Mosso:

Have you tried increasing the I/O Buffer Size in Logic's Preferences? (Logic/Preferences/Audio/Drivers) Using a buffer of 512k or 1024k I get no pops and clicks on my G5 (after the first plaback run) - - although it may also help that I am using a MOTU PCI 424/2408 audio interface rather than built-in audio.