Posted: Sat Mar 25, 2006 3:04 pm
Well, just to see what would happen, I finally borrowed my brother's RME Hammerfall HDSP 9652 that was in his PowerMac G5. I couldn't remotely get MME or MIDI to show up in Windows, but ASIO worked OK. So I ran my testing projects...and found, to my great surprise, that it worked no better than Scope's ASIO -- in fact, in some cases it performed worse!! Here's a breakdown of my results:
-System setup: I played all projects at 2.9ms latency. Sonar 5.01, Scope 4.0, Windows XP SP2. Since the Hammerfall is ADAT only, the Scope software had to be constantly running to provide a D/A converter. I simply switched Sonar between the Hammerfall's ASIO drivers and Scope's. I don't know if Scope running would present a problem, but I don't think so. I have a Scope card, a Pulsar card, and a Powersampler, plus the Hammerfall for testing. The Powersampler shares an IRQ with my video card (moving the card to a different slot didn't fix anything, so that isn't the problem -- and anyway, it isn't the master board). Hammerfall ended up sharing an IRQ with Scope.
-Project A: Many Scope synths loaded, light CPU load - 10% or less average. Hammerfall produced no ticks or pops at all, Scope does only when running dualcore.
-Project B: No Scope devices other than a basic mixer. Medium CPU load - 50-60% average. Some ticks and pops with Scope, MANY ticks and pops with Hammerfall -- almost to the point of crackling. Problem was even more apparent when I would scroll through my Sonar project. Dualcore only, CPU load too high for singlecore without raising latency, which negates comparison.
-Project C: Same basic configuration as Project B. Problem not so severe, but unacceptable playback.
-Project D: My project I use for testing when I'm running many Sonar envelopes at 1.5ms latency. This one is worse with singlecore. 10-20% CPU load, no Scope synths. Clicking problem exactly the same with both Scope and Hammerfall.
The bottom line? I've eliminated almost every suspect that could be causing all these problems. I've reinstalled Windows from scratch, I've changed motherboard/CPU/memory, I've done IRQ troubleshooting, I've played around with PCI latency, I did diagnostic mode to remove almost all services/background programs, including Windows themes and fancy graphics. What's left? The presence of Scope itself? No, no one else has had a problem. The only remaining item is my video card, which has been constantly present in my system since 2002. It's an ATi Radeon 8500 128MB AGP. Hammerfall made it even more obvious that the audio performance grew worse when there was visual movement. I don't have another video card to test, though, and no money to buy another one.
So, anyway, for all I know, Scope is completely compatible with multiprocessing/dualcore/AMD x2/whatever. DON'T PAY ANY ATTENTION TO MY RESULTS ANYMORE!
Shayne
-System setup: I played all projects at 2.9ms latency. Sonar 5.01, Scope 4.0, Windows XP SP2. Since the Hammerfall is ADAT only, the Scope software had to be constantly running to provide a D/A converter. I simply switched Sonar between the Hammerfall's ASIO drivers and Scope's. I don't know if Scope running would present a problem, but I don't think so. I have a Scope card, a Pulsar card, and a Powersampler, plus the Hammerfall for testing. The Powersampler shares an IRQ with my video card (moving the card to a different slot didn't fix anything, so that isn't the problem -- and anyway, it isn't the master board). Hammerfall ended up sharing an IRQ with Scope.
-Project A: Many Scope synths loaded, light CPU load - 10% or less average. Hammerfall produced no ticks or pops at all, Scope does only when running dualcore.
-Project B: No Scope devices other than a basic mixer. Medium CPU load - 50-60% average. Some ticks and pops with Scope, MANY ticks and pops with Hammerfall -- almost to the point of crackling. Problem was even more apparent when I would scroll through my Sonar project. Dualcore only, CPU load too high for singlecore without raising latency, which negates comparison.
-Project C: Same basic configuration as Project B. Problem not so severe, but unacceptable playback.
-Project D: My project I use for testing when I'm running many Sonar envelopes at 1.5ms latency. This one is worse with singlecore. 10-20% CPU load, no Scope synths. Clicking problem exactly the same with both Scope and Hammerfall.
The bottom line? I've eliminated almost every suspect that could be causing all these problems. I've reinstalled Windows from scratch, I've changed motherboard/CPU/memory, I've done IRQ troubleshooting, I've played around with PCI latency, I did diagnostic mode to remove almost all services/background programs, including Windows themes and fancy graphics. What's left? The presence of Scope itself? No, no one else has had a problem. The only remaining item is my video card, which has been constantly present in my system since 2002. It's an ATi Radeon 8500 128MB AGP. Hammerfall made it even more obvious that the audio performance grew worse when there was visual movement. I don't have another video card to test, though, and no money to buy another one.
So, anyway, for all I know, Scope is completely compatible with multiprocessing/dualcore/AMD x2/whatever. DON'T PAY ANY ATTENTION TO MY RESULTS ANYMORE!
Shayne