| View previous topic :: View next topic |
| Author |
Message |
wmountney Jammer

Joined: 06 Mar 2005 Posts: 36
|
Posted: Sun Mar 06, 2005 3:49 am Post subject: Possible denormalization issue? |
|
|
First let me just say: I'll all over this thing! This is one of the coolest programs I've seen in a while and will go along just great with BFD (which is the other coolest program).
However, I have encountered a problem, running under SONAR 4 Producer, that smells a lot like a denormalization problem to me. It seems that when the velocity drops to zero for five seconds or so, the CPU usage shoots up dramatically. When I'm using the internal drum sounds, it jumps from around 7 - 10% up to 30 - 33%. However, if I'm using BFD, it jumps up from around 15% all the way to overload where the audio engine stops. Once it happens, you can sort of reset things by disconnecting Jamstix in the Synth Rack and then reconnecting it, but it never really seems quite right again until you quit SONAR and go back into it again.
This problem can be reproduced by simply sliding the "Velocity" slider all the way to the left in Jamstix while it's playing. However, it's most problematic when you are using the "Jam With Audio" feature. In that case, it's kind of like a musical version of the plot from that movie "Speed" but instead of driving a bus, you're playing a guitar, and if you stop, it blows up.
I've tested this using both Cakewalk's VST adapter and DirectiXer (which is my preferred host, since it is the only way to get the VST version of BFD to have the proper combination of mono and stereo outputs in SONAR). I suppose since the denormalization problems are primarily an Intel issue, you'd need a Pentium processor to reproduce the problem. In my case, I have a Pentium 4 2.4GHz.
On an unrelated postitive note, using BFD with DirectiXer under SONAR always had an extremely annoying problem where every time you saved your project in SONAR it would somehow cause BFD to reset and reload its sample data, which isn't exactly a quick operation. However, using BFD through Jamstix under DirectiXer does *not* have this problem. It seems that Jamstix magically filters out whatever weird signal DirectiXer was sending to BFD in this case. I'm just mentioning this to help ensure that this magic filter doesn't ever get unintentionally removed from future versions of Jamstix.  |
|
| Back to top |
|
 |
Ralph [RZ] Site Admin

Joined: 02 Jan 2008 Posts: 13332
|
Posted: Sun Mar 06, 2005 3:42 pm Post subject: Re: Possible denormalization issue? |
|
|
| wmountney wrote: |
However, I have encountered a problem, running under SONAR 4 Producer, that smells a lot like a denormalization problem to me.
|
We're on it. Sure does sound like a denorm issue.
| Quote: |
I'm just mentioning this to help ensure that this magic filter doesn't ever get unintentionally removed from future versions of Jamstix.  |
It won't and if it does let us know and we'll fix it  _________________ Ralph Zeuner
Rayzoon Technologies LLC
http://www.rayzoon.com |
|
| Back to top |
|
 |
Ralph [RZ] Site Admin

Joined: 02 Jan 2008 Posts: 13332
|
Posted: Mon Mar 07, 2005 9:05 am Post subject: |
|
|
wmountney was gracious enough to help us track down this bug. It's a denormalization issue with Pentium-4s in the Jamstix compressors.
We will fix this in the next update but for now just slide the min velocity slider a bit to the right or switch off the compressors _________________ Ralph Zeuner
Rayzoon Technologies LLC
http://www.rayzoon.com |
|
| Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2002 phpBB Group
|