Re: Core audio driver
Re: Core audio driver
- Subject: Re: Core audio driver
- From: Lubor Prikryl <email@hidden>
- Date: Fri, 18 Feb 2005 10:18:30 +0100
Hello Guys,
For audio, I suggest using FireWire in either case. USB needs always much more of CPU "attention". Use either Physical Memory Space or isochronous streaming, if you can accept its timing.
Lubor
Lubor Prikryl
DSound (CZ)
Audio Software Developer
email@hidden
On 18.2.2005, at 8:13, pwicker wrote:
On Feb 18, 2005, at 12:40 AM, Herbie Robinson wrote:
At 10:04 PM +0100 2/17/05, pwicker wrote:
On Feb 17, 2005, at 9:17 PM, Mark Cookson wrote:
Be careful in your terminology here. There is NO USB Audio 2.0 Device specification (that's released to the public). There is only a USB Audio 1.0 Device specification. There is of course USB 2.0, and you can put USB Audio 1.0 Device functionality on a device that connects to the host over USB 2.0 and runs at high speed, but that doesn't make it a USB Audio 2.0 device, and you will find a couple of short comings in the USB Audio 1.0 Device specification that limit the device's usefulness.
The problem that you are seeing is that there is a bug in the USB stack that prevents the USB Audio 1.0 driver from working correctly on the USB 2.0 stack. Until that is fixed, there is nothing that the driver can do to talk to a USB 2.0 device.
Could you please elaborate on this point? I'm currently working on a project that involves the connection of an audio/midi external USB 2.0 device (high speed needed). Until now, our plan was to rely on the generic Mac OS X audio and midi drivers. IIRC this could be not possible and we may have to develop our own driver :(
You are probably already aware of this, but the erata posted on AMD's web site (http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_9004,00.html) for the AMD-8111 chip claims that the USB 2.0 port doesn't work in high speed mode. If your testbed has this in it, you may need to use a PCI card.
No, I wasn't :/ Thanks for the info. So If I sum up: high speed USB is not working on the Mac because of a USB system software bug, and on PC because of host controllers bug. Maybe we should reconsider a Firewire connection instead of the USB.
Also, the AMD-8131 PCI-X chip has severe performance problems with the pass-through port to the 8111. A number of users are reporting they cannot run a single 8 channel audio interface and a hard drive on the same 1394 bus. They are also buying PCI cards to get around this (or buying slower machines that don't have PCI-X slots). Given that the USB 2.0 port will be in the same place in the bus architecture, it's probably going to be crippled on those machines, too.
--
-*****************************************
** http://www.curbside-recording.com/ **
******************************************
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
Philippe
email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden