Not quite true - Mac desktops are marketed as audio processing units. They even come with the software preinstalled. They took pains to make the mic-speaker loop low-latency so you could listen to yourself as you record.
Since other products are all about the audio experience, it does seem strange that the iPod, the device which most of humanity uses to experience music, does such a poor job. And Apple doesn't care. And neither do the people listening. Strange.
The inbuilt microphones do work very well. The whole loop is, as you say, heavily adjusted to compensate for the resonances of the unit. I wouldn't be surprised if there was some quite fancy DSP powering that.
I thought that was for making facetime and skype etc work well though? Musicians are not using the built-in audio interface, regardless of quality considerations, it lacks the necessary functionality.
Everyone I know has a sub $100 USB interface hanging off their machine and swears constantly at the latency and wishes they'd just stuck with ASIO on Windows...
Personally I still have a workstation...
As for the iPods they sound horrible. The latest line of nano has noticeable noise on the outputs.
You're complaining about WDM audio. There's built in WaveRT and external ASIO stacks as well. The latter has become the standard on Windows for professional audio and is direct hardware access.
USB audio devices are and aren't polled. The cheap ass speakers and stuff are but anything $20+ are buffered raw device access.
I can get 8ms out of a Windows 7 laptop with ASIO over USB that isn't even tuned. My Korg Triton Studio manages about 4ms and that's seriously high end kit.
Right, ASIO isn't Windows its 3rd-party to work around the Win7 issues.
WaveRT is what WDM uses, right? With the associated delays going user/kernel mode. How does that help?
And USB devices are polled, at three levels as I recall - 'control', 'interrupt' including 'isochronous' and 'bulk transfer'. The frame interval depends on the USB speed - and I see they are sub-millisecond. SO very adequate for audio sampling. I used to use it for serial protocols, and that frame rate was entirely inadequate for signalling so I formed my low opinion of USB at that time.
Comments
Not quite true - Mac desktops are marketed as audio processing units. They even come with the software preinstalled. They took pains to make the mic-speaker loop low-latency so you could listen to yourself as you record.
Since other products are all about the audio experience, it does seem strange that the iPod, the device which most of humanity uses to experience music, does such a poor job. And Apple doesn't care. And neither do the people listening. Strange.
The inbuilt microphones do work very well. The whole loop is, as you say, heavily adjusted to compensate for the resonances of the unit. I wouldn't be surprised if there was some quite fancy DSP powering that.
I thought that was for making facetime and skype etc work well though? Musicians are not using the built-in audio interface, regardless of quality considerations, it lacks the necessary functionality.
I don't buy that.
Everyone I know has a sub $100 USB interface hanging off their machine and swears constantly at the latency and wishes they'd just stuck with ASIO on Windows...
Personally I still have a workstation...
As for the iPods they sound horrible. The latest line of nano has noticeable noise on the outputs.
Since Windows Vista there has been 100ms latency added to the audio loop. So you're experience was Xp? Keep that machine around.
USB interfaces are a terrible idea - they are polled and add latency every time.
That's basically rubbish and misinformation.
You're complaining about WDM audio. There's built in WaveRT and external ASIO stacks as well. The latter has become the standard on Windows for professional audio and is direct hardware access.
USB audio devices are and aren't polled. The cheap ass speakers and stuff are but anything $20+ are buffered raw device access.
I can get 8ms out of a Windows 7 laptop with ASIO over USB that isn't even tuned. My Korg Triton Studio manages about 4ms and that's seriously high end kit.
Right, ASIO isn't Windows its 3rd-party to work around the Win7 issues.
WaveRT is what WDM uses, right? With the associated delays going user/kernel mode. How does that help?
And USB devices are polled, at three levels as I recall - 'control', 'interrupt' including 'isochronous' and 'bulk transfer'. The frame interval depends on the USB speed - and I see they are sub-millisecond. SO very adequate for audio sampling. I used to use it for serial protocols, and that frame rate was entirely inadequate for signalling so I formed my low opinion of USB at that time.