3 ms·

"Audio and MIDI stacks on Windows were a mess compared to those in OSX" What do you mean? That sounds like its harder to write drivers for soundcards for window

"Audio and MIDI stacks on Windows were a mess compared to those in OSX" What do you mean? That sounds like its harder to write drivers for soundcards for windows.. But the user is not going to be writing their own drivers. Any user who needs to perform audio with any computer will need a soundcard in order to have any sort of decent quality, for #1 Decent quality inputs and outputs #2 The correct input and output types like XLR (or fibre optic or whatever) #3 Acceptable latency for live performance. Soundcards are available for mac or windows, the makers provide drivers which deal with the "Audio and MIDI stacks". From a users perspective theyre both fine. After that, you care about the machines performance, CPU, how much RAM, how fast is the RAM, how fast is the SSD.

ratww 6y ago
It means that Windows/Microsoft doesn't provide anything to driver writers or DAW authors. MS tried a bunch of tech in the past (WinMM, MCIWnd, DirectSound, WaveOut, WASAPI, XAudio2, etc), but none of them ever worked for professional Audio. The proper solution was always to use ASIO, which is third-party technology by Steinberg. It works but it's not integrated with Windows: it goes directly to the audio interface, bypassing stuff in the Kernel. This bypass causes some limitations, such as not being able to use the system mixer (which prevents from using media players or multiple audio apps), or sometimes having different audio-interfaces not work well with each other. There are workarounds to those issues, but they have to be handled by the interface manufacturer when writing the drivers. Also, you can't have low-latency with built-in soundcard unless you use something like ASIO4ALL, which is not super stable IME. This sucks when you want to work on-the-go with headphones. Of course, it can be very stable when you use the right combination of DAW, drivers and sound interfaces, but when you don't you have problems. On macOS and iOS? CoreAudio is native and it has super low latency by default. It's mostly plug and play and all apps use it. It even provides APIs to use Audio Plugins or reroute Audio, so DAW writers don't even have to write it themselves (unless they want to). Do you have multiple Audio/MIDI interface? In macOS there's a built-in app to "link" them together.

ratww 6y ago

It means that Windows/Microsoft doesn't provide anything to driver writers or DAW authors. MS tried a bunch of tech in the past (WinMM, MCIWnd, DirectSound, WaveOut, WASAPI, XAudio2, etc), but none of them ever worked for professional Audio. The proper solution was always to use ASIO, which is third-party technology by Steinberg. It works but it's not integrated with Windows: it goes directly to the audio interface, bypassing stuff in the Kernel. This bypass causes some limitations, such as not being able to use the system mixer (which prevents from using media players or multiple audio apps), or sometimes having different audio-interfaces not work well with each other. There are workarounds to those issues, but they have to be handled by the interface manufacturer when writing the drivers. Also, you can't have low-latency with built-in soundcard unless you use something like ASIO4ALL, which is not super stable IME. This sucks when you want to work on-the-go with headphones. Of course, it can be very stable when you use the right combination of DAW, drivers and sound interfaces, but when you don't you have problems. On macOS and iOS? CoreAudio is native and it has super low latency by default. It's mostly plug and play and all apps use it. It even provides APIs to use Audio Plugins or reroute Audio, so DAW writers don't even have to write it themselves (unless they want to). Do you have multiple Audio/MIDI interface? In macOS there's a built-in app to "link" them together.