Idea for a new "stream" plugin: ASIO bridge
-
- KVRist
- Topic Starter
- 110 posts since 15 Sep, 2005
Sometimes I've thought it would be useful to have a virtual ASIO driver, which would stream its input and output to a VST plugin inside a "real" ASIO/VST host application. The "encapsulated" ASIO application would appear as a VST plugin in a program.
Another operation mode would be an ASIO-to-ASIO bridge, so that the virtual driver would stream the output of one ASIO application to the input of another. For instance, I could get the audio output from Sibelius's Kontakt Player or any other stand-alone program directly into Cubase's mixer.
Yet another use would be to create virtual "external device" loops for Cubase SX, or a virtual "patch cable" from an output buss to an input channel.
Or the virtual driver could report having a dozen 24-bit outputs and inputs, even though the machine is a crummy laptop with integrated AC-97 chip and nothing else. Just to run some ASIO hosts in special circumstances.
You've got the Stream Boy thing working, so this wouldn't need to be written completely from scratch. (only 95% from scratch?;))
Or is this something I should propose to the maker of the ASIO4ALL drivers?
Another operation mode would be an ASIO-to-ASIO bridge, so that the virtual driver would stream the output of one ASIO application to the input of another. For instance, I could get the audio output from Sibelius's Kontakt Player or any other stand-alone program directly into Cubase's mixer.
Yet another use would be to create virtual "external device" loops for Cubase SX, or a virtual "patch cable" from an output buss to an input channel.
Or the virtual driver could report having a dozen 24-bit outputs and inputs, even though the machine is a crummy laptop with integrated AC-97 chip and nothing else. Just to run some ASIO hosts in special circumstances.
You've got the Stream Boy thing working, so this wouldn't need to be written completely from scratch. (only 95% from scratch?;))
Or is this something I should propose to the maker of the ASIO4ALL drivers?
-
- KVRian
- 1214 posts since 2 Jun, 2004 from Québec, CANADA
Emu patchmix does that..... great feature.Kitchen Sink wrote: Another operation mode would be an ASIO-to-ASIO bridge, so that the virtual driver would stream the output of one ASIO application to the input of another. For instance, I could get the audio output from Sibelius's Kontakt Player or any other stand-alone program directly into Cubase's mixer.
- KVRist
- 352 posts since 8 Jul, 2003
ESI has something like this called DirectWire (http://www.esi-pro.com/directwire.php).
jouni - www.markvera.net - Stardrive Studio - Orionology
-
- KVRAF
- 5629 posts since 22 Sep, 2005
Uhh hello? ASIO STREAM not MIDI.. duh! You'l have the guy trying to stream SYSEX dumps with those tools..Elakhna wrote:Wouldn't Hubi's MIDI Cable or MIDI Yoke do what you want??
Last edited by Lagrange on Thu Aug 17, 2006 2:59 pm, edited 1 time in total.
-
- KVRAF
- 3378 posts since 27 Feb, 2004 from Paris (france)
-
- KVRAF
- 5629 posts since 22 Sep, 2005
-
- KVRist
- Topic Starter
- 110 posts since 15 Sep, 2005
That DirectWIRE thing seems great - if it just wasn't tied to the ESI stuff. Though I think we would still need more than one virtual ASIO device, i.e. more of those ASIO columns in the screenshot. It think it would work best if each application had one virtual ASIO device all to itself.
Sibelius has another problem too, but it cannot be fixed with any device drivers. Namely, their MIDI timing engine has its own timing system, and it drifts away from the ASIO clock. It syncs the video player to its own position every now and then, but the drift is so large, after ten bars the clock difference can be something like a second, and then the video clip just jumps back in time. The MIDI timing calculates "delta times" between events and waits for that time before outputting the next event... So actually there is no "clock" at all.
Sibelius has another problem too, but it cannot be fixed with any device drivers. Namely, their MIDI timing engine has its own timing system, and it drifts away from the ASIO clock. It syncs the video player to its own position every now and then, but the drift is so large, after ten bars the clock difference can be something like a second, and then the video clip just jumps back in time. The MIDI timing calculates "delta times" between events and waits for that time before outputting the next event... So actually there is no "clock" at all.
-
- KVRian
- 951 posts since 11 Jan, 2004 from Netherlands
-
- KVRAF
- 1981 posts since 29 Feb, 2004
too much latency
- KVRAF
- 35249 posts since 14 Sep, 2002 from In teh net
Can do this with Creamware Scope cards too.
-
- KVRAF
- 5629 posts since 22 Sep, 2005
Agreed even with a small buffer..asseca wrote:too much latency
-
- KVRist
- 46 posts since 24 Jul, 2003
maybe the VB-Audio ASIO Bridge could help : www.vb-cable.com