tlw
ASIO4ALL is difficult to configure and can cause all kinds of problems itself and with genuine ASIO drivers as well.

... And if you get hooked, baby. It's nobody else's fault, so don't do it!

while evryone's moaning about asio4all again, some slighty worrying info on their forum: http://tippach.business.t-online.de/asio4all/phpbb/viewtopic.php?f=4&t=1231&sid=e9112bd7ead6bef1360711830cbadd3c

pwalpwal
while evryone's moaning about asio4all again, some slighty worrying info on their forum: http://tippach.business.t-online.de/asio4all/phpbb/viewtopic.php?f=4&t=1231&sid=e9112bd7ead6bef1360711830cbadd3c

Yup I brought this up a while ago in the Windows 10 threads. They are mainly concerned with optimizing Windows driver models for tablets, touchscreens and whatever...  it will probably make ASIO4ALL even less relevant then because why would you still need a ASIO wrapper when you have these options?

MS said though that ASIO will still be the "best experience for DAWs". So nothing to worry about...

tlw
For softwRe synths I just track the MIDi then bounce the tracks. Never looped one back through the interface, never needed to.

Remember that the software synth is entirely controlled by MIDI (as are all the functions on quite a few hardware synths) so the recorded MIDI is as complete a record of the performance as the output of the synth is.

As for ASIO4ALL if your interface manufacturer supplies an ASIO driver use that not ASIO4ALL. ASIO4ALL is a hack to try and fool the Windows WDM drover into acting like an ASIO driver. It's intended as a last resort for when the sound card has no other ASIO driver available, such as the sound chips on many motherboards or built into laptops, ASIO4ALL is difficult to configure and can cause all kinds of problems itself and with genuine ASIO drivers as well.

This is my line of thought too.  And I would think this is why real time recording of a soft synth is not offered.  If you are able to capture the audio out of a soft synth in real time...you are not going to capture anything different than what the midi is transmitting anyway.  In other words....you can't capture any nuances or anything else in real time that is not recorded with midi to begin with.
 
Recording a hardware synth via analog outs is a totally different animal IMO than trying to record the soft synths in real time.  Personally...I would rather record the midi first and able to do any editing if needed before I bounce it down.
 
Edited to add:
I'm not dismissing anyone's preference on wanting to be able to do this...I just personally don't think it would be any different that what you record via midi first. :)

alex, we shall see! i've had the w10 preview kicking about for a while just not around to installing/playing with it

Teds_Studio

tlw
For softwRe synths I just track the MIDi then bounce the tracks. Never looped one back through the interface, never needed to.

Remember that the software synth is entirely controlled by MIDI (as are all the functions on quite a few hardware synths) so the recorded MIDI is as complete a record of the performance as the output of the synth is.

As for ASIO4ALL if your interface manufacturer supplies an ASIO driver use that not ASIO4ALL. ASIO4ALL is a hack to try and fool the Windows WDM drover into acting like an ASIO driver. It's intended as a last resort for when the sound card has no other ASIO driver available, such as the sound chips on many motherboards or built into laptops, ASIO4ALL is difficult to configure and can cause all kinds of problems itself and with genuine ASIO drivers as well.

This is my line of thought too.  And I would think this is why real time recording of a soft synth is not offered.  If you are able to capture the audio out of a soft synth in real time...you are not going to capture anything different than what the midi is transmitting anyway.  In other words....you can't capture any nuances or anything else in real time that is not recorded with midi to begin with.
 
Recording a hardware synth via analog outs is a totally different animal IMO than trying to record the soft synths in real time.  Personally...I would rather record the midi first and able to do any editing if needed before I bounce it down.
 
Edited to add:
I'm not dismissing anyone's preference on wanting to be able to do this...I just personally don't think it would be any different that what you record via midi first. :)

ted, tim, some synths add randomised stuff that is different each playback, so there's a use, but if your premise were true generally why do other hosts offer this? (this = recording the synth out into another audio track) cakewalk's previous explanation was to prevent rubbish users creating feedback loops...

Ahhhh.....understood.  Did not realize that.

The dreaded feedback loop has been mentioned more than once over the years in the context of why this and other similar MIDI routing issues have never seen the light of day. If it is impossible to create a feedback loop, then even the least attentive of users will not be able to cut off that limb while sitting on the wrong side of the saw. Various workarounds involving physical and digital loopback have been adopted as a result. So the wisdom of protecting your users from doing something stupid inside your program system and forcing them to do so outside using kludges that your system can not detect is questionable. If user safety is the issue, would it not be better to offer the routing options, and provide the logical analysis of how the user is hooking up the inputs and outputs so that if a feedback loop has indeed been created in Sonar, there will be an error message and a halt so the damage will not be done. This kind of thing is done routinely in a spreadsheet to tell you you have created a circular reference in a calculation. The technical problem of doing something like that in Sonar should not be insurmountable.

bitman
If it were a simple thing, it would have been added to the icing on the cake in Foxboro instead of a (midi cable review?)
 

One more time...I don't write code, and Cakewalk has nothing to do with the content I create (or write the eZine). My activities have zero impact on SONAR's development. Well, except provide a bonus that makes some people happy...more happy people, more people buy SONAR, Cakewalk can hire more developers to make even more people happy. 

pwalpwal
 
ted, tim, some synths add randomised stuff that is different each playback, so there's a use, but if your premise were true generally why do other hosts offer this? (this = recording the synth out into another audio track) cakewalk's previous explanation was to prevent rubbish users creating feedback loops...

Although I understand this is something some people want, it doesn't get in the way of my working with soft synths for the following reasons.
 
1. If you want to record real-time control tweaks, in most synths any parameter you can tweak in real time is recordable as MIDI or VST automation.
2. If the playback truly does something random, then you won't know whether you like what it does until you hear it do its thing. But that's also the case if you render and listen back. The difference is whether you hear the change being generated in real time, or hear the change being generated in real time after rendering. So the only advantage of real-time recording is it would allow you to decide whether a part was a "keeper" or not right after it played, instead of rendering first and then evaluating.
3. You can always use the external insert to do a physical loopback. It requires going out of the box, but it works.
4. Jack Audio supposedly has a 64-bit version under development so you can do what Soundflower does on the Mac, which is more sophisticated than simply recording an instrument out.
 
I'm not arguing with anybody who wants it; if you want it, you want it. I just don't think it's a big deal, given there are ways to do what you want to do.