Can Sonar use both the Roland Octacapture and Roland FA-06 interfaces simultaneously ?

 
Owners of Roland FA-06 or FA-08 can possibly help me with this.
 
I am seriously studying and considering the Roland FA-06 or FA-08 for purchase sometime in future to work in conjunction with Sonar as they have DAW integration with Sonar.
Being a studio music producer and not a live performer, one of the important aspects for me is the audio interface that is part of the FA-06/08. 
I already have my Roland Octacapture all hooked up. Now suppose I need to record audio directly from FA-06/08 into Sonar, an obvious way would be to connect audio cables from the FA-06/08 audio outs into inputs of the Octacapture.
 
But what if I want to directly use the interface of FA-06/08 (connected to my laptop through USB) ? That is, say, Sonar will detect the 2-in of FA-06/08 as stereo input in an audio track. I know I can do this. All I want to know is whether I can do this simultaneously with being able to use the Octacapture inputs as well for other inputs like say mics, other acoustic instruments, guitar, etc.
 
If say, on choosing the FA-06 interface, the Octacapture would have to get deactivated, then that would be a huge disappointment, for me rendering the use of the FA-06 interface as useless.
The reason I wish to go this way is because the audio recording from FA-06 directly into Sonar through its own interface (connected by USB) is most likely to be of better quality than recording the audio through audio cables routed to inputs of Octacapture. This avoids additional audio routing. 
 
A related question would be : Does FA-06 audio interface work with ASIO as the driver ? As I use ASIO for the Octacapture.
 
Hope my queries are clear. Thanks in advance.
 

The ASIO spec allows the host to use one DRIVER and a time. How many devices are managed by the driver is the question. I see nothing in the FA-06 or Octa-Capture documentation that suggests these two devices may share the same driver or otherwise link together. That said,  I will have to defer to Roland or someone with both devices for the definitive answer.

I have an FA 06.  I can either use it as the interface to record FA via USB, OR I can have another ASIO interface enabled.  It's basically like every other cool piece of gear I have that has a USB interface built in. 
 
Believe it or not, the easiest way (for me) is to take the SD card out of the FA and put a guide track on it from my PC.  SD card back in the FA and assign the guide track as a sample.  Trigger the sample via the FA sequencer and play/record whatever accompaniments I want in the FA via MIDI.  Bounce the tracks to audio in the FA, then move the SD card back to the PC to pull them into a DAW project.
 
It would be easy enough to run the audio from the FA to a proper interface, but I use the main audio outs on the FA directly to my matrix for direct play/practice.
 
It is also possible to use the FA sounds by using it as an external sound module.
 
This is a limitation of Windows and ASIO, not necessarily anything wrong with the FA design.
 
Despite all of this ... I love my FA

 
Thanks so much everyone for your inputs. Does help a lot.
 
Only thing is, does WDM perform as good as ASIO ?
Can the same low latency be expected in WDM to the level that is usually observed with ASIO ?
Will all older projects open the same if set to WDM without any problems ?
 

It's too bad neither has ADAT capability, you would be all set otherwise. Best bet as others have said is to use WDM, but make sure the clocks are in sync.
 

Much more than drivers involved in using two different devices. The sample rates of each device have to be identical which requires synchronising of the 2 devices. Either by dedicated Clock inputs and outputs or at a pinch you can get away with using SPDIF connections. Failure to provide synch means that the clock from each device can drift from the other. When you think that the most basic and common sample rate is 44,100 samples per second you can see that it doesn't take much for two independent clocks to drift apart. This creates all sorts of horrible garbled unmusical sounds.

That's a very basic coverage of the topic.

Might I ask a question concerning your screen shot? Was, or were...all those tracks recorded from the same source material? I see the head of some tracks look identical while others appear a bit different. I'm assuming you didn't record all tracks at the exact same time,or did you?
Sorry, call me stupid...
I know what I'm looking at are the different latency values of a given track...I don't know, perhaps I'm not understanding latency quite as much as I probably should, but what seems to me is...any latency is bad, no matter how minimal.
Please...bust my chops where you will, but...
If track one starts on dead zero, then every overdub/added track afterward will have 'some' latency.
Then I might assume that all my tracks will look like your representation, except for the fact that all tracks recorded after the initial one will be 'dead on' to each other more than they would be to track 1.
Then, instead of nudging all the added tracks to track 1, would you move track one to match tracks 2~.
I've read a lot about...''x' much latency is acceptable'.
Is it?
so given even the acceptable amount, still, is it true, if you zoom in enough on even the lowest latency, you'll find it still looks like your screenshot?

Sorry if my question hijacked this thread