TASCAM FW-1884 as a control surface

[lpgasparotto]

Hi,
I'd love to be able to use the FW-1884 as a control surface, as far as I know it's possible with Ardour with the Mackie Emulation Mode, it would be great to add a feature which allows to switch between the available control surface modes (Native, Mackie Emulation Mode and Mackie HUI). I guess it should be possible.

Thank you very much in advance,
Luis Pablo

[takaswie]

Hi @lpgasparotto ,

Thanks for your request. Against you expectation, TASCAM (TEAC and Frontier design, precisely) implements the functionality of Mackie HUI emulation in the driver software. Therefore even if receiving any MIDI messages in Mackie HUI protocol directly, the control surface does not work as expected.

You can see current design of snd-firewire-tascam-ctl-service in https://docs.rs/firewire-tascam-protocols/0.2.0/firewire_tascam_protocols/#process-events-for-control-surface. If you implemented both machine state and message converter for the emulation, it would be achievable. At present, the service program interprets and generates the usual control messages for both the control surface and ALSA sequencer application.

[lpgasparotto]

Thank you very much for your prompt reply! What I understand is the current service is able to achieve some control surface functionality in the native mode, am I right? How could I use that functionality in Ardour?

I'd like to help you improve the features of this software, please let me know if there's something I could do.

Best regards,
Luis Pablo

[takaswie]

am I right?

Exactly.

When running snd-firewire-tascam-ctl-service, you can see an extra port named as Control Surface in ALSA sequencer system. The port emits ALSA sequencer event when any fader and buttons are operated in actual control surface during any PCM substream runs. Technically, the hardware transfer the operated event in IEEE 1394 isochronous packet which delivers PCM frames for ALSA capture direction.

As a first step, let you execute aseqdump to connect to the port, then operate any faders. I expect the aseqdump receives and dumps the corresponding ALSA sequencer event. Please take care that any of ALSA PCM applications runs to receive the isochronous packets.

[DrungSturm92]

Hi @takaswie and @lpgasparotto I've been following this issue and other posts, mailing list, etc. that include the TASCAM FW1884 since I own the little brother, the FW1082, and they are pretty much the same.
I've been able to compile and run the "snd-firewire-tascam-ctl-service" and effectively, the new port appeared. As you said, in order to interact with it I had to open another app that uses the FW1082 as ALSA backend (CARLA or Ardour in my case). After this, I run aseqdump, connect to the port and see the MIDI messages being sent from the TASCAM.
I could interact with Ardour selecting the Generic MIDI protocol (in Control Surface setup) and using the midi learn feature.
Despite my limitations, I'm available for testing and will happily contribute with feedback. In fact, I am facing an issue with my TASCAM that refuses to connect with the host pc. It seems that the 'gap count' is involved but that's another topic.
Thanks for your work and this amazing community.

[takaswie]

Hi @DrungSturm92 ,

using the midi learn feature.

It is what I expected in user side, since I'm unfamiliar with Mackie HUI protocol and can not program the service program to emulate it.

In fact, I am facing an issue with my TASCAM that refuses to connect with the host pc. It seems that the 'gap count' is involved but that's another topic.

Indeed. It seems that the issue is what I'm facing and blocks my work to solve #188 ...

[DrungSturm92]

Hi @takaswie according to the bug I mentioned before, I bought a FireWire pci card with a Texas Instrument chip (Belkin f5u503) and problem solved. In fact, I bought another "new" cable (a retractable one) and the communication between the host and the FW1082 was really unstable. I think that although the cable had 6 pins connectors in both side, the cable was in fact a 4 pin one. So... good old cable and TI chip card, shows no error in dmesg.
I also continue testing the TASCAM with Ardour and found some "problems". When I assign an Ardour button to the FW1082 button, the communication is bidirectional (possibly not the appropriate word), hit software button and led light up, and viceversa... but with faders, things are different. Moving Ardour faders don't move motorized faders. Doing some research I also noticed that moving the TASCAM faders send MIDI messages with values that exceeds the 0-127 typical range. I think some type of conversion in the user side would be necessary, but I'm not expecting you to work on that too.
Finally, using qasmixer I could change the sample rate from 44,1 to 48 but choosing 88 and 96 didn't light up the appropriate leds in the surface. Nothing urgent but I thought it should be mentioned.
Thanks again!!