Thanks for your reply, MC.

Nope - I'm barely kissing the DSP (less than 5% for now).  FYI - I'm running via the Sonnet PCIe card if that has any bearing.

It's certainly not a show-stopper, just very annoying if you are expecting the "realtime feel" of a tactile control surface.

I also just noticed the ARC Video I saw on YouTube shows the ARC and TMFX (from the UFX) in the same frame.  The ARC starts off with perfectly acceptable lag (:22 seconds in / looks like about 50ms-100ms of lag from encoder movement to TMFX responding), but later seems to lag by a second or more (1:32)?  The first shot (0:22) is a "live" shot of both units and the later shots are superimposed picture-in-picture - so maybe the later shots drifted during editing or something?  It gets close to 5 seconds behind at the end (1:50):

http://www.youtube.com/watch?v=HgfGrWRkYPg

Anyways, let us know what you find out.

Have a great weekend!

Grazie

I can try to use USB temporarily (both USB2 and USB3 on an X79 based LGA2011 system) and see if I notice the same issues - but it will be a few days until I want to go mucking with the currently configured PCIe setup (and I obviously prefer PCIe )...

PS - Did you check the ARC Video RME posted on Youtube (linked in my above post)?  Is that just a post-production video editing/drifting issue, or is that really how TMFX is responding to the ARC (close to 5 seconds of "lag" at the 1:50 mark on the video!)?

That's what I was hoping - The end of the video makes it seem unusable.

OK - I have some updates that should help track this down.  I only tested with PCIe at this point.  I held off on testing over USB as I won't ever use USB on my main DAW - and these PCIe results seem to offer a sufficient starting place to track this issue down:

The lag only seems to occur if the MADIface-XT is powered-up with the ARC connected and/or if the ARC is connected when the PC loads TMFX.  .

If I power-down the entire system and disconnect the ARC, then power-up and boot the system, the MADIface-XT's front-panel encoders work instantaneously in TMFX as they should.  I can even connect the ARC AFTER booting into Windows and the instantaneous response remains (for both ARC and front-panel encoders)!!!

However, if I shut down the PC completely and reboot - the lag is back (since the ARC is still connected).  I must power-down the MADIface-XT and unhook the ARC to be able to boot up without lag.

"Hot" Un-plugging the ARC after booting with the laggy response has no effect (lag remains after unplugging the ARC).

During one of my dozen (or so) reboots while testing, I observed a very strange behavior in TMFX (only once):
ARC was unhooked during power-down state.  I powered up and booted the system and left ARC unhooked.   The front-panel encoders responded instantaneously, but TMFX would actually move TWICE for each encoder movement!  Once in realtime, and a second reaction with the "lag" about 750ms later!  It was like seeing an instant-replay of the fader move.  Maybe that will shed further light on this bug?

Last note:  The front-panel encoders always seem to match the front panel display (no lag from moving the encoder to seeing the front-panel-display change) - The lag is ultimately seen and heard in TMFX on the PC.  But strangely, when moving a fader in TMFX with my mouse, the front-panel-display will respond instantaneously (even when there is lag in the other direction).  The reporting/controlling of TMFX's state only seems to be an issue in the direction from MADIface-to-PC/TMFX, and not in the direction from PC/TMFX-to-MADIface (if that makes any sense at all).

Enjoy

To confirm - are you connecting the ARC to the MF-XT with the MF-XT and PC powered-down completely?

And you have kept the ARC connected and attempted to reboot (without changing between USB/PCIe in the meanwhile)?

The ARC pins are all good.  I have been very careful with the pins as suggested in the manual (lots of crawling around with a flashlight involved on my part ).  The ARC itself and the front-panel encoders always cause an immediate response on the front-panel display, and all ARC functions & lights work - so it appears the ARC and MF-XT are communicating properly.

What does the MF-XT do differently during boot-up if there is an ARC connected/detected?  Does the MF-XT "ping" the ARC on boot-up (or upon loading TMFX) and alter anything at all?

No clue if this will help, but it seems like whatever is happening is happening right after the MF-XT is powered on and does not know which state it will be in (no active status colors on the front display to indicate PCIe or USB is being used).  Once the MF-XT discovers it will be in PCIe mode (right after the E-PCIe card is powered on with the PC and it re-displays the "SW Version 30" message), it will "remember" that it was in PCIe mode even when I shut-off the PC and E-PCIe card (the front panel display still shows "PCIe" unless I power the MF-XT off and back on).

So once the MF-XT shows "PCIe" in the front-panel display, powering off the PC and disconnecting the ARC and rebooting the PC has no effect on the lag (it remains through a reboot).  The MF-XT must be powered-down, ARC removed, and powered back on in the initial state (doesn't know which mode it is in yet).

This observation suggests switching between PCIe and USB might cause different behavior during boot-up even if you leave the ARC connected.

Any chance you have the USB cable connected but are using the PCIe interface during your testing (I assume PCIe will override USB if both are present)?  I'm solely using the Sonnet PCIe connection and PCIe (MADI-FX) drivers - no USB connection or drivers at all.

That's all I have for now.  I will try to stay after the session on Thursday and play with this some more if I'm not too exhausted.

Thanks for your time!  I know we will get this worked out

Cool!  I figured the PCIe mode might come into play...

One thing though - In my setup I did not observe any lag without the ARC (at least on the first reboot after a complete powercycle), and ALWAYS got the lag with the ARC connected - even on the first boot after a complete powercycle.

RME is the best!

I'll be heading to the studio later today and will give it a shot.

Wasn't expecting a fix this quick - very impressed!