HD11 - Coreaudio. Why are hardware inserts corrected by 1230 samples by default?

[LDS]

When going through the procedures in the manual regarding manual offsets for hardware inserts in the I/O menu for coreaudio interfaces, I have hit a wall.

The resulting audio when running the test (track 1 with hardware insert routed to track 2. Arm track two, record and measure the difference in time between the two - Yes, delay compensation is enabled as per the manual's instructions)...

I get a result of negative 7 samples. That is to say, the audio recorded to track 2 is 7 samples ahead of the original audio on track 1.

This makes the I/O menu hardware offsets absolutely redundant as it assumes the resulting audio on track 2 will be behind track 1. Any offset you enter will drag track 2 forward by that amount. Wrong direction!

On a side note, running this test using Protools HD 11.1.2 resulted in track 2 being over 200 samples in front of track 1. Updating to the latest version reduced the result to 7 samples.

What is going on? Is the new AAE in HD hardwired to compensate for HDx/HDn only? As you can, channel 1 is being compensated automatically by 1230 samples despite there being no hardware offset values entered in the I/O menu. It appears its 7 samples too long.

Would a feasible conclusion be that my converters are 7 samples faster at DA/AD than Avid's, hence not being compensated correctly? If that is the case, why can't the I/O Hardware Offset Delay affect the audio in both directions?

While its a non-issue with HD/HDx/HDn, this aspect of protools is incredibly poorly implemented compared to other coreaudio/asio compatible systems.


Thoughts? Has anyone managed to set these offsets correctly?

[LDS]

Nobody has any experience with this?

[LDS]

Nobody has anything to add on this??

Surely this indicates a bug somewhere?

Either in Protools ADC or the latency being reported by the Lynx USB card?