MTRX STUDIO + Thunderbolt 3

[ntp-jl]

Originally Posted by uptheoctave

Here is reading using Digital Loopback via AES, rather than just the TB to TB patch in DADMAN.

Sorry if it was confusing for people.

I suspect you have enabled SRC on the AES card. This will add some latency and explains why the measured RTL is higher than the reported RTL.

[ntp-jl]

The whole issue of RTL is quite complex, as can be seen in this thread so here's a quick comment.

The actual RTL (measured) depends on a number of factors. Apart from the obvious one which is the Coreaudio buffer size, it also depends on which type of interface (analog, MADI, AES, ADAT, etc.) is used for the loopback.

Another factor affecting the RTL is the internal delay in the AX64/MTRX/MTRX Studio. By default the internal delay is set to "minimum" which is 2 samples. If you enable the monitor profile, the internal delay is increased to 7 to allow for summing, SPQ, etc. The internal delay remains at 7 even if you subsequently disable the monitor profile. It can be set back manually if needed.

We have made some quick measurements on an AX64 with the minimum system delay and various interfaces. This shows the difference in RTL between the different types of interfaces.

It is also worth noting that the headphone outputs on MTRX Studio have a "short delay" mode which reduces the latency by applying a different digital filter. The amount of reduction in latency is dependent on sample rate.

I am sure this will raise a lot more questions so we are working to get a more elaborate explanation in a few weeks. Please be patient with your questions until then as I will not be answering questions about RTL in the mean time.

[uptheoctave]

Originally Posted by ntp-jl

I suspect you have enabled SRC on the AES card. This will add some latency and explains why the measured RTL is higher than the reported RTL.

Ah yes indeed, I have.

[uptheoctave]

Originally Posted by ntp-jl

The whole issue of RTL is quite complex, as can be seen in this thread so here's a quick comment.

The actual RTL (measured) depends on a number of factors. Apart from the obvious one which is the Coreaudio buffer size, it also depends on which type of interface (analog, MADI, AES, ADAT, etc.) is used for the loopback.

Another factor affecting the RTL is the internal delay in the AX64/MTRX/MTRX Studio. By default the internal delay is set to "minimum" which is 2 samples. If you enable the monitor profile, the internal delay is increased to 7 to allow for summing, SPQ, etc. The internal delay remains at 7 even if you subsequently disable the monitor profile. It can be set back manually if needed.

We have made some quick measurements on an AX64 with the minimum system delay and various interfaces. This shows the difference in RTL between the different types of interfaces.

It is also worth noting that the headphone outputs on MTRX Studio have a "short delay" mode which reduces the latency by applying a different digital filter. The amount of reduction in latency is dependent on sample rate.

I am sure this will raise a lot more questions so we are working to get a more elaborate explanation in a few weeks. Please be patient with your questions until then as I will not be answering questions about RTL in the mean time.

Hi Jan,

Thanks for this very useful information.

[Frank Kruse]

When using the new TB3 option can the system still be gen-locked to external video Ref with a SYNC HD or X or is that only possible via Digilink (since the serial connection to the SYNC is missing)

Thanks!

Frank.

[smurfyou]

Originally Posted by Frank Kruse

When using the new TB3 option can the system still be gen-locked to external video Ref with a SYNC HD or X or is that only possible via Digilink (since the serial connection to the SYNC is missing)

Thanks!

Frank.

Still only possible with Digilink as the playback engine. I asked Mark Corbin about that in the main MTRX II thread but no reply. I try to point this out to Avid every chance I get. PT needs to change on the software side and accept other sync sources.

As of this moment if you want frame-edge sync that's a $9,000 USD option