No LFE Output through HDMI on M1 Mac / Monterey 12.1

[natebot88]

Running REW 5.20.4 on a 2021 M1 MacBook Pro, Monterey 12.1

I have confirmed my AVR (Denon AVR-S750H) is receiving multichannel input via HDMI because playing Dolby Digital test files successfully isolates the LFE channel. The AVR display also shows "Multi In".
When running a measurement in REW with LFE selected as the output, REW simply outputs a normal, full-spectrum sweep through sub+mains, which does include the Sub until it passes the crossover point. It's like it's running a regular stereo sweep with one of the mains selected as the output.

MIDI output is set to 6ch, 24 bit, 48 kHz. Confirmed that is compatible with my AVR.
SoundScape SoundSource running / not running does not affect anything.

Trying "Check Levels" from the Preferences page simply outputs nothing with either setting (to subwoofer or mains).
Note: I usually use a UMIK-1, just testing outputs here.

Not sure what else to try... and hopefully I'm posting this in the right place

 

[John Mulcahy]

The HDMI connection is multichannel PCM, uncompressed - playing DD does not use the multichannel PCM path. Only the processor or the OS can convert a signal sent from REW on the LFE channel to one that plays through mains+sub (e.g. if the processor is set to Stereo), so the AVR processing mode would be the first thing to check.

 

[natebot88]

The HDMI connection is multichannel PCM, uncompressed - playing DD does not use the multichannel PCM path. Only the processor or the OS can convert a signal sent from REW on the LFE channel to one that plays through mains+sub (e.g. if the processor is set to Stereo), so the AVR processing mode would be the first thing to check.

Thanks for the reply John. Understood. Dug through the AVR settings and could not find a setting that resolves this. Working with Apple to see if we can determine what is folding the multichannel audio into stereo. Will update the thread with my findings!

 

[natebot88]

Hi John, I was able to confirm my AVR is correctly decoding Multichannel PCM over HDMI by trying a 6ch, 24 bit, 48 kHz .wav file. Apple's built-in MIDI Setup application also correctly sends an isolated signal to the sub. Neither Apple support nor the folks on the Apple forms could come up with any working suggestions. I have tried every relevant setting on my AVR and many other small troubleshooting steps within my MacBook to no avail.

The only other problems I could think of are: Possible cable issue (it is a nice cable, but who knows, will try another one later this week), or that REW is not outputting at 48 kHz, which (oddly) is still the only sample rate I can select within the OS. This is proving to be quite a tricky case to solve. Any thoughts?

 

[John Mulcahy]

Cable length is a common issue for HDMI, best kept under 5m. The only other thing I can think of is the channel mapping, sometimes the channel order is different to standard. If that were the case the signal sent on the channel currently labelled LFE could be going to another channel, but if you have speakers on all the channels you would still hear it, just not coming from where you expected.

 

[PyriteGLI]

I too am having a similar problem - no matter what settings I try I can only get 2ch out of my MacBook Pro to my Denon x3770h. I am using a rather long HDMI cable so I will try a short one to see if that helps.

 

[PyriteGLI]

For anybody else having this problem - I had thought my Mac was not working correctly as it would never output 8ch on hdmi out. I had tried multiple cables of different lengths and tried it on 2 different AVR's - same result.

It ended up being the USB-C to HDMI adapter that was at fault!!!!!! As soon as I tried a different usb-c hub I instantly got 8ch as an option.

Hope this helps somebody in the future save time.

 

[natebot88]

As of 7/31, REW 5.20.9 and Monterey 12.5, this is still not working. Thankfully, isolating output to SR or SL does work, and EQ'ing the sub individually can at least be done by running a full sweep on either L or R channels and disregarding data above the crossover point. Still an odd glitch though.