Multi-Channel PCM not the same as real passthrough!

After using the Shield and Xbox Series X for some time with audio passthrough on a new proper 7.1 setup, I switched back to my Apple TV + Infuse yesterday as news about audio passthrough for MacOS had been circulating. To my surprise the Apple TV (or Infuse) definitely does not process the audio of TrueHD or DTS-HD MA correctly to my ears! The difference was slightly noticeable on my decades old 5.1 home theatre setup, but on a 7.1 Dolby ATMOS and DTS:X capable setup? The difference is VERY noticeable!

So I feel theirs a bit of misinformation regarding the difference between audio passthrough and Infuse Multi-Channel PCM on the forums. Typically people say “oh you only lose height channels for Atmos and DTS:X and everything else is the same” and that is just not true. You also lose the original mix of the audio tracks as it was intended to be heard. Especially from movies you rip from your own blu-ray or UHD disk.

It sounds like the original audio mix does not translate properly on the Apple TV at all. Dialogue is much lower than the loudest parts (with audio passthrough everything is balanced and cohesive) and the Apple TV (or Infuse) has some kind of Mid Range-High Frequency push/saturation going on.

So while Infuse can technically play TrueHD and DTS-HD MA codecs, it’s a bit misleading to suggest they can play them properly or “as intended”. This section listed below on Infuse FAQ is definitely not accurate.

Am I losing any quality by using LPCM?

No. Since LPCM is a lossless format, using it will result in no loss of quality. What your ears hear will be exactly the same. The only difference is your receiver will recognize the audio stream as PCM instead of Dolby/DTS."

While you may not be losing 7.1 channels or 48/24 bit depth, you are not hearing the elements of the mix correctly or “as intended” (dialogue, sound effects spacing and placement etc). So saying “what your ears hear will be exactly the same” is not accurate.

Hopefully we can push Apple to allow real passthrough, and I hope Infuse can clarify that on the FAQ because it is a bit misleading and definitely was one of the reasons I purchased an ATV and Infuse thinking the difference was not there.

Cheers and sorry for the long winded post!

1 Like

Correct. Infuse is processing the lossless core for those formats and loses the metadata needed to recreate object-based. But for non object-based formats, there is no loss in quality. If there is a sound difference for non-Atmos audio, then it is your receiver settings and how it is processing the audio.

1 Like

This is what I thought as well, but that is not true unfortunately.

For instance just a standard blu-ray DTS-HD MA 5.1 track (no DTS:X) does not sound the same as the Multi-Channel PCM 5.1 from Infuse. I have no audio enhancements going on at the receiver, even the EQ is flat with no additional processing. But the dialogue on the DTS-HD MA track is noticeably clearer in comparison to the multi channel PCM, and the overall sound and dynamic range of the audio is more balanced on the DTS-HD MA passthrough track. No additional processing happening at the receiver.

So it seems some mix information is either lost by going to PCM (not bit depth to be clear that is the same) or the Infuse or Apple TV decoder is not replicating the mix properly.

I’m a musician, audio engineer and music producer so my ears are very sensitive to volume and frequency changes down to even the smallest .1db changes, so for me this is very noticeable now with a good setup.

If I knew how to do it I would love to import the audio tracks into my DAW to compare the waveforms of audio passthrough vs PCM. Even the volume levels are not the same between the two.

1 Like

I have to agree with @SpaceRanger33 here. Playing an actual Blu-ray Disc vs playing a ripped file with TrueHD audio via Infuse(while keeping the exact same settings on the receiver in both cases) the experience is quite different.

The audio levels, especially the dialogues are too low as if the LPCM conversion by Infuse is doing a disservice to the audio.

I have to say though this is my experience with the recent version of the Infuse. Previously (7.5/7.6 versions) the audio levels were almost the same if not exactly.

Not all receivers are able to work well with decoded audio in PCM. A well-known fact. If you can hear the difference even in DTS HD MA, then you have exactly such a receiver. Write to the manufacturer’s technical support so that they release a fix for this problem.

Has nothing to do with my AV setup. Even older PCM 5.1 tracks from older blu-ray movies I have sound a bit different in passthrough even though they technically shouldn’t?

It’s definitely something up with Apple TV OS or Infuse audio decoders. Would be curious to hear from James if the audio decoder adds anything extra into the signal that could cause audio not to be decoded properly.

I thought LPCM audio is naturally lower in volume compared to Dolby and DTS due to the other formats having boosts applied.

Are you saying when comparing a Blu-ray Disc to Apple TV with Infuse and both playing PCM? Have you checked to make sure the settings are the same for both inputs in your AVR or pre/pro?

And if you’re really an audio engineer how are you clueless on how to load the audio into your DAW?

It likely varies by movie and depends on the mix. Some LPCM tracks can be mixed loud, while others can be mixed a bit softer.

As I said above, yes this is with all sound/settings the same with no additional processing happening on the AVR. No EQ no boost no enhancement when comparing all devices are just outputting the audio.

You misunderstood me perhaps. It would be easy to strip the mkv source file audio out with mkvtoolnix and load it into my DAW. That’s easy, but is pointless to compare how the Apple TV decodes the audio vs another device. I would need to find a way to capture the raw decoded audio of the Apple TV, then compare that to the source file audio track or even an audio passthrough file from a Shield or Ugoos AM6B+ to see where the differences are.

Perhaps I’ll do a bit of digging today on my free time and see what I can find out.

This post might be useful:

And you checked the levels? They can vary by input.

As long as you didn’t cheap out and your AVR has pre-outs then you route them into your ADC and record. You can compare both sources and the original.