I stopped watching at the point where he goes into speaker design. I don't have enough background in that to comment with any sort of authority. But I did pay attention to the first half of the video and made some notes.
He mentioned recording at 96 or 192 kHz to avoid the brick-wall filter. I do completely agree with this. Real time, brick-wall filtering is difficult to get correct, and when it is done wrong, it causes audible artifacts. Low-pass filtering can be done better with a dedicated down-sampling process. So yeah, recording at higher sampling rates in the studio makes sense.
But then he goes on to say that the higher sampling rates give more detail at even the audible frequencies. He gives the example of having more samples at 15 kHz. Here he's wrong. I used to make the same mistake, until I actually started doing the math, and understanding how PCM really works. It's simple, the only thing that higher sampling rates give you is the ability to reproduce higher frequencies. Period.
The host brings up inter-modulation distortion from high sampling rates. This is a real problem, but the Allen says he's never noticed it when asked. I'm guessing he has really high quality gear in every location he has the ability to playback 96 or 192 kHz audio. It's less of a issue with well designed gear, because it's purposely worked around, usually by filtering ultrasonics out. Cheaper gear doesn't attempt to do anything, and you can get strange, harsh harmonics in the audible range. If you don't expect everyone to spend 1000s of dollars on studio grade gear for playback, don't give them 192 kHz audio. Do a good down-sampled output, and give them something that doesn't have ultrasonic content.
The story about having different pressings of discs sound different seems familiar. I wonder if I heard it from him before. Back in the 80s there were issues with some pressing houses doing bad jobs at reproducing the master to disc. With today's duplication equipment it doesn't happen anymore. His comment about copies of CDs he's made himself sounding different is likely his personal bias making him hear something that's not there. I'd like to see him prove this with a blinded study. The host is also showed obvious incredulity.
In addition to that Allen didn't answer the question. The person writing in asked why a demo pop song distributed with Logic didn't sound like the mastered CD. He wanted to know what mastering magic is performed. It seems like Allen is a mixing engineer, not a mastering engineer, so it wouldn't be a topic he'd have much to say about. Plus it is a bit larger than can be covered in an one hour podcast.
Toward the end of the first half Allen talks about dynamic range. He mentions recording two tracks one 10 dB hotter. He says the louder track will clip during the louder parts, so he'll use the quieter one there, but uses the hotter to use the higher bits. Again he's wrong. This is a belief left over from analog days, when the noise floor of the media was a real thing. So you'd want your signal to be as loud as possible to give you the largest signal to noise ratio. SNR is fixed with digital, there's no benefit of "using the upper bits." The only place that the a hotter signal matters is in the analog chain, but studio gear has such a low noise floor that it isn't worth mentioning.
So of the five big topics of the talk he was right once. For the most part he doesn't know how digital audio works, and is hearing what he's expecting to hear based on his misunderstanding. He probably has a great ear for mixing, and obvious lots of connections from the number of names he drops, and his good business sense got him where he is. None of those things needs an understanding of how the bits are working behind the scenes.
