It's so cute looking. I just want to hug it.
It uses a nextgen XMOS XU-216 USB chip (1) which is capable of receiving 32 bit / 768khz PCM input. It also supports MQA and DS512 of course but those formats are already obsolete. The DAC is an ES9038Pro DAC chip with an ES9311 dedicated power chip that provides clean, low-noise power to the DAC. As icing on the cake it sports a brand new self-developed CK-02 clock processing circuit that reduces clock jitter.
So yeah. 32 bit / 768Khz uncompressed or GTFO. The dynamic range that can be represented by a 32-bit (floating point) file is 1528 dB. WOW. I'm speechless (one slight problem. The greatest difference in sound pressure on Earth can be about 210 dB. Not something you want to experience).
I hope you have enough space on your NAS or fileserver to store your entire collection. The bitrate is 368.64 Megabyte per minute. So for a CD of 74 minutes you need 27.28 GB of HDD space. It's not a big deal. According to Seagate's roadmap (a US HDD manufacturer)
50+ TB HDDs will be available sometimes in 2026. You don't need to worry about running out of space.
I'm waiting for the day we start doing some ABX testing with those files like MasterApex is doing on a thread on this wonderful forum. (2)
I'm joking of course. Can you hear the difference between a 24/96khz and 32/768khz PCM file? Don't be silly. Of course you can't.
I already have difficulties hearing the difference between a FLAC file and the very crude Low-complexity sub-band codec (SBC) bluetooth is using. I only use it in my car. But still I ripped my entire CD collection and captured some of my more obscure vinyl stuff to FLAC just because I can and I have plenty of HDD space on my NAS. I can access the thing from the internet (IT nerd) of course. But in case my apartment complex burns down (it happens you know) I also have a backup in the cloud. The problem is that I only have 2 TB of cloud space at the moment. So I also converted my collection of 100.000+ files with a program called dBPowerAmp to AAC (Apple encoder) at maximum quality (true variable bit rate quality 127 ~335kb) which is already overkill because it's already supposed to be transparent at 128kb.
There are several reasons for this. I'm not trained. I live in a city with a train station at walking distance and parallel to that is a highway. On the other side of my apartment there's a main road which leads to that highway. As you can imagine this generates a lot of noise. Normal SPL levels here in my living room are about 38dB and add a few db more at rush hour. I have a few closed headphones for monitoring purposes and they do isolate somewhat of the background noise. But unfortunately not enough. For this kind of work you need an environment with < = 25 dB of noise (this also depends on the frequency of the noise. Car noises are predominately low frequency). This is about the same as a typical recording studio.
To spot tiny artifacts in music is hard work which will require a lot of training and requires your complete attention. I don't envy expert listeners. It's kind of a tedious and time consuming job. You also have to do blind tests over and over again. (statistics is a bitch. You want a 95% confidence interval or not? Let's do some more) Some people have a knack for it (like Amirm). I'm also getting older and that doesn't help (vintage 1969). There's also the fact that different expert listeners are sensitive to different artifact types. I name a few: Pre-echo, unmasked granular (quantization) noise (Amirm is good at that) and high frequency cut or boost.
In the early days of MP3 (the format was released on 6 December 1991 by Fraunhofer institute. So the format is more than 30 years old). I was fascinated with how this was even possible and experimented with encoding myself. I thought it was a kind of magic. (see my signature) It wasn't. It's psychoacoustics. In the year 1999 a new phenomenon emerged on the internet It was a software program called Napster. You could share music files for free. That was totally illegal of course. And I apologize (Admission of guilt. please don't sue me) to the copyright holders.
Most stuff was encoded at a constant 128kb bitrate. A popular encoder at the time was the Xing Encoder because it was fast. What it did was cutting off high frequencies > 16 Khz. I could immediately spot the difference between the Xing encoder and the Fraunhofer encoder files even with my crappy computer speakers. At about the same time there was an open source project named L.A.M.E (LAME Ain't an MP3 Encoder) It wasn't very good at first but they kept on tinkering with the psychoacoustic model until they got it right. The MP3 spec only specifies the decoder and the format of the file so it can be played with your Ipod or whatever. The encoder isn't specified. It's the encoder where "the magic" happens. I can go on and on about Fletcher–Munson curves, critical bands, BARK scales, ERB (equivalent rectangular bandwidth) and stuff like that but this can get overly technical for the uninitiated. It's magic..
I listen to music to relax and unwind after a stressful day at work and for that my AAC files are good enough. But I have FLACs on my NAS so why not listen to them instead? I know my limitations and I'm good with that. I have no pretentions at all of hearing differences between this amplifier or that amplifier. But I strongly believe in the science because it's the only thing that's verifiable. Opinions can't be verified or tested so not valid in science. I'm also interested in psychology and love reading the comments here on the this forum. Happy new year everybody.
Opinions are like buttholes. Everybody has one.
(1) Datasheet-porn:
https://www.xmos.ai/download/XU216-512-TQ128-Datasheet(1.16).pdf
(2) highly recommended:
https://www.audiosciencereview.com/...ic-amazon-hd-sound-quality-differences.29450/ I'm afraid i won't make the cut so I won't be able to participate unfortunately.
If you want to know more about this Amirm made some excellent video's about the subject. I Highly recommend them
Audio Blind Tests and Listener Training:
Audio Blind Testing - You Are Doing It Wrong!