Well I've nothing more to say, I've exhausted all explanation so I'm out of the thread from this point onward.
Hopefully future readers will learn something interesting from it no matter what their take on the topic.
Both sides have been put forward, but only one is correct.
I'm stickin' to real-world physics and common sense 
One reason why I dont write too much in this post, cause I think it is semi-pointless.... AND it is a debate that have been discussed thousands of times on every audio forum around. And ALWAYS end up in the same place; everyone is disagreeing and putting their personal thoughts in and ends up semi fighting over somthing that is a personal choice........ And I think that is actually the REAL point here...... Do what YOU prefer to do... And dont puts limits on your creativity because of made up rules.
If it works for you, use it. If not, then don't. But I would like to point out that eliminating digital circuits from a studio is impossible. I mean a lot of analog synths use digital circuits to control the analog oscillators. Digital is everywhere also in analog synths. I think there are very few analog synths that are truely 100% analog these days. I couldnt live with out granular synths and wavetables.
I agree, but the problem a lot of the time (and you'll notice it in this thread), is that people often dismiss the fact that the one arguing the side for analogue is not against digital.
I love digital every bit as much as everyone else here does. I wouldn't want to be without digital wavetables and sampling technology either. I plan to design and market a sample-based instrument myself, so it would be majorly hypocritical of me to be against digital.
blind listening tests show that beyond 48k sample rate most people can't hear any difference between digital and analog.
Your point is valid but it's largely based on mythology and placebo effect I'm afraid.
You have probably never heard a truly analog recording anyway.
Imagine for a moment that digital did not exist as a technology…
We could have spent this entire thread arguing about whether air or electrons are superior. One side would be arguing that air "is the daddy" and is superior. Any conversion of sound - which we hear only through the medium of air - into another format like electrons (Voltage) is inferior to "the real thing."
In other words, any electronic representation of sound is fake, and is inferior to the real thing. Embrace the air, fellow humans, because it is the only way we can hear sound. Every other medium must be translated into movement of air before we can hear anything.
Anyone with experience in electronics will understand why analog electronics are inferior. There is noise, there are unwanted radio waves that are picked up and mixed in with the audio. A reference, usually a ground voltage, is always needed to interpret any electronic waveform, and yet analog electronics suffer from ground loops and many other destructions of that reference, putting hum and other noise into our audio. I could go on for days about how electronics are inferior to acoustics, even conceding the various improvements like balanced audio, and still would be convinced that acoustic "air" is the daddy.
Everything else (everything that is not acoustic movement of air) is a pale, inferior copy, and is not real.
Now imagine that we include digital in the discussion…
Forget for a moment that digital is quantized and sampled. Forget what you see on the screen when you edit a digital recording. None of that matters because you can't hear it. All that you can hear is the result of those digital numbers first being converted into electronic signals, and then those resulting electronic signals being converted into acoustic vibrations of air. My point here is that the quantization no longer exists in a proper conversion with an appropriate reconstruction filter. The reconstruction filter that is required after a Digital-to-Analog converter is analog, and it fills in the gaps of the digital steps. There is no longer any instantaneous jump from one sample value to the next, and there is no longer any evidence that the time domain has been sampled, except for the lack of frequency content above a certain point, but that is equivalent to an analog low-pass filter.
Here I say that digital and analog are equally inferior to acoustic. Each has problems and advantages, but as time goes on, analog electronics have fewer and fewer advantages. Ideally, only a very small part of the non-acoustic representation would be analog, with the vast majority being handling in the digital domain. That would allow us to dispense with nearly all of the shortcomings of analog.
In the real world, where acoustics are superior to both electrons and digital numbers, there is no definitive superiority between analog electronics and digital electronics. Both are vastly inferior to the real sounds made and heard in the air. However, as many in this thread have pointed out, we humans are not all happy being limited to the sounds that can be produced acoustically. Since we desire to hear sounds that can only be produced electronically, then we must put up with the shortcomings of both analog and digital electronics.
If you want to understand more about digital, then I recommend reading Audio Engineering Society papers from Stanley P. Lipshitz and John Vanderkooy. Until you can understand their research on quantization and sampling, then the crude criticisms of digital waveforms that you see on your computer screens will continue to be grossly misguided.
Brian Willoughby
Sound Consulting
@strum
I've read a few times that 48 is the sweet spot before deminishing returns. I'm inclined to agree with it as well cause I had an external Steinberg audio interface where it actually had better noise performance and sounded better running at 48 than it did at 96. I could tell the difference in the audio quality when feeding it with audio and switching back and forth between the two. I definitely preferred it at 48.
Regards the "placebo" thing, nope, definitely not. It's not just about how it sounds, it's about how it behaves. I don't just mean physically in response to knobs been tweaked, I mean how it behaves when levels are pushed, and when it's being coloured by components and recording media. That's not plecebo, it's very real, sounds different (analogue characteristics), much nicer, more solid, more width, more balls.
The amount of 'sound satisfaction' I get from my old 80s Hi-Fi system playing Tape or Vinyl is leaps and bounds above anything I get out of CD or MP3 or whatever - there's no comparison. The all analogue Hi-Fi sounds big, lush, thick, soft, totally gorgeous in comparison.
@rsdio
You're talking about a different thing there. Sure, an analogue recording of an instrument is going to be inferior to the sound you heared live, but the same is true of digital. The point is that sound travels, and causes vibration, completely in the analogue domain, so obviously it's better to capture those vibrations directly to an anlogue format than it is to quantize it, process it, then convert it back to analogue so you can actually hear it.
As I said earlier, it would go against physics to suggest that converting something from analogue to digital and then back to analogue is better than just letting it remain analogue in the first place. With analogue tape the vibrations are stored magnetically, and with analogue vinyl, through the groove cut by a vibrating stylus.
Analogue is the daddy - as pure and direct as it gets.
Except that it's not direct. Air, electrons, and magnetic particles can each be described using multiple terms - "analog" is only one part of the description. You're making the mistake of thinking that Voltage is the same thing as air pressure just because they're both analog. You've oversimplified the comparison, and then convinced yourself that they're the same when they're not.
You're not literally recording air. There is a transducer that converts air movement into Voltage fluctuations. Another transducer converts Voltage fluctuations into magnetic fields for tape, but slower tape speeds end up having fewer magnetic particles. Tape is even "digital" in some sense, because there's a threshold below which it doesn't work, due to hysteresis. Below the threshold, tape records and plays a "0" - above the threshold it's basically analog. Tape bias works around this semi-digital limitation of tape by mixing in an ultrasonic frequency. But if you think about it, this means that tape is sort of sampled in time, because every time the bias frequency crosses zero, there's a chance that the analog input signal is not getting recorded because the tape stores a zero. Some of the time, the signal without the bias will be strong enough on its own to exceed the threshold, but that is not true all of the time. So, between the problems of tape speed and tape bias, analog recording is neither pure nor direct. It's manipulated in many ways, some of which could be called digital or binary.
You've chosen to ignore all of the details that make electricity or magnetics unique, and decided that calling them analog makes them the same. Unfortunately, they're not. Conversions are needed between every analog medium, and these conversions are neither completely better nor completely worse than digital conversions. Maybe a wax cylinder recording is direct, but that doesn't allow any processing and I'm sure none of us would prefer a wax cylinder recording to a digital recording.
Keep in mind that everything is quantized. Air vibrations involve a finite number of molecules. Voltage fluctuations involve a finite number of electrons. Tape recordings involve a finite number of magnetic particles. Time may be continuous (the theory that time is discrete was shot down, I believe), but some of these "analog" media involve discrete time. Saying that analog is better than digital is just saying that the quantization is finer with analog, if you can detect it in the presence of noise.
You're explaining in a similar fashion to Sputnki, and just like Sputnki, you continue to dismiss the main fact that is pointed out time and time again. You cannot take an analogue signal, convert it into computer speak, then convert it back to analogue and expect that it would be superior in any way whatsoever, to keeping it analogue in the first place and avoiding two conversion stages in the process.
The stuff Sputnki and yourself speak of also dismisses the fact that analogue is a completely different thing to digital. Trying to lump analogue under the "quantized" hammer is completely pointless. Digital is a man-made quantization system, analogue is not. Humans do not listen in Digital, they listen in Analogue. The cheap DIY turntable video I posted earlier demonstrates this perfectly. You ain't listening to any CD or MP3 in such a "direct" no-nonsense fashion.
So that's it in a nutshell (for me anyway). Analogue sound reproduction is better primarily because it is the purest way to store and replay sound. That's the main aspect. The second aspect, is that is simply sounds better. Components, choice of media etc, it all makes a difference, and the beauty of it all is that every component in the sytem is reacting to the way it's designed, giving any analogue device it's characteristic sound. Poor design gives poor performance, good design gives good performance, exceptional design gives exceptional performance - which is why NEVE and SSL consoles cost what they do - it's expertise in analogue design.
The attempts to lump analogue under the quantization hammer will not work*
Analogue is analogue, digital is digital - very different systems.
*That's not to suggest I don't appreciate the interesting stuff I read here, I love it actually. But you're never going to convince anyone with any common sense that conversion is better than no conversion, or that analogue is a quantized system. Like I said, that seemingly innocent video of the DIY-turntable using cardboard, either it will trigger something in people, maybe help them understand all they need to understand, or it won't.
And if it doesn't, then I really feel for those who don't get it.
Would it be possible to build a machine that senses the seperate quantized digital states of the analog audio output of the Axoloti? Lets also give as a caveat to this audio Turing Test that the output is coupled acoustically to the air via speaker. A reverse Axoloti, able to recreate the exact digital code of the now acoustic waveform (without give aways as to the sample rate and bit depth). I would guess no, that the jaggedness of the waveform would be sufficiently filtered, but I am a bicycle mechanic and not an engineer.
I love analog cause you can touch it and twist it and find the sweet spot. I love digital because of state saving, bit crushing insane manipulations you can do to a signal through time.
by the way.
human hearing is digital.
because sound is transmitted to the eardrum via very large, but finite number of discrete air molecules. 
@Dolphinwolf
What you describe sounds like a DAC, something like demonstrated in this video, basically a bunch of resistors, but I'm not sure you'd be able to build one for Axoloti as there's a DAC already fitted to the board, situated before the audio out. Axoloti is already doing what you describe as part of the hardware spec, unless I'm misunderstanding you. There might be a way to bypass it and use your own, but that's a bit beyond me at the moment 
@chaocrator
My hearing is not digital, nor is yours, cause digital is a man-made system (it's computer speak) whereas your hearing system is both organic in nature, and analogue in operation. I sometimes wonder whether it's just the names used in Digital technology that draw people to it.
Fact is though, Analogue technology has much cooler terms, such as 'Electron Gun'
I mean come on, "Electron Gun", how bloody cool does that sound? And they're real, not science fiction, Electron Guns were used to display the image on the screen of analogue display equipment like this:
Ladies and Gentlemen, I give you a box with an actual Electron Gun inside of it. Make no mistake, if you were to remove the back panel and touch one, you'd be looking at 'Death by Electron Gun' ... but still ... 