NOS vs OS: An Audiophile Culture War?

There are no transparent ADs or DAs. NOS is much further from the top tier ds chip DACs but there aren't a lot of top tier ones honestly. Most of them are crazy overbuilt 2 channel pro shit for thousands of dollars and they're not totally transparent, eg the Crane Song, Prisms, Lavry, and better Apogees. Ones beneath them (Lynx then RME) have less detail but are flatter, both tonally and sonically for whatever reason. If you do NOS well enough with a good enough clock, it's wildly inaccurate but it doesn't suck ass. It sounds like an oversampled dac with weird staging and dulled sound. Unfortunately audiophiles love mud and distortion so most of them are junk.

All of your big time popular records back when music sounded good had a massive build up of sound color from using the same mic pres and console circuitry on everything. The frickin SSL 4000s with underspecced tantalum caps and NE5532/NE5534 with old school lexicon reverb (distorted as hell) into some Studer with a bunch of mediocre opamps is on most big budget 80s and 90s records. some steve hoffman forum troll is only getting rid of it with an incredibly distorted setup. Old school rap guys were flying in vocal lines from 12-bit samplers. Can't get lower fidelity than that. But those records still sound better than most of what is made now in DAWs.

 

While I agree with a majority of what you say, (since most hi-fi purists would be appalled at how recording actually happens), but you know as well as I that nobody, and I mean nobody, used the mic pres in the SSL 4000. Lest maybe some jazz recordings you might hear on the Weather Channel. Seriously, those things were clean, bright, and had no color to them at all. Bright and flat. SSL 4000s are the choice to mix down on though. Then run everything through that brick wall compressor in the master section.

Most of the small studio's I saw in town in Nashville had some run of the mill console and tons of outboard preamps. Most of their stuff they recorded would be farmed out by someone else to mix anyway, or they would book time themselves in a real studio mixdown to use that crazy Lexicon reverb or other effects and compressors they did not have on hand.

It is also funny you mention clocking. I remember when the Apogee Big Ben was the thing and doing an A/B comparison with DAC's with that thing. It was crazy how even the worst sounding DAC could sound really close to the most expensive DAC just with that clocking box. So I can imagine what you say about good clocking and NOS.

 

Now you can buy clones of the 4000 circuitry for the distortion! The current “Super Analogue” design ones around since the 9000 are quite good though but they do not sound “SSL”

Yeah they used some small PCB consoles that most of the time, sounded just as cool as the SSL, just more distorted. And you don’t hear the lol ssl channel strip comp.

Apogee does not mess around sound wise.

 

This has been my experience as well with my Dante Data Delivery System, using FS725 rubidium time reference disciplining Antelope LiveClock providing word clock to Focusrite D16R. Most DACs in my collection demonstrate both audible and measurable improvements when fed from the DDDS.

 

At least for me, I prefer modest oversampling. By modest I mean up 4X PCM oversampling with various filters either from my Roon server, HQPlayer (sinc-M, NS9, PCM), or Gryphon.

Any of my sources/DACs offer NOS, but I find that I enjoy the ability to adjust the sound to my preferences. Maybe it's because I'm a grey muzzle with damaged hearing, maybe I wasn't hugged enough as a child, or that I just have awful taste in music, but I have found that oversampling and filters makes the genres of music I listen to sound great to me.

 

https://pinknoisemag.com/pink-papers/pink-paper-002

Is a good primer for those interested in the history of external clocks. Of course people are often just crippling very well clocked digital hardware by slaving it to an external clock.

the high end 2 channel converters mostly have great clocks. And the Apogee Symphonies of course are still awesome but not geared toward audiophiles at all.

 

Keep in mind my Dante Data Delivery System does not discipline the DAC directly. No word clock input to the DAC. Rather it serves to provide the most stable, jitter free digital data stream such as AES/EBU or spdif to the DAC digital input.

 

Switching back to NOS after a few days there is something that feels more "musical" about the sound, a little more "unfiltered/raw". OS (HQPlayer) sounds a bit more exciting, but I think the filters definitely have a certain "flavor" that is very noticeable. Will be interesting to compare the Holo Spring 3 to Yggdrasil LIM, I'll have both in house next week to compare.

 

Finally got a DAC with a native NOS mode, the Teac UD501, and I've been playing with the filtering for the past few days.

I find that on 44.1khz music, I notice and am bothered by the rolloff towards 20khz. However, on actual hi res files 88.2/96khz or higher, NOS sounds more detailed than either of the filtering modes even though the FR difference is minimal and generally in OS's favor anyway.

What I ended up doing was set up 8x software oversampling for only 44.1 and 48khz files in Foobar, and leaving the DAC filter turned off (I find retroarch does a better job than the 1795's internal sharp filter anyway). Still have to figure out some solution for Tidal that doesn't involve paying a bunch for Roon, tho :|

 

Another vote for what Erroneous said. If it sounds good, it sounds good. Who cares if it's NOS or not.

Holo Spring 2 is in NOS - I was cycling through modes and for most tracks found the NOS to be slightly better so I ran NOS. I had no idea at the time that the general consensus was Holo = run NOS. Same with when I tried the WD or WL - didn't realize the consensus was OS is better, I just knew I liked one of the OS modes better in that DAC. So on and so forth.

I'm sure we could also examine all of this from a signal theory perspective blah blah blah - which I don't see here thank goodness. But IME the theory often doesn't translate into how things will actually sound in terms of enjoyment or if one induces headaches for any one person or not, etc.