SoX Optimizations: Direct USB versus UPnP/DNLA

Studio
1

Audirvana SoX Optimizations: Direct USB Performance

After my recent deep dive into Audirvana Studio’s SoX upsampling optimizations and the comparison between my UltraRendu and Yamaha receiver, I got curious about how well those settings would translate to a direct USB connection. Some forum members suggested I should retest the Mac mini’s USB output (which I had been using before I got the UltraRendu years ago), so I decided to give it a shot.

I connected my Mac mini’s USB directly to the IsoRegen (USB reclocker and galvanic isolator) which is connected to the Benchmark DAC3, using the same carefully tuned Audirvana upsampling settings. I was prepared for a significant dropoff in sound quality, but was surprised by how good my system still sounded. The Benchmark DAC3 maintained much of the clarity and refinement I’d achieved through the Audirvana optimizations. The soundstage remained impressively “free”, and the tonal characteristics achieved with the improved upsampling settings were largely preserved.

When I switched back to the UltraRendu, however, the differences still became clear. The UltraRendu sounds similar to the Mac mini-USB path, but still results in a noticeably blacker background, deeper bass, even more depth to the soundstage, better instrument separation, warmer harmonics (notably in acoustic guitar) and more natural decay.

But here’s the interesting part: the performance was closer than I expected. My SoX optimization seems to have provided substantial improvements that mostly hold up even without the UltraRendu in the chain. Re-introducing the UltraRendu reminds me of the improvement achieved when I added better power cables to my amps and DAC: Frequency response remains pretty much the same, but music and soundstage become more natural and deep.

Side note: During the weekend I compared the UltraRendu setup to my Yamaha receiver’s DAC using Audirvana’s upsampling. To my ears, the Mac mini → USB → IsoRegen → Benchmark DAC3 with DSD64 still outperforms the Mac mini → Ethernet → Yamaha Receiver setup, even without the UltraRendu. The Yamaha sounds quite clean and revealing, but lacks some soundstage depth and results in less of a black background. The IsoRegen-Benchmark DAC combination really is great.

My takeaways:

  • Audirvana’s SoX optimizations resulted in “new magic”
  • Much of this magic was maintained when I removed the UltraRendu
  • The UltraRendu still provides significant improvements
  • Don’t underestimate a well-optimized direct USB path

(Perhaps also worth noting: The Mac mini [specs: late 2012, 16GB RAM, SSD, MMK fan kit, running macOS Monterey 12.7.6 using OpenCore Legacy Patcher, connected over 1m Supra ethernet to EtherRegen] is powered by an LPS with a medium-end power cable. When I did that upgrade I already had the UltraRendu. But the cleaner power probably also had a positive effect on the Mac mini’s USB performance.)

3 Likes
2

You will notice an appreciable difference when you replace the ancient Mac with the layer of intermediary software, between Audirvāna and the OS/hardware APIs, with a basic Apple Silicon M4 MacMini with more RAM, running the Audirvāna audio-engine running natively… The difference in memory bus throughput and computational power makes your 2012 MacMini platform look pathetic in comparison… The Apple silicon platform advancements in the context of digital-audio reproduction are significant in scale. :wink:

Again, the DAC comparisons are better left to other forums that foster these sorts of subjective assessments.
:notes: :eye: :headphones: :eye: :notes:

3

When I got the MMK kit (last year, at a discount) I compared my Mac mini (then still running Audirvana 3.5) to my M1 MacBook (Audirvana Studio). The old Mac mini with LPS sounded better, so my investments have gone into further power cleanup rather than a new Mac.

Oh and your understanding of OpenCore Legacy Patcher seems to be mistaken. Apart from replacing the boot sequence so that newer macOSes will boot (which has no effect on system performance), OpenCore re-introduces and replaces some drivers with patched ones. AFAIK Audirvana does not go through any extra layer (and this is born out in the magical sound quality I now achieve that goes beyond what I ever experienced with 32-bit Audirvana 3.5).

1 Like
4

I replaced my 2016 MBP 2.7GHz i7 with 16GB RAM with a M2 Max Mac Studio with 64GB RAM…

There is a bias of expectations that needs to be removed in the assessment of the power-supply swap and the employment of OpenCore Legacy Patcher software… I don’t argue that you perceive an audible difference… However, this is only relevant to the Mac platform you are employing, nothing else.

You are mistaken to presume that your software enabler does not affect performance… it is not native… it is not Rosetta… Audirvāna addresses low-level operations that affect the signal integrity at the Core Audio level, so this software interferes with the Core Audio level of Audirvāna operations and requires more CPU operational overhead.

And please spare us of another subjective ‘review’ of your playback system in this context.

:notes: :eye: :headphones: :eye: :notes:

5

You seem to be speaking on behalf of everyone here in the forum. But I must admit that (although subjective) I enjoy reading the reviews of @mhsmit

So as far as I’m concerned there is no need for him to stop posting his own experience here.

6 Likes
6

I expect some find value in the unverifiable subjective assertions… Some watch motor-car racing to just see the crashes and whether or not the driver survives… :wink:

Many of us have been indoctrinated by the ‘audiophile’ media community, to accept such subjectiveness…
Gordon Holt (Stereophile Magazine)

“Audio as a hobby is dying, largely by its own hand. As far as the real world is concerned, high-end audio lost its credibility during the 1980s, when it flatly refused to submit to the kind of basic honesty controls (double-blind testing, for example) that had legitimized every other serious scientific endeavor since Pascal. [This refusal] is a source of endless derisive amusement among rational people and of perpetual embarrassment for me…”

:notes: :eye: :headphones: :eye: :notes:

7

I hope you realize that any of your comments are as subjective as other posts here. Unless the moderators disapprove of my input for other audiophiles, IMO it is inappropriate for you to speak on behalf of others. If you don’t like subjective experiences, then I suggest ignoring them. I personally like reading other people’s subjective experiences. Imo a healthy forum community requires focusing on sharing our own experiences rather than disqualifying the personal experiences of others.

I don’t like boasting about my background (as it has nothing to do with musical enjoyment or the capability of hearing improvements to a sound system) but you can rest assured that I have in-depth knowledge about macOS architecture and macOS in general. The choice of moving to faster Mac (or a higher-spec DSD-upsampling DAC) versus optimizing the system in other ways is entirely up to the individual. I have achieved huge improvements in other ways. Thank you for the tip that I could buy a newer Mac and that it might increase performance, but that is not my priority at this point. Even a 2012 Mac mini is nearly idle when Audirvana is just serving its upsampled music over DNLA and 16GB RAM is more than enough :)).

8 Likes
8

I do… :sunglasses:

System optimization is relative to the amalgamation of components involved in the playback experience and the electro-mechanical and acoustic environment in which the system propagates sound-waves… all are subjective decisions which may or may not be quantifiable otherwise… I personally have not described all of my system power/ground/earthing enhancements with you which are esoteric and extensive, as they are specific to my subjective decisions and technical insights… all are extensions of fundamentally best practices, at the macro level and at the platform circuit topologies…

:notes: :eye: :headphones: :eye: :notes:

9

Now that’s some funny shit coming from you :rofl::face_with_hand_over_mouth::joy:

2 Likes
10

When have I produce a ‘review’? :roll_eyes: :smirk:
:notes: :eye: :headphones: :eye: :notes:

11

Exactly correct, I’m finding his comments and observations refreshing. Damn it he’s got me thinking about grabbing a USB cable and firing up 3.5…….

5 Likes
12

I would rather see these types of ‘reviews’ submitted to the “Lounge” :wink:

13

As it’s about experiences with Audirvāna Studio I don’t have a problem with these kind of posts here.

This is a forum and OP is well within the rules that apply in my opinion.

If the admins deem it necessary that we should refrain from these kind of posts, then I guess they will take appropriate action.

1 Like
14

I’m expressing my opinion… anybody on the forum can disagree or agree with my perspective… this is important… the balance of opinion is essential to the determination of value for the individual and for the determination of contextual veracity of assertions.
It is my opinion these sorts of reviews are better posted in the “Lounge”.

:notes: :eye: :headphones: :eye: :notes:

15

That’s just fine. I did the same. So let’s say we agree to disagree on where these kind of posts should best be posted.

16

I see there is opportunity to argue technical elements when these things can be corroborated by anybody reading these ‘reviews’… Otherwise the Lounge is limited to a demographic of Audirvāna users with greater insight than somebody just popping onto the site to view some threads… or a bot…

:notes: :eye: :headphones: :eye: :notes:

17

Your reasoning may not be correct. The Lounge is limited to a demographic of Audirvāna Forum users that are active. It may very well not be representative for Audirvāna users with a lot of insight. In my experience the majority of users/customers in general aren’t (very) active in forums.

1 Like
18

This then makes these reviews less valid, due to the lack of user insight/opinion in concert with the inability to corroborate subjective assertions.

19

As it may help others to engage in trying adjustments I think there’s enough value for reviews like this. And these others can then decide wether it is an improvement or not.

So I absolutely see benefits in it. I see your reasoning but for me personally the balance goes the other way.

And by no means I want to sound rude, but I have to go to sleep at this moment in time. Duty calls tomorrow :wink:

3 Likes
20

No argument there… I see simpler and less biased ways to compel this action… :wink: Thank you for your investment in time and insights, I personally appreciate this. :sunglasses:

2 Likes