I’ve been using a macbook pro 2011 with high sierra for a while with audirvana. In the past there was a script that would optimize older Mac OS for music playing. I still have it, but it stopped working well a few systems before high sierra. Is there any “official” optimization path for macs, both old and new using audirvana? Of course we can do it item by item going through the system preferences and a few other things, but I wonder if something more effective regarding the excessive number of processes could be done.
Audirvana manages the macOS Core Audio API’s very well… it is a high-performance audio engine and optimizes audio playback on the hardware topology… A simple common-sense approach, without superfluous background operations will deliver an un-compromised signal to output buses… Careful consideration of Audirvana settings pertaining to macOS background operations and careful attention to DAC settings, playback pre-load buffer RAM allocation, etc, in Audirvana will provide you with a very high level of music playback appreciation…You are better served to manage your playback system component configuration and power/ground/earthing, RF and EMF influences on the digital-audio bit-signal, that you ultimately assess in the final audition.
Not related directly to Audirvana but I used this to optimize my Macbook with High Sierra:
Thanks, buddy. You again coming to rescue, appreciate it.
The ground and electrical noise are obviously crucial aspects of any step of an audio system and one thing (optimization) doesn’t exclude the other, does it?
Just so you know, I have bought the 2019 MBpro with 64gb ram and configured in a radically optimized way to be used exclusively with audirvana. Let me tell you right away that my old macbook pro 2011 with 8gb running High sierra and without any particular configuration sounds more musical. There’s more detail with the 2019 MB, but my guess is that either the USB-c (and the nasty adaptors needed) or the second graphics card (probably both), hurt the sound pretty badly. The older mac is simpler and I think will sound clearly better than the newer one if I change the memory for 16gb, which I’ll do.
By the way, you had suggested that using the AC supply sounds better than the battery. I thought it was a strange observation as with the older models the battery sounds no-comparison better. However, since, I believe, Catalina, they have indeed changed the battery management in the system, having mainly introduced an option to let the computer automatically select the graphics card according to the situation in order to save battery. For sound it’s a disaster. Having that option checked when on the battery makes indeed the sound worse than the ac supply. Unchecking it and configuring other sets properly make again the battery a superior option. There’s a script to turn off the discrete card when on batteries. I have tried it and it doesn’t work: when you use it, the automatic graphics option gets automatically selected. It’s a mess if you want to optimize things for music and one is better off almost in every possible way with a simpler/older machine.
Also, the argument regarding future updates and compatibility can be at the limit avoided by using the linux version and controlling it through an iphone connected by ethernet. Whenevr I have more time I will try that too.
I can only guess that the M series with the SoC chips are even more problematic in that regard and perhaps will compare even worse. In those machines the aforementioned option is not available, but I suppose much more battery management interference goes on there. I might get one for the office, try it just to confirm this and post my findings here.
It’s been my experience that major system changes take time to get a grip-on and tweak to satisfaction… Cables that worked previously don’t work for the new system, etc, etc… What you are describing is obviously subjective and centric to your system configuration… It will be interesting to see what you ultimately arrive at in these regards…
Thank you, that’s a good effort on their part. Other than the scripts I have been doing all that for years. It’s mostly a common sensical approach, with the exception of suggesting using bluetooth keyboard and mouse… So one does all that and leaves bluetooth on…right. If one can’t hear much of a difference when either bluetooth or wifi are on (not only on the computer, but ANY where in the house) they have a huge amount of noise circulating in the system masking those differences, and a much bigger problem than optimizing the OS…
The way people talk about subjectivism is a bit silly to me. All experience is subjective, by definition. You can/t feel what I feel or think what I think. And not only art is like that, but any field of human knowledge. Objectivity only exists when we have a way of measuring the phenomenon in a way accepted by the others involved. But the experience that you are perceiving with your mind cannot be measured, only tiny aspect of it. At the same time, there must be objectivity “within subjectivity”, something that stopped being discussed since the Renaissance, otherwise human being would not be able to communicate. Otherwise, they would not be human.
The second comment I have is this. Whenever I read something from someone I don’t know that describes what one can think as his sincere testimony of something, I give the person the benefit of the doubt and actually act as if what is being said is entirely true, even if completely against my first inclination. That’s what I believe to be the right scientific method or approach.
The key to discernment is understanding the potentials of Naïve Realism… 
I like that, but can you expand it a little more?
I believe a naive approach to things is the ONLY way to live. Especially in philosophy and difficult books of the past. Always secretly take the position of the dumbest student in the class…
The construct of objectivity must entail an understanding of the non-linearities involved in the observation or assessment… Otherwise we are bound by Naïve Realism that has us believing that our perceptions are common among all observers, when they are not.
oh so you were being sarcastic… I see.
No, not at all… Maybe in corroboration, but also adding a perspective relevant to objective methodology and the assessment of the results of the objective mechanism…
To be continued. I must go to bed.
When you wake up, you might consider this: Installing Linux, mpd, and upmpdcli on a Raspberry Pi is an inexpensive way to see whether a low power computer running stripped down software connected to your DAC would provide better sound quality. Then you can actually use the computing power of the newer Mac to run Audirvana sending the signal via UPnP without worrying that it’s transmitting noise directly to the DAC.
For various reasons this may or may not be something you want to do, but I thought I would throw it out as an alternative.
Thank you, Jud. But if the USB-c ports and adapters plus the two graphics cards and battery management are making more noise than in a smaller computer (either a rpi or an older macbook) using it with linux would not make it quieter, would it? Or perhaps it would, I have no idea. Right now I have no time to go after that answer, maybe during the summer.
I suspect from my comparisons so far that the extra computing power advantage is completely shadowed by those other aspects. Either the newer OS or the newer machines or both seems to make more noise.
Right, if you did this it would be an arrangement where the MacBook sat in a remote location and you controlled it with a tablet or phone. Just the Pi would be in the listening room connected to the DAC, and that would be running an extremely light load so it wouldn’t be noisy.
The issue (if it is one) is that @Jonas believes his 2019 MacBook Pro (Intel) and the M series SOIC platform topology is some-way precipitating noise related jitter, and this will be presented to the output buses (USB and Ethernet), We all can agree that platform topology ‘noise’ will influence the bit-signal and packet-signal as compared to the bit-signal that is encoded on and lifted from the storage medium… Without some sort of objective measurement we are bound by our biases that influence our perceptions and juxtapositions to prior experiential assessments of the final output we audition… In many ways this is a bit of a snipe-hunt…
Jitter being precipitated by electronic noise through a wire may be possible. Not saying it is happening, but it’s rationally possible. That’s why you remove the source of electronic noise and connect the DAC to an electrically quiet computer that isn’t doing any heavy lifting.
It is all about the ‘intrinsic jitter’ of the computer platform topology related to component slew-rate, RF migration, power buffer slew-rate as it affects gate speed and comparator read/write accuracy, and how these influence the software registers (buffer) slew-rate and wait times, along the clocked bit level signal-path to the output buses…
You can provide a perfect square-wave power signal, but this is subjugated by the regulator circuitry and the power distribution topology of the computer platform and component topology in concert with the layout of the signal-paths and the quality of the component topology slew-rates and tolerances.
We can expect that at higher CPU operational speeds, these things will become more influential on the encoded digital-audio bit-signal. In a SOIC the signal-paths are extremely short and wait-time latencies are inconsequential.
I agree reducing the OS operational demands on the CPU will reduce ‘noise’… My perspective is associated with how the bit-signal is handled by both the OS and the computer platform topologies… You will find that a vertically integrated computing platform + OS will provide the least possible level of operational noise influences… There is a synergy of CPU and memory throughput and available operational RAM… especially in our realm of digital-audio playback and subjective sound-quality assessments.
These key non-linearities are always at play:
From the Dolby.io article linked below:
Using Deep Learning for Understanding the Perception of Audio Quality
Arijit Biswas, Lars Villemoes, and Guanxin Jiang
(1) Audio has a high dimensionality
It is possible to identify the content of an image from a 64×64 image, but the substance of audio may not be recognizable from an equivalent number of samples.
(2) Audio is temporal
With audio, there is a multi-scale temporal dependency on the interpretation of a frequency and the order in which it occurs.
(3) Perception of sound isn’t uniform
Two audio waveforms may look completely different, but may sound exactly the same.
Using Deep Learning for Understanding the Perception of Audio Quality - Dolby.io