Jump to content
  • The Computer Audiophile
    The Computer Audiophile

    Audiophile Optimizer Raises The Bar

    Hi Guys, just a quick update from my travels this week. I’m currently flying over northern California on my way home to Minneapolis. I’ve been in the Bay Area for four days working on a couple music servers. My task was to install Windows Server 2012 R2 and Audiophile Optimizer to squeeze every ounce of sound quality from a computer. The end result surprised me very much. I had some reservations about the install and the efficacy of going to such great lengths tweaking a PC. I figured the sound would improve, but I didn’t figure it would improve by so much. This week I heard the best digital playback I’ve ever heard. Period.

     

     

    I’m not at liberty to divulge the entire hardware formula used this week, but I will talk about the software and the final outcome. What I found was similar to what many computer audiophiles have been saying for quit a while, the combination of Windows Server 2012 R2 and Audiophile Optimizer is fantastic.

     

     

     

    I started the project by getting a copy of Windows Server Essentials 2012 R2 and downloading Audiophile Optimizer. I prepared the hardware with a new SSD for the operating system and a couple 1TB SSDs for music storage. Installing 2012 R2 was pretty simple and nearly identical to installing any new Windows OS. On the other hand, installing and configuring Audiophile Optimizer is a different story. This software isn’t for those who can’t or won’t read the 52 page setup guide. Fortunately the setup guide is thorough and provides enough information for users to at least get their systems up and running.

     

    After installing AO and going through all its options a few times I got the hang of the program. One problem I ran into revolved around AO’s Core Mode. Core Mode pretty much turns the PC into an appliance with nothing but a command line or a playback application running. The issue I had was that it took about 30 minutes for AO to get the PC into core mode and another 30 minutes for AO to get the PC out of core mode. This is likely because the low horsepower of the computer I was using rather than a real issue with AO. For most people this may not be an issue because they will put their PCs into Core Mode and call it a day. However, I needed to go into Core Mode and come out of Core Mode several times while I tested different configurations and made software configurations changes that were only possible in AO’s GUI Mode. Again, this isn’t a big deal but readers should consider the speed of their hardware and try to complete all tasks (other than playing music) before setting their PCs into Core Mode.

     

     

    The final configuration of the PCs I setup was AO Ultimate Mode, Core Mode, and JRiver Media Center set as the Shell. In this config the PC booted right into JRMC in an appliance-like fashion.

     

     

    Once configured, the PCs were placed in a system with Berkeley Audio Design and Constellation Audio components, and Magico M Project loudspeakers. The sound I heard when the AO optimized PC was played for the first time, and throughout my stay, was stunning. We even had an identical computer setup, but running Windows 7 and without AO and its optimizations, for comparing sound quality. This enabled us to do A/B comparisons between the Windows 7 computer and fully optimized 2012 R2 / AO computer with the only difference being the optimization of software. The sonic differences were easily apparent. The optimized computer made it possible to hear the smallest details in recordings at incredibly low volume levels. For example, we had a microphone placed half way between the loudspeakers and the listening position. The microphone was connected to a small display that showed us the decibel level during our listening sessions. It wasn’t uncommon for the display to read volume levels around 60 dB (keep in mind that the listening position was at an even lower level) and for us to hear everything a recording had to offer. Turing up the volume didn’t increase our ability to hear details, it just made the music louder.

     

     

    We continued to listen to all kinds of music and continued to hear incredible sound quality. Everything from soundstage to reverb trails to the ambiance of a concert hall was improved with these software optimizations. On one track the drummer struck a bell and the sound seemed to hang in the air for ever while at the same time being completely distinct from the rest of the sounds. Every instrument had its place and its space. The listening experience was truly something to behold.

     

     

    I’m sure this experience will raise more questions than answers for many people. They will want such an optimized system compared to every server under the sun and I don’t blame them. Who wouldn’t want that information? However, we must be careful. Just because a system is great for one person doesn’t mean anything for someone else. The Windows Server 2012 R2 / Audiophile Optimizer system isn’t for the faint of heart. Depending on how far one optimizes the system, there may be no Ethernet connection and it may require a keyboard, mouse, and monitor for operation. Plus, installation is much more involved than purchasing a server like an Aurender and having one’s dealer visit to set it up and get everything running smooth.

     

     

    Based on my experience this week, the sound quality bar has been significantly raised.




    User Feedback

    Recommended Comments



    As we are discussing differences between computer related sources and asking ourselves how significant these differences are in fact... this is my experience, based on what I have or had:

     

    Source-1: Auralic Aries connected via Audioquest Diamond to Weiss DAC1-Mk3

    Source-2: PC (Win7, Jriver v20, foobar2k) connected via USB to BelCanto Reflink then via Weiss Chiron Glass-1 to Weiss DAC1-Mk3

    Let me add: Source-2 easily violates any hardware related rule for an audiophile PC: it's a watercooled very silent PC with a pump, lots of (slow) fans, fan controller unit... you see, it contains lots of stuff which should be avoided, though it is really well isolated by the Reflink and the Weiss Glass-1 from the DAC and the rest of the chain.

     

    Other equipment:

    Preamp: Ayre KX-5

    Amp: Pass Labs X250.8

    Speakers: Apogee Duatta Sig. (fully refurbished 2015)

     

    Comparing Sorce-1 with Source-2:

    if I fully concentrate, switch as much as I want between then sources, then I identify in most cases (ca. 10 out of 15) the Auralic Aries as the better source.

    If my wife activates one source and I enter the room, then it's... just a guess.

     

    Opposed to:

    - I had a much easier time to distinguish the Ayre preamp from my previous preamp (Headamp GSX-MK2) using Manger MSMc-1 active speakers.

    - I had hardly any difficulties to identify the differences between a Weiss DAC202 and the DAC1-Mk3.

    - Another magnitude of sonic differences exist between my old active speakers and Apogees driven by a good amp. These are just different worlds.

     

    IMO: the different approaches are interesting and one or the other can help to improve overall sound quality in the listening chain. I am sure, I would not be able to distinguish blindly between three or four high quality sources (Aries, CAPS/AO, Aurender)

    Share this comment


    Link to comment
    Share on other sites

    Even better would be a *blinded* music server shootout between (for example) Aurender ?N100, Auralic Aries, CAPS + AO. Could start with trials determining if the "obvious sonic differences" of PC + AO and other servers were confimed blinded. With Chris' experience I would be thrilled to see one of these solutions come out a clear winner *blinded*. If one came out a blinded winner, I am very likely to go down that solution.

     

    It's always possible to wish for the perfect trial and comparison. Unfortunately those conditions rarely exist in the real world, and certainly didn't exist in this instance. Time, money, and lack of availability of equipment almost always come into play.

     

    Chris was listening to a system with a DAC he is very familiar with, and I dare say he has heard Magicos before. In addition, he did do a comparison with 2 identical systems, one Running 2012/AO with optimizations, and one running just W7. So his basis for saying the AO system sounded better are highly credible.

     

    BTW, to the other poster, nowhere does the report say he was listening to an old fashioned computer; my guess is the situation was just the opposite.

    Share this comment


    Link to comment
    Share on other sites

    BTW, to the other poster, nowhere does the report say he was listening to an old fashioned computer; my guess is the situation was just the opposite.

     

     

    I meant a traditional computer motherboard etc. as opposed to something like an Aries or Aurender which is purposely designed for audio.

    Share this comment


    Link to comment
    Share on other sites

    If one is able to improve sound quality by shutting down processes in a computer OS, does that not mean that since a computer OS can never be pared down to the level of a CD player, that anyone really seeking the highest in sound quality would have to use a disc player to achieve it?

     

    Does that not mean that computers are inherently detrimental to sound quality?

    Share this comment


    Link to comment
    Share on other sites

    If one is able to improve sound quality by shutting down processes in a computer OS, does that not mean that since a computer OS can never be pared down to the level of a CD player, that anyone really seeking the highest in sound quality would have to use a disc player to achieve it?

     

    Your premise is incorrect. The CD player is electrically "noisy," and adds mechanical noise and vibration to that with the disc spinning mechanism.

     

    Does that not mean that computers are inherently detrimental to sound quality?

     

    Mechanical and electrical noise and vibration are, so the idea is to minimize the amounts of these going on in the DAC.

    Share this comment


    Link to comment
    Share on other sites

    If one is able to improve sound quality by shutting down processes in a computer OS, does that not mean that since a computer OS can never be pared down to the level of a CD player, that anyone really seeking the highest in sound quality would have to use a disc player to achieve it?

     

    Does that not mean that computers are inherently detrimental to sound quality?

     

    A physical motor is much simpler than a computer OS, yes. So is a lawnmotor motor. What is your point? Noise is not the byproduct of sophistication. AO, and other worthy attempts to reduce "unncessary" processes, are trying to reduce possible sources of noise produced by electricity, friction and heat, to name a few. But those sources begin as orders of magnitude less noisy than a physical motor spinning a plastic disc (or cutting a lawn). Just because the goal is to reduce irrelevant processes and prioritize the good ones doesn't equate to "get as simple as possible, use a motor".

    Share this comment


    Link to comment
    Share on other sites

    A physical motor is much simpler than a computer OS, yes. So is a lawnmotor motor. What is your point? Noise is not the byproduct of sophistication. AO, and other worthy attempts to reduce "unncessary" processes, are trying to reduce possible sources of noise produced by electricity, friction and heat, to name a few. But those sources begin as orders of magnitude less noisy than a physical motor spinning a plastic disc (or cutting a lawn). Just because the goal is to reduce irrelevant processes and prioritize the good ones doesn't equate to "get as simple as possible, use a motor".

     

    Given that, would it not mean that linux is then by far a superior platform? You can install a very barebones install of CentOS or one of the Raspberry Pi distros that have far fewer processes than Windows.

     

    And by that definition a "good" process creates just as much noise as an irrelevant one.

     

    In fact, since a Raspberry Pi only needs a 5V adapter to run, isn't it then the ultimate music playback device? Very barebones OS, very little current, very few onboard devices...

    Share this comment


    Link to comment
    Share on other sites

    Given that, would it not mean that linux is then by far a superior platform? You can install a very barebones install of CentOS or one of the Raspberry Pi distros that have far fewer processes than Windows.

     

    And by that definition a "good" process creates just as much noise as an irrelevant one.

     

    In fact, since a Raspberry Pi only needs a 5V adapter to run, isn't it then the ultimate music playback device? Very barebones OS, very little current, very few onboard devices...

     

    Yes, this is why Linux is the ultimate as you could even go down to kernel-level tweaks if you want to, and also why the HQ Player client-server mode with a smaller Network-attached audio device is a good solution too, as are well-designed dedicated players.

     

    The cPlay website has a lot of good info.

    Share this comment


    Link to comment
    Share on other sites

    Given that, would it not mean that linux is then by far a superior platform? You can install a very barebones install of CentOS or one of the Raspberry Pi distros that have far fewer processes than Windows.

     

    And by that definition a "good" process creates just as much noise as an irrelevant one.

     

    In fact, since a Raspberry Pi only needs a 5V adapter to run, isn't it then the ultimate music playback device? Very barebones OS, very little current, very few onboard devices...

    You must also take hardware into account. Linux is a great platform for customization and tweaking. However, devices such as the Raspberry Pi have subpar USB implementations that share the same physical chip as the Ethernet controller.

    Share this comment


    Link to comment
    Share on other sites

    Yes, this is why Linux is the ultimate as you could even go down to kernel-level tweaks if you want to, and also why the HQ Player client-server mode with a smaller Network-attached audio device is a good solution too, as are well-designed dedicated players.

     

    The cPlay website has a lot of good info.

     

    That's what I was thinking. If turning off of processes makes a big difference in sound quality, instead of spending $800 for Windows server and SO, you could install Linux for free and actually have even fewer processes running, have a system that's even more barebones.

    Share this comment


    Link to comment
    Share on other sites

    Again...there's a real danger in reaching single threaded conclusions. Barebones does not equal sound quality; it is only part. I had a Linux MPD Alix box, run off an SD card. The sq was....guess what, lightweight with no heft or balls. Likely had some thing to do with OS, parts, power supply reserves, etc.

    Share this comment


    Link to comment
    Share on other sites

    That's what I was thinking. If turning off of processes makes a big difference in sound quality, instead of spending $800 for Windows server and SO, you could install Linux for free and actually have even fewer processes running, have a system that's even more barebones.

     

    I've actually done this with both Linux and FreeBSD, going so far as to run the (minimal) OS and music player from the command line.

     

    Players like HQPlayer, XXHighEnd (which has facilities for minimizing Windows) and Audirvana Plus sound far better. It's not close. That's because there's far more to playback than minimizing hardware and software demands, though that plays a part.

    Share this comment


    Link to comment
    Share on other sites

    Given that, would it not mean that linux is then by far a superior platform? You can install a very barebones install of CentOS or one of the Raspberry Pi distros that have far fewer processes than Windows.

     

    And by that definition a "good" process creates just as much noise as an irrelevant one.

     

    In fact, since a Raspberry Pi only needs a 5V adapter to run, isn't it then the ultimate music playback device? Very barebones OS, very little current, very few onboard devices...

     

    Less processes certainly help, but the clock accuracy, USB implementation, and memory quality all play a role in SQ, with the result that simple Linux boxes can not compete in SQ with good full mobo's running WS2012 or Win 10

    Share this comment


    Link to comment
    Share on other sites

    Yes, this is why Linux is the ultimate as you could even go down to kernel-level tweaks if you want to, and also why the HQ Player client-server mode with a smaller Network-attached audio device is a good solution too, as are well-designed dedicated players.

     

    The cPlay website has a lot of good info.

     

    Well the ultimate... ok... Have actually done much of what you have mentioned. Linux on different platforms and a number of devices and have not achieved near the sound quality I have with

    Server 2012 R2, AO and JPlay 6.1 Streamer.

    Share this comment


    Link to comment
    Share on other sites

    Less processes certainly help, but the clock accuracy, USB implementation, and memory quality all play a role in SQ, with the result that simple Linux boxes can not compete in SQ with good full mobo's running WS2012 or Win 10

     

    So why is it that Auralic, Aurender and others who make custom CPU boards for their products haven't figured out how to make a renderer/server that sounds better than a Windows system?

    Share this comment


    Link to comment
    Share on other sites

    Chris, Ive read enough pro and cons to give this a go. Will you tell us what hardware was used or recommend something to run this on?

    Thanx

    Share this comment


    Link to comment
    Share on other sites

    That's what I was thinking. If turning off of processes makes a big difference in sound quality, instead of spending $800 for Windows server and SO, you could install Linux for free and actually have even fewer processes running, have a system that's even more barebones.

     

    And that is why some enjoy a NAS (linux) through optical network to a streamer/DAC combo. No USB, no SPDIF, no Mac, no Win. But I guess I have wandered off topic.

    Share this comment


    Link to comment
    Share on other sites

    Or will optimiser for Windows Nano server be out next week?

    Share this comment


    Link to comment
    Share on other sites

    Less processes certainly help, but the clock accuracy, USB implementation, and memory quality all play a role in SQ, with the result that simple Linux boxes can not compete in SQ with good full mobo's running WS2012 or Win 10

     

    But you can take that same motherboard and instead of running Windows on it, run Linux. So not the clock, USB, and memory are all the same. You have a barebones OS with fewer processes than a Windows box can ever have.

     

    If sound quality is greatly improved by limiting processes, I'm wondering why Linux on the same hardware wouldn't always sound better.

    Share this comment


    Link to comment
    Share on other sites

    But you can take that same motherboard and instead of running Windows on it, run Linux. So not the clock, USB, and memory are all the same. You have a barebones OS with fewer processes than a Windows box can ever have.

     

    If sound quality is greatly improved by limiting processes, I'm wondering why Linux on the same hardware wouldn't always sound better.

     

    In that example., yes, it's a good candidate for Linux. However, Linux brings its own hurdles, such as lack of DAC drivers (if DAC is not UAC2 compatible), lack of users Linux experience (and I'm not talking expertise, just understanding and overall familiarity with simple troubleshooting), lack of real world guides like those that come with AO. Linux still, unfortunately, remains very much a DIY or serious hobbyist platform for many of us. But yes, it's ability to run a dedicated audio OS is pretty beneficial. Aurender is still the best Linux platform I've heard/owned (Alix, Aries, Auraliti...and other A words).

     

    Net/net, you can clearly skin this music server cat a thousand ways, with Windows optimized OS's, with Linux kernels, with OSX mods, with NASes running Minimserver, with dedicated (usually some form of Linux) Aurender or other preconfig'd server products.....etc. If there was one best way, we'd have nothing to debate about. :)

    Share this comment


    Link to comment
    Share on other sites

    But you can take that same motherboard and instead of running Windows on it, run Linux. So not the clock, USB, and memory are all the same. You have a barebones OS with fewer processes than a Windows box can ever have.

     

    If sound quality is greatly improved by limiting processes, I'm wondering why Linux on the same hardware wouldn't always sound better.

     

    limiting Process, disabling services, etc. helps a lot to avoid dpc spikes. It helps to keep latency down etc, helps to prevent hickups and dropouts and all that.

     

    what really matters for SQ are many other things... I told and wrote many times, all the disabling of services etc. is only a VERY small percentage of what AO does for a system.

    Share this comment


    Link to comment
    Share on other sites

    But you can take that same motherboard and instead of running Windows on it, run Linux. So not the clock, USB, and memory are all the same. You have a barebones OS with fewer processes than a Windows box can ever have.

     

    If sound quality is greatly improved by limiting processes, I'm wondering why Linux on the same hardware wouldn't always sound better.

     

    Since folks have already answered this same question several times, rather than repeating the answer yet again I'll just say go for it, and I hope you achieve a system that brings you enjoyment.

    Share this comment


    Link to comment
    Share on other sites

    I am using a dual streamer setup with PPA components/cables, JPlay, AudiophileOptimizer and Tidal for over 2,5 years now. Paired with a dCS Paganini it is on par with the top of disc players. I could imagine having this hobby without these streamers. One thing I really like is the fact that every now and then there are updates for JPlay or AO. The improvements in SQ is like getting free new hardware every now and then.

    Share this comment


    Link to comment
    Share on other sites

    Nobody remembers this huh? Call me very skeptical (and Chris, it is not clear to me you actually level matched when you did an A/B, perhaps you did but it isn't clear).

    Share this comment


    Link to comment
    Share on other sites




    Create an account or sign in to comment

    You need to be a member in order to leave a comment

    Create an account

    Sign up for a new account in our community. It's easy!

    Register a new account

    Sign in

    Already have an account? Sign in here.

    Sign In Now




×
×
  • Create New...