Jump to content
  • The Computer Audiophile
    The Computer Audiophile

    The Computer Is Back, But Really Never Left

     

    Audio: Listen to this article.

     

     

    When collecting my thoughts for this article, it dawned on me that I've been using a computer for my audio needs much more than I did in the previous couple years. By computer, I mean something with a mouse, monitor, keyboard rather than a canned music server. Both styles are really computers, but canned music servers are usually not called computers because of the connotation that comes with a computer. No worries, perception is important and I completely understand. 

     

    It feels like the early days of computer audio here at Audiophile Style headquarters. The flexibility, power, and features found in computers, versus canned music servers, are unbeatable. The most critical item in that list is features (keep in mind that sound quality must be there, or none of this matters). When computer audio started, there weren't any canned music severs that could do anything close to what a computer could do. Then high end companies caught up and released some fantastic products that not only had the features, but also simplicity, sound quality, and great customer support. 


    m2_pro_memory__ey38zvrqkfiq_large_2x.jpgI've always used a little bit of everything, but, for the most part, prefer to use a ready-made solution that people can purchase. This coming from the guy who built CAPS Twenty may sound a bit comical, but I built that computer to fill a void in the market. Right now, there are additional voids that need filling, so I'm using my computers much more than in years past. Among the features I just can't get in a high end music server, but would love to have, are support for 12 channel immersive audio, convolution and enough power to run it at 24/352.8 with hundreds of thousands of filter taps, and power for applications such as HQPlayer upsampling to the limits of today's DACs. 

     

    Aurender supports Ravenna, and has enabled 12 channel support on my ACS10. It's really nice to have everything that comes with an Aurender, and this high channel count, but I still need to pipe the audio through convolution elsewhere because the system needs frequency correction and time alignment. No immersive system can be optimized without this capability. I say no system, but I suppose a room could be built where all 12 speakers were placed exactly at the same distance from the listening position and the dimensions of the room were perfect for audio playback. Until such room exists, I'll stick with my statement. 

     

    It may seem like I'm a bit down on high end music servers at the moment, but that's far from true. I look at what needs to be done in addition to what's been done in the past, and I think the future is really bright. When the last hole in the market was filled, high end server companies created, and continue to create, some truly stunning components. I hope the current void can be filled by some of the same companies because the solutions would take my system to new heights. 

     

    I'm focused on the computer aspect of the industry right now, but there are other voids that need filling. Perhaps the topic of another discussion at a later date. A twelve to sixteen channel DAC, with volume control, is needed to compete with or surpass the DACs offered by companies such as Merging Technologies. The current crop of multichannel high end DACs stops at eight channels. That's enough for a 5.1.2 Atmos music system, but not for 7.1.4 or 9.1.6. 

     

    While I'm on this topic, I'll briefly touch on a question I was asked this week.

     

    "If you could create or have a single audio device that you believe the market wants or needs, what would it be?

     

    I love questions like that. They allow for creativity, evaluating needs and wants, and the possibility of improving the music listening experience. Answering this one was easy for me. It's a product I’ve wanted for a few years, and a product that tens of thousands of audiophiles can use, even if they don't know it. Not that I no better than tens of thousands of audiophiles, but tens of thousands of audiophiles are just unaware of the possibilities. 

     

    At the most rudimentary level, the single audio device I want is a DSP box that sits between a music server or streamer, and a DAC. This box would have all the digital inputs of a DAC or streamer, including UPnP, and all the outputs needed to feed a DAC. Internally, it would be capable of running convolution and/or upsampling. 

     

    Ideally this box would have a simple web interface to upload convolution filters to it. My preference would be for it to run @mitchco Mitch Barnett's Hang Loose Convolver, but technically it could run Dirac, Focus Fidelity, Home Audio Fidelity's Room Shaper, etc... The device could be similar to a dCS Upsampler, but with the power and in/outputs for many more channels and convolution. 

     

    Because there is not a single best filter or target curve (Harman curve, EBU curves, Atmos curve, etc…), user selectable filters that switch immediately would also be part of this component, including the ability to disable filters for certain music one may want to hear without correction. A box like this would work with existing music servers, just like they work with a DAC, either via USB, AES, UPnP, etc...

     


    MacBook Pro M2 Pro

     

    spatial_audio__dipr8iji32uu_large_2x.jpgI purchased a new MacBook Pro with 12 core M2 Pro chip, 1TB of storage and 16GB of RAM, last week. This is the best computer I've ever owned, by a long shot. I use it mainly for music playback and music library curation. Using my previous Intel Core i7 MacBook Pro, I could play 12 channel 24/352.8 discrete immersive audio using Audirvana. However, when I tried to play it using convolution, the laptop's fan went into space shuttle mode and the audio suffered from sever dropouts. On my new MacBook Pro I can play 12 channel DXD with 65,000 tap convolution filters, at the same time as doing whatever else I want, and the fan doesn't even turn on. The audio is smooth as silk. 

     

    Adding this MacBook Pro to my Thunderbolt dock, previously written about here, is a simple and perfect solution. I can't recommend it highly enough. A single cable with power, 10GbE, 1GbE for Ravenna, and room to attach any drive I may need.

     

     


    Pearl Jam's Yield Goes Immersive

     

    Today I awoke to the news that Pearl Jam's 1998 album Yield was released in Atmos. This is one of my favorite Pearl Jam album because it features drummer jack Irons. Jack's style is unmistakable and sounds glorious in Atmos. I've had the album on repeat since I returned home from dropping my daughter off at school. 

     

    I can't find who mixed Yield for the immersive mix because this information can be ridiculously hard to find. Josh Evans mixed the previous PJ albums for Atmos, and I wouldn't be surprised if he is responsible for this terrific mix. 

     

    The tracks use all the Atmos channels in a way that serves the music perfectly. It's adventurous at times and conservative when it needs to be. The soundstage is terrific. I can perfectly place guitarists Mike McCready and Stone Gossard, and bassist Jeff Ament as they come and go on each track. This placement is mostly in the 180 degrees in front of the listening position, but extends to other areas when it "works." 

     

    I have a feeling my weekend will be full of immersive Pearl Jam, and I love it :~)

     

     

    Pearl Jam Yield Immersive.jpg




    User Feedback

    Recommended Comments

    I use my year old or so MacBook with the M1 Pro chip as my music server and get great results. I have 16 go ram and a 512 SSD internal drive. I use a USB C external 4 and 5 TB drives. I am sure the data speeds of USB C helps, but the SQ of this laptop smokes my old Mac. It helps also that I found a spiffy DH Labs Silver Sonic Mirage USB C to B cable to run into my LIM which yes does sound appreciably better than my old USB cable connected via the Apple sourced USB C to A dongle. I hope to get a new laptop when the M3 Quantum chip is out. 😎

    Share this comment


    Link to comment
    Share on other sites

    4 minutes ago, SQFIRST said:

    The intermediate device sounds great. Wasn't sure if you implied this, if so I didn't catch it, but how about having a connection for a calibrated microphone input to build the convolution filters? 

    I hear ya. Baby steps :~)

    Share this comment


    Link to comment
    Share on other sites

    I want an IP based published non-proprietary standard protocol for transferring "music, video, audio."  We can then toss out USB, SPDiF, TOSLINK, AES, etc.  This is the enabler for your box in the middle.  OR, it is the elimination of it because a software-based solution should be able to run on the computing platform of choice.

    As @The Computer Audiophile listed above, a standard way of locally and remotely controlling volume with ONE volume control in the chain.

     

    I wanted to talk about clocking between devices, but I am unsure if we understand where and how that should be done.  Nonetheless, the "DAC" needs to be comfy with the data being fed.

     

     

     

    Share this comment


    Link to comment
    Share on other sites

    3 minutes ago, bobfa said:

    I want an IP based published non-proprietary standard protocol for transferring "music, video, audio."  We can then toss out USB, SPDiF, TOSLINK, AES, etc.  This is the enabler for your box in the middle.  OR, it is the elimination of it because a software-based solution should be able to run on the computing platform of choice.

    As @The Computer Audiophile listed above, a standard way of locally and remotely controlling volume with ONE volume control in the chain.

     

    I wanted to talk about clocking between devices, but I am unsure if we understand where and how that should be done.  Nonetheless, the "DAC" needs to be comfy with the data being fed.

     

     

     

    AES67 is a good start, but is pretty limited. That’s why Ravenna was created as a superset of AES67. 
     

    I get it though, everything gets much easier with a network interface routing it all. 

    Share this comment


    Link to comment
    Share on other sites

    The choice of a complete server in a box versus a computer is a difficult one.

     

    I've looked at JCAT, Innuous, Antipodes, Wolf and they all are great machines, when it comes to size of hard drives for storage though, they come up very short. 

    For library that's now a modest 8TB, it means to have external drives connected by USB. Over the years, portable external drives are bad news for servers that are on all the time. The drives get hot, there's poor ventilation, and the controllers die far too soon. 

     

    To move data from storage to a media server isn't my cup of tea, as choice is music is spontaneous, if an album is played, I might play half, then switch to similar genre or totally different, then having to wait for the copy files to some through ruins the flow.

     

    So for now, music is stored on a Windows LTSC XEON platform , with two 16TB drives to arrive this week, one is the mirror of the other, not RAID, just a backup NTFS volume. Backups are separate again. I've tried Linux, if you need to search with 'how to' so many times, and going nowhere, it's not for me.  

    That server has audirvana, controlled by the remote app on an iPad. JPLAY for iOS is good, but Audirvana sounds better and update time of new music is very fast.

     

    I thought about NAS often, I'm used to the way Windows works, and more or less made a NAS that's more of a server anyway. Power consumption is pretty low and the size of drives is not restricted. The server lives in the garage, it's not that loud, but it's in a better spot than the listening room. The drives are cooled with the slow speed case fans operating, some say it causes EMI/vibration, there's DX filters and JCAT NET XE to clean up electrical noise. For vibration, it's more important in the analog world to keep this to a minimum, in a computer, there's far more electrical noise to worry about.

     

    SQ wise sounds very good on a modest office secondary system, even better with renovations to start in March on the new listening room, can listen again with 'real' audio equipment again. These last few months are very much trial and error, more errors than success, but that's half the fun with audio.

    Share this comment


    Link to comment
    Share on other sites

    4 hours ago, One and a half said:

    I thought about NAS often, I'm used to the way Windows works, and more or less made a NAS that's more of a server anyway. Power consumption is pretty low and the size of drives is not restricted. The server lives in the garage, it's not that loud, but it's in a better spot than the listening room. The drives are cooled with the slow speed case fans operating, some say it causes EMI/vibration, there's DX filters and JCAT NET XE to clean up electrical noise. For vibration, it's more important in the analog world to keep this to a minimum, in a computer, there's far more electrical noise to worry about.

    What do you see as a downside to this option?

    Share this comment


    Link to comment
    Share on other sites

    Quote

    ...but I still need to pipe the audio through convolution elsewhere because the system needs frequency correction and time alignment. No immersive system can be optimized without this capability. I say no system, but I suppose a room could be built where all 12 speakers were placed exactly at the same distance from the listening position and the dimensions of the room were perfect for audio playback. Until such room exists, I'll stick with my statement.

     

    Do you know Blackbird Studio C? :)

    https://theblackbirdacademy.com/campus/studios/studio-c/

    I believe there are a few more rooms that have been professionally treated to a level where digital room correction can't improve the sound any further. However, there are millions of people who don't have the opportunity to listen in such a room and still want the best sound quality they can afford, and I dare say all of them can benefit from DRC. So I totally agree with you that there's a void in today's market. It seems like DRC is used by some casual listeners when it's integrated in an all-in-one speaker like the new Apple HomePod and by a few enthusiasts like the folks round here. But most music lovers don't even know what they're missing.

    I have an idea that could fill this void right away: What if DRC processing happened right at the streaming service provider? Imagine you could upload a convolution file or Dirac filter or whatever in your Qobuz settings and configure it the way that every time you stream with your Aurender, the filter gets applied. Everyone would be free to use any hardware. I'd be totally ready to pay 50% more for my subscription if it had this feature, but I'm afraid it'll be unlikely to become a reality, let alone for multichannel... Anyway, I just wanted to share my thoughts inspired by your reasoning.

    Share this comment


    Link to comment
    Share on other sites

    11 minutes ago, mForMusic said:

    It looks like a great studio for Atmos. Really cool design. However, I think they still need DSP for time alignment of the loudspeakers. If the speakers were in a circle they might have a chance at being equidistant from the listening position, but the shape of the room and placement looks like some channels are farther away than others. Hard to tell for sure with the photos though.

    Share this comment


    Link to comment
    Share on other sites

    I think you're looking for a DEQX box, probably one or two HD-Active Preamp Processors

     DEQX Products

     

    These do digital crossover, driver correction, time alignment and room correction.

    Share this comment


    Link to comment
    Share on other sites

    17 minutes ago, billg said:

    I think you're looking for a DEQX box, probably one or two HD-Active Preamp Processors

     DEQX Products

     

    These do digital crossover, driver correction, times alignment and room correction.

    Hi @billg thanks for the link to DEQX. These look extremely limited in capability, but very easy to use for a consumer without any help. I hadn't looked at DEQX in a while, but just took a look at the specifications available. Perhaps I'm not looking at this correctly, but it appears that the EQ only has 7 + 3 bands. To quote from my review of my immersive system, with respect to the type of filters I want in a component and the type of filters I currently use, "65.536 filter taps which is the equivalent to a graphic equalizer with 32,768 eq sliders. That is 1000 times the frequency resolution compared to a 31 band graphic equalizer."

     

    I don't believe the DEQX products are powerful enough to do anything close to what I'd like. I could be wrong though and would be happy to be corrected. It also appears that their DSP is fixed at 96 kHz, no matter the input signal, and the digital output is 96 kHz on coaxial S/PDIF only. A bit underwhelming, but very easy I'm sure.

     

     

    Screenshot 2023-03-03 at 2.46.08 PM.png

    Share this comment


    Link to comment
    Share on other sites

    Great retrospective and overview. I remember building the CAPS 2 back in the day. Things have moved on a bit since then and it's an evolving field, despite the laws of electronics and acoustics remaining the same. Debates re. high res remain. Physical room treatment and DSP to correct any remaining aberrations are definitely worth exploring. There are a lot of 2-channel purists, and I've not ventured into more channels yet. I wonder if multichannel audio is a better solution? 

     

    I'd agree with renaming the site to "Computer Audiophile". That is after all what the core of the site is about, and where its roots are. Having "style" in title suggests form over substance; aesthetic over function; superficial focus on current trends as opposed to underlying principles and ultimately audio quality and enjoyment. 

    Share this comment


    Link to comment
    Share on other sites

    Amen to this, @The Computer Audiophile!

     

    This is one of my favorite Pearl Jam album because it features drummer jack Irons. Jack's style is unmistakable and sounds glorious in Atmos. I've had the album on repeat since I returned home from dropping my daughter off at school.”

    Share this comment


    Link to comment
    Share on other sites

    I just run roon on a M1 MacBook Air with 8gb ram.  You can run convolution filters with this. I think I saw people even use Dirac with Roon. 

     

    But I'm more into headphones so I didn't look much into that yet.

     

    HQPlayer will be my next meaningful upgrade that I'm going to do.

     

    There are plenty of nice streamers with DACs out there that work as roon endpoint.. some support NAA. (Gustard A26 for example).

     

     

     

    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...