I am new to PC audio so need to pick the best brains on here. It seems that there are 2 different ways of approaching it - METHOD 1 is to put your resource into the PC itself then output via USB to SPDIF converter to DAC or METHOD 2 use network-based player such as Squeezebox feeding DAC, with basic PC used to control it. What is the best approach for best audio ??? I am sure they both have their advantages/disadvantages but I have yet to hear any real expert opinion on which sounds best (for similar budget). What do you reckon, guys ???
Consider the Squeezebox, e.g. SB Touch, a PC running Linux with
special focus on Audio. It's a PC and high quality audio interface
in one box.
It'll take quite some effort to achieve the SB Touch performance on a comparable audio PC.
It's much easier to tweak the SB. E.g it requires just a little linear 5V supply upgrade or you can try my toolbox.
Try that on a PC.
Here you'll find some useful information what to to with a normal PC: Cics CMP2
The real problem is that the audio interfaces (external DACs) can't cope with source originated jitter and noise.
If all those designers would do a better job, we should be able to hook up any bit-perfect stream from
any digital source without facing source related distortions and differencs.
Slowly but surely manufaturers start taking measures. E.g. galvanic isolation, multi-stage de-jittering, intelligent reclocking.
However. IMO one reason why there is a lack of movement in the market: There are only certain manufacturers who rule
the market with their propriatery licensed interface SW. This you'll find implemented in e.g. many USB audio interfaces.
You bet those suppliers are not interested to move on as long as they can sell their license.
special focus on Audio. It's a PC and high quality audio interface
in one box.
It'll take quite some effort to achieve the SB Touch performance on a comparable audio PC.
It's much easier to tweak the SB. E.g it requires just a little linear 5V supply upgrade or you can try my toolbox.
Try that on a PC.
Here you'll find some useful information what to to with a normal PC: Cics CMP2
The real problem is that the audio interfaces (external DACs) can't cope with source originated jitter and noise.
If all those designers would do a better job, we should be able to hook up any bit-perfect stream from
any digital source without facing source related distortions and differencs.
Slowly but surely manufaturers start taking measures. E.g. galvanic isolation, multi-stage de-jittering, intelligent reclocking.
However. IMO one reason why there is a lack of movement in the market: There are only certain manufacturers who rule
the market with their propriatery licensed interface SW. This you'll find implemented in e.g. many USB audio interfaces.
You bet those suppliers are not interested to move on as long as they can sell their license.
Last edited:
I went PC because, in addition to acting as music server, it also functions as crossover, equalizer, and ambiophonics decoder - try that with a squeezebox. 😉
Squeezebox does equalizer and ambiophonics decoder using Inguz DRC plugin. as for the crossover you can alway try minidsp 2*8. If you wanna go PC route please knock yourself out, but in the end you'll come back to non-PC based solution, and I am talking from experience here.
When you talk about the audio quality achievable with a PC, do you mean an arrangement where SPDIF is taken from a PC internal soundcard and fed to an outside DAC? If this is your comparison, then yes, I tend to agree.
However, if you take a PC USB-feed to a well implemented, galvanically isolated, asynchronous converter with an independent, low-jitter master clock, then this is a different story. Such a scheme should have a potential at least equivalent to the SB-Touch. Any experience based on actual comparison would be very helpful.
Well my problem was that my PC was connected to the same power outlet as my audio equipment and I can hear it on my speaker. No audio cable connected and the PC was not running. The only remedy is to run an extension from a different outlet.
My setup was running a JRiver MC with touch screen Acer monitor to mimic Meridian sooloos system. It was a 4 way active setup with 3 way floor stander and stereo sub. Everything was passive i.e. no fans and Lynx card. I invested around $2k on the setup.
The good thing is that I re-used the PC as a Media Center in other room.
Now I am running SqueezeBox Transporter with DRC and couldn't be happier crossover duties are with MiniDSP at the moment but I will be moving to a passive speaker system and removing MiniDSP from the chain.
For $2000 you can get DEQX Express and Tact (Used). There are so many options out there that it is really not worth to take the PC route, and explain your spouse how everything works in the end.
Best option right now is to get a MiniDSP 2x8 apply room correction and cross over filters using biquads and call it a day.
My setup was running a JRiver MC with touch screen Acer monitor to mimic Meridian sooloos system. It was a 4 way active setup with 3 way floor stander and stereo sub. Everything was passive i.e. no fans and Lynx card. I invested around $2k on the setup.
The good thing is that I re-used the PC as a Media Center in other room.
Now I am running SqueezeBox Transporter with DRC and couldn't be happier crossover duties are with MiniDSP at the moment but I will be moving to a passive speaker system and removing MiniDSP from the chain.
For $2000 you can get DEQX Express and Tact (Used). There are so many options out there that it is really not worth to take the PC route, and explain your spouse how everything works in the end.
Best option right now is to get a MiniDSP 2x8 apply room correction and cross over filters using biquads and call it a day.
I tried a few things and finally went with a PC and soundcard. Just a cheap reconditioned Dell I got on eBay for about $90. I did add a USB hard drive and a sound card that has optical out (and a SPDIF header).
JRiver is the player and I can control that from any computer in the house with a choice of interfaces. The music files could reside anywhere, but it's convenient to have them plugged into the Media PC. A mirror is on my main PC where I do the ripping.
A great system for a low price.
JRiver is the player and I can control that from any computer in the house with a choice of interfaces. The music files could reside anywhere, but it's convenient to have them plugged into the Media PC. A mirror is on my main PC where I do the ripping.
A great system for a low price.
I'm talking about any PC based audio chain incl. Squeezeboxes - Touch in particular , which are nothing else than special purpose computers.
I havn't found any audio interface yet which, would be immune against distortions caused by those computers.
If it is SPDIF or USB or internal cards or network streamers, it doesn't really matter, they all pick up distortions. And not always through the obvious front door.
Don't get me wrong. I'm not saying that the todays achievable audio quality from a computer based audio system is bad by the way.
I do believe or better know that nowadays a lot more people have a chance to enter audiophile spheres at a fragment of budget as it used to be.
It's very hard to compare interfaces such as "USB" vs. "SPDIF" alone. A lot of other factors need to be looked at. USB interfaces can be implemented in
a thousand different ways. Add another thousand ways of implementing SPDIF, recklocking schemes, isolations, powering, clocking, board layouts, asf asf.
I used to run an isolated USB DAC and heavily tweaked PC. Now I'm
fiddling around with a Squeezebox Touch and SPDIF.
The SBT I consider - without applying modifcations - as poor as any other basic PC based Transport.
It's even worse. The network has quite an audible impact on its performance.
You don't have that problem on a standlone PC based machine. With these you start up your RAM player and you're set.
However. With a bit of tweaking the situation can change radically - on the SBT at least.
And - forget those marketing phrases about asynchronous USB DACs and master clocks.
Those asynch USB DACs pick up jitter as any other DAC, it's the amount which is usually slightly less. You find a million posts about how great
Pure Music/Amarra works and that people experience improvements on their top end asynchronous audiophile USB-DACs with slightest changes made to that player software or operating system or HW of that transport.
And that's what the marketing message of async USB DACS actually promisises not to happen. Forget it.
The key subjects to be looked at in all these discussions are IMO the
audio-interface receiver stage, it's reclocking or jitter suppression
scheme, avoidance of intermodulations, it's power supply and isolation.
The key question really is - How well does the interface section isolates from the source and how well does it feed the actual DAC.
The EXAU2I guys with their quite nice propriatary (and Windows only🙄) multichannel asynch USB interface implemented I2S isolation as you might know. Great approach.
However. That isolation chip itself was and probably still is causing 250ps jitter (under lab conditions) on the I2S bus. That's not what I call high-end.
Though there is no free lunch. At one point you have to go for the better compromise , the lesser of two evils. And at that decision point jitter defeated noise, which is fair enough. Let's hope they get this resolved with Rev 2.
FPGAs and PICs on the actual audio boards are also pretty tricky to handle. These devices are usually sources for different distortions and can interfere with the sound critical parts.
And not to forget the SW. You wouldn't imagine how complex the whole SW
handling (OS/drivers/player/firmware) is. I read about a pretty generic problem, where quite some SW-players were doing faulty float2integer conversions - a very basic subject causing non-transparent bit streams.
I'll leave it for now. Gotta run.
I can just advise you - watch out. It's getting more an more complex to really see what's going on out there. Meanwhile I do know
quite some people who are piling up dead audio interface bodies.
That we should avoid.
And one thing I experienced. If you think your systems peforms etremely
well at ony point. Don't touch it anymore. The next SW update can easily change that situation - usually into the wrong direction.
Hi Soundcheck
Thanks for your detailed reply. Yes, I fully agree that this is very complex, each part done by different people, difficul to impossible for anyone to have a total overview and knowledge about all the parts involved.
And yes, I was thinking about an EXAU2I feeding a Buffalo DAC. Music would be stored on the PC, no need to to even connect to the Internet. The PC would be fully silent (except the HD) and control would be via a large touchscreen. Actual ripping would be done on my office PC to an external HD, from which the files would be transferred to the music-PC in the living room. The external HD would also serve as a backup. So this would not even involve a network. But as you say, there are a zillion different possibilities. The really nice thing about your solution is that you can control it via an iPad. And I could imagine that you have a lot less interference and noise in a SB-Touch than in a PC. Even though the SB-Touch really is a PC, its functionality is reduced to the bare minimum required, and this helps to cut noise, etc.
With the SB-Touch, I don't see how I could avoid a wireless network, and that worries me a bit. So I`m not sure at this point which way I will finally go.
Thanks for your detailed reply. Yes, I fully agree that this is very complex, each part done by different people, difficul to impossible for anyone to have a total overview and knowledge about all the parts involved.
And yes, I was thinking about an EXAU2I feeding a Buffalo DAC. Music would be stored on the PC, no need to to even connect to the Internet. The PC would be fully silent (except the HD) and control would be via a large touchscreen. Actual ripping would be done on my office PC to an external HD, from which the files would be transferred to the music-PC in the living room. The external HD would also serve as a backup. So this would not even involve a network. But as you say, there are a zillion different possibilities. The really nice thing about your solution is that you can control it via an iPad. And I could imagine that you have a lot less interference and noise in a SB-Touch than in a PC. Even though the SB-Touch really is a PC, its functionality is reduced to the bare minimum required, and this helps to cut noise, etc.
With the SB-Touch, I don't see how I could avoid a wireless network, and that worries me a bit. So I`m not sure at this point which way I will finally go.
I ended up doing both, sort of. I couldn't come up with a PC playback interface that worked well headless via remote control. So, I feed the spdif output from the squeezebox into the PC which then does the xover/eq and then outputs via a Steinberg MR816.
I use a hardware SRC unit slaved to the word clock of the MR816 so that I don't end up using a PLL recovered clock for output. Not entirely ideal, but I 'need' to support at least 44, 48 and 88 rates (and probably 96 eventually as well), it was more practical to run the PC at a fixed rate. (88k for SACD converted to PCM via an Oppo)
IMHO for all the annoyances of the server software, the squeezebox is a pretty darn nice setup. Despite being a long-time PC-based audio guy, I'd find it pretty much impossible to justify a PC based system over the squeezebox if you don't need xover capability, and as it turns out maybe even if you do 🙂
Potentially interesting options, albeit with a different set of trade-offs from a PC-based system. I'd be willing to bet that my current system with the MR816 at least equals and likely exceeds those, and ran about $1k for the pc hardware and interface. It still leaves the dsp/filter generation process though, which is where a lot of problems can crop up. I'm pretty happy with where I ended up - 'external' sources and a single-purpose headless (more or less) PC - although that's definitely not where I thought I was going to start with.aazeez1975 said:
Not in the least. I run a pro soundcard (m-audio 1010LT)in asio mode for input of two Turntables, a bluray player, input from the satellite receiver, streaming from the net etc.; dsp processors in a foobar vst wrapper (equalizer, ambiophonics DSP - the Moscovitz version and the Hotto program for TV and analogue input signal) and I find the sound quality comparable to a standalone cd player (I had the micromega stage 2 and a pioneer 59AVI before I went the PC route).
I transfer and listen to Lps with my unit, record streaming video and rip Cd's.
I have ripped > 1TB in flac by now, and I use foobar Upnp as my server for other machines in my home network.
I run SPDIF from the soundcard in 48kHz to Kef 104/2 with A 2.100 hypex amps.
I would not even consider going back to a single purpose unit.
audio kraut, why would you bypass m-audio DA converters? they don't suck. Certainly not the best, but better than a consumer product which is likely to be 1bit delta sigma or cheap 24bit (especially the onboard clock) and cheap, noisy preamps. . . .BTW I use a digidesign 192 I/O 16 channel with an external Lucid clock (that is damn near the Best available at $7000) in the Studio I work in. . .if you are converting 44.1 to 48k to stream your output your are potentially adding error as you are sample converting to an odd numbered sample rate...most audio you did not personally create is going to be 44.1 sampled...I record at 88.2 so I get a smooth, easy sample conversion to 44.1 (they are multiples of each other) for CD audio....48k-96k (multiples again) are for DVD output and 192 is a pointless format that is intended yet again to be outputted to DVD formats, it is a multiple of 96k . . . I would have all your DA conversion done in your interface. . .
Last edited:
Hillbear neither of your methods considered are normal or standard for professional recording engineers and recording musicians. That is all that I am
familiar with as I am a recording engineer.
In the studio Analog goes into the "Interface" and digital to the computer via USB, Firewire and Proprietary systems used by Avid/Digidesign (Protools HD Systems).....we do all our processing and storage in the computer and use the interface for DA conversion and preamp duty. . . .
In the studio I use Protools HD3 which is a system that uses two Accel cards and a controller card (6 TDM DSP processors) in an apple pro with 8 Xeon processors. So I have 14 processing cores handling my audio and DAW...then it is sent to a Digidesign 192 I/O with an external Lucid Clock . . .this is the best way to handle audio ( PC versions available ) .. . .now you are not a recording engineer. . .
I will tell you what I do at home for entertainment and mastering my mix downs from the studio:
Anything computer based stays there. mp3/ internet streams/ DVD movies. ( I don't own a TV or DVD player ). . . I output everything via USB to my interface ...a Line 6 UX2 (I will only use the ASIO Drivers ) . . .which handles all my clocking, AD/DA conversion and preamp/headphone...this goes to my premium home stereo on one of the line inputs. I could use the preamp outputs and go straight to my amp. How ever I listen to Vinyl, tape, FM and have a ultra premium Yamaha CD player so my computer based audio goes to my Sansui c2101 preamp. That is the best way to run a computer based system...get yourself a professional interface ( be careful the cheap stuff is cheap ) I recommend Presonus firewire based FireStudio and Emu 0404USB. . . there is not reason to use spdif for anything other than using your receivers DA converters or as an input ( or DAT recording ). . . beware of clocks. there is a good outboard ART product that is only $100 and the EMU has a good clock...but anything capable of using outboard clocks already has good clocks in it. External clocking is for studio recording/monitoring when we break out $2000 clocks.
And with a pro product you get input sampling up 96k or 192k...once a recording is made the sampling is set and does not concern you as long as your gear can handle the format.but nothing is being sold that is not a minimum of 24bit/48k (DVD minimum) . . .JetPLL sounds like it might be marketing hype...somewhere I saw that it is noise shaping...not actually a better clock...noise shaping is great (MASH and Yamaha Pro-S ? ? ?)....
All this other stuff I read about on hear is unfamiliar to me and questionable...hub on your PC, a single output to your "interface"
And hey with a pro interface you get instrument inputs, mic inputs, line inputs, and headphone outputs that do not have to go through other gear....
keep it simple and clean and you will be happy . . BTW my home system is PC/Vista Premium . . there is nothing Apple only out there except Apple software
Logic/Garage Band. I crash the studio's $21,000 Apple daily...it is a turd- but I push it hard.
later,
~e`~
familiar with as I am a recording engineer.
In the studio Analog goes into the "Interface" and digital to the computer via USB, Firewire and Proprietary systems used by Avid/Digidesign (Protools HD Systems).....we do all our processing and storage in the computer and use the interface for DA conversion and preamp duty. . . .
In the studio I use Protools HD3 which is a system that uses two Accel cards and a controller card (6 TDM DSP processors) in an apple pro with 8 Xeon processors. So I have 14 processing cores handling my audio and DAW...then it is sent to a Digidesign 192 I/O with an external Lucid Clock . . .this is the best way to handle audio ( PC versions available ) .. . .now you are not a recording engineer. . .
I will tell you what I do at home for entertainment and mastering my mix downs from the studio:
Anything computer based stays there. mp3/ internet streams/ DVD movies. ( I don't own a TV or DVD player ). . . I output everything via USB to my interface ...a Line 6 UX2 (I will only use the ASIO Drivers ) . . .which handles all my clocking, AD/DA conversion and preamp/headphone...this goes to my premium home stereo on one of the line inputs. I could use the preamp outputs and go straight to my amp. How ever I listen to Vinyl, tape, FM and have a ultra premium Yamaha CD player so my computer based audio goes to my Sansui c2101 preamp. That is the best way to run a computer based system...get yourself a professional interface ( be careful the cheap stuff is cheap ) I recommend Presonus firewire based FireStudio and Emu 0404USB. . . there is not reason to use spdif for anything other than using your receivers DA converters or as an input ( or DAT recording ). . . beware of clocks. there is a good outboard ART product that is only $100 and the EMU has a good clock...but anything capable of using outboard clocks already has good clocks in it. External clocking is for studio recording/monitoring when we break out $2000 clocks.
And with a pro product you get input sampling up 96k or 192k...once a recording is made the sampling is set and does not concern you as long as your gear can handle the format.but nothing is being sold that is not a minimum of 24bit/48k (DVD minimum) . . .JetPLL sounds like it might be marketing hype...somewhere I saw that it is noise shaping...not actually a better clock...noise shaping is great (MASH and Yamaha Pro-S ? ? ?)....
All this other stuff I read about on hear is unfamiliar to me and questionable...hub on your PC, a single output to your "interface"
And hey with a pro interface you get instrument inputs, mic inputs, line inputs, and headphone outputs that do not have to go through other gear....
keep it simple and clean and you will be happy . . BTW my home system is PC/Vista Premium . . there is nothing Apple only out there except Apple software
Logic/Garage Band. I crash the studio's $21,000 Apple daily...it is a turd- but I push it hard.
later,
~e`~
Last edited:
I actually use the converter, but only to upsample to 48kHz. I feed the spdif directly into the appropriate input of the hypex amp, which internally converts any input to 48kHz anyway for the dsp and then converts to analogue.
I use the dsp of the hypex as a 2nd order crossover, which then feeds a subwoofer via single analogue cable (bash amp). Eventually I might use a PC based crossover and assign output channels to the m-audio outputs, but the present setup is rather convenient. Only one cable per speaker and less sensitive to noise.
I run windows & on my system, upgraded last year from XP.
Cannot think of any problems with the OS. In xp I had some soundcard crashes, and had to reinstall. No problem in W7. Love it.
- Status
- Not open for further replies.