[325xi]

Of course cool expensive CD player with very thick brushed aluminium front panel (a must!) would beat those cheap no-name PC! The brushed aluminium panel I have is cryogen treated, which seriously improved sound of my CDP. But what I'm yet to try is to apply that green CD lacquer on hard drive plates, I just don't know how to open it, but I'll find out! HDD based sources suck! And please don't tell me all that stuff about bits and bytes, thick front panel protects me from all those disturbances!

 

[BrianS]

someone tell me a nice storm digital d02 and a lian-li computer case wouldn't look stunning together, in fact, stunning enough to make up for the lack of audio quality !!!...

 

[Bosk]

I also find it very telling that many former PC-as-source users have later become converts to the CD-as-source or Vinyl-as-source camps, yet the opposite happens much less frequently.

There's no doubting PC's are cheap and convenient transports, but are they really the very best sonically?

 

[325xi]

You won't perceive any difference in sound quality if you place both DAC and CD transport on a good jitter-isolation platform. Transport does better with another platform placed on its case. But don't forget to spray both with antijitter spray before you try the platforms first time, or they will be much less effective!

BTW, please don't say you don't keep your SPDIF cables in the fridge when you're not listening!

 

[BrianS]

Quote:

Originally Posted by 325xi You won't perceive any difference in sound quality if you place both DAC and CD transport on a good jitter-isolation platform. Transport does better with another platform placed on its case. But don't forget to spray both with antijitter spray before you try the platforms first time, or they will be much less effective! BTW, please don't say you don't keep your SPDIF cables in the fridge when you're not listening!

microwave duhh....it's fact that electrons move faster when they're hot!! remember those nordost specs!

 

[325xi]

Quote:

Originally Posted by Bosk There's no doubting PC's are cheap and convenient transports, but are they really the very best sonically?

Seriously, this is not about PC, this is all about hard drive vs. CD as a source. Technically it's much easier to get closer to perfection with hard drive then with CD. It doesn't mean any hard drive based source will be better then any CD transport though, any can be crap if design is no good.

But if you know what you're doing, you can send cleaner data stream from a cheap PC then from hi-end CD transport.
(My personal belief is that either clock link or asynchronous outboard network/USB receiver is required, I don't believe not in good EMI/RF isolation of sound cards, nor in iso-/synchronous USB solutions)

 

[Akathriel]

There is no way in hell any kind of antivibration platform is going to have a correlation with the amount of jitter. 190 pps and 140 pps are not very good jitter measurements either, Consonance Opera claims to measure less than 14 pps and with i2s and an external synchronous clock you can get well under 10 or even possibly 5.

EDIT: Oh I get it, it was a joke hehe, sorry 325xi, I should stop skimming. Its really not easier to get closer to perfection with HD than CD though. When you show me the hard drive that shoots out straight i2s w/o a PCI interface, then I would agree with you.

 

[philodox]

Quote:

Originally Posted by 325xi Thinking again... It is not that Squeezebox should be able to outperform any regular CD player, but rather SB's concept. Imagine Squeezebox designed by Lavry...

Ok, I'm with you there... a high end device made with the same ideas as the Squeezebox would be awesome.

 

[mojo]

I'm a computer technician who has a virtually silent PC which I use as a transport, so I thought maybe I'd offer my humble oppinion.

As was said a long time ago in this thread, a CD is basically a series of 1s and 0s on a disc. A computer using EAC can rip those 1s and 0s perfectly, with as much error correction as needed. A CD player gets once chance to get it right. There is limited error correction, but it's not perfect.

So, once you have your 1s and 0s, you need to transport them. You need an accurate clock, with a good PSU to feed it. Computer PSUs can actually be pretty good. Consider the speeds at which computer components operate at, compared to the speeds required for audio. On a good sound card, jitter should really be zero. Also, don't forget there is USB, which does not use a clock as such. It sends a chunk of data and error corrects it if necessary, where it is then clocked by the receiver. If the receiver is also the DAC, then you end up with only one clock instead of the usual two (transport+DAC), thus reducing the chances of jitter even more.

As for PC noise levels, these days I don't think they are a major problem. My PC is water cooled, but now you can easily get close to absolute silence on air cooling alone. AMD's Cool 'n' Quiet is a useful technology here. See http://www.silentpcreview.com/ for more info, but basically with a 2.5" laptop HDD, good case and some work you can build a PC that is silent at 1m. If that isn't enough, then get a USB extenion cable, a longer coax lead or a Squeezebox.

One caviat with the PC is that you need to make sure you are getting the best sound from it. Most people here will already know about kernel streaming/ASIO, but it's pretty essential. Also, if your sound card does have optical out, make sure it's bit perfect. I don't think the Audigy is, for example, so it might not stand up to a £2000 CD player.

As for blue screens in Windows, it shouldn't happen. If it does, it's probably a dodgy driver or broken hardware (RAM is a good place to start). My computer never, ever blue screens. Well, it did once a few days after I built it, but that was bad RAM. A lot of other computer issues are down to spyware. In my job, it's unusual to see a PC that doesn't have any spyware on it - it really is that bad. My tip is use Firefox instead of IE.

At the end of the day, £500 spend on a PC will give you a really good spec system, virtually silent and with the potential for at least as good quality optical out as a £2000 CD player. Plus, it will store thousands of tracks ready for instant access.

 

[Ruppin]

Quote:

Originally Posted by Bosk I also find it very telling that many former PC-as-source users have later become converts to the CD-as-source or Vinyl-as-source camps, yet the opposite happens much less frequently. ?

I don't think it is anywhere close to fair to lump "cd as source" advocates together with "vinyl" fans.

Vinyl fans (unless it was digitally recorded like some DG classical records in the 80s), are dealing with straight analog. This is a different and arguably better source. If a vinyl fan prefers a CD player, to a computer to DAC, it is only because the player is tweaking the sound in its DAC to fake an analog sound.

As Mojo just said, files on a CD or files on a hard drive are just zeros and ones.

Although, I would add to mojo's point to suggest that the easiest way to arrange for quiet, is to send digital music files over a network to another room (apple airport express, optical out---there are others). My MacBook Pro is also dead quiet and has an optical out.

 

[325xi]

Quote:

Originally Posted by mojo So, once you have your 1s and 0s, you need to transport them. You need an accurate clock, with a good PSU to feed it. Computer PSUs can actually be pretty good.

Good for what? To beam EMU/RF - not treated mass market switching PSU?

Quote:

Originally Posted by mojo Consider the speeds at which computer components operate at, compared to the speeds required for audio. On a good sound card, jitter should really be zero.

Why would it be zero? There is no clock with zero jitter. Add here highly polluted environment inside of PC case...
Speeds are irrelevant here, as unlike SPDIF transfer, computer doesn't have to deal with clock embedded into data stream. There is no data loss in computer, but there is no data loss in SPDIF either (well, I know, but it's too little to mention here) - just timing distortion,which makes all the harm. Even if you manage to isolate your sound card perfectly (which I never seen), you still have bare connectors...

Quote:

Originally Posted by mojo Also, don't forget there is USB, which does not use a clock as such. It sends a chunk of data and error corrects it if necessary, where it is then clocked by the receiver. If the receiver is also the DAC, then you end up with only one clock instead of the usual two (transport+DAC), thus reducing the chances of jitter even more.

I thought the same, but it's wrong. USB drives work this way. But all existing USB DACs use isochronous, not asynchronous mode, which still requires clock feedback to be "jitter-less". Current measurements I found show much stronger jitter on USB connection then on 'old bad' SPDIF.

Quote:

Originally Posted by mojo At the end of the day, £500 spend on a PC will give you a really good spec system, virtually silent and with the potential for at least as good quality optical out as a £2000 CD player. Plus, it will store thousands of tracks ready for instant access.

Well, I agree, except optical out. It is (was) supposed to be an ultimate transport, but for some reason its jitter level typically measures few times higher then coaxial...

My point is: take your sound card outside of the box, stream your SPDIF from there and not directly from PC, and you'll be fine. And if you find a way to provide wordclock link between that external "sound card" and DAC - sky is your limit.

 

[325xi]

Quote:

Originally Posted by BrianS microwave duhh....it's fact that electrons move faster when they're hot!! remember those nordost specs!

That's obvious, everybody does it - but only if you don't have Bybees in the signal path. It may cause overspeeding of electrons, which is dangerous as:
- they can get large speeding tickets
- they'll have troubles cornering where your cable isn't staight - you'll have synchrophasotron out of your audio system!

BTW, have you noticed Fusical Midelity came up with a new line of audio microwaves called mW (megawatt)?

 

[mojo]

Quote:

Originally Posted by 325xi Good for what? To beam EMU/RF - not treated mass market switching PSU?

Have you taken a look inside a PC lately? Typically, the PSU is right next to the CPU. If a PSU was beaming out EMU/RF, surely it would have some affect on a component that needs to run perfectly (zero bit errors) at 3GHz+ on low voltage with nanometer scale transistors.

A 1GB memory stick has to store 8589934592 bits of data without errors for indefinate periods of time, but be able to change any of them at 400MHz or more with total accuracy.

Quote:

Originally Posted by 325xi Why would it be zero? There is no clock with zero jitter. Add here highly polluted environment inside of PC case...

Okay, fair point. What I meant to say is that jitter only matters if it is perceptable. Any descent DAC will reclock the digital stream it receives, so jitter below a certain level (i.e. the level that causes the reclocking to produce errors) is irrelevent.

Consider the CPU again. Running at 3,000,000,000Hz with maybe +-100 variation per second. Realisticly, I can't see how any half descent audio card could be so poor that the clock causes enough jitter to mess things up at the DAC end where it is reclocked. Okay, so you have external connectors (hay, don't CD players have them too?) but again - consider external SATA. 300,000,000Hz clock, cables that cost 10p. There just can't be enough noise at audio speeds to cause problems.

This must apply to CD players too. Once you get past a certain point, all the jitter you are hearing must be in the DAC. Even then, a simple 16 bit FIFO buffer should allow you to use a very accurate osscilator. Can you near nanosecond scale jitter?

Quote:

Originally Posted by 325xi I thought the same, but it's wrong. USB drives work this way. But all existing USB DACs use isochronous, not asynchronous mode, which still requires clock feedback to be "jitter-less". Current measurements I found show much stronger jitter on USB connection then on 'old bad' SPDIF.

That's true, I read up on it and you are seemingly correct. Firewire is supposed to be better, but SPDIF from a good sound card is probably the best you are going to get.


Quote:

Originally Posted by 325xi Well, I agree, except optical out. It is (was) supposed to be an ultimate transport, but for some reason its jitter level typically measures few times higher then coaxial...

My mistake, I meant to say digital out, not optical.

 

[325xi]

To make this long story short, do you mind to read this? http://www.tnt-audio.com/clinica/diginterf1_e.html

 

[mojo]

Quote:

Originally Posted by 325xi To make this long story short, do you mind to read this? http://www.tnt-audio.com/clinica/diginterf1_e.html

Read and understood. However, I don't really see what it adds to your argument.

The fact still remains that, unless the jitter is really heavy the signal can always be reclocked, especially after a buffer. So, it comes down to whatever the DAC is doing, the quality of the source clock not being nearly as important.

Which is not to say that soundcards can't have good clocks, of course.

It's pretty much the same with CD-ROM drives. Consider that they need to get a 100% perfect "rip" of the data on the disc. Of course, the computer buffers it, and everything is effectively reclocked to go over the various busses in the system. So, why not just make a CD player with a tiny buffer and reclock the jittery output from the CD reading laser unit? This is, I suspect, what most CD players actually do.

I covered reclocked in A Level Electronics - I wish I still had the textbook so I could come up with some maths.

BTW the Squeezebox apparently supports externally clocked I2S.