EAC encoded music lossless off HDD better than any transport?

[dknightd]

Quote:

Originally Posted by philodox If a bit perfect high quality digital output can exist in my $25 sound card [chaintech av710], I don't see how it is such a big leap to say that it can exist in my mp3 player.

Have you tried the chaintech digital output into the same string you used for the others? Curious how it compares.

 

[philodox]

Quote:

Originally Posted by dknightd Have you tried the chaintech digital output into the same string you used for the others? Curious how it compares.

Actually I havent as my Flute switches between optical and coax internally and the AV710 only has optical out. I can perform the test, but then we are introducing another variable. [coax vs. optical]

Maybe if I have some time this weekend I will try it out. I can fufill most of the requirements that Jon L asked for on my computer setup... with the exception of the power conditioner and platform. I have a ONEAC power conditioner, but I doubt that it has enough amps for my computer and there is no way I am putting my computer on my audio rack for the test.

 

[philodox]

In Shane Brinkman-Davis' review of the AV300, he lists the "Digital-out quality Perfect" under AV300 Pro's... but I'm not sure if he means bit-perfect, nor does he say how he arrived at this opinion. I'll keep looking.

 

[breadnbutter]

philodox: I just read the article you linked to. I am not sure if the little experiment described there is really adequate to reproduce the effect of real jitter (and yes, I know Fourier analysis

).

Does anybody have a link where _real_ jitter of a digital signal (e.g. the one comming out of a soundcard) is actually measured somehow? With some special hardware it should be possible to determine the actual quality of the digital signal. Then it might be interesting to measure (in an analogue way?) the analogue output from a normal DAC connected to different digital sources?
Maybe (hopefully) this has been already done, in which case I would be happy to get the links.

Just to prevent any misunderstandings: I do not doubt the existence of jitter, I am just not sure of its effects in the audio domain. And I have not yet seen any objective measurements of the actual magnitude of it.

 

[Jon L]

Quote:

Originally Posted by philodox Actually I havent as my Flute switches between optical and coax internally and the AV710 only has optical out. I can perform the test, but then we are introducing another variable. [coax vs. optical] Maybe if I have some time this weekend I will try it out. I can fufill most of the requirements that Jon L asked for on my computer setup... with the exception of the power conditioner and platform. I have a ONEAC power conditioner, but I doubt that it has enough amps for my computer and there is no way I am putting my computer on my audio rack for the test.

Now that I think about it, they sure don't make audiophile-grade PC audio easy to implement. Currently, it's very difficult to buy an off-the-shelf PC audio USB and/or spdif interface that can be called high-end.

Basically ALL current devices that have toslink spdif out are using very crappy optical sender/receiver modules. One has to at least demand a properly-built digital pulse transformer for coax spdif output, but I'm not aware of any off-the-shelf units that has such quality coax output, either.

This is why I swallowed my tears and spent my hard-earned $$$ on Empirical Audio modded USB/spdif box ($500-950). Even these finely-tuned devices are not optimal, though. Somebody really needs to market USB converter that does asynchronous digital transfer with I2S output to a DAC. I'll buy such an animal..

To make the long story short, maybe the average consumer IS better off using traditional CD transports and enjoy better sounds than consumer-grade PC audio. Not me, though. There's no way I'm giving up instant access to all my songs via wireless LCD screen from my couch. It takes a lot of $$ and extra effort, but it was worth it for me.

 

[sgrossklass]

Quote:

Originally Posted by breadnbutter Does anybody have a link where _real_ jitter of a digital signal (e.g. the one comming out of a soundcard) is actually measured somehow?

Since I haven't kept bookmarks for ages, I always misplace interesting pages. I used to have a link to one where someone measured the output of his modded CD player with both the original output stage (5532 etc.) and DAC (TDA1541) and a DIY DAC/output stage. Ah well, this one will have to do for now:
http://www.pcavtech.com/soundcards/t.../JIT/index.htm
In practice, you'll see a few peaks to the sides of the main signal, typically -100 to -120 dB (or more) down.

 

[Solude]

Quote:

Originally Posted by breadnbutter Does anybody have a link where _real_ jitter of a digital signal (e.g. the one comming out of a soundcard) is actually measured somehow?

Stereophile does, the review for the RME has the graphs for jitter with coax and optical, electrical jitter vs. data jitter. Basically on coax its utter garbage electrically but near perfect data, on optical obviously no electrical and near perfect data.

http://www.stereophile.com//digitals...99/index5.html

And the difference wasn't minor like the article pointed to up top. The electrical jitter was huge.

 

[philodox]

Quote:

Originally Posted by breadnbutter I just read the article you linked to. I am not sure if the little experiment described there is really adequate to reproduce the effect of real jitter.

I thought AOS' little experiment was good, sorry it doesnt match your criteria for 'real jitter'.

Quote:

Originally Posted by Jon L Now that I think about it, they sure don't make audiophile-grade PC audio easy to implement.

Thats what it seems like to me... just way more factors to consider than with a CD based setup. Quote:

Originally Posted by Jon L This is why I swallowed my tears and spent my hard-earned $$$ on Empirical Audio modded USB/spdif box ($500-950).

Nice toy! How are you liking it so far?

 

[breadnbutter]

philodox: I think the article nicely explains the effect jitter can have, in that respect I found it quite interesting and I actually learned something from it. But as the artice says:

"While theoretical analysis of jitter requires considerable math and engineering skills, and practical measurement requires very precise instruments that cost a lot of money, you can still get a taste of what it's like with just a (very good) soundcard."

So I was asking myself if this artificial jitter is comparable to real jitter. After all periodically shifting the signal by one bit implies (from my naive point of view) quite a big effect.

Thanks to the stereophile article and measurements I am now willing to believe that the real jitter can be indeed be that strong.
I am not sure if they added the artificial jitter (the shifting of the last bit) for comparison. Indeed the graphs nicely shows that the amplitude of the "real jitter" comes pretty close to the "artifical jitter" (good, I really should find better terms for that...).

For me that is quite staggering, since it means that the effect of not-so-good cabling is almost equivalent to a full bit shifted with various periodicities. Well, I am quite happy that I learned something new, after all that is one of the best reasons to come to this forum

. Thanks.

 

[Jon L]

Quote:

Originally Posted by philodox I thought AOS' little experiment was good, sorry it doesnt match your criteria for 'real jitter'. Thats what it seems like to me... just way more factors to consider than with a CD based setup.Nice toy! How are you liking it so far?

The modded Transit coupled with great DAC gives stunning results. I'm a bit broke right now, but I'm looking for a great DAC all the time, maybe EMM Labs or Dodson 218

But the line-out of modded Transit has been very surprisingly robust and detailed. Initially it sounded like cr&*(, likely due to blackgates, but now I can listen to the line-out into SuperMacro or SI all day without the urge to go back to line-out of my Lynx 2B card.

 

[philodox]

Quote:

Originally Posted by breadnbutter So I was asking myself if this artificial jitter is comparable to real jitter.

Ah, I see... sorry if I misunderstood you at first.

Quote:

Originally Posted by Jon L The modded Transit coupled with great DAC gives stunning results. I'm a bit broke right now, but I'm looking for a great DAC all the time, maybe EMM Labs or Dodson 218

Cool, it does sound like an interesting way to handle computer audio. I prefer to keep my computer and my music seperate myself, but I think I am in the minority there.

 

[breadnbutter]

After some more reading I have to correct my previous post. Please correct me if I am wrong:

It is not possible that the actual audio data in the S/PDIF is shifted by up to a bit by the jitter. With such an amount of jitter the signal would simply not be decoded, one reason being the validity bit.

So I guess what really introduces the jitter is the synchronisation to the signal clock? If the incomming S/PDIF is slightly shifted this will result in a variation in the clock feeding the DAC as well. The effect of this will be a frequency modulation, resulting in the sidebands (for a pure sine wave).

Of course it may still happen that a single bit from S/PDIF signal is so malformed that it is accidently inverted. This would be a data induced "jitter". Due to parity checks and stuff this should happen only very seldom.

Oh my, I really want to finally understand this

 

[Scrith]

The key to a computer being the ultimate transport is a high-end DAC with an asynchronous digital input. I am still waiting for such a device to arrive, but when it does all CD players will instantly become obsolete, for many of the reasons mentioned in this thread (jitter between CD transport and DAC, errors that occur when reading CDs in real time, inconvenience of using one CD at a time, etc.).

Wavelength Audio makes a couple of USB DACs now, but I've never read a review comparing them to something like a DAC1 (the current favorite amongst computer users, due to jitter tolerance, relatively reasonable price, and high-quality digital-to-analog conversion).

PS Audio is working on a DAC with USB input which they claim is better than anything they've heard (and will sell at around the $1000 price point). Nobody who doesn't work for PS Audio has heard one yet, however.

As I've mentioned in another thread on this topic, eventually the high-end audio manufacturers will realize that the way to make their DAC sound best is to support an asychronous digital input, so one day soon we will probably see some very, very good DACs connected (asynchronously) to computers via a connection like USB, Firewire, or ethernet.

 

[sgrossklass]

Here's the article that was supposed to be linked in the AOS article but for whatever reason wasn't:
http://www.tnt-audio.com/clinica/jitter2_e.html

 

[breadnbutter]

Thanks, again a very interesting read.

This reminded me of a previous discussion on this board about modding the internal cable of the EMU 1212m, even though its I2S transport is much less affected by jitter.

I can only agree with Scrith on the solution to all this jitter mess. Do I remember correctly that the new Headroom lineup does not use this asynchronous technique for its USB DAC module? Considering their major push of PC audio that would really be an opportunity to shine (but I guess the engineering effort would be quite substantial?).