[daerron]

FYI, USB audio does NOT feature error correction. It is not like your printer that will re-read the corrupted data (USB bulk mode).
Does it matter? Probably not, unless there's a heavy activity on the computer, there should never be any data dropout (I read a paper that said it would be in the order of one audio sample every few hours / days).

If you think analog cable make a different, consider how much more bandwidth a digital signal requires. Theoretically, it is infinite (square signal). We're dealing with transmission lines at this point: much more critical than a 5-30kHz audio signal.

 
Yep, fair enough, it is one of the modifications required to make USB stream audio better e.g like UDP and not TCP/IP. This is why the playback software, FIFO and buffering strategy matters. Ideally you'd like to buffer the entire song in memory and optomise the flow to the audio receiver without interruption, where you would have a small memory buffer that is will be continuously filled with blocks. The buffer will be emptied using the on board receiver side clock which should reduce the impact of jitter from the transmission side. Yes, a square digital signal requires more bandwidth (a square wave contains lots of unwanted EMI, therefore most signals have a trapezoidal like look), but USB uses differential signalling which as long as your cable meets the USB 2.0 specification should not matter.

 

[Argo Duck]

^ nice discussion. It's possible what 2222440 describes is as real as anything else we experience here in head-fi land but has different explanations than s/he has put up. If so, there is a different model (theory) needed to account for 'the listening experience' than just strictly the electronic/audio model, something that bridges the objective/subjective 'divide'...

But that comment is probably going OT for this thread!

 

[2222440]

Daerron has very good knowledge of digital theory.
 
Based on the current digital theory, I don't quite undertand why there are analog characteristics on digital cables/devices as well.
 
Instead of denying the facts that exist, I tend to believe that there is something beyond the confines of the current knowledge, there are still something unknown in this area.  It is like human being sees the natural lightings first, then after, find out the electricity theory behind it.
 
In practical, digital audio media is not as consistent as it should be. A good example is XRCD and normal CD, they are both red book spec CDs, but XRCD sounds more clean and clear. Some other red book spec CDs like HQCD,AQCD,all sounds better than normal CD. Some record companies are selling studio master direct burnt CD-R at a higher price than massive producted CDs.
 
I always find a slight difference of 2 copies of a same CD release.
 
If you burn the same wav/flac files to different blank CD-R, you will find different brands of CD-R has its own typical sound, even different factory batch CD-R under same label sound different. 
 
[size=15.833333015441895px]It is not a subjective/objective matter, the physical part of the digital audio media is not that  unimportant as the protocols suggest.[/size]

 

[highstream]

2222440 - While the specific focus is turntables, the following article by system setup guru Jim Smith speaks to many of the questions you raise about CDs: http://thehighfidelityreport.com/big-wheel-keep-on-turning/

 

[daerron]

  2222440 - While the specific focus is turntables, the following article by system setup guru Jim Smith speaks to many of the questions you raise about CDs: http://thehighfidelityreport.com/big-wheel-keep-on-turning/

 
Thanks for the link, interesting it raises a lots of points I've been having with my turntable setup that I got since the beginning of the year. I am always thinking that the TT could sound better, though I am quite pleased with the sound quality. Looks like I still have a ton of optimisation to do on mine!

 

[Happy Camper]

http://www.audiostream.com/content/draft Part 1

There's no such thing as digital: A conversation with Charles Hansen, Gordon Rankin, and Steve Silberman

http://www.audiostream.com/content/theres-no-such-thing-digital-conversation-charles-hansen-gordon-rankin-and-steve-silberman-p Part 2

 

[daerron]

http://www.audiostream.com/content/draft Part 1

There's no such thing as digital: A conversation with Charles Hansen, Gordon Rankin, and Steve Silberman

http://www.audiostream.com/content/theres-no-such-thing-digital-conversation-charles-hansen-gordon-rankin-and-steve-silberman-p Part 2

 
Interesting article, and agree with some points made, it has been my experience as well that digital systems usually get mucked up by analogue problems. If you get the analogue right, you've got a solid foundation to work from. Been there myself, mucking up the analogue design.
 
With digital transmission, analogue design and cable problems typically manifest itself in intermittent reliability, protocol breakdown and data corruption and will reflect in the bit error rate and result in slow throughput. For example if you end up with a distorted or variable signal, you won't be able to set a PLL with certainty to read the data correctly. The bits also need time to settle, so you will always see settling time specified where clocking is involved. If you need to regenerate a clock from the distorted analogue signals (USB being a point in case) you would end up with a clock with lots of jitter which would affect audio if used in the conversion process. Apart from clock regen, digital transmission should not result in subtle differences in audio playback, but more as garbage and broken/no playback. Check how many people with USB DAC issues are caused by people using their USB 1.0 cables that came supplied with their printer...
 
There is nothing stopping DAC makers from reading the entire WAV into physical memory on the DAC and doing a CRC check before and after and see whether the data integrity is still valid, but it would be impractical, better to focus on the DACs control of the flow of data from the computer. It would be like the interesting point made on the CD laser to bits conversion process, when ripping CDs to disk and the accurate rip checks done thereafter. But those checks are usually necessary because the rips are made at high speed and not all CDs are of the same quality. If you read them 1x the chances of errors would be very minute.

 

[2222440]

Good article. It somewhat explained lots problems found in digital audio.
"Anyone who feels it's only "1" and "0" is missing a ton more variables that need to be addressed."

 

[Happy Camper]

^ nice discussion. It's possible what 2222440 describes is as real as anything else we experience here in head-fi land but has different explanations than s/he has put up. If so, there is a different model (theory) needed to account for 'the listening experience' than just strictly the electronic/audio model, something that bridges the objective/subjective 'divide'...

But that comment is probably going OT for this thread!

Knowledge is found in many OT comments. Mining them is the fun part.

 

[2222440]

 
As more and more studio masters are available in the market  for downloading, it might become a main audio source in near future with its competitive value for money. 
With digital studio master tapes or HDtracks or SACDs as source, what Ciunas can dilever is already amzing.  It let me get much more involved into the music,  much closer to the performer, much more information from the performance.  It is as if the performers are standing in front of me in the room, I can feel every subtle action of the singer's mouth and face. The music instrument placement feels like 3D.   I even start to like those music that I didn't pay attention to, just beacause of the liveness of the sound.
 
[size=15.833333015441895px]Though I am quite happy with the current Ciunas, I still expect a future Ciunas that can handle raw DSD and DXD 24/384. [/size]

 

[drez]

Best explanation of why computer and digital cabling matters:

http://www.computeraudiophile.com/f8-general-forum/new-nordost-heindall-2-universal-serial-bus-industry-standard-cables-connectors-and-communications-protocols-between-computers-and-electronic-devices-19782/index5.html#post327648

 

[preproman]

Has anyone compared the Ciunas converter to any other TOTL converter?

 

[Clemmaster]

I own it and the Off-ramp 5 (coax + i2s upgrades + TeddyPSU).
The Ciunas is a much better value but has limited connectivity.

The OR5 makes everything sound a bit darker but has better separation and focus. It is not as convenient to use (bigger box, separate PSU), much more expensive... But it is the better converter, so...

The Hydra-X+ looks very interesting feature-wise (i2s + battery) and 6moons rated it's performance up there with the 2-box SOtM solution.

 

[preproman]

Yeah - I'm not so much worried about the connectivity part.  I would also imagine the difference is like splitting hairs.  I guess the value part says it all.  Thanks Clem.

 

[antonyfirst]

I might add some inputs too.
I have compared the Ciunas to the Audiophilleo and the former is more precise and detailed. It's almost unsensitive to the USB cable used.
On the other end, it depends a bit on the coaxial cable, whereas the Audiophilleo doesn't need one.
 
I have also used the Ciunas with the Eximus DP-1, and it makes the sound a good notch better, compared to connecting the USB cable to the internal receiver of the Eximus (a Stello U3), which is too dark. Separation increases and the tonality doesn't get screwed up.
I will try the JCAT usb cable with the Eximus to see if the combo gets as good as the Ciunas + Eximus.