[mikeaj]

Quote:

From 1986 through 2001, I was a programmer working (almost) solely on digital communications, so it's often trivially easy to recognize silly claims about cables, etc. OTOH, 100% of this work was with cables with didn't include a separate power conductor. I had considered only the recovery of the digital signal in my claim.
 
Thanks. It's good to learn something new.

 
I see, a good background and perspective then.  Well, I hope spelling out some of the concepts and writing the extra words weren't wasted, as there are other readers too.
 
 
But in this kind of situation, do we really call it the digital communications that's making the difference?  It really isn't.  It's the analog considerations of the cable (maybe also taking a loose definition of a "cable" possibly including passive circuit elements like ferrite beads) used for the digital communications.
 
Usually when we think about digital communications, we're just interested in digitized data getting from point A to point B, maybe in a timely fashion or with some reliability, etc.

 

[Currawong]

I suspect mikeaj is right in this. My own impressions of the effects of different equipment along with asking questions of people who design digital audio gear suggest that a lot more is going on. With something like an ODAC (which everyone should be talking about rather than derailing the thread) it has very little in the way of power circuitry and has a much higher noise floor than many more expensive DACs so I don't think it wouldn't be a surprise if the cable, power or whatever had some effect on the shape of that noise. It was designed to be cheap, however, so it's rather the wrong place to talk about it.

 

[mikeaj]

That's not really what I said, but anyway...
 
Noise levels should be an aspect that almost everybody can agree on as something that's cleanly measurable, directly can be mapped to some measurement—unlike some facets of sound quality and audio perception.
 
Anybody recall seeing any noise measurements of ODAC?  Even some "guy" out there never got around to posting those explicitly, though noise floors can be found on some of the other graphs.  Maybe somebody can ask Tyll to do it eventually, though he's a busy guy.  Or we could maybe ask John Seaber of JDSLabs, since they're selling ODACs and also now have lab-grade audio benchmark gear.  For all that's been said on the subject, I want to see the magnitude of changes evidenced by say a powered USB hub vs. a noisy computer USB port, ferrite bead vs. no ferrite bead, etc.
 
If the differences are as great as some subjective descriptions have been, then they should be readily measurable without much trouble.  If they're not so much (that would be my guess), then that result might fall out as well.  Either way, it would be interesting to see.

 

[adamlr]

Quote:

 
I believe you're referring to a ferrite bead. It's used to prevent interference from the devices. I have no experience with the odac so I don't know if it's prone to interference but I know most of the time, USB ports on computers are pretty noisy themselves. MIght be safer to grab one with a bead.
 
E: Just about any micro cable will work, just use the one that sounds clean to you.

 
Quote:                                                                                                                                  

There are a lot of claims and conjecture out there with little to no supporting evidence of high credence.  It's hard to say what really counts without some stringent testing, but you should be able to make some good guesses otherwise.
 
The "plastic 'bottle'" thing near the end would be a ferrite bead, which mainly filters out some high-frequency noise.  Seeing as the DAC is powered from USB, this is a good thing; you could possibly even think of it as part of the power supply.  That said, there is filtering on board and I don't think it should really make much of a difference to not have one.  If there's any difference, it would manifest itself in background noise levels (mostly; you probably wouldn't get anything else significant without any increase in noise), so if you don't hear any hiss then you're good to go.
 
So it's almost certainly not important in the sense of working or in terms of using it and performance, but it probably helps benchmark results in some small but measurable way.  Depends on the source computer.

                                                                                                                                                 
 
thank you both very much. i guess ill have to try it out myself and see

 

[adydula]

There is a lot of noise data and measurements on the ODAC on the designers blog.
The first version of the ODAC, despite following the reference design, only had about 98 dB DNR.
On the fourth version this went down to over 111dB.
 
There are many obvious advantages to USB power but it often degrades performance due to noise. To get around this, the ODAC uses split digital and analog power supplies each with their own filtering and regulator. The analog supply has additional filtering and the critical reference voltages, and negative supply for the DAC chip, are further optimized.
 
Also his transpanecy guidelines states that all noise components should be below -110db and the total sum should be below -100dBFS.
 
Also read where the lab guys state that anything below 100DNR in the real world relates to being transparent...and anything beyond this is well....
Anything beyond -110dB is well past the point of diminishing returns...
But for us spec guys ...great.
 
Take some time over there..lots of good info in his blogs under the ODAC, updates and details..
 
Alex

 

[autoteleology]

The "plastic 'bottle'" thing near the end would be a ferrite bead

 
Ha! I always wondered what those things were.

 

[JakeJack_2008]

Quote:

There is a lot of noise data and measurements on the ODAC on the designers blog.
The first version of the ODAC, despite following the reference design, only had about 98 dB DNR.
On the fourth version this went down to over 111dB.
 
There are many obvious advantages to USB power but it often degrades performance due to noise. To get around this, the ODAC uses split digital and analog power supplies each with their own filtering and regulator. The analog supply has additional filtering and the critical reference voltages, and negative supply for the DAC chip, are further optimized.
 
Also his transpanecy guidelines states that all noise components should be below -110db and the total sum should be below -100dBFS.
 
Also read where the lab guys state that anything below 100DNR in the real world relates to being transparent...and anything beyond this is well....
Anything beyond -110dB is well past the point of diminishing returns...
But for us spec guys ...great.
 
Take some time over there..lots of good info in his blogs under the ODAC, updates and details..
 
Alex

 
Very interesting, thanks. Thanx to mikeaj and Currawong as well.
 
***************************************
The following sounds so convincing to a non-expert like me.
 
AUDIOQUEST - FOREST USB MINI CABLE
Price: $29.00
 
 
[size=medium]Surprising Performance, Affordable Price[/size]
 
[size=small]Chuck Those Generic USB Cables and Step Up To Forest USB![/size]
Priced only a few dollars more than generic USB cables, Audioquest Forest is a substantial improvement. Forest offers a far smoother & richer tonal balance, much greater resolution and a clearer sonic picture. The secret is in Audioquest's unyielding commitment to quality; every aspect of Forest's construction from the type of connectors used to the direction of the conductors has been scrutinized, auditioned, revised and perfected to create a cable that offers outstanding performance for a fraction of the cost of most audiophile designs.
 
"The AudioQuest Forest is very affordable by standalone cable standards, and it offers all-round improvements over bundled cables.
Detail levels are markedly improved, there's greater precision and punch, and a general upgrade in musical expressiveness."
-- What HiFi?
 
Metallurgy
Forest uses high-quality Long Grain Copper solid-core conductors. Long Grain Copper has far fewer imperfections than OFC copper for lower distortion and greater conductivity. LGC copper conductors are used in many Audioquest analog cables, providing a far smoother, richer and more dynamic sound than the terrible materials used in generic USB cables. LGC's lower distortion characteristics allow Forest to deliver greater resolution and musical energy with a noticeably lower noise floor.
Forest is a directional cable, AQ listens to every primary signal conductor used in Forest to determine proper conductor orientation,
this simple test yields a surprising improvement to sonic performance.
 
Digital Audio over USB
There are significant, audible differences between USB cables. We're not sure how this is possible, since USB is purported to be a purely digital interface, however, the sonic differences are repeatable and consistent from system to system. A better USB cable makes a better digital audio cable, period.
 
Termination
All Audioquest Indulgence series USB Cables are terminated by Wave Solder, a very high-precision soldering method originally developed for surface-mount applications in the computer industry. Wave Soldering is fast, efficient and has no human interaction, eliminating all conventional solder problems, creating a strong, reliable and electrically consistent connection. Wave Soldering is just one feature that makes Indulgence Series USB cables a far better choice than conventional USB Cables
 
 
From http://www.musicdirect.com/p-58290-audioquest-forest-usb-mini-cable.aspx
 
*********************
 
So, what do you think about the Forest   directional cables???
 
Should I follow their advice and  chuck

 the USB cable (with a bottle neck) which came with the ODAC from the JDS Labs???
 
 
*********************************************************************************************
 
P.S.
Edited to add this:
 
AudioQuest - Forest - Mini USB Cable  ($29)
 
Listed in The Absolute Sound 2012 High-End Audio Buyer's Guide
 

High-Performance Entry-Level USB Cable

Forest is the high-performance entry level cable in AudioQuest's 'Indulgence Series' of USB cables. Boasting solid Long-Grain Copper (LGC) conductors, Forest delivers low jitter, low distortion audio for superior performance. Enjoy Forest USB cables with PCs, DACs, and other digital gear, and experience how good digital music can sound.

AudioQuest Forest USB cable was honored by being included in The Absolute Sound 2012 High-End Audio Buyer's Guide. According to the magazine, "For those on a tight budget, Forest is a large upgrade over a generic USB cable that was never designed for audio. Compared to standard USB cables, Forest offers a larger and more dimensional soundstage, more liquid and lifelike timbres, and greater transparency."
 
....
...
 
The Next Evolution of USB Cable Performance
...
There are huge differences in the sound of USB cables. As part of AudioQuest's re-examination of USB cable design, specific attention was paid to maximizing the sonic performance of USB. AQ's Indulgence USB series pushes USB sonic performance to new levels. While test equipment is excellent at tracking control errors or ensuring data pass through, no computer can compete with the human ear when it comes to perceiving the subtle but crucial audio errors and distortion that can reduce the enjoyment of your music and video files via USB.
 

The Engineering and Construction Edge

 
AudioQuest begins with Long-Grain Copper (LGC) and adds increasing amounts of silver-plate culminating in Diamond USB, whose conductors are made using solid Perfect-Surface Silver (PSS). Solid Polyethylene insulation ensures stable conductor geometry while minimizing insulation induced phase distortion. On the two top USB cables, Coffee and Diamond, AudioQuest's patented Dielectric-Bias System (DBS) polarizes the insulation, providing a drop in the perceived noise floor and an increase in dynamic contrast.
 
From http://www.audioadvisor.com/prodinfo.asp?number=AQMUSBFOR%20.75

 

[HamilcarBarca]

Quote:

Forest is a directional cable, AQ listens to every primary signal conductor used in Forest to determine proper conductor orientation,
this simple test yields a surprising improvement to sonic performance.

 
There's no such thing as a "directional [USB] cable" or "conductor orientation" for electrical cables. I guess they couldn't find any other way to justify charging $30 for a $1 cable.

 

[mikeaj]

I mean, it's your money and you can try it if you're curious, but a lot of the claims are dubious at best and so you're not going to find that many positive responses here.  Let's even forget that we're talking about a digital data (well also supply lines) cable.  Lots of previous audio system measurements of "directional" analog audio cables shows pretty much 0 difference no matter which way they're hooked up.  And here you're talking about hooking up something that is steps removed from any analog process.
 
Something like those HDMI cables with Redmere tech in them are directional, but that's because they're cables with chips (active electronics) in them that actually do things.

 

[adydula]

OMG....
 
Well if you have the money, buy both and test and let us know!
 
But if I were you I would 'RUN"...
 
Alex

 

[tomb]

Quote:

There is a lot of noise data and measurements on the ODAC on the designers blog.
The first version of the ODAC, despite following the reference design, only had about 98 dB DNR.
On the fourth version this went down to over 111dB.
 
There are many obvious advantages to USB power but it often degrades performance due to noise. To get around this, the ODAC uses split digital and analog power supplies each with their own filtering and regulator. The analog supply has additional filtering and the critical reference voltages, and negative supply for the DAC chip, are further optimized.
 
Also his transpanecy guidelines states that all noise components should be below -110db and the total sum should be below -100dBFS.
 
Also read where the lab guys state that anything below 100DNR in the real world relates to being transparent...and anything beyond this is well....
Anything beyond -110dB is well past the point of diminishing returns...
But for us spec guys ...great.
 
Take some time over there..lots of good info in his blogs under the ODAC, updates and details..
 
Alex

OK - I have issues with much of this thread, including the above.

 
Quote:

 
Very interesting, thanks. Thanx to mikeaj and Currawong as well.
 
***************************************
The following sounds so convincing to a non-expert like me.
 
AUDIOQUEST - FOREST USB MINI CABLE
Price: $29.00
 
 
[size=medium]Surprising Performance, Affordable Price[/size]
 
[size=small]Chuck Those Generic USB Cables and Step Up To Forest USB![/size]
Priced only a few dollars more than generic USB cables, Audioquest Forest is a substantial improvement. Forest offers a far smoother & richer tonal balance, much greater resolution and a clearer sonic picture. The secret is in Audioquest's unyielding commitment to quality; every aspect of Forest's construction from the type of connectors used to the direction of the conductors has been scrutinized, auditioned, revised and perfected to create a cable that offers outstanding performance for a fraction of the cost of most audiophile designs.
 
"The AudioQuest Forest is very affordable by standalone cable standards, and it offers all-round improvements over bundled cables.
Detail levels are markedly improved, there's greater precision and punch, and a general upgrade in musical expressiveness."
-- What HiFi?
 
<snip>
blah blah blah ...

However, this is just pure bunk ...
 
Having sold USB-powered DACs for several years, I have told everyone that it's important to ensure that you have a shielded USB cable - not all of them are (mouse, keyboard USB cables), but other than that - it's all 0's and 1's, period.  There may be some incremental differences/improvements to be had if power is an issue (USB-powered DACs), but even then, it doesn't mean squat compared to a good DAC design with onboard regulation (LDO's, capacitors) or good quality USB power (hubs, quality PC power supplies, etc.).  Just MHO ...

 

[adamlr]

just "clipped" my O2. i used my swiss army knife to cut the resistors and it all went very smoothly. my m-100 is still delayed in the mail unfortunately, but i have found the unity gain is quite enough to drive my dt 770s, even with heavy eqing. i cant say ive noticed any improvement of sound quality though, but i was never a "golden ears" type. 
a note for anyone planning on doing the same with their O2: jds labs apparently glue the board to both the front, and back plate. i had to remove both in order to get the board out.
 
just wanted to thank everyone in this post for helping me understand what i have to do. big ups 

 

[lorriman]

just "clipped" my O2. i used my swiss army knife to cut the resistors and it all went very smoothly. my m-100 is still delayed in the mail unfortunately, but i have found the unity gain is quite enough to drive my dt 770s, even with heavy eqing. i cant say ive noticed any improvement of sound quality though, but i was never a "golden ears" type. 
a note for anyone planning on doing the same with their O2: jds labs apparently glue the board to both the front, and back plate. i had to remove both in order to get the board out.

Unity gain gets people past 9 o'clock on the volume control/pot, which is desirable since just before 9 o'clock the volume control has channel imbalance. Other than that there shouldn't be any audible improvement since even with 6.5x gain distortion etc is under audible levels. It's popular to clip the 6.5x gain leaving 1x and 2.5x but 6.5x is needed for connecting low-voltage mp3 players and other low-voltage sources. Unless one is certain that one is never going to do that then it's better to clip the 2.5x. Many high fidelity types wouldn't even dream of using a DAP/mp3-player as a source, but since the clip+ effectively measures as hifi (other than for super-sensitive iems), there is reason even for them to use a DAP; even the IPOD is just about hifi.

 

[adamlr]

Quote:

Unity gain gets people past 9 o'clock on the volume control/pot, which is desirable since just before 9 o'clock the volume control has channel imbalance. Other than that there shouldn't be any audible improvement since even with 6.5x gain distortion etc is under audible levels. It's popular to clip the 6.5x gain leaving 1x and 2.5x but 6.5x is needed for connecting low-voltage mp3 players and other low-voltage sources. Unless one is certain that one is never going to do that then it's better to clip the 2.5x. Many high fidelity types wouldn't even dream of using a DAP/mp3-player as a source, but since the clip+ effectively measures as hifi (other than for super-sensitive iems), there is reason even for them to use a DAP; even the IPOD is just about hifi.

 
yes your right, thats why i "clipped" it in the first place, because the m-100 are sensitive and 2.5x would definitely be too much for them, so unity gain it is.
and im using the O2 as a desktop amp. theres no way ill ever take it with me portably, so no 6x gain for me. infact, looks like 2.5x is more then i need too...
btw, while hifi is a somewhat... questionable term, i agree that different sources, atleast to my ears, dont make as big of a difference as their sometimes made out to.

 

[lorriman]

yes your right, thats why i "clipped" it in the first place, because the m-100 are sensitive and 2.5x would definitely be too much for them, so unity gain it is.

and im using the O2 as a desktop amp. theres no way ill ever take it with me portably, so no 6x gain for me. infact, looks like 2.5x is more then i need too...
btw, while hifi is a somewhat... questionable term, i agree that different sources, atleast to my ears, dont make as big of a difference as their sometimes made out to.

That doesn't make much sense. I use my O2 as a desktop amp also. Clipping the 2.5x would have given me unity gain for my ultra-sensitives and left me the option of playing from my mp3 player or HD650s (serious desktop headphones).

I've never seen a justification for clipping that 6.5x that adds up.

For some reason people have got this idea of clipping the 6.5x in to their heads (justified by the higher distortion at 6.5x) and encouraged others to do so too, but all it does is limit your options. That higher distortion isn't audible and is irrelevant.