Distance between music server and listening room

[mp]

I am trying to build a HiRez MAC music system using an iMAC which is located approximately 50 feet from my listening room. I want to output from the iMac to a USB DAC but understand there are distance limitations with USB cables.

 

I am considering several options and am curious whether any are viable or if there are others that are better.

 

1. Place Apple TV in listening room connected to iMAC via ethernet and use USB output from Apple TV to USB DAC. Will output be bit perfect?

 

2. Same as above but using Airport Extreme instead of Apple TV?

 

3. Same as above but using MAC mini instead of Apple TV or Airport Extreme?

 

4. Other options??

 

 

 

[The Computer Audiophile]

Hi mp - You have a common problem. How to get music from here to there?

 

 

 

1. Place Apple TV in listening room connected to iMAC via ethernet and use USB output from Apple TV to USB DAC. Will output be bit perfect?

 

The Apple TV will not output audio over the USB port. I have tested it using the optical Toslink output and it is bit perfect.

 

2. Same as above but using Airport Extreme instead of Apple TV?

 

Airport Extreme is not an audio device. It is a wireless router with the capability to attach a hard drive or printer. Unfortunately no audio :-(

 

3. Same as above but using MAC mini instead of Apple TV or Airport Extreme?

 

A Mac Mini could work via USB output as long as it could connect to your disk via USB, FireWire, or Ethernet. Or, have local disk.

 

4. Other options??

 

Get a DAC that supports Toslink and use a simple Airport Express. Then you're done.

 

[mp]

Thank you,

 

I am replacing a very high end CD player with the Server/DAC combo. My understanding is that USB provides the highest quality connection between the Music Server and DAC. If true, then the MAC mini seems like the right solution. If not true, please advise.

 

That said, in the configuration I referenced, should itunes lossless music files be loaded on the MAC mini, or can they remain on my iMAC 50 feet away with the signal simply passed through the USB output of the MAC mini having travelled from the iMAC to MAC mini over ethernet?

 

 

 

[The Computer Audiophile]

Hey mp - I would load the music on the Mini. No reason to keep it on the iMac and sharing the drive or library.

 

[starcat]

Place the USB interface box next to the computer and get yourself something like the Weiss DAC2 or a Trends UD-10.1 interface both having AES/EBU output you can run a cable 100m in lenght without a problem. Even up to 1km. Don't try to extend the USB or Firewire cable.

 

It is not the "USB provides the highest quality" it is more like the USB or Firewire is able to provide the best digital output depending on how well the USB or Firewire to digital out converter is built, stages like current regulation, jitter reducing, etc, etc have impact on the digital out quality. These interfaces are much better than the optical out of any Mac Mini (or other Apple computer), so that you need to consider an interface for serious listening.

 

[mp]

Thanks

 

So to clarify, running iTunes on a MAC (iMAC or MAC Mini), I can output to a USB interface box (e.g. Weiss DAC2 or Trends UD-10.1) connected to the MAC via USB or Firewire, run a 50 foot AES/EBU cable (presumably with XLR connectors) to a DAC (e.g. Benchmark Pre) in my listening room then to my stereo system.

 

And .. this configuration is as good or is possibly better than a MAC located in the listening room with USB output directly to the DAC?

 

[starcat]

Yes, exactly!

 

AES/EBU cables are always symmetrical terminating with XLRs connectors. 100m is a default max length, you can even run them for 100m but then you have to do something more. And yes,

 

[ASB]

Chordette GEM is using Bluetooth A2DP protocol and you can move your computer gear up to 30m away from the DAC.

 

The quality is the same as the USB connection. I have tested it with one wall in between the PC and GEM, and it worked without any problem.

 

No software needed (plug & play), choose MAC or Windows, any player is good and you can play Hi-Rez files up to 24bit/96khz. In Windows sample rate change is automaticly done.

 

http://www.chordelectronics.co.uk/products_detail.asp?id=53

 

[starcat]

Please do yourself the favor and laydown the 30m AES cable. It doesn't cost much.

 

[audioengr]

MP - If you are replacing a high-end CD player, then devices like the UD10 will not cut it. Too much jitter I'm afraid. If you use a device like the AirPort Express, even with a good glass toslink, same thing too much jitter. If you want to get really low jitter, you will need something better. And IMO you are getting some bad advice here, such as 30-foot AES cables. This is NEVER a good idea.

 

Here are some USB and other converters compared:

http://www.positive-feedback.com/Issue39/ramblings_computer.htm

 

email me at:

[email protected]

I have some ideas for you.

 

Steve N.

Empirical Audio

 

[starcat]

Steve N., I have had them all, Sony 7ES and Meridian CD players and I can assure you that you indeed can replace them all with the UD-10.1 and then running even 50m AES cable is not a problem at all. And yes, I have some serious class A mono blocks driving my B&W Nautilus 800 speakers to judge. By all respect but *some* proprietary mods and links are just a waiste of money. If the pal is going to throw $2k then please, please let him get a Weiss DAC2 that can do *all* freqs between 44.1 and 192kHz and act as a transport and/or DAC. And then still run that AES cable (not a standard XLR cable but a dedicated 110ohm cable).

 

Excuse me Steve, but you can't be serious on that link too? You are posting a review praising your own units? And the same review states that different USB cables "sound" different? C'mon, you must be kidding!! Give us a break!

 

[audioengr]

"I have had them all, Sony 7ES and Meridian CD players"

 

Now you have to be kidding right? Unless you have a Zanden 5000, a Bremen or one of my DAC's you have not heard a decent DAC yet IMO. The technology has come a long way. I have been going to shows for almost 10 years and I hear the Meridian stuff over and over. Not impressed, sorry. I have modded some of the Meridian DAC's fo customers BTW.

 

As for the Weiss, I'll gladly do a shootout with my Spoiler or my Overdrive. The Spoiler beats the Zanden 5000 at a fraction of the price. Just read the enjoy the music and 6moons reviews.

 

And I posted an article, not a review, that compares a lot of different audio converters. Mine were included and they just happened to come out on top. This is a fair post because it is not a review of my products per se. These types of comparisons are fairly rare in the audio writings. And USB cables can sound different in systems that are fine-tuned and resolving enough to hear the differences. If you need an engineering explanation of why they sound different, I can provide this. I have over 30 years experience designing high-speed digital interfaces for some of the largest computer manufacturers on the planet.

 

Steve N.

Empirical Audio

 

[The Computer Audiophile]

Hi Guys - It looks like we've come to a disagreement on this one (to say the least). Keep the discussion laid back and respect everyone's opinions and it's cool with me. Remember this is a hobby that's suppose to bring us enjoyment :-)

 

Also, I encourage manufacturers to post their opinions and even links such as the one Steve posted. As long as manufacturers identify themselves, like Steve has done here, the readers of Computer Audiophile are smart enough to consider that when reading the linked article. What it all means is up to each reader. I personally don't mind Steve's link and I don't mind starcat's opinion about the link. We're not talking life and death here. Just remain respectful.

 

Thanks guys.

 

[starcat]

I would be very interested to hear your engineering explanation on how an USB cable transmitting coherent and lossless computer data may sound different? If it ever could then our computer data would come once this and then that way. Coherent means the bits are transfered intime and lossless means that no bit can get lost.

 

Excuse me, but an article about different USB cables sounding different is (at least to me) of no use.

 

You should be familiar with this excelent engineerng explanation of some things:

http://www.lavryengineering.com/documents/Sampling_Theory.pdf

 

Oh, Zanden... really cool, but for some reason serious people use Weiss and EMMlabs to master and listen to SACDs and CDs. I have heard the UD-10.1 through a Meither DAC8 and believe me the USB cable didn't matter, even the lenght of 1m or 5m. Probably the listening room wasn't resolving enough or the phase of the moon wasn't correct.

 

Cheers

 

 

 

[The Computer Audiophile]

Hi starcat - Check out this link where Gordon Rankin from Wavelength Audio explains his reasoning why USB cables sound different. Let me know what you think. I'm interested in your opinion.

 

http://www.audioasylum.com/forums/pcaudio/messages/2/28679.html

 

[starcat]

I forgot to mention: when referring to USB cables I was thinking about properly shielded USB 2.0 cables and not the older USB 1.1 thinner cables designed for keyboards and mice. Don't even have the older ones, but they still might exist. And of course, the USB output on the host computer must be a proper implementation. I am only using Macs and have to see a single problem. Besides, I kind of prefer a firewire interface for a various reasons but today there are quite a lot of proper USB transports and interfaces out there.

 

[audioengr]

The difference is USB cables is the same as the difference in S/PDIF or AES cables. The effect is the addition of jitter. There are several phenomenon at work that cause this:

 

1) loss in the cable that reduces bandwidth - this slows the edge-rate, which makes is difficult for the receiver to switch at repeatable time intervals

 

2) reflections on the cable due to impedance discontinuities - this will cause a phenomena called inter-symbol interference which pushes edges in time

 

3) Dielectric absorption that causes dispersion - this is a flattening of the pulses due to stored charge that is not released at the same rate that the charging occured. More prevalent in cables with high dielectric constants. Teflon has a relatively low dielectric constant.

 

These effects are real and measurable.

 

Then the issue becomes whether the device, whether it is a DAC or converter, is insensitive to incoming jitter. I can say that after modding 15 different DAC's over the last 6 years, that they all are sensitive to incoming jitter, some more than others. If you read the posts on asylum you will find lots of anecdotal evidence for this as well. Here are some recent USB cable threads:

 

http://db.audioasylum.com/cgi/m.mpl?forum=pcaudio&n=27332&highlight=synergistic+USB&r=

 

http://db.audioasylum.com/cgi/m.mpl?forum=pcaudio&n=30687&highlight=synergistic+USB&r=

 

http://db.audioasylum.com/cgi/m.mpl?forum=pcaudio&n=28869&highlight=synergistic+USB&r=

 

If you dont understand jitter, do some searches on the forums. If you dont believe it's real, then I wont waste any more of my time on this.

 

Steve N.

 

 

 

[starcat]

Next time I have to copy data over USB I will check different cables to see which one copies them most accurately. Just to make sure *all* bits get copied.

 

Btw, while we are on this: what is the impact of using a better network cable running from the NAS to the player unit? I mean, we should care about the whole chain. Should we exchange cat5e for better cat6 or 7 cables, get them out of the walls and lay them on cable isolators just to remove magnetic interferences from the earth? And what about the length, more length means more signal loss??

 

[Gag Halfrunt]

Steve,

 

Unlike S/PDIF, USB is asynchronous. It sends data in packets not as a datastream. So jitter on a USB signal would only occur with one or more damaged controller chips; 'jitter' in a USB context would manifest itself as music cut with 0dB bursts of white noise, not subtle temporal changes to the performance.

 

Jitter is a well-documented phenomenon... in S/PDIF. It is not – and cannot be – a factor in USB, which is inherently bit-perfect. If there is some quantifiable performance improvement to USB data delivery brought about by expensive USB cables, it has nothing whatsoever to do with jitter. Given you have modified DACs, I'd suspect you are altering the current and voltage feed to the controller chip in the process. This is the one reason why an expensive USB cable could benefit - if it's very low impedance, it won't tax a compromised controller chip.

 

Whatever... it's not jitter.

 

[The Computer Audiophile]

Starcat - Steve provided very specific and measureable reasons why USB cables can sound different. Your disbelief disguised with sarcasm is very rude and shows more about you than anything else.

 

Please remember this is a laid back site. If comments don't help readers increase their enjoyment of our wonderful hobby, they should be published on other websites where typical sophomoric behavior is the rule.

 

Thanks for your understanding and your very informative comments up to this point.

 

[starcat]

but there is NO jitter associated with USB cables, period. I understand you have to be on your sponsors side, this is ok. Sorry, but arguments valid for cables in general and especially for analog cables are of no interest with asynchronous data cables.

 

If USB would alter data so would network cable alter data too. Of course there are different qualities cables, like shielded or not shielded, thicker AWG, rated for higher speeds, better insulation, installation quality, etc, but everything else like the "audiophile" nonsense associated to USB and network cables is really only to mislead consumers buying expensive "high-end" USB cables. Noone in the pro-industry *mastering* the music we are listeing to would accept this s***t.

 

And Chris, I am by no means sarcastic, I was just into understanding what is the impact of audiophile grade network cable as I actually care for the whole reproduction chain and if I was going to swap the USB cable I have to swap the network cable too.

 

[The Computer Audiophile]

Starcat - Your suggestion that I have to be on my sponsor's side is preposterous. I don't have a single advertiser that that sells a USB cable and in fact it would be better for all my advertisers if I said USB cables did not matter. This would allow more disposable income to be available to purchase their equipment. Empirical Audio is not a sponsor here at Computer Audiophile and I have not interested in siding with Steve if the facts don't warrant this.

 

I suggest you try several USB cables, measure them and listen to them, and report back your findings. I would be interested in your opinion after that. But until such time your accusations of pandering towards advertisers your discrediting of someone's engineering experience really is rather unfounded.

 

[audioengr]

Data copies??? I guess you missed the point entirely Starcat. Streaming audio data over USB or S/PDIF is comprised of both data and real-time timing. Printing is only concerned with data.

 

As for networked data, this is packetized, so the real-time timing is of little concern, except for the possibility of drop-outs due to contention or interference on the network. It is only the networked end-point device that must have low -jitter. This is where the real-time clock domain is created.

 

Steve N.

Empirical Audio

 

[audioengr]

Gag - there are only two companies on the market that are using Async USB. The others are sensitive to jitter to varying degrees because these USB interfaces are synchronous and use some type of analog or digital PLL or DLL to synchronize the local clock to the USB timing. BTW, there is a lot of anecdotal evidence (read the linnks) that even the Async interfaces are sensitive to incoming jitter from USB.

 

It is jitter.

 

Steve N.

Empirical Audio

 

[Gag Halfrunt]

Steve,

 

I think I see where we are arguing at cross-purposes. What you are calling jitter is what I think of as packet delivery variation.

 

Regardless of the different terms, how does jitter effect a signal passed across USB? I get how jitter undermines S/PDIF signals because there's no block-accurate addressing, but USB is delivering a series of data payloads. How does the variation in packet delivery alter the performance, when the data itself is undamaged?