2. digital
  3. 133904-4132

Firewire immune to jitter?

I wonder if a firewire connected DAC is as problematic at rejecting jitter as a USB is and if there are reclockers available for firewire.

I currently have a MacBook pro running Audirvana+ feeding a Weiss DAC202 to it’s firewire input with satisfying sonic results. I really like the DAC202. Firewire cable is an Audiquest carbon.
But, would I be best to use the USB out of the Mac using a USB to S/PDIF converter into the rca input of the Weiss. I use to have one of those (a Wavelength I think) feeding an ARC DAC7 that could not process HR files through its USB input. I have since sold the DAC7 and gotten the Weiss.

I was also considering replacing the Mac with an Aurender N100h (Or similar) but many people are telling me that I must be prepared to spend mega bucks to better my current digital set up. 
Cost of an Aurender N100h or N100c
Cost of a new DAC (USB capable) but not necessary if using a N100c 
Cost of a reclocker (Synchromesh or similar)
Cost of USB, S/PDIF/BNC cables

I barely have enough money for an Aurender let alone all the peripherals mentioned above.

Note, I am not into MQA and DSD and will never be. If I want better sound than 16/44 or 24/96 I just put a vinyl on my $30k analog front end. 

Any thoughts much appreciated. And if the best solution is to leave everything as is then that’s cool too.



smoffatt
13 responses Add your response
audioengr
3,282 posts
 

As far as I know, there are no reclockers for Firewire.  I have heard that someone did a reclocker for USB though and it is battery powered and part of the cable.  Rather than buying this, I would recommend the Berkeley USB converter.  USB is not my first choice however.

The connection that delivers the best jitter performance now is actually Ethernet, either DLNA/uPnP or RAAT using Roon.  Here is a study I did of my own XMOS USB versus my own Ethernet:

http://www.audiocircle.com/index.php?topic=155232.0

It is for this reason that I am not doing any new USB interfaces.

If you were to go Ethernet, you could buy an external renderer and drive S/PDIF to your Weiss DAC.  This is the best way to get the full benefits of hi-res up to 192. You may need to use Jriver rather than Audirvana+ however.  I have not had good luck running DLNA Ethernet using Audirvana+.  Jriver works perfectly and sounds great. 

The converter I sell for Ethernet is identical to my Off-Ramp, but has Ethernet input.  It is called the Interchange.  It will deliver sample rates up to 192 with jitter in the 16psec range, my lowest jitter solution.  It must be used with Jriver. $2999.00. A WIFI adapter is also available for $1K.  I believe this to be lowest jitter solution on the market.

If you got a micro-Rendu or UltraRendu from Sonore, this would allow you to go Ethernet and then USB to the DAC. The jitter will be dependent on the clock in the USB interface in the Weiss DAC, so this is not my preferred choice.  These are compatible with DLNA (Jriver), Audirvana+  and Roon.

If you go Ethernet or use the Berkeley USB converter, I would highly recommend using a good S/PDIF cable:

http://www.audiocircle.com/index.php?topic=154425.0

No sense in throwing away low jitter by using a poor cable.

BTW, the cheapest way to getting really low jitter is to use Toslink from your MacBook Pro to the Synchro-Mesh and my Reference BNC cable from Synchro-Mesh to your Weiss DAC. You could compare this to your firewire connection starting at $1198.00 (with OTL option).  30-day money back, less shipping. This would work with Audirvana+.

Steve N.

Empirical Audio

shadorne's avatar
Ag insider logo xs@2x
shadorne
7,467 posts
 
I am not sure why you think your DAC202 has a problem. This is one of the best DACs on the market probably only superseded by the latest Weiss. Weiss has excellent jitter rejection so why are you worried?
smoffatt OP
205 posts
 
Shadorne, I do not think there is something wrong with my Weiss. To your point it is an incredible DAC and would not want to part with it as I would likely go sideways (or worst) with something else. I was only throwing a few ideas around to lower jitters from the source and Steve proposes a few solutions although expensive in some cases.

I have yet to see comparative measurements/graphs about jitter rejection (USB vs Firewire) showing which is the worst of 2 evils.

Steve, forgive my ignorance but how do you connect a Toslink optical cable on a Mac when there is only a USB and Firewire port.
Does the quality/Cost of this Toslink cable going into the mesh paramount. No BNC input port on the Weiss. RCA yes. 

My Mac is a mid-2009 but would like to replace it with a mid-2012 which is the last model year that you can mod memory cards and drive. Beyond 2012 everything is soldered onto the boards. You can’t even replace the battery I don’t think. Anyway, I digress. 

Thanks for your detailed informative response....as typical of you.
Sly
audioengr
3,282 posts
 

Steve, forgive my ignorance but how do you connect a Toslink optical cable on a Mac when there is only a USB and Firewire port.

I believe the earphone jack has Toslink output.  It just needs an adapter.  If you play music and look into the jack, there should be a red light.


Does the quality/Cost of this Toslink cable going into the synchro-mesh paramount? No BNC input port on the Weiss. RCA yes.


Yes, it is important, but I can recommend an inexpensive Toslink cable that I use as reference.

My Reference BNC cable has RCA adapters with it, 75 ohms.

Steve N.

Empirical Audio


shadorne's avatar
Ag insider logo xs@2x
shadorne
7,467 posts
 
@smoffatt 

This is what JA of Stereophile says


“Finally, the Weiss DAC202 offers the best rejection of datastream jitter I have encountered. I have shown the spectrum of the processor's output when fed FireWire data representing 16- and 24-bit versions of the J-Test signal (fig.12). The AES/EBU and TosLink spectra are identical. With 16-bit data (cyan and magenta traces), the harmonics of the low-frequency, LSB-level squarewave were at the residual level, and were not accentuated or modified by the DAC202. With 24-bit data (blue, red), all that is visible is the central spike of the Fs/4 tone, this sharply reproduced, and with very little spectral spreading at its base. The jitter was too low for the Miller Analyzer to measure. Again: Wow!
Read more at https://www.stereophile.com/content/weiss-dac202-firewire-da-converter-measurements#DuQYUGqLYrkoyitU...

Daniel Weiss prefers FireWire as the communications protocol is less likely to induce LIM due to the way it streams data.

I think you have nothing to worry about but if you find buying more gear or extra cables will alleviate your fears then why not....

bsme85
151 posts
 
What shadorne says. It's unnecessary to purchase additional gear that's been hawked here.
smoffatt OP
205 posts
 
Interesting read. Thanks shadorne.
In light of that, I may just replace my firewire cable for a better one. A Diamond or similar. Will look at the used market first.
peter_s
1,313 posts
 
 Regarding Steve’s argument about ethernet being the best connection… I have a PS audio direct stream dac, and I wonder if getting an ethernet bridge would bring it to that optimal performance? 
kijanki's avatar
kijanki
4,394 posts
 
In synchronous S/Pdif , even when source and the DAC clocks are different DAC adjusts its clock to follow average rate of the source.  This is usually done with PLL, that often introduces its own jitter.  In properly executed asynchronous USB clocks are completely independent.  Source is sending data in frames at about 1 kHz rate while DAC places them in the buffer and sends back buffer under/overflow signal, to which source adjusts number of samples in the next frame.  With this scheme amount of jitter depends only on the quality of internal DACs clock.  I suspect that Firewire uses similar method.
erik_squires
15,643 posts
 
I had the ARC DAC 8 - I would not judge most modern DAC's based on them at all.

Up to and including the DAC 8 they did not have "great" clocks, but only good clocks. There are a number of DAC's which outclass them by far in the USB realm.

Also, the issue with USB is as often power supply noise from the PC, including ground noise issues. If you have a laptop it is worth experimenting on battery power vs. not, and also consider a USB isolator.

Most top-tier DAC's now do much better in isolating the USB from the possibility of ground loops, but it's not guaranteed.

Also, tools like Wyred4Sound's Remedy reclocker work great with older generation DACs.

In the modern DAC realm I would encourage you to listen to Mytek and Schiit as excellent USB capable DAC's, especially in Redbook.

Best,

E
audioengr
3,282 posts
 
the Weiss DAC202 offers the best rejection of datastream jitter I have encountered.

Here is an experiment that will show just how well it rejects jitter:

Change the digital source or the cable or both. If the sound quality changes, then it will benefit from a reclocker and a better digital cable. The reclocker will make a bigger difference however. The DAC that is totally immune to incoming jitter is a rare item, no matter what is advertised.

In synchronous S/PDIF , even when source and the DAC clocks are different DAC adjusts its clock to follow average rate of the source. This is usually done with PLL, that often introduces its own jitter. In properly executed asynchronous USB clocks are completely independent.

Certainly true, although good S/PDIF receiver chips add extremely low amounts of jitter. I can compare a direct I2S feed and a S/PDIF feed from the same device and the difference between the two is barely audible and only on certain tracks, at least with my DAC.

Interesting read. Thanks shadorne.
In light of that, I may just replace my firewire cable for a better one. A Diamond or similar. Will look at the used market first.

If the Weiss rejects all jitter, then why would you waste your money on an expensive Firewire cable? Jitter is the only thing you are trying to minimize, nothing else. The reason is that it doesn’t reject all jitter.

Like I said, jitter always matters and the best way to minimize it is a good reclocker that will give you 20psec of jitter like the Synchro-Mesh and secondarily a good cable like my Reference BNC. You can try other reclockers until the cows come home. None of them will beat the Synchro-Mesh:

http://www.audiocircle.com/index.php?topic=154498.0

I’m not trying to make a lot of money here. The SM sells for $599. I don’t need a sale here. I’m just trying to help you optimize your system and give you good advice.

Steve N.

Empirical Audio

audioengr
3,282 posts
 

Up to and including the DAC 8 they did not have "great" clocks, but only good clocks. There are a number of DAC's which outclass them by far in the USB realm.

Also, the issue with USB is as often power supply noise from the PC, including ground noise issues. If you have a laptop it is worth experimenting on battery power vs. not, and also consider a USB isolator.

Most top-tier DAC's now do much better in isolating the USB from the possibility of ground loops, but it's not guaranteed.

All very true. These are some of the reasons why I do Ethernet now and not USB anymore.  Here is a direct comparison of Ethernet to isolated XMOS USB, apples to apples:

http://www.audiocircle.com/index.php?topic=155232.0


Steve N.

Empirical Audio

smoffatt OP
205 posts
 
Thanks Steve. Advice well taken.
Log in to respond
More to discover
80x80
McIntosh MCD550
80x80
Chord Company Hugo TT2
80x80
McIntosh MCD-1100