Article: Weiss Engineering DAC202 Review

[cfmsp]

<br />

<br />

"I would add that you are constantly polluting threads here with MH products."<br />

<br />

I'll be the 'judge' of that! ;0<br />

<br />

"That would be very nice to try and pollute only one thread out of two :D"<br />

<br />

nice!<br />

<br />

hope all is well, when this thread winds down, you'll hear a lot less from me. <br />

<br />

clay

[Audio_ELF]

Maybe what we're learning is that there is more than one way to skin a cat!! Or more than one interface design which can (probably legitimately) be described as async.<br />

<br />

I do remember than there was an outcry about dCS's statement when they (in an FAQ) stated that FireWire couldn't be used in the same way as USB async. Maybe they were right?

[Lars]

I acquired my Wavelength Crimson DAC at the end of 2006. This version was adaptive USB. In 2007, Gordon developed the asynchronous USB and sent me the update chip. There was a significant improvement in the sound. If one does a search in Audio Asylum for Mercman you can read my initial comments. <br />

<br />

I agree with Charles that the computer interface is only one variable to consider when designing a DAC. I have heard the MH ULM-8, Weiss DAC 2 and Berkeley Alpha on my system and preferred the sound of the ULM-8. I also feel that the Wavelength Silver Crimson with ESS Sabre 32 board (Denominator) sounds significantly better than the ULM-8. Is it solely based on the asynchronous interface? I don't think so.<br />

<br />

The one fabulous feature of Gordon's Crimson has been the modular design allowing me to update the DAC with minimal effort. <br />

<br />

Hopefully, I can find someone in the San Diego area that would be willing to bring his Weiss DAC 202 over for a listening session.<br />

<br />

Steve<br />

<br />

<br />

<br />

<br />

[cfmsp]

Here is Bj's response to my questions, which he replied to promptly despite that he was out of the office yesterday and today.<br />

<br />

My comments begin with '>'<br />

<br />

> I hate to bother you, but everyone is in a dither (bad pun intended) over at Computer audiophile as to how Metric Halo implements the Firewire interface on the ULN/LIO-8. It's always been assumed to be "Asynchronous" (as defined by Gordon and others at CA) in nature. Gordon upset all that yesterday. He [...] apparently exchanged emails with MH yesterday and then declared that [like the Weiss DAC202] neither do Metric Halo use an Asynchronous interface for Firewire (based on info in your FAQ, and possibly on emails).<br />

<br />

<br />

> Couple of questions, do the LIO-8 / ULN-8 DAC use any sort of PLL when using Internal Clocking mode via Firewire?<br />

<br />

No.<br />

<br />

> I believe that your DACs use fixed Frequency oscillators as the master (in internal mode).<br />

<br />

Yes.<br />

<br />

> How does the 8 series (in master clock mode) control the flow of data coming from the source Firewire node during Firewire-based audio playback?<br />

<br />

By sending feedback signal to the computer source so that it sends data at the correct pace.<br />

<br />

...<br />

<br />

By the CA definition of Asynchronous, the ULN-8 (and all MIO interfaces) are Asynchronous, and have been from the start (in fact, I think we were the first company to implement this way back in 2001).<br />

<br />

Our boxes transport audio via Isoch packets (as this is the most reliable way to do it on FireWire), but we do not use 61883-6 timing and we do not recover audio clock from the stream -- the stream from the MIO drives the computer's sense of clock asynchronously and that is the way it works.<br />

<br />

I don't know who Gordon may have talked to, but it was not me (I am out of the office) and I am sure that the person he talked to did not understand the peculiar usage of "asynchronous" that the Computer Audiophile's use (which has nothing to do with the physical transport layer on the bus), and gave him the standard answer of asynch vs. isoch packets.<br />

<br />

BTW, Gordon does have a point with the Weiss product since they use the DICE chip. We don't and our transport is Async as far as the CA term is used.<br />

<br />

<br />

end of BJ's response<br />

<br />

<br />

well, I'm no longer gobsmacked, am happy that my original understand that MH's Firewire is indeed as we all thought, but am really embarrassed at my contribution to the confusion in this thread. Hopefully, BJ's response clears up some of it. <br />

<br />

Hey, ELP, how do you like my new Mantra? "Metric Halo - Asynchronous Firewire interfaces since 2001!" <br />

<br />

later,<br />

<br />

Clay<br />

[Elprior]

<i><br />

Hey, ELP, how do you like my new Mantra? "Metric Halo - Asynchronous Firewire interfaces since 2001!" <br />

</i><br />

<br />

Cool, I like this one (since the answers were quite good).<br />

All in all, this is a very informative thread.<br />

<br />

Just try not to introduce the MH dacs to someone asking for an amplifier then <br />

<br />

Elp<br />

<br />

[EDIT]<br />

<br />

<i><br />

hope all is well, when this thread winds down, you'll hear a lot less from me.<br />

</i><br />

<br />

[Charles Hansen]

>> I am not from mainland China but but your comments of "Chinese prison camp" shows your ignorance and insensitivity. <<<br />

<br />

Insensitivity -- sure.<br />

<br />

Ignorance -- nope.<br />

<br />

Have you ever visited the places where A/V equipment is made in China? I have, starting over 10 years ago. Our Hong Kong distributor wanted us to take advantage of the then-new boom in loudspeaker cabinet factories. I visited three facilities in Guang-Zhou and PanYu. Only one was a "prison camp" in the sense that it had 10' high chain-link fencing with razor wire around the top. This fencing surrounded the compound that included both the factory space and the dormitories. In other words, the workers could not leave. Kind of like prison.<br />

<br />

The other two did not have the razor wire. But all three had abominable working conditions. The worst was the finishing rooms. People would work 12-hour shifts spraying speaker cabinets with no breathing protection. The fumes just about knocked me over just stepping into the room. I know the chemicals used in these paints, and I have no doubt that the workers are slowly rotting their brains, livers, and kidneys.<br />

<br />

Then to top it off, at the end of the day they would clean out their spray guns with solvent and dump the mess either down the drain (where it would go to the Pearl River or dump it it the back lot (where it would contaminate the ground water). There are probably thousands of Love Canals brewing in China:<br />

<br />

http://en.wikipedia.org/wiki/Love_Canal<br />

<br />

After that experience I swore I would never have anything to do with that system of production. It turns out that I was right. The problems have only gotten worse:<br />

<br />

http://www.telegraph.co.uk/finance/china-business/7773011/A-look-inside-the-Foxconn-suicide-factory.html<br />

<br />

So "insensitive", yes. But what would you call that? I think "prison camp" fits the bill pretty well.

[rlodad]

I have to agree with Charlie here. Not everything built in China is done under poor labor conditions and pollutes the environment, but many do. Before I buy a Chinese product, or a product made there, I try to check out the conditions the product is made under and the manufacturer's commitment to environmental pollution. Simply put, I have no Chinese made products in my system. If they can't produce kids toys without heavy metals, how are they going to produce audio kit without them? It is not insensitive, it is factual and important.

[labjr]

I just believe in patronizing companies who make stuff here. I know too many people who are out of work. If it costs more than so be it. Better to pay more and keep people working here. The quality is almost always better to boot.

[barrows]

Thanks for getting that confirmation from BJ, it is good to hear that the MH products are truly async in the CA understanding of the technique.<br />

It appears that we are mostly in agreement, but I did want to clear up one thing:<br />

<br />

"Barrows, perhaps you missed my point here. I was referring ONLY to the (gratuitous, imho) use of the word "Asynchronous" as a description of the nature of the PWT and Digital Lens products themselves (Paul was using one year ago to describe their buffering scheme), and not the PWD used alone or with the PWT/Network bridge. I only made this comment due to your inclusion of the PWT and Digital Lens in your comment about the use of Async"<br />

<br />

Not sure what you are referring to here? The PerfectWave Transport, as a stand alone product, is asyncronous in nature, the same way that the Halide Designs Bridge and ART Legato are asyncronous.<br />

The PWT works like this: A DVD ROM drive is used to read CDs or DVD data discs with wav files (like HRx), and this drive is controlled by powerful custom firmware running on a FPGA (BTW this reads the same way as EAC, going back and doing re-reads until the data is correct, so the process produces a bit perfect read, just like ripping a disc into a computer). The digital data is read from the drive, and the data alone (no clock) is dumped into a the memory, the control code running on the FPGA monitors the amount of data in the memory, to insure there are no overuns or underuns. At the output of the memory there are two fixed frequency oscillators (one for 44.1 base, and one for 48 base) and one of these clock out the data providing a very low jitter output. The clocks at the output of the PWT are asyncronous, they are fixed, and do not adjust their speed to sync to anything else. The Digital Lens (will) works the same way, with fixed frequency oscillators at the output. The output of both of these products produces a digital data stream with jitter levels very near the inherent level of the clock itself-same as an ART Legato, or Halide Designs Bridge.<br />

Now, if one uses the SPDIF output of the PWT into, say, a Berkely Alpha, one is getting the best possible SPDIF (or AES if preferred) data stream for the DAC. But the Alpha is going to increase the jitter a little, through the use of an SPDIF receiver and PLL for clock recovery. But, when the PWT is used with the PWD and connected by I2S, these problems do not exist-the clock in the PWT becomes the master for the DAC, and no PLL or SPDIF receiver is necessary, so the jitter level stays very low.<br />

<br />

I hope that this clears up any tecnical questions you might have as to how these products work.<br />

<br />

I suspect that fairly soon Chris will be getting a PWD with the bridge for review? I am interested to see how he ends up liking them. I have not heard the PWD with the bridge as a network player yet-theoretically this combo should produce excellent performance (I do know that the PWT-I2S-PWD produces superb sound, and if they did it "right" the network bridge should be at least as good, if not better).

[barrows]

I must note, I applaud any company that makes the committment to producing their products in the US. As a consumer, I am definitely willing to pay more for products made in the US when possible. At some point the Feds are going to have to level the playing field a little somehow, and hold China accountable for working/environmental conditions for products to be exported to the US.

[Pere Barceló]

Well,<br />

<br />

Maybe we should try to move this conversation to the forum. Don't know if a review is the space for this type of discussion. I'm thinking if I was the owner of Weiss Engineering I'd be disapointed.<br />

<br />

Can't resit a little comment though. I don't know of a single developed country that hasn't broken all rules environmentally and labor wise while it was "emerging". Let's keep that in perspective.<br />

<br />

Furthermore, I've heard many times that production quality is better in china than in mainland UK or US...and this from owners of very well respected companies.

[barrows]

General Forum for further China discussion.

[cfmsp]

<br />

<br />

Not to worry, it's for a good cause, I need to setup and fine tune an entirely new system in a new house, with much lower noise floor and much better AC (which I can modify to my hearts' content).<br />

<br />

I'll report progress!<br />

<br />

clay<br />

<br />

PS, perhaps I'll even get a pair of horn speakers and eliminate the power amp in my setup, driving them from the headphone out of the LIO-8. ;0<br />

<br />

<br />

<br />

<br />

<br />

[cfmsp]

<br />

Barrow,<br />

<br />

thanks for clarifying my misguided understanding of the word Asynchronous as it related to the PWT and Digital Lens.<br />

<br />

Yes, we're mostly in agreement. As I said, even when I thought that Gordon was correct in his belief that MH boxes were not asynchronous, I do believe there is a theoretical advantage.<br />

<br />

clay<br />

<br />

<br />

[Wavelength]

Clay,<br />

<br />

Thanks for the info on the MH stuff... Look I got that quote directly from their FAQ page. Someone should go back and update that so to reflect what they are doing.<br />

<br />

http://www.mhsecure.com/faq/index.php?action=artikel&cat=1&id=37&artlang=en<br />

<br />

I sent BJ a nice message and a link to the FAQ on their site that is wrong.<br />

<br />

Thanks<br />

Gordon

[Bob Stern]

Q for Daniel or someone who has their hands on a DAC202:<br />

<br />

In the Weiss Control Panel, what are the choices listed in the Operation Mode popup menu, and what do they actually do? Does the Operation Mode have any effect when the Sync is Internal?<br />

<br />

If the Operation Mode controls the length of an input buffer that determines latency, then presumably the maximum value would be OK if you are just listening to audio without video and don't care about latency.<br />

<br />

If the Operation Mode controls the time constant of the PLL, then presumably a slow time constant would give you the lowest jitter even though it may lose lock with high jitter sources.

[cfmsp]

Nice gesture on your part, but I'm sure that Bj knows as well as I do that you had no mal intent. From offline conversations with you, I've learned of your considerable fondness and respect for BJ and Metric Halo.<br />

<br />

<br />

And yes, I knew you go it off their FAQ page. As BJ said, the audiophile definition of Asynchronous (due to being new perhaps?) is different than how it is viewed WRT the physical Firewire protocol.<br />

<br />

I only wish, for both our sakes, that I had challenged your comments.<br />

<br />

There is another interesting tidbit on that FAQ:<br />

<br />

"A properly designed 1394 device may or may not implement isoch transfers, but *must* support async transfers." <br />

note: this is referring to the non-audiophile specific definition of async.<br />

<br />

<br />

<br />

Q: will my Proton ever work with the current iPad? <br />

Right now I'm using a Devilsound.<br />

<br />

cheers,<br />

clay<br />

<br />

<br />

<br />

[cfmsp]

"I do remember than there was an outcry about dCS's statement when they (in an FAQ) stated that FireWire couldn't be used in the same way as USB async. Maybe they were right?"<br />

<br />

I think that was marketing speak, which is disappointing to me coming from a firm like dCS. As I recall, it sounded like rationale for their choice of USB rather than Firewire. <br />

<br />

Contrarily, Gordon, who INVENTED USB DACs and developed the first Async USB DAC, doesn't make statements like that.<br />

<br />

<br />

clay<br />

[cfmsp]

<br />

<br />

"But, when the PWT is used with the PWD and connected by I2S, these problems do not exist-the clock in the PWT becomes the master for the DAC, and no PLL or SPDIF receiver is necessary, so the jitter level stays very low."<br />

<br />

So in this instance the PWT's Async Clock is the only clock? Is there not another clock in the DAC?<br />

<br />

As stated earlier, I believed the Async clock in the transport to be gratuitous. If it is the ONLY clock in an i2s scenario, i'll need to change my views. <br />

<br />

although then I'll have another question - how does having a clock several feet from the rest of the DAC NOT cause issues?<br />

<br />

listening and learning!<br />

<br />

clay<br />

[Bob Stern]

I think it would be helpful to adopt a term that more clearly suggests what audiophiles mean by async computer audio.<br />

<br />

My suggestion is "non-adaptive", "non-adaptive word clock", or "non-adaptive sample clock".<br />

<br />

By "non-adaptive" I mean that the DAC does not include any means for adapting to the sample rate of the source such as a PLL to extract the word clock from the audio data or ASRC (asynchronous sample rate conversion).<br />

<br />

I think Daniel, BJ and Gordon have clarified that the audiophile usage of async is misleading because a non-adaptive DAC need not use an async communication protocol for audio data.

[Bob Stern]

From the PS Audio literature, when the PWT transport and PWD DAC are connected via their I2S ports (HDMI cable), the transport sends the master clock to the DAC. Therefore, this is a true non-adaptive design like the so-called async designs.<br />

<br />

Because the master clock is in the transport rather than the DAC, the engineering challenge for the DAC to attenuate the jitter between its I2S input and the D/A conversion stage is similar to that when a conventional DAC uses an external word clock.

[jhbpa]

Reading all of you expert textperts (Chas H excepted) has left me in a state of non stop shaking in my boots, jitter if you will.<br />

<br />

Ahhhhhh, but I have a solution.....I'll grab a slab of vinyl and crank up the tubes.<br />

<br />

Carry on soldiers!

[chrille]

Chris how can I stop everything discussed here from also popping up in my private mailbox!???<br />

<br />

I have un-ticked both boxes for messages and still everyday my mailbox is full of this discussion!<br />

<br />

I only want to read this or any other thread here.<br />

I want my private mailbox to remain private.<br />

<br />

Quite often when I try to change the default setting of which order posts appear here my changes take no effect there either.<br />

Is it my computer or what's up?<br />

<br />

<br />

Help please!

[agentsmith]

Charles,<br />

<br />

Being from the west you and I know this is an easy target. I felt the same way as well before coming back to Hong Kong. Then I realize this is a complicated matter. I am sure traveling from pristine Colorado to a factory 10 years ago in China must have been a shock to your system. On the other hand do not forget that this hyper competitive production environment is largely originated by companies from the West. Surely Steve Jobs must at least know a little about the Foxxcomm modus operandi. But what did he do after the outbreak of bad publicity, he defended them, and you wonder why? Where else would they churn out those cheap and excellent iphones/iPads? Those cheap products has to come from somewhere.<br />

<br />

The whole issue is a lot more complicated than you'd think. Not all factories are like prison camps, compared to a factory in the US of course they are a long way from there but they are improving. They were put back by years of isolation from the rest of the world and they are playing catch up very quickly. <br />

<br />

In terms of environmental issues we are much worse than we think when it comes to treating the third world. As you know asbestos were banned from Canada for 10s of years, and yet Canada has been mining and exporting to third world countries and claiming that it is safe.<br />

<br />

http://www.minesandcommunities.org/article.php?a=9206<br />

<br />

[4est]

You need to hit "save" at the bottom .