Review and Measurements of Wyred4Sound DAC-2v2 SE DAC

[GeekyBastard]

This is the main concern, the analog output board:

View attachment 21100

Those plate things are foil resistors. Precision is only 0.1% here (they can have a precision as low as 0.005%). The good thing is they are very stable when temperature changes. They are expensive, too. About $20-30 each. 10 foils are used here. They are worth of $200 to $300 alone!

Now here comes the issue. Looks like it's a discrete based IV. Through hole transistors/FETs? here. Are those transistors perfectly matched? Thermally coupled? And they should have used SMD parts for better signal quality. No NFB to control errors? No wonder this I/V can not match OP based designs, and you have left/right channel mismatch problem.

At least we can see other than the booming and harmonics, the DAC output is generally OK. The bass is still pretty good. You may lose a bit details and pin pointed images yet in general it will have a pleasant sound which is a bit different from OP based design. This is why people will still buy this DAC for its different sound signature.

Great technical details, thanks to you, Sir!

 

[finneybear]

Great technical details, thanks to you, Sir!

You are welcome! Oh, one more thing. Two sets of foils are used here. This means they cost $400-$600 already.
High end audio gears usually have a part cost around 20% to 30% of the retail price. For a $4000 DAC, a lot of budget is already spent on those foils.

 

[EJ Sarmento]

Hello all,
Just wanted to drop a note how we have many things going against us here. For one, understand that these DACs have discrete output stages. The ones you rate so highly and compare it to are Op-Amp implementations which have an incredible amount of local and global feedback to provide good numbers. This doesn’t come free and that is why we opt to not use them where possible. Im not posting here to debate on the legitimacy of Femto clocks, naked Z-Foil resistors, discrete output stages etc. We all have opinions and if we dont agree, I understand that and we can move on. Given all this, we must remember that a system will only be as revealing as the weakest link!

I would like to point out a few things. First is that measurements matter and are good guidelines but sometimes zeroing out distortion doesnt always yield the same result.

*Most of our DACs have discrete output stages (meaning no op-amp in the audio signal). Noise figures that you are looking for would only be possible when using ICs in the design – such as the ESS Sabre reference design or the Oppo products. In my experience, harmonic distortions have always been more inherent in discrete designs because there aren’t 100s of parts involved which we try to avoid. There are some downsides to this but we feel the upside is worth it.

*Our DAC-2s have a digital volume control. The product is optimized around a volume setting of 53 which is 2v out on this particular unit –enough to clip our most power amplifier. As it is, customers use this DAC with other manufactures amps and some have far less gain which we have to work with as well. As the volume goes down there is a level of digital truncation happening depending on the level selected so we try to aim for a usable range that is still in the sweet spot. Its hard to be a product so flexible without any compromises. If you set it in different settings you will see different results – I’m not claiming 120db just pointing out that it is reliant on the level. I believe 97db is a more accurate number. If you look at noise by itself it is even better.

*The Femto clock is an ultra-precision and low jitter clock – this is the clock that runs the DAC and dictates its internal timing that paves the path for its operation. If you look at the reporting of the Oppo, you will see it is very similar once you consider the scaling differences – actually, our noise floor of the measurement appears to be about 15db better!

*Remote – I have the very first remote we ordered with the rubberized coating which is still not sicky. What we have found is that cleaners with chemicals to disinfect happens to also break down the rubberized coating. We have some without this coating for those who are interested.

*The cost of the unit surely isn’t cheap and we don’t claim to be the least expensive. How the price is formulated is directly proportional to the cost. This unit has particularly high cost in the naked Z-foils that are hand made in Texas, the ESS chip as we know it, the Crystek clock as noted, Dale copper leaded/non inductive resistors (if not naked Zs), and most of all – labor to build the unit. The metal work is sourced local from our county as well. All of this adds up to a much more expensive product than what can be had from others who import from overseas.

If there was a direct correlation to what we hear vs what we measure then we would all be able to produce the exact same sounding products. I believe that there are things we can hear but not measure, and measure that we cannot hear.

 

[EJ Sarmento]

I would like to point out the Oppo 205 measurement (Green) (rated best on the chart) superimposed on the DAC-2v2se (Blue). Note that the scaling had to be altered to show the Oppo graph in similar form and not sure why it was measured at a different frequency but take note of the noise floor between the two. "Poor performance"? we are talking -145db+ from a non integrated circuit output stage! Poor-really?

 

[Blumlein 88]

Hello all,
Just wanted to drop a note how we have many things going against us here. For one, understand that these DACs have discrete output stages. The ones you rate so highly and compare it to are Op-Amp implementations which have an incredible amount of local and global feedback to provide good numbers. This doesn’t come free and that is why we opt to not use them where possible. From what I read, it is very clear to me that we are not considering differences in sound such as timbre, texture, delicacy, attack, decay, emotional connection, etc. because we can get that from a $100 box right? Im not here to debate with (and I won’t) on the legitimacy of Femto clocks, naked Z-Foil resistors, discrete output stages etc. We all have opinions and if it’s not in your religion, I can understand that and we can move on. Given all this, we must remember that a system will only be as revealing as the weakest link! To an extent I feel like my time here is wasteful based on some of these posts, however there are some that might find it helpful.

I would like to point out a few things. First is that measurement numbers are not always what matter the most..

*Most of our DACs have discrete output stages (meaning no op-amp in the audio signal). Noise figures that you are looking for would only be possible when using ICs in the design – such as the ESS Sabre reference design or the Oppo products. In my experience, harmonic distortions have always been more inherent in discrete designs because there aren’t 100s of parts involved which we try to avoid. There are some downsides to this but we feel the upside is worth it.

*Our DAC-2s have a digital volume control. The product is optimized around a volume setting of 53 which is 2v out on this particular unit –enough to clip our most power amplifier. As it is, customers use this DAC with other manufactures amps and some have far less gain which we have to work with as well. As the volume goes down there is a level of digital truncation happening depending on the level selected so we try to aim for a usable range that is still in the sweet spot. Its hard to be a product so flexible without any compromises. If you set it in different settings you will see different results – I’m not claiming 120db just pointing out that it is reliant on the level. I believe 97db is a more accurate number. If you look at noise by itself it is even better.

*The Femto clock is an ultra-precision and low jitter clock – this is the clock that runs the DAC and dictates its internal timing that paves the path for its operation. If you look at the reporting of the Oppo, you will see it is very similar once you consider the scaling differences – actually, our noise floor of the measurement appears to be about 15db better!

*Remote – I have the very first remote we ordered with the rubberized coating which is still not sicky. What people don’t understand is that cleaners with chemicals to disinfect happens to also break down the rubberized coating. we have some without this coating for those worried about it but for the most part is not a problem.

*The cost of the unit surely isn’t cheap and we don’t claim to be the least expensive. How the price is formulated is directly proportional to the cost. This unit has particularly high cost in the naked Z-foils that are hand made in Texas, the ESS chip as we know it, the Crystek clock as noted, Dale copper leaded/non inductive resistors (if not naked Zs), and most of all – labor to build the unit. The metal work is sourced local from our county as well. All of this adds up to a much more expensive product than what can be had from others who import from overseas.

If there was a direct correlation to what we hear vs what we measure then that magic test box would allow us all to produce the exact same sounding products. I would like to think you all realize that there are things you can hear but not measure, and measure that you cannot hear. Glad to see that there are a couple of people mentioning they enjoy this dac and that it doesn’t sound as bad as the pink panther suggests - Thank you to those of you who did! I also talked to the individual who sent this unit in who noted to me after this review that he still has it in his main system just that he wishes there was some good numbers behind what he was hearing. Rightfully so but what really matters is what we hear and most of what we are talking about here is so minuscule on the measurement side that it should be accepted for benefit that it has rather than the meter it does not peg.

I own an amp of yours. This ridiculous post makes me want to sell it and get something from a reputable supplier. I'd have been more diplomatic usually, but by the time I finished your post I didn't see the point in it. There so much disingenuous in it. Or worse perhaps you believe it.

So IC's are a hindrance despite lower noise, and lower distortion, but some quantity missed by the measures are important to sound? The IC's are incapable and your discrete units are capable even while measuring worse.

 

[Blumlein 88]

I would like to point out the Oppo 205 measurement (Green) (rated best on the chart) superimposed on the DAC-2v2se (Blue). Note that the scaling had to be altered to show the Oppo graph in similar form and not sure why it was measured at a different frequency but take note of the noise floor between the two. "Poor performance"? we are talking -145db+ from a non integrated circuit output stage! Poor-really?

View attachment 21109

Do you understand how to read noise levels on an FFT? And the effect of bin size on the numbers you read? Or do you hope that many reading your post don't?

The raw noise level of your product wasn't the big notable. A higher though likely inaudible third harmonic distortion and the raising of the noise floor around high level tones with your femto clock were a bit suspect.

 

[Thomas savage]

Cardinal Dennis has spoken.

 

[rebbiputzmaker]

Cardinal Dennis has spoken.

 

[restorer-john]

There are some downsides to this but we feel the upside is worth it.

What exactly is the 'upside'? Can you articulate it in a non-nebulous way?

Remote – I have the very first remote we ordered with the rubberized coating which is still not sicky. What people don’t understand is that cleaners with chemicals to disinfect happens to also break down the rubberized coating. we have some without this coating for those worried about it but for the most part is not a problem.

My personal experience with any and all rubberized satin-feel coatings is they all go sticky. Some quickly, others take years. Everything from electric beard trimmers to Nikon and Canon DSLR camera body grips, torches, remote controls, keyboards and computer mice.

The stuff is dreadful and it has absolutely nothing to do with chemicals in cleaners. I have had several sticky remotes, in fact I have a Philips one here that went sticky new and unused in its box. We made a rule in our house never to buy anything with a satin-feel rubberized coating ever again.

what we are talking about here is so minuscule on the measurement side that it should be accepted for benefit that it has

Again, specifically, what is the actual 'benefit' such objectively poor measured performance brings to the table? Seriously, 1st generation CD players in 1983 tested better than this device.

Do you feel the reviewed sample is not a representative sample of the typical performance or is the design just inherently faulty?

 

[EJ Sarmento]

I own an amp of yours. This ridiculous post makes me want to sell it and get something from a reputable supplier. I'd have been more diplomatic usually, but by the time I finished your post I didn't seen the point in it. There so much disingenuous in it. Or worse perhaps you believe it.

So IC's are a hindrance despite lower noise, and lower distortion, but some quantity missed by the measures are important to sound? The IC's are incapable and your discrete units are capable even while measuring worse.

You are correct, my appologies. long day and kids screaming at me and I'm out of line. I will edit what I didnt mean to be offensive.

 

[Purité Audio]

Amir’s reviews really are incredibly valuable aren’t they?
Didn’t anyone measure anything from 1974 ...
Keith

 

[Veri]

Amir’s reviews really are incredibly valuable aren’t they?

Worth their character count in gold

 

[miero]

@EJ Sarmento do you perform in-house listening tests against your previous or competitor's products? if yes, are the volume matched and listeners don't know what they are listening to? Also it would be interesting to know how do you decide if these or other components will be used - is it based on listening or also practical reasons. For example I know that thin film SMD resistors are almost unobtainable for small manufacturers in these days...

 

[VintageFlanker]

Now... femtoclock? What was that brand that made a rubidium word clock again?

RME proved a FS clock can bring improvements in jitter treatment. See improved measured jitter performance ADI-2 DAC (FS) vs PRO (NON-FS).

Anyway, the only conclusion is: Wyred4Sound doesn't seem to know how to properly design a DAC and should be ashamed to ask 4K$ for it.

 

[Killingbeans]

My personal experience with any and all rubberized satin-feel coatings is they all go sticky. Some quickly, others take years. Everything from electric beard trimmers to Nikon and Canon DSLR camera body grips, torches, remote controls, keyboards and computer mice.

Got the same experience. But the worst is those easy-grip screwdrivers where the rubber/polyurethane(?) starts smelling strongly like vomit when it breaks down. Nothing like struggeling with an unruly screw then smell your hand and go: "Mmmm, vomit..."

 

[graz_lag]

... Anyway, the only conclusion is: Wyred4Sound doesn't seem to know how to properly design a DAC and should be ashamed to ask 4K$ for it.

Sure, however how many of us in here have been adding a W4S's DAC in their short list for their new DAC ?
(Based upon the excellent subjective reviews on quite reputable sites this W4S has got over the years ...)
Personally, twice at least, and because I buy only second-hand, I never jumped on the right occasion, luckily ... I can say it now after this gem of review by @amirm
As a matter of fact, reviews - from any source, are worth nothing, if not backed by engineering-based measurements.

 

[maty]

*Most of our DACs have discrete output stages (meaning no op-amp in the audio signal). Noise figures that you are looking for would only be possible when using ICs in the design – such as the ESS Sabre reference design or the Oppo products. In my experience, harmonic distortions have always been more inherent in discrete designs because there aren’t 100s of parts involved which we try to avoid. There are some downsides to this but we feel the upside is worth it.

DACs, without tubes, must have the harmonics at least below 90 dB. A good/very good DAC, below 100 dB. $3800, more below.

Here, balanced, with 4 V -> +3 dB

H2: 90+3= -93 dB, OK

H3: 85+3 = -88 dB

And H3 is higher than H2! The world upside down.

 

[maty]

WesionTEK Khadas Tone Board DAC

https://www.audiosciencereview.com/...ents-of-wesiontek-khadas-tone-board-dac.4823/

And the harmonic profile has to be ALWAYS monotonically decreasing. DAC, preamp, amplifier...

 

[VintageFlanker]

Found these RMAA (16bits) measurements by HD-Fever. This was the DAC-2 DSD SE, back in 2016.

 

[graz_lag]

Yes, that was one of the review that convinced me to put that particular DAC into my short list in 2016, with some doubt though 'cause the 16/44.1 sampling mode ...
Went for the NAD M51 instead in that occasion.
It might well be "It is possible they just slapped the new chip on their 9018 boards to make it look sexier" for this DAC-2v2 SE version as well said by @gvl