Weiss DAC501 Streamer and DAC Review

[KSTR]

^ Good points.
I think the problem in this case is an uncommonly large extra latency inside the DAC on top of the total ASIO USB driver latency. If that extra latency is stable and predictable (for each sample rate) then indeed the ASIO driver could report the correct total latency to millisecond precision and the AP and its operator would be happy ;-)

 

[Billy Budapest]

Did anyone reviewed here the sophisticated EQ menus based on Weiss intellectual property?

@MikeSQ Found one for you—Michael Lavorgnia addressed EQ in his review. I suggest you read that article for an examination of the Weiss 501 features and Amir’s review for the measurements.

Review: Weiss DAC501: Beauty With Brains - Twittering Machines

Michael Lavorgna's Twittering Machines: HiFi with a twist

twitteringmachines.com

 

[AdrianusG]

We do know it uses TI DACs though, and they don’t have any models that have SOTA performance. So one has to come with a whole lot of tricks to actually improve these by at least another order of magnitude to get even close. Now, the box is big enough… so who knows

For me it's strange that some of these high-end companies keep using Dac Chips that long have been surpassed, specs-wise (and prove-able through measurements!) by newer products from AKM, ESS and even Cirrus, it's weird.

I mean, if you are gonna sell a product for 200K$ the price of a Dac chip could hardly be a consideration for them now could it.

 

[muslhead]

I mean, if you are gonna sell a product for 200K$ the price of a Dac chip could hardly be a consideration for them now could it.

That is absurd ... of course it does if you are trying to optimize your profit margin. Everything on the bill of material for this device matters if you are in the business of making a profit.
Last time i checked, Weiss is not a non-profit
If you switch hats from that of an engineer to that of a CEO, you would come to a different conclusion

 

[Sokel]

For me it's strange that some of these high-end companies keep using Dac Chips that long have been surpassed, specs-wise (and prove-able through measurements!) by newer products from AKM, ESS and even Cirrus, it's weird.

I mean, if you are gonna sell a product for 200K$ the price of a Dac chip could hardly be a consideration for them now could it.

Yep,I can see the marketing appeal of an R2R scheme for example but using inferior parts (even daisy-chained,paralleled or whatever) seems odd.
There are arrays of resistors in some of these DACs (I'm not talking about the R2R ones) that each one of the resistor costs more than the DAC chip itself (I have repeatedly see nice MP930 Caddocks in some of them)

 

[IAtaman]

For me it's strange that some of these high-end companies keep using Dac Chips that long have been surpassed, specs-wise (and prove-able through measurements!) by newer products from AKM, ESS and even Cirrus, it's weird.

I mean, if you are gonna sell a product for 200K$ the price of a Dac chip could hardly be a consideration for them now could it.

You need to redesign the PCB unless the new chips are pin compatible. With the DAC chip, components also change most likely. You need to retest and validate everything works fine together and there are no weird bugs from different input / outputs combinations You need to update the UI probably for new filters and/or features so that is a firmware update. These are all the things I can think of - in reality there might be other considerations as well. And all that for what, 2db unnoticeable improvement? Don't change a working system trumps in this case in my opinion.

 

[voodooless]

That is absurd ... of course it does if you are trying to optimize your profit margin. Everything on the bill of material for this device matters if you are in the business of making a profit.
Last time i checked, Weiss is not a non-profit
If you switch hats from that of an engineer to that of a CEO, you would come to a different conclusion

Imagine the cost of milling 150 kg of aluminum… the cost of a DAC chip is nothing in comparison. If you pretend to make high performance DAC, the chip should not be the thing to skimp on. Never mind all the electronics that you’ll need to put around it to tickle out a bit more performance. That does add up quickly. No, a SOTA DAC chip is way more cost effective than the nonsense they offer.

 

[AdrianusG]

You need to redesign the PCB unless the new chips are pin compatible. With the DAC chip, components also change most likely. You need to retest and validate everything works fine together and there are no weird bugs from different input / outputs combinations You need to update the UI probably for new filters and/or features so that is a firmware update. These are all the things I can think of - in reality there might be other considerations as well. All that for what, 2db unnoticeable improvement? Don't change a working system trumps in this case in my opinion.

Sure,

But they could have implemented all that long ago, it's not as if ESS or AKM chips came available last week for the first time, they have been available many many years.

 

[Sokel]

Imagine the cost of milling 150 kg of aluminum… the cost of a DAC chip is nothing in comparison. If you pretend to make high performance DAC, the chip should not be the thing to skimp on. Never mind all the electronics that you’ll need to put around it to tickle out a bit more performance. That does add up quickly. No, a SOTA DAC chip is way more cost effective than the nonsense they offer.

I think that some of them just keep their backs covered.
Their warranties included everything,some few (or even some more,depends on the dealer) include home pick-up and return,etc.
I imagine that this cost is considerable and they would prefer to avoid it no matter what.

So,a proven rock-solid device that has worked without problems must be their way to go.

 

[AdrianusG]

Sure,

But they could have implemented all that long ago, it's not as if ESS or AKM chips came available last week for the first time, they have been available many many years.

Its possible however that they want to stay with a specific "house sound' though, similar to IFI which sticks with Burr Brown chips no matter what.

 

[Purité Audio]

The more arcane the product the greater the allure , you all know that fantastic quote from Nelson Pass, ‘it was rare exotic and poisonous the perfect audiophile component’

Keith

 

[Purité Audio]

Its possible however that they want to stay with a specific "house sound' though, similar to IFI which sticks with Burr Brown chips no matter what.

There is no ‘house sound’ between chips there are better and poorer measuring, the manufacturer can contrive either knowingly or unknowingly to mess things up.
Keith

 

[Sokel]

There is no ‘house sound’ between chips there are better and poorer measuring, the manufacturer can contrive either knowingly or unknowingly to mess things up.
Keith

Do you offer home pick-up and return Keith?
And if so,to all your products,some of them or only by extra charge?

 

[AdrianusG]

There is no ‘house sound’ between chips there are better and poorer measuring, the manufacturer can contrive either knowingly or unknowingly to mess things up.
Keith

True, the final product and it's performance is of course much more than the Dac-chip used.

 

[Purité Audio]

Do you offer home pick-up and return Keith?
And if so,to all your products,some of them or only by extra charge?

Generally I ask interested parties to visit us first, to make sure they are serious and then to narrow down between D&D and Kii say, then I take the preferred speakers to the customers home install/measure and then just leave them to it. I don’t charge.
I firmly believe that ultimately you have to hear everything in your space to make a really valid comparison.
Keith

 

[IAtaman]

Sure,

But they could have implemented all that long ago, it's not as if ESS or AKM chips came available last week for the first time, they have been available many many years.

Upgrade for sake of upgrade do have an appeal indeed - people want the latest product with the latest chip, not because they were sweating all the 16 billion transistors of the A16 Bionic but because why get the old one if there is a newer and better one. And manufacturers who rely on that do launch new products every year, with newer features and chips and ideally a revised price tag. There is definitely a market for such consumption habits. I personally don't think super high-end audio is such a market.

 

[Oldasdrt]

This is a review, listening tests and detailed measurements of the Weiss DAC501 balanced DAC, headphone amp and network renderer. It is on kind loan from a member and costs US $9995.
View attachment 316239
The case is fine enough but not nearly so for what it costs. Stamped steel should have no place in this price range. The LCD is decent size but not very high resolution. It is touch sensitive but wasn't easy to navigate using that. It uses inverse scrolling in that to move the menus up, you rotate the knob to the left, not right. Fortunately a remote control is provided which allows quick random access to inputs and such. Back panel shows what you would expect as far as connectivity:
View attachment 316242

What distinguishes this streamer becomes visible once you connect to it using a web browser. In there, you see a suite of very useful signal processing components including parametric EQ, vinyl simulation, Room EQ (just a set of filters), etc. Here is the UI for the DSP:
View attachment 316247

The interface was easy to use and much nicer than screwing around with an app. Here is for example the parametric EQ for Room EQ (you program them manually):

View attachment 316248

For some odd reason though, when you select headphone out, you loose all that and only get a set of preprogrammed Audeze settings:
View attachment 316249

Fortunately the generic EQ remains with three filters:
View attachment 316250

You can create multiple presets and select them on demand using the remote.

There is a DAC502 which is the same but is wider and has balanced headphone out. Otherwise it is the same.

If you are not familiar with the tests that are about to follow, please watch my video on understanding DAC measurements.

Weiss DAC501 DAC and Streamer Measurement
Let's adjust the volume control such that we get 4 volts and measure the output using USB input:
View attachment 316251
This is good performance but I am left wondering why there are so many distortion spikes. That seems to do with its output buffer stages as lowering the gain there, does improve things:
View attachment 316252
But as you see, you can no longer reach 4 volt output even at max volume. Using the higher value, the DAC501 lands in our "Excellent" category:
View attachment 316253
Zooming in:
View attachment 316254

We have under $100 DACs outperforming it. That said, distortion is at threshold of hearing so likely transparent to the source. Here is the RCA performance with 0 dB Trim:
View attachment 316256

Since output can go over 4 volts, I ran my sweep of distortion+noise relative to digital level:
View attachment 316260
As noted, the curve has those ups and downs which I have not seen before. Same thing happened when I ran this test but used the front panel volume control. We will circle back to this later.

Since I know one of you rascals would ask , here is the vinyl emulation:
View attachment 316257
I got a kick out of having the noise floor artificially increased in addition to incredibly high amount of distortion.

I was impressed with the noise performance:
View attachment 316258

Somehow that didn't translate to better performance in IMD vs level:
View attachment 316259

I liked that multitone distortion was totally frequency independent:
View attachment 316261

There is no option to change the filter:
View attachment 316262
Attenuation is good enough to give us a good wideband THD+N vs frequency:
View attachment 316263
Usually in audio gear distortion increases with frequency but here is slightly backward. A bit strange.

I was surprised (again) that despite low noise floor, the jitter measurement did not reveal the low order bits of the J-test signal on the left:
View attachment 316264
We also have a series of spikes that should not be there in an audio product in this price level and from an engineering company.

Now check out the linearity measurement:
View attachment 316265

As Spock would say, "fascinating!" This test simply varies the digital input to the DAC, expecting corresponding change in analog level. A perfect level would produce a line at 0 dB. What we see is that at precise digital input levels, we have a 1 dB error! Cleary the low order bits of digital audio samples are getting corrupted. Yes, I have all effects disabled which should cause the unit to pass the PCM samples through but clearly it is stepping on them. It is likely a software bug that should have been caught during design verification.

EDIT: further investigation shows the above is due to possible data loss/corruption/latency using USB driver. Toslink produces perfect response: https://www.audiosciencereview.com/...01-streamer-and-dac-review.48377/post-1737069

I have reported my findings to Danie Weiss, hoping a fix will be provided.

Finally, for compatibility with Stereophile DAC measurements, here is the performance of 50 Hz tone into 600 ohm load:
View attachment 316267
It beats an MSB DAC by nearly 18 dB so not bad:

Weiss DAC501 Headphone Measurements
The dashboard once again shows competent performance but not where it should be relative to its class:
View attachment 316268View attachment 316269

I was disappointed in noise performance as it was not great in absolute and was worse in one channel:View attachment 316270

View attachment 316271

There are four trim levels. I tested three of them:

View attachment 316272
View attachment 316273

These are OK power levels but again, not befitting of the class.

Output impedance is zero which good but there is just not a lot of current available:
View attachment 316274

Volume control is digital so no need to run channel imbalance test. But recently people have been doubting this so here it is:
View attachment 316275

DAC501 Listening Tests
As usual I start with my closed back Dan Clark Stealth headphone. This is a low impedance headphone that requires fair bit of drive. The DAC501 could only get it to moderate levels above which bass notes got distorted. Push harder and distortion kept increasing. It is useable with this headphone for background listening but no kind of foot tapping will follow.

I then switched to Sennheiser HD650 which is high impedance and much easier to drive. Here, the DAC501 managed to produce quite high dynamics to levels that you don't want to listen to for more than a few seconds. I listened to few tracks and sound was quite enjoyable but I did not detect anything euphonic or special about the DAC501. For comparison, I plugged the HD650 into my everyday ADI-2 Pro headphone out and I was now able to play much louder with even more fidelity (strictly due to higher output level).

Conclusions
I did some initial testing of the DAC501 a week ago (basically the dashboard) and then looked at all the processing it provided. I was completely sold on it and was willing to ignore its price and give it very high mark. Today though, those hopes were dashed. There is no reason for a DAC to not do the basics of what it is asked to do: convert digital samples to corresponding analog samples at accurate levels. This test ignores all noise and distortion, and I only measure to 20 bits so a DAC needs to be able to nail it. As noted in the review, the DAC501 clearly has an implementation bug that a simple test would have caught. I am hopeful that the company will fix it with a firmware update. If not, the hardware should be recalled.

The other less than good news is the headphone output performance. While better than a checklist item, it is anemic when it comes to driving lower impedance headphones. It is also rather noisy with one channel being worse that the other.

Finally, going into the review somehow I thought this was a $2,000 DAC/streamer. Today I double checked and was shocked that it was $10,000! I don't mind paying that much if it came in a 20 pound case with machining that would make me drool. And a high resolution display akin to a mobile phone. None of that is provided so I don't see how that price was ever justified. We have higher performance streamers selling for less than $1000 with all of these features.

The one thing this streamer has going for it is its signal processing. With a bit of modification (such as allowing Room EQ filters to be used for headphones), it could be a major advantage over its competitors at lower prices. But they need to fix the DAC before this matters.

I am very saddened as to not be able to recommend the Weiss DAC501.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Crazy price tag for lousy performance,wow
you definitely don't get what you pay for here.
Thanks for the great review

 

[FWLarson]

I’m not certain all that we hear is measured, or we know all the performance measurements that represent what we hear.

 

[Ken1951]

I’m not certain all that we hear can be measured, or we know all the performance measurements that represent what we hear.

Asked and answered. For about 1,000 times.

 

[Billy Budapest]

So what is the final verdict—bug in the Weiss firmware and measurement shortcoming of the AP?