Topping D50 III Balanced DAC with EQ Review

[Sokel]

Does that show the RME also with EQ on? Interesting....

Yes,in the linked thread you can see that the comparison is with identical filters.
What's more interesting than the noise is the fact that D50 filter is not peaking at the given 100Hz but at about 105Hz.

 

[Sokel]

This February, sarieri published his tests of the miniDSP Flex, and he has been answering members' questions in that thread. He may be able to let you know whether the EQ in the Flex produces troublesome levels of noise.

MiniDSP has partially addressed some issues with it's DSP down low.
You can see the difference between the first page measurements and the ones at about the end of the thread here.

 

[Henreid]

MiniDSP has partially addressed some issues with it's DSP down low.
You can see the difference between the first page measurements and the ones at about the end of the thread here.

That's very helpful. His conclusion following the tests:

"Clearly engaging filters in the SHARC based miniDSPs will result in measurable degradation at the DAC output. However, I think it will be very difficult/impossible to actually hear this as the increase in noise is all in lower frequencies and level dependent. I also do not see anything that will still not clear 16 bit fidelity."

 

[ah-ra]

MiniDSP has partially addressed some issues with it's DSP down low.
You can see the difference between the first page measurements and the ones at about the end of the thread here.

Puh... Thanks a lot for pointing me to that thread, but I can't say I really understand the technical/theoretical part behind the discussion.

I understand that there seems to be a problem with the sharc DSPs in the miniDSP units and that it can be avoided reducing the input level by 3db in the digital domain... and there's no real solution for that if you are using CD player or Streamer as source via toslink/spidf...

I also understand that most people believe the effect on Sinad introduced by EQ does not really matter as it's inaudible.

So I'm more confused than before... what are my options now?

 

[Sokel]

Puh... Thanks a lot for pointing me to that thread, but I can't say I really understand the technical/theoretical part behind the discussion.

I understand that there seems to be a problem with the sharc DSPs in the miniDSP units and that it can be avoided reducing the input level by 3db in the digital domain... and there's no real solution for that if you are using CD player or Streamer as source via toslink/spidf...

I also understand that most people believe the effect on Sinad introduced by EQ does not really matter as it's inaudible.

So I'm more confused than before... what are my options now?

I wouldn't worry much if I was you.
If you test it and no noise is heard at your listening position you're good to go,there's not much else to this.

It's us with the measurements obsession who usually debate such stuff.
Otherwise it should be inaudible.

 

[ah-ra]

I wouldn't worry much if I was you.
If you test it and no noise is heard at your listening position you're good to go,there's not much else to this.

It's us with the measurements obsession who usually debate such stuff.
Otherwise it should be inaudible.

You are probably right. Thing is that I'm currently using a 2x4HD between pre and power amps and wanted to upgrade to the minidsp Flex, partly due to it's better measurements and to skip the additional AD-conversion (still not sure whether this has any negative effect...) with the 2x4HD. I wanted to keep EQ in the digital domain.

I'm using minidsp for bass management (two subs) plus IIR and also FIR filters for phase correction.

The test of the Topping DAC now got me thinking... I would go for a solution like Camilla DSP or Acourate, but it's not as convenient as miniDSP and always requires a PC within the setup.

 

[Henreid]

So I'm more confused than before... what are my options now?

There are a couple of approaches you might take to resolve your concerns. When I am considering a major purchase and have lingering questions about some aspects of the product's performance, I will usually scour the Internet for reviews and forum discussions to see if the issues I'm wondering about are being encountered by others. If you find little to no complaints about signal degradation from the miniDSP Flex Digital, you may be able to buy with more confidence. Another option would be to purchase from a seller like Deer Creek Audio, offering a 30-day return window. Unfortunately, that benefit will cost you extra. Their price is $54 higher ($549) and if you decide to return the Flex you'll need to pay for shipping, plus a 15% restocking fee.
Like you, I am often mystified by the charts and graphs that are produced - as Sokel admits - by those who are obsessed with measurements. But after their tests are complete and they tell me that the device contributes no audible noise or distortion, I tend to have faith in their conclusions.
[Since the subject of this thread is the Topping D50 III we're going a bit off-topic with the miniDSP Flex. It's probably better to address questions about the Flex on this thread:
Minidsp Flex Review (Audio DSP)]

 

[ah-ra]

There are a couple of approaches you might take to resolve your concerns. When I am considering a major purchase and have lingering questions about some aspects of the product's performance, I will usually scour the Internet for reviews and forum discussions to see if the issues I'm wondering about are being encountered by others. If you find little to no complaints about signal degradation from the miniDSP Flex Digital, you may be able to buy with more confidence. Another option would be to purchase from a seller like Deer Creek Audio, offering a 30-day return window. Unfortunately, that benefit will cost you extra. Their price is $54 higher ($549) and if you decide to return the Flex you'll need to pay for shipping, plus a 15% restocking fee.
Like you, I am often mystified by the charts and graphs that are produced - as Sokel admits - by those who are obsessed with measurements. But after their tests are complete and they tell me that the device contributes no audible noise or distortion, I tend to have faith in their conclusions.
[Since the subject of this thread is the Topping D50 III we're going a bit off-topic with the miniDSP Flex. It's probably better to address questions about the Flex on this thread:
Minidsp Flex Review (Audio DSP)]

Thanks for your thoughts, but as I live in Germany this is no option for me...

I don't think it's okf-topic, as it might also apply for other DSP solutions. The first time I read about this "problem" was here in this thread about the D50III, that's why I asked here, but I will go through the Flex-thread again.

 

[nanook]

Did anyone have the chance to have a look at the PCB of the Topping D50 III ?
Toppings marketing material suggests, that they might have implemented the I/V- stage that is shown in the ES9039pro datasheet. To my opinion this is the smartest circuit topology I've seen so far.
--> In case someone has pictures of the PCB (particularly the I/V- stage) I'd be curious.

It should look similar to this:

 

[Henreid]

Did anyone have the chance to have a look at the PCB of the Topping D50 III ?
Toppings marketing material suggests, that they might have implemented the I/V- stage that is shown in the ES9039pro datasheet. To my opinion this is the smartest circuit topology I've seen so far.
--> In case someone has pictures of the PCB (particularly the I/V- stage) I'd be curious.

It should look similar to this:
View attachment 383531

That exact component is shown on the product page of the Topping D50 III, as featured by this retailer:
Apos Audio

 

[Big White]

Which is brighter sound signature E50 or D50 III ?

 

[Bleib]

Which is brighter sound signature E50 or D50 III ?

Neither. Unless early roll off filters are used.

 

[Henreid]

Which is brighter sound signature E50 or D50 III ?

They're DACs. Their purpose is to convert a digital audio signal to analog audio without adulterating the sound. Topping's high-quality DACs are not meant to have a sound signature of their own. However, the D50 III differs from the E50 by including an equalizer that allows users to adjust the frequency response to suit their needs. So if desired, the user can adjust the D50 III to "sound brighter" by boosting the treble frequencies. The E50 does feature 3 different high-frequency filters allowing attenuation of the signal above 18kHz (though I wonder how many of us who are middle-aged or older would be able to hear those adjustments...! )

 

[nanook]

DACs have reached a level, where I personally can hardly believe that there are audible differences.
Measurement - wise there are differences and the ES9039xxx (Hyperstream IV generation) is a very well behaved Sigma-Delta DAC chip. ESS obviously has addressed some of the weaknesses of the SD-technique with this generation compared to Gen II (ES9038 / ES9068) .
Measurement- wise the Topping D50 III is likely the cleanest implementations of this chip generation (dual ES9039q2m) that I have seen (see review on ASR) .
On top you have a multi-band parametric equalizer when using the USB input (afaik parameters can be changed via windows only).
I had just ordered another SMSL D-6s - otherwise I would have tried the D50 III for sure.

 

[trungdtmc]

Did anyone have the chance to have a look at the PCB of the Topping D50 III ?
Toppings marketing material suggests, that they might have implemented the I/V- stage that is shown in the ES9039pro datasheet. To my opinion this is the smartest circuit topology I've seen so far.
--> In case someone has pictures of the PCB (particularly the I/V- stage) I'd be curious.

It should look similar to this:
View attachment 383531

I'm also interested in this circuit part of the d50iii, but currently there is only 1 video, no one has actually tested it. There are a lot of sot23-6 components here, I can't determine specifically what they are

 

[herr.jemine]

this board contains opamps for I/V conversion

 

[trungdtmc]

this board contains opamps for I/V conversion

Yes, I know that, but there are many ways to design I/V circuits, and the D50iii doesn't seem like a classic I/V circuit, like the one used with SMSL's DACs.

 

[Henreid]

Did anyone have the chance to have a look at the PCB of the Topping D50 III ?
Toppings marketing material suggests, that they might have implemented the I/V- stage that is shown in the ES9039pro datasheet. To my opinion this is the smartest circuit topology I've seen so far.
--> In case someone has pictures of the PCB (particularly the I/V- stage) I'd be curious.

It should look similar to this:
View attachment 383531

I'm also interested in this circuit part of the d50iii, but currently there is only 1 video, no one has actually tested it. There are a lot of sot23-6 components here, I can't determine specifically what they are
View attachment 384057

The exact circuit you have shown is included as part of the promotional material for the D50 III. Here is a direct quote from that material:
Brand New I/V conversion

"A brand new I/V conversion circuit is designed for better audio reproduction. The new circuit design provides lower distortion and lower noise while occupying less space. Designed to approach physical limits in order to maximize the full performance of the DAC chip."

Below is the photo from the D50 III promotional material. If you require more information it may be helpful to contact Topping with your questions.

 

[nanook]

Yes, I know that, but there are many ways to design I/V circuits, and the D50iii doesn't seem like a classic I/V circuit, like the one used with SMSL's DACs.

I had reverse engineered the I/V - and unbalanced stage of the SMSL D-6s. SMSL uses a fairly straightforward topology with the Vref for the I/V-stage derived from AVCC using a resistive divider with some LP-filtering. This resistive divider needs to be adjusted to obtain 0V at the unbalanced outputs (trim pots).
The disadvantage is that the common-mode content that is present on the outputs of the DAC chip shows up on both legs of the balanced output.

The topology shown in the ES9039pro datasheet uses one more OpAmp that picks up the average common mode signal on balanced output and acts as an integrator (with a very fast time constant) that provides the Vref that is required in the I/V stage.
- no offset adjustment required
- near zero Common-Mode noise on the balanced output

Sure, the differential input stage of the connected preamp or active speaker will subtract this CM-noise and eliminate it to a large extent (just like the RCA output stage inside the DAC does), but I would rather avoid having this CM-noise at the output of the DAC.

 

[nanook]

Attached is the IV-stage of the SMSL D-6s. I have not de-soldered the capacitors, so their values are unknown.

For comparison the I/V-stage proposed in the ES9039Pro datasheet. It has been discussed here:

Let's talk about the I/V circuit with common mode servo as shown by ES9039pro datasheet

I recently came about the I/V circuit proposed by ESS's ES9039pro datasheet. Firstly let's show it here: I will try to describe how I think it works, let's see how many misconceptions I have about it... There is a "deglitching" cap (C18.1) directly around the DAC outputs. I remember that...

www.diyaudio.com