Topping D900 DAC Review

[RichB]

Chip implementation certainly measure wonderfully but not necessarily better on all measurements.
Filtering looks better on the D900.

Topping D900.

OCTO Dac8 similar in what seems to be a linear phase fast filter 6.

- Rich

 

[Vovgan]

Topping pleeese do AV processor with this level of DAC performance, with no forced internal 24 bit 48Khz conversion for Audyssey curves as Denon does, and we’ll pay for them, I promise!

 

[RichB]

Topping pleeese do AV processor with this level of DAC performance, with no forced internal 24 bit 48Khz conversion for Audyssey curves as Denon does, and we’ll pay for them, I promise!

The upcoming MiniDSP Tide16 has the 48kHz limitation, but their published measurements are excellent. Coming in April and hopefully, to be reviewed on ASR. Here is the thread:

miniDSP Tide16 - Holy Grail with 16 Channel Atmos/DTS:X, high SINAD

Just received the newsletter, sounds like the one to rule them all (rather expansive at 3500 USD, but ok) https://www.minidsp.com/products/ht-series/tide16 Hardware Features Processor: Quad Core ARM processor / 1.8GHz HDMI 2.0 + EARC/ARC interface (3x1) Included License: Dirac Live® 3.x full...

www.audiosciencereview.com

- Rich

 

[welwynnick]

What's the point of this device? What makes it "custom"?
Why would it be worth so much more over the other Topping dacs, and wth does upsampled to 1 bit mean?

It's more expensive BECAUSE it has a proprietary converter instead of an off-the-shelf delta-sigma DAC chip.
Some subjectivists say they prefer R2R or 1 bit DACs to delta-sigma DACs, and this is Topping's attempt to meet that market.
Topping also make an R2R DAC, the Centaurus.

 

[Blockader]

What's the point of this device? What makes it "custom"?

Why would it be worth so much more over the other Topping dacs, and wth does upsampled to 1 bit mean?

Seems very expensive for, really, nothing but high voltage to reduce amp gain and thus a small amount of noise that's probably already imperceptible.

I think this device is a definitive answer to everyone who doubted Topping's engineering expertise. If you've spent any time on audio forums, you know Topping has been dismissed for years as a company that simply sources the latest chip and slaps on −100 db of negative feedback. That was always the go-to criticism: "That's not real engineering. They just grab the newest DAC chip, pile on negative feedback with cascading op-amps, and call it a day. They don't actually know what they're doing. They're not a real engineering company. They don't have the expertise to design a 1-bit DAC or an R-2R DAC."

Basically this is a big fuck you from Topping to haters and showing others(who sell their dacs for 14000$) how it's done. I don't think you are the target audience here. They beat Chord in their own game while also making the dac 8 times cheaper.

 

[respice finem]

It's more expensive BECAUSE it has a proprietary converter instead of an off-the-shelf delta-sigma DAC chip.
Some subjectivists say they prefer R2R or 1 bit DACs to delta-sigma DACs, and this is Topping's attempt to meet that market.
Topping also make an R2R DAC, the Centaurus.

Exactly. I can't blame them for "harvesting" both "camps". A business must make profit to survive.

 

[phoenixdogfan]

It's a wonderful piece-- it comes in a beautiful case, and its performance is exemplary. But it's $1800, and it's $1800 because it's a 1 bit rather than a standard delta sigma converter that comes on an ESS or AKM chip. And the latter are every bit as good if not better and available at a fraction of the cost of all "handcrafted" circuit Dacs. Well at least it's not an FPGA like the Chords. Maybe it's like a kind of "halfway house" for the people coming over here from Positive Feedback, offering them "the best of both worlds' for a price, of course. Because they just "can't quit" Gizmo!

Also, it really cranks out the voltage which is great for those who have low gain input stage amps.

Amir, as always, a truly impeccable review. Thank you.

 

[Roland68]

What's the point of this device? What makes it "custom"?

Why would it be worth so much more over the other Topping dacs, and wth does upsampled to 1 bit mean?

Seems very expensive for, really, nothing but high voltage to reduce amp gain and thus a small amount of noise that's probably already imperceptible.

The price of such devices is comprised of the development costs, which are likely many times higher for this particular device, the cost of each unit produced, including components, housing, packaging, etc., and the company's share of the costs.

Up to this point, the price per unit is primarily determined by the anticipated number of units sold, because the fewer devices sold, the higher the cost per unit must be.

Then there are the usual percentage markups, such as handling and transport costs, which are often underestimated and depend heavily on the device's size, profits for distributors/distribution, profits for local retailers (who also bear the costs of warranty service), platform fees such as those charged by eBay, Amazon, AliExpress, etc., and, of course, customs duties, taxes, and VAT, depending on the country.

Fortunately, only the manufacturer decides whether to develop such a device, bring it to market, and take on the associated risk. Customers decide whether the device is worth the asking price or not, and whether it has a reason to exist. Ultimately, that's the only measure that matters.
The device has apparently sold well so far, and after this test, sales should improve even further.

If only it weren't for the problem with warranty claims (which depends on where the device is purchased) and the lack of repair guarantees for the next 10-20 years.....

 

[Bleib]

What's the point of this device? What makes it "custom"?

Why would it be worth so much more over the other Topping dacs, and wth does upsampled to 1 bit mean?

Seems very expensive for, really, nothing but high voltage to reduce amp gain and thus a small amount of noise that's probably already imperceptible.

The high voltage is possible with E50 II obviously..

 

[El Baron]

Topping pleeese do AV processor with this level of DAC performance, with no forced internal 24 bit 48Khz conversion for Audyssey curves as Denon does, and we’ll pay for them, I promise!

Have a look to miniDSP

 

[jazbrokr]

The D900 offers crossfeed too. This is a big deal for those who listen to older recordings with hard panned instruments. I think only a couple of other Topping DACs and the Luxsin offers this.

 

[respice finem]

The D900 offers crossfeed too. This is a big deal for those who listen to older recordings with hard panned instruments. I think only a couple of other Topping DACs and the Luxsin offers this.

Like the RME ADI-2 DAC, for less than half of its price - provided you can still get one. Bonus: Excellent headphone amp.

ADI-2-DAC FS - RME Audio Interfaces | Format Converters | Preamps | Network Audio & MADI Solutions

RME Audio Professional Live, Studio, Recording and Broadcast Solutions. Unrivalled Quality, Performance & Stability MADI Interfaces, Converters & Preamps.

rme-audio.de

 

[gvl]

Filtering looks better on the D900.

Does it with only 7db attenuation at half sampling rate aka Fn?

 

[NDRQ]

Kinda disappointing.
From this company, we have seen better stuff for way less money.

 

[TurtlePaul]

What's the point of this device? What makes it "custom"?

Why would it be worth so much more over the other Topping dacs, and wth does upsampled to 1 bit mean?

In case these are asked seriously:

In the opamp world, there are integrated circuits that cost a few dollars which internally have dozens of transistors. Nonetheless, companies like Muses and Sparkos have made devices on PCBs with separate discrete transistors, capacitors and resistors. These are called discrete opamps.

DACs typically use integrated circuits from TI, AKM or ESS. These chips have hundreds of transistors and capacitors inside them, including digital logic, the DAC bits buckets, output filters, etc. In this device, Topping has designed their own (very large) DAC circuit board with all of these components broken out instead of using a single chip implementation.

There is no reason that discrete proprietary devices should sound better but some audiophiles like it.

Regarding single-bit: DACs get fed information at a certain sample rate (e.g. 48 khz) and bit depth (e.g. 16 bits). However, if you have a 15-bit DAC working at 96 khz, you can send the same data. A 15 bit DAC wouldn’t be able to output the even numbers from a 16 bit DAC (where the least significant bit is not zero) but you are operating at double the clock rate so you output the number rounded up one cycle and rounded down the second cycle. This makes a -90 dB noise at 48 khz, but the output filter will quash that.

It is quite easy to demonstrate that each doubling of frequency can remove at least a bit of required sample depth. The math is complex (as all delta-sigma designs rely on noise shaping), but by operating a DAC at a few mhz, the full dynamic range and frequency range of a 16/44.1 signal can be achieved with a single bit DAC.

The advantage of a 1-bit DAC is that for a multi-bit DAC you need precision matched resistors to be able to make the various voltages. A 1 bit DAC uses a single voltage across a single transistor feeding a single capacitor. This single device will always match itself. There is also a cost advantage for single-bit designs but this is canceled out by Topping doing a discrete circuit.

 

[Nopiano]

Maybe a silly question but -

Does it do the old de-emphasis flag which some early and still playable discs have?

I wish you hadn’t reminded me that I’m also old enough to remember, and own some of, those. You’ll be asking about HDCD next!

 

[DSJR]

I wish you hadn’t reminded me that I’m also old enough to remember, and own some of, those. You’ll be asking about HDCD next!

Nah, never liked that , as the discs processed that way all seemed to have added reverb on them - highly subjective and well over twenty five plus years ago now...

 

[Thomas_A]

Perhaps not related to this particular review, which is excellent as usual. My question is more general and that is if there are any DACs that do not drop like 0.2 dB at 20 kHz for 44.1 kHz? Not sure if it has been consistently measured like that. Not that it is audible, but I was just wondering if this is typical between these quite good DACs or not, since those drops add to other potential drops (e.g. 0.2-0.5 dB @ 20 kHz) of various components in the chain.

 

[antcollinet]

Perhaps not related to this particular review, which is excellent as usual. My question is more general and that is if there are any DACs that do not drop like 0.2 dB at 20 kHz for 44.1 kHz? Not sure if it has been consistently measured like that. Not that it is audible, but I was just wondering if this is typical between these quite good DACs or not, since those drops add to other potential drops (e.g. 0.2-0.5 dB @ 20 kHz) of various components in the chain.

Not sure where you are seeing that drop. Both this DAC, and the Octo Dac8 whose filter plots are posted above (lilic/default) both look flat to 20kHz to me.

 

[Thomas_A]

Not sure where you are seeing that drop. Both this DAC, and the Octo Dac8 whose filter plots are posted above (lilic/default) both look flat to 20kHz to me.

@amirm said and shows in the plot: "With the tiniest bit of droop at 20 kHz (0.2 dB):"