Topping Centaurus R2R DAC Review

[NTTY]

Interesting, I saw a very similar issue with the SMSL D200 in "No filter" mode. The internal DAC does not offer that functionality so it might be a forced deactivation of the oversampling filter. I failed to understand the root cause and that was my comment when I reviewed the DAC. If the same is seen from very different DACs, then it’s worth investigating.

 

[rocksteady]

There's a reason no major manufacturer (Sony, Philips) has ever released an R2R DAC.

I do understand the urge to design and build one oneself, I'm a tinkerer too. But performance and audio-wise the IC DAC's are a much better solution.

And then there's SABRE...

Say what? Both Philips and Sony have released several TDA1541A DACs, especially the latter. The TDA1541A WAS a R2R Chip.What do you think R2R means? Denon also made R2R DACs, only in this case with BB chips, usually.

 

[Jimster480]

Say what? Both Philips and Sony have released several TDA1541A DACs, especially the latter. The TDA1541A WAS a R2R Chip.What do you think R2R means? Denon also made R2R DACs, only in this case with BB chips, usually.

I think it's also important to note that those chips are indeed integrated circuits as well, and this guy mentions that the IC DACs are just way better, and it's obvious that he doesn't understand that there were, and are, R2R integrated circuit DAC chips for over 30 years.
Every major company use them at one point in time.
While to me it's pretty obvious that modern day Delta Sigma designs have surpassed the performance of the age-old ladder resistor style DAC, it's important to know that we do still use these exact types of chips for volume control today.

 

[Roland68]

I think it's also important to note that those chips are indeed integrated circuits as well, and this guy mentions that the IC DACs are just way better, and it's obvious that he doesn't understand that there were, and are, R2R integrated circuit DAC chips for over 30 years.
Every major company use them at one point in time.
While to me it's pretty obvious that modern day Delta Sigma designs have surpassed the performance of the age-old ladder resistor style DAC, it's important to know that we do still use these exact types of chips for volume control today.

These volume control chips have just as little to do with R2R DACs as discrete, relay-based volume controls.
It's only because some idiots started calling these relay-based volume controls R2R that they're also labeled as such. Here, resistors are simply inserted into the signal path as attenuators or as ground connections (voltage dividers), and these are static. These components also exist in purely mechanical form as SMD-based potentiometers.

 

[cany89]

TDA1541A WAS a R2R Chip

Except it’s not an R2R…

 

[Roland68]

Except it’s not an R2R…

Are you saying that the first DAC ICs from Philips and Sony, including the 1541, did not use the parallel method and that the data was processed with a resistor network?

 

[cany89]

including the 1541, did not use the parallel method and that the data was processed with a resistor network?

Parallel yes. Resistor network (for R and 2R), no. The TDA1541 architecture is segmented current-steering, not a resistor ladder. You can check it out yourself.

PS: I’m talking about TDA1541A because that’s a common misconception. Even DAC manufacturers such as SW1X use it and market as R2R which is false.

 

[VientoB]

Parallel yes. Resistor network (for R and 2R), no. The TDA1541 architecture is segmented current-steering, not a resistor ladder. You can check it out yourself.

PS: I’m talking about TDA1541A because that’s a common misconception. Even DAC manufacturers such as SW1X use it and market as R2R which is false.

How is the current steered?

 

[cany89]

How is the current steered?

Each bit controls a switch that routes the current source to either the output node or to a dummy node. The sum of all active current sources forms the analog signal.

I had some bookmarks about it. There is a good discussion on diyAudio about R2R discrete networks and ICs, and how most ICs are actually not R2R (the old fashioned way with resistors and ladder...)

For those interested, additional information about TDA1541A can be found here.

 

[Mnyb]

What they probably want to market with the moniker R2R is some kind of true multibit architecture 16 bits or more , however that is defined ?

 

[cany89]

Well, multibit doesn't mean it's R2R. You can also argue about what is true R2R, since most discrete or IC R2R DACs are segmented. Meaning they don't use the resistor ladders for both MSB and LSB. However, chips like the AD1862/AD1865, for example, utilize a resistor ladder. Contrary to TDA1541A...

 

[Billy Budapest]

Besides the “if it’s what’s audiophiles want, why not offering it?” reason, I still wonder how much of this is driven by the AKM factory fire aftermaths, when OEM’s realized their extreme dependence upon a couple of chip suppliers.

That could be it. Maybe too it’s an irresistible engineering challenge and they knew it would sell to a certain segment who perhaps would not have considered Topping before.

 

[CedarX]

That could be it. Maybe too it’s an irresistible engineering challenge and they knew it would sell to a certain segment who perhaps would not have considered Topping before.

I wonder if the Topping implementation and HiBy’s Darwin-III are similar—@Joe Bloggs, the HiBy rep., had an interesting post on Head-Fi, about their Darwin-III solution:

https://www.head-fi.org/threads/new-hiby-flagship-rs8-ii.977742/post-18833501/

If I understand it right, they are not seeking unobtainum precision resistors. Instead, they measure each individual R2R ladder, and adjust the parameters of a correction model running in the FPGA. Still an R2R DAC, but with its imperfections compensated in the digital domain.

 

[mocenigo]

I wonder if the Topping implementation and HiBy’s Darwin-III are similar—@Joe Bloggs, the HiBy rep., had an interesting post on Head-Fi, about their Darwin-III solution:

https://www.head-fi.org/threads/new-hiby-flagship-rs8-ii.977742/post-18833501/

If I understand it right, they are not seeking unobtainum precision resistors. Instead, they measure each individual R2R ladder, and adjust the parameters of a correction model running in the FPGA. Still an R2R DAC, but with its imperfections compensated in the digital domain.

The correction is an interesting idea, but at the end it is overengineering to match results that can be obtained with much less expensive components. If one wants one of those pieces of jewellery (such as HiBy’s Darwin-III), I am not gonna stop them, but I personally not interested — unless somebody can show a concrete (and audible) SQ advantage.

 

[bigpilot]

Say what? Both Philips and Sony have released several TDA1541A DACs, especially the latter. The TDA1541A WAS a R2R Chip.What do you think R2R means? Denon also made R2R DACs, only in this case with BB chips, usually.

Let me rephrase that: "Philips and Sony have never released discrete R2R DACs"

 

[agnostic]

I dont think anybody says vinyl is more detailed.

There are many people who advocate that analogue is lossless but digital is not.

 

[antcollinet]

There are many people who advocate that analogue is lossless but digital is not.

That took you a while

 

[Roland68]

There are many people who advocate that analogue is lossless but digital is not.

These are people who don't understand the technical or metrological background of analog and digital data transmission and believe any old nonsense.
Analog data cannot be transmitted without any change occurring; you just have to scale the measurement up far enough.
Digital data consists of 0s and 1s. As long as this remains intact or can be restored, there can be no change to the data.
This applies regardless of the area where it's used, not just audio.

You do realize the enormous amounts of digital data that are transmitted worldwide every single day without anything being lost or changing, right?

 

[Haskil]

R2R had a poor domestic audiophile reputation at the start. The earliest non-oversampling CD players were R2R and critics called them bright, shrill, gritty, glassy, shouty, lacking in bass etc.

Now, according to audiophiles, these bright, gritty DACs are smooth and "analogue". So, "audiophile status" is valueless.

Excuse me for reviving such an old thread, but no, when the first CD players R2R came out, the prevailing opinion in the press was that they were a major improvement over LPs, as I have read several times in these exchanges.

I attended the initial pre-release presentations when the Marantz was displayed on a table with a skirt, its electronics hidden underneath. Later, in my office at Le Monde la Musique, I had the Philips and Sony number 1 CDs with the first CDs released by Sony and Philips-DGG-Decca... mostly remasters of analog tapes. So I owned the LPs of most of them (Beethoven and Schubert by Bruno Walter and George Szell), but also Lorin Maazel's digital Stravinsky, Gould Goldberg digital, and Vladimir Horowits' analog recording (incidentally, the same tapes would later be used for SACD, which was single-layered at Sony).

And the comparisons were unanimous: the CD crushed the LP in terms of treble smoothness, stereo image clarity, and bass tightness, and when you reached the end of the LP side, the comparison became terribly unfair. The turntables were Thorens 125 with a long SME tonearm and an Ortofon MC cartridge, and Sony with a tangential tonearm and a Shure V 15V MM cartridge... And at the time, all the music critics and hi-fi journalists were pro-CD... I was member of a jury hifi in this time.

It was only a few months, even a good year, later that criticism started appearing in hi-fi magazines. These magazines were no longer buoyed by Philips and Sony's CD-related advertising, but instead reverted to the usual ads from small hi-fi brands unable to market CD players, yet clinging to their LP turntables, cartridges, and so on. That's when people started talking about the coldness of digital and the grainy sound of the treble. Some imaginative people even heard the dots they saw on the published curves... What a rip-off! I've recently heard several CD 104s (second or third generation Philips), Toshiba models from 1984-1985, Denon OEM models from another Japanese manufacturer, and the first Yamaha in the series, and they still deliver excellent sound today—not cold or grainy at all.

 

[milosz]

There are many people who advocate that analogue is lossless but digital is not.

Analog is NOT lossless. If you take the original master tape and compare them electronically to vinyl playback, you will find there is loss between the master and the vinyl.