[jimmychan]

I have just bought my R2R7 recently. I upgraded from Reference 1.
The R2R7 seems has a better DSP, that I tried various inputs and the differences are not very minimal.
The Transports that I have tried are dCS Network Bridge, Aurender N10, Naim Uniti Core, RaspberryPI USB & RPI/Hifi Berry Digi+

 

[Currawong]

What were the differences with the different transports? How would you rate them?

 

[barbas]

I have just bought my R2R7 recently. I upgraded from Reference 1.
The R2R7 seems has a better DSP, that I tried various inputs and the differences are not very minimal.
The Transports that I have tried are dCS Network Bridge, Aurender N10, Naim Uniti Core, RaspberryPI USB & RPI/Hifi Berry Digi+

Hi Jimmy, can you please elaborate a little bit your findings regarding inputs and sources differences? Many thanks.

 

[jimmychan]

I found dCS NB has most of the energy, its energetic, fast and punch.

Aurender N10 is very close to dCS, only when playing some music like symphony, it lose a little bit on the depth of image. Overall is good.

RPI/Hifi Berry Digi+ and RaspberryPI USB are good at casual listening, you won't easy to spot the differences right away when you switched between dCS and RPI. You have to listen very careful to find its weakness. When you compared RPI with dCS, it is not day and night. RPI USB can play DSD too.

Note:
The above comments based running on Roon Server and set the transports to Roon end points.
This is just a causal quick test and not done formally.

 

[Chopin75]

HI ALL,
i am new to Head-Fi and like to join you guys in this discussion of R2R 7 which I am quite interested in. I have been reading this interesting thread for the last few days but since it stretches all the way back many months with 64 pages, I am have missed quite a bit. For a while I have been thinking of getting R2R DAC (right now i have ESS based LH Geek Pulse) but I can't really decide which one to get. I have been thinking seriously about the Metrum Pavane or its Junior Onyx. Has anyone done or known of anyone who has done comparison between the Metrum DAC and Audio-GD R2R 7? Interestingly they both use some form of FPGA correction technique. Metrum uses "FPGA-corrected 24-bit Transient modules" that someone uses at least 2 DAC chip and split signal into 1/2 (12+12 bit) and stick them together again and apparently Audio-GD uses FPGA for every bit of the R2R ladder process (if I understand correctly, sorry I am not electronic engineer).
Would be interested to know how R2R sounds compared with Metrum.
And I just discovered this ANK (audio Note) DAC 4.1 or 5.1 which supposed to be an awesome tube output R2R DAC too!
Too many choices of R2R now!

 

[FredA]

HI ALL,
i am new to Head-Fi and like to join you guys in this discussion of R2R 7 which I am quite interested in. I have been reading this interesting thread for the last few days but since it stretches all the way back many months with 64 pages, I am have missed quite a bit. For a while I have been thinking of getting R2R DAC (right now i have ESS based LH Geek Pulse) but I can't really decide which one to get. I have been thinking seriously about the Metrum Pavane or its Junior Onyx. Has anyone done or known of anyone who has done comparison between the Metrum DAC and Audio-GD R2R 7? Interestingly they both use some form of FPGA correction technique. Metrum uses "FPGA-corrected 24-bit Transient modules" that someone uses at least 2 DAC chip and split signal into 1/2 (12+12 bit) and stick them together again and apparently Audio-GD uses FPGA for every bit of the R2R ladder process (if I understand correctly, sorry I am not electronic engineer).
Would be interested to know how R2R sounds compared with Metrum.
And I just discovered this ANK (audio Note) DAC 4.1 or 5.1 which supposed to be an awesome tube output R2R DAC too!
Too many choices of R2R now!

In the case of the r2r 7, you're talking about a sign magnitude balanced design. Discrete. High-end approach really. The r2r 7 is very accurate. Mine a keeper for as long as i can foresee. My friend who bought my Master-7 is a very happy camper too. i also have a r2r 11 at the office. It is worthy of being used in a main setup.,it's great as a dac.

 

[christianvogel]

HI ALL,
i am new to Head-Fi and like to join you guys in this discussion of R2R 7 which I am quite interested in. I have been reading this interesting thread for the last few days but since it stretches all the way back many months with 64 pages, I am have missed quite a bit. For a while I have been thinking of getting R2R DAC (right now i have ESS based LH Geek Pulse) but I can't really decide which one to get. I have been thinking seriously about the Metrum Pavane or its Junior Onyx. Has anyone done or known of anyone who has done comparison between the Metrum DAC and Audio-GD R2R 7? Interestingly they both use some form of FPGA correction technique. Metrum uses "FPGA-corrected 24-bit Transient modules" that someone uses at least 2 DAC chip and split signal into 1/2 (12+12 bit) and stick them together again and apparently Audio-GD uses FPGA for every bit of the R2R ladder process (if I understand correctly, sorry I am not electronic engineer).
Would be interested to know how R2R sounds compared with Metrum.
And I just discovered this ANK (audio Note) DAC 4.1 or 5.1 which supposed to be an awesome tube output R2R DAC too!
Too many choices of R2R now!

The Metrum Pavane is a really interesting option. My dream Pavane is the Adagio (https://metrumacoustics.com/adagio-det/), so smart to have the volume control in the DAC itself! I considered the Pavane lvl-1 along with the Holo Spring Lvl 3 and the R2R-7. I lusted most for the Pavane, but keep in mind that has three levels, and even the entry level Pavane is considerably more expensive than the R2R-7. One of the reasons to go for the Pavane for me was that the R2R-7 felt like a bit of a gamble, there were hardly any reviews. What I read about the Pavane was that is was an amazing NOS DAC, like a more refined Holo Spring, with a sound that grows on you. I considered going 2nd hand on a Pavane lvl 1, but even then, you would pay the price of a new R2R-7.

Inside they look so different, and I love the way the Pavane is built. Very clean and thoughtful design. In comparison, the R2R-7 feels more hardcore, and impressive in its own way.

A friend and I tend to validate every new purchase on its merits, and we've burned down expensive own purchases before, we tend to have a reverse version of "buyers positive bias". However, now I have the R2R-7, and we agree that the gamble paid off. I cannot tell you if it is better than the Pavane, but I can tell you it is very good indeed.

I got the impressions from reviews that the Onyx, while very good, just is not at the same level as the Pavane and R2R-7. However, I heard there may be a version coming with volume control built in, like the Adagio, which saves you a preamp. Also, if you like MQA, go for Metrum, as they are offering MQA support now.

 

[christianvogel]

In the case of the r2r 7, you're talking about a sign magnitude balanced design. Discrete. High-end approach really. The r2r 7 is very accurate. Mine a keeper for as long as i can foresee. My friend who bought my Master-7 is a very happy camper too. i also have a r2r 11 at the office. It is worthy of being used in a main setup.,it's great as a dac.

I would love to see a comparison between the R2R-11 and the Chord Mojo. They seem to be in a similar price range: R2R-11 380 euro vs Mojo 500 euro.

The Mojo seems a little more practical, since it is very mobile, and the possible expansion with a Poly is a nice option to have.

However most curious about the sound comparison.

 

[Chopin75]

Hi @Wynnytsky , I read in several places that while DSD is 1-bit signal, the implementation detail in most chips is that they process the digital signal to actually use 6 to 8 bits to reduce the DSD-to-analog noise floor.

I am no expert on this, so just guessing here, but perhaps there is similar digital processing happening in the FPGA, which then gets fed to some of the steps of the ladder. Or, perhaps the R2R7 has a separate implementation path for this, like the Holo Spring DAC.

Perhaps mr @Kingwa can tell us?

I have been trying to figure this out too. According to 6 moon, it uses the PFGA technique
"This DAC uses a proprietary digital filter simply called DSP-1 written to an Altera Cyclone IV processor. This FPGA performs all the oversampling calculations and is responsible for the processing of DSD. The R2R 7 is thus two DACs in one. DSD is not converted to PCM but processed as DSD all the way". Someone put this quote too.
I wish Audi-GD can confirm this officially.
The interesting thing is PS-Audio uses a similar FPGA as well ! Though for PS-Audio they also allow PCM signal to go through same pathway. I am not aware of R2R way of doing DSD , as apparently Holoaudio is doing that kind of DSD decoding, mentioned somewhere on this thread. (maybe I misunderstood). Now if R2R 7 is employing similar technique as the $4000-6000 PS audio DAC then this is great bargain. The question is, does sound gets compromised with all these circuits in same small box? Also, how does DSD sound on R2R 7, ? much better than Delta-signa? Apparently Holoaudio's DSD is also really amazing, suggesting DS is not ideal for DSD especially when only 2ch is needed. Can R2R 7 be adjusted to output only as DSD no matter what the input files are (using jumpers) ?? Or can it be modified that way, just curious about its flexibility

 

[Wynnytsky]

I have been trying to figure this out too. According to 6 moon, it uses the PFGA technique
"This DAC uses a proprietary digital filter simply called DSP-1 written to an Altera Cyclone IV processor. This FPGA performs all the oversampling calculations and is responsible for the processing of DSD. The R2R 7 is thus two DACs in one. DSD is not converted to PCM but processed as DSD all the way". Someone put this quote too.
I wish Audi-GD can confirm this officially.
The interesting thing is PS-Audio uses a similar FPGA as well ! Though for PS-Audio they also allow PCM signal to go through same pathway. I am not aware of R2R way of doing DSD , as apparently Holoaudio is doing that kind of DSD decoding, mentioned somewhere on this thread. (maybe I misunderstood). Now if R2R 7 is employing similar technique as the $4000-6000 PS audio DAC then this is great bargain. The question is, does sound gets compromised with all these circuits in same small box? Also, how does DSD sound on R2R 7, ? much better than Delta-signa? Apparently Holoaudio's DSD is also really amazing, suggesting DS is not ideal for DSD especially when only 2ch is needed. Can R2R 7 be adjusted to output only as DSD no matter what the input files are (using jumpers) ?? Or can it be modified that way, just curious about its flexibility

And once the basic DSD questions are answered, my next question would be ...
When the jumpers are set to 2/4/8 OS, does that have any effect on DSD playback? (I'm assuming it doesn't correlate to DSD 128/256/512)

I wouldn't suggest that the psaudio direct stream and r2r7 are similar in design. The r2r7 has the coveted discrete resister ladders that I would think are only of use with PCM, and I read years ago that the PS Audio internally over/up-samples PCM 10 times, then converts that to DSD. Until Kingwa speaks up I'll always be wonder if the r2r7 is perhaps converting DSD to 8xPCM to make use of that ladder.

 

[DACLadder]

Remember DSD is a very simple approach to digital audio. Pulse density modulate 1-bit (full on and off) and then to play back run the resulting pulses through a low-pass filter to reconstruct an audio signal. So, all you need to do is use the most significant bit in a ladder dac as the DSD 1-bit. No conversion to PCM is needed. The DSD low-pass filter has to be in the analog stage immediately after 1-bit D/A (not at the DSP.as with PCM).

Not sure how DSD is truly implemented in the R2R 7. But the digital board Altera FPGA used does not have enough internal resources to convert DSD to PCM on the fly plus store all the PCM functionality - filtering, OS/ NOS, etc..

 

[Chopin75]

Remember DSD is a very simple approach to digital audio. Pulse density modulate 1-bit (full on and off) and then to play back run the resulting pulses through a low-pass filter to reconstruct an audio signal. So, all you need to do is use the most significant bit in a ladder dac as the DSD 1-bit. No conversion to PCM is needed. The DSD low-pass filter has to be in the analog stage immediately after 1-bit D/A (not at the DSP.as with PCM).

Not sure how DSD is truly implemented in the R2R 7. But the digital board Altera FPGA used does not have enough internal resources to convert DSD to PCM on the fly plus store all the PCM functionality - filtering, OS/ NOS, etc..

Thanks for the reply! I also wonder if running in DSD, any oversampling done for DSD? Are any jumpers needed to manipulate the DSD at all? I guess I should write directly to Audio-gd about this. It does not appear R2R would convert DSD to PCM at all nonetheless.

 

[Currawong]

I have been trying to figure this out too. According to 6 moon, it uses the PFGA technique

An FPGA is equivalent to a computer. It doesn't have software on it at all. Just because two DACs use an FPGA, doesn't say anything about the software that has been written for it. The reason for using an FPGA is that you can have a lot of computing power in a device with your own software, without having a full-blown computer.

 

[Chopin75]

(love how you can resume a headfi msg on desktop after starting on mobile)

I'm getting as much bass as the Gustard X20 served up, but the integration with the mids (transition?) is the best I've heard. When I see a DAC with this much attention to the power section I expected it to be an over achiever with bass and to have controlled highs, but these things are supposed to take time to mature. I don't even have 20 hours -- must be the 300 at the factory.

The mids match the tone/texture of a great tube DAC. There's also lot's of extension which one would think would render confusion in complex recordings, yet my live recordings have never sounded this involving.

I guess I'm saying, if [like me] your coming from a sub-$1000 DAC (and your I2S interface is on point) then you'll have plenty to be entertained with from day 1.

Look forward to taking this on the road next month and hearing how the presentation changes for other speakers (Analysis Omega, Magico S5 Mk2, Odyssey Kismet Liquid, Revel F206)

HI Wynnytsky, do you mind telling me what source you are using ? The thread is so long it is hard to figure this out. Do you use I2S too either directly from streamer or via a DDC? Thanks!

 

[Chopin75]

An FPGA is equivalent to a computer. It doesn't have software on it at all. Just because two DACs use an FPGA, doesn't say anything about the software that has been written for it. The reason for using an FPGA is that you can have a lot of computing power in a device with your own software, without having a full-blown computer.

Thanks for clarifying, Currawong! (On a side note, hope you don't mind me asking why is your name is partly "Wong" and you have Chinese Calligraphy on your profile, but from your video of reviews, clearly you do not appear to be Chinese!)