GUSTARD DAC-R26 Balanced Decoder R2R+1Bit Dual Native Decoding Music Bridge
Feb 4, 2024 at 10:22 PM Post #8,371 of 8,835
@BShaw It is a bit out there /mad scientisty, that's for sure. 😅

Btw having had another quick listen today and doing a bit more experimentation incl damping tweaks outside the R26 and elsewhere in the system, I don't want to overstate the magnitude of the internal shielding etc improvements. The benefits are real and of the character I described, but easy to get carried away in the moment with the relative scale... there's many ways to denoise one's system, many of which are much simpler but also very effective so have a greater effort/return ratio. Ultimately though they're all satisfyingly additive/cumulative.

I'll keep tweaking and report back with any further insights in coming weeks, bring a bit more perspective, what's best bang for buck/effort.

For now though just one more observation from playing around wirh HQplayer settings:
- reconfirmed my optimum PCM bit setting* definitely 18 bits for my unit as present.. resolution and dynamics, incl richness of timbre and sense of space drops off a cliff at 19 bits and a bit less -a touch softer - but still very good at 17. Similar incremental delta to 16, 15 and 14 which I listened to
- *there's a pretty interesting exception case I discovered where an 18 bits setting is too high (beyond the linearity range of the R26's R2R ladders incl FPGA) and loses resolution relative to 17 bits, when 10mhz ext clock is set to OFF, so R26 is using internal oscillator. Found 17 bits is optimal there, after comparing to 16 & 15 bits etc
- DSD, yes has a low level background white noise hiss that I can't hear while playing music, but definitely a notch more resolving and dynamic than even 18 bits PCM.. though it's getting close. PCM has a slightly warmer and richer midrange, terrific for vocals. My preference will vary by music genre methinks, to my ear DSD tends to better suit acoustic instrumental music and electronic music, but I reserve the right to change my mind tomorrow...

Main track used for comparisons was Nenad Vasilic Gavrilo's Prinzip (don't judge it from YT, you gotta hear it lossless), superbly recorded double bass, a great test of the speed of attack on the leading edge of a plucked string, then the richness of its timbre and length of decay, and expansiveness and precision of soundstaging. Not to mention bass weight, impact and texture.

 
Last edited:
Feb 4, 2024 at 10:44 PM Post #8,372 of 8,835
@BShaw It is a bit out there /mad scientisty, that's for sure. 😅

Btw having had another quick listen today and doing a bit more experimentation incl damping tweaks outside the R26 and elsewhere in the system, I don't want to overstate the magnitude of the internal shielding etc improvements. The benefits are real and of the character I described, but easy to get carried away in the moment with the relative scale... there's many ways to denoise one's system, many of which are much simpler but also very effective so have a greater effort/return ratio. Ultimately though they're all satisfyingly additive/cumulative.

I'll keep tweaking and report back with any further insights in coming weeks, bring a bit more perspective, what's best bang for buck/effort.

For now though just one more observation from playing around wirh HQplayer settings:
- reconfirmed my optimum PCM bit setting* definitely 18 bits for my unit as present.. resolution and dynamics, incl richness of timbre and sense of space drops off a cliff at 19 bits and a bit less -a touch softer - but still very good at 17. Similar incremental delta to 16, 15 and 14 which I listened to
- *there's a pretty interesting exception case I discovered where a 18 bits setting is too high (beyond the linearity range of the R26) and loses resolution relative to 17 bits, when ext clock is OFF, so using internal oscillator. Found 17 bits is optimal there, after comparing to 16 & 15 bits etc
- DSD, yes has a low level background white noise hiss that I can't hear while playing music, but definitely a notch more resolving and dynamic than even 18 bits PCM.. though it's getting close. PCM has a slightly warmer and richer midrange, terrific for vocals. My preference will vary by music genre methinks, to my ear DSD tends to better suit acoustic instrumental music and electronic music, but I reserve the right to change my mind tomorrow...

Main track used for comparisons was Nenad Vasilic Gavrilo's Prinzip (don't judge it from YT, you gotta hear it lossless), superbly recorded double bass, a great test of the speed of attack on the leading edge of a plucked string, then the richness of its timbre and length of decay, and expansiveness and precision of soundstaging. Not to mention bass weight, impact and texture.


Stupid question if you don't listen via loudspeakers but via headphones, how is it possible to transmit sound if the room is mostly silent?
I don't quite understand that, I could just about understand it via loudspeakers, even if I think that there is no microphony via transistors, I would absolutely understand it with tubes.
 
Feb 4, 2024 at 11:04 PM Post #8,373 of 8,835
Stupid question if you don't listen via loudspeakers but via headphones, how is it possible to transmit sound if the room is mostly silent?
I don't quite understand that, I could just about understand it via loudspeakers, even if I think that there is no microphony via transistors, I would absolutely understand it with tubes.
Not sure I entirely get your question, but assume you read my damping & shielding refs as relating to audible spectrum sound/noise? If so, that's not what I was meaning and my understanding is - and I'm no technical expert - the problem I'm endeavouring to tackle is high frequency electrical noise from a variety of sources both within and outside the system. It is well beyond the audible hearing range into the upper khz, through into MHz and GHz. Mitigation can be to reflect it (not so great internally within a component!), to absorb them and/or to offload it to ground or dissipate em by converting some of it to kinetic energy and then heat locally (piezo-electrics, of which I employ a few).
 
Last edited:
Feb 5, 2024 at 3:49 AM Post #8,374 of 8,835
even if I think that there is no microphony via transistors, I would absolutely understand it with tubes.
Ceramic capacitors are microphonic, electrolytes much less. Transistors are not microphonic for acoustic noise, but FET transistor's principle of operation is "field effect" (as a name suggest), react on electromagnetic external noise. Therefore most of digital integrated circuits using MOSFET technology, convert it to jitter.

The idea of shielding of R2R ladder also makes sense, as it is a large surface area on which EMI convert to electricity.
 
Feb 7, 2024 at 5:39 PM Post #8,375 of 8,835
Just ordered a Holo Audio Cyan 2, looking forward to comparing with my R26 when it arrives in 6 weeks or so.
Possibly the only R2R DAC to really rival the R26 in terms of performance per $.

FWIW the rated dynamic range on the Cyan 2 is 127dB which lets me go up to 21 bits in HQPlayer vs 116dB (19 bits) on R26. So there's at least one objective benefit to the switch. As for whether that's audible remains to be seen. Holo May is rated at 130dB so that's quite impressive that their "budget" DAC at 1/4th the price does 127dB.

Also regarding EMI shielding, I can quite literally hear the noise floor lower when I throw EMI blankets over certain components. Specifically the RCA to XLR converter box I use for my speaker setup was picking up some interference (I think from my WiFi router) recently and it instantly goes away when I put an EMI blanket over it and then instantly comes back when I remove it. I can even hear noise from my wireless mouse sometimes and what's really crazy is if I increase the polling rate on my mouse from 1kHz to 4kHz I hear the tone of the noise increase too. Wild stuff.
 
Last edited:
Feb 8, 2024 at 2:31 AM Post #8,376 of 8,835
I'm also trading in my R26 now. It's been a fantastic year with it, sound wise, but now it's time to say goodbye, as a Tambaqui is on it's way.
I've been a disciple of Bruno Putzeys PWM products for more than a decade now, so I already know I'm gonna love the sound of the Tambaqui 😊
It's been great fun following this thread, and who knows, I might stick around for longer time 😁
 
Last edited:
Feb 8, 2024 at 2:39 AM Post #8,377 of 8,835
I'm also trading in my R26 now. It's been a fantastic year with it, sound wise, but now it's time to say goodbye, as a Tambaqui is on it's way.
I've been a disciple of Bruno Putzeys PWM products for more than a decade now, so I already know I'm gonna love the sound of the Tambaqui 😊
It's been great fun following this thread, and who knows, I might stick around for longer time 😁
Now that is an upgrade and a half... You're not mucking around. Awesome!
 
Feb 8, 2024 at 3:02 AM Post #8,380 of 8,835
I'm also trading in my R26 now. It's been a fantastic year with it, sound wise, but now it's time to say goodbye, as a Tambaqui is on it's way.
I've been a disciple of Bruno Putzeys PWM products for more than a decade now, so I already know I'm gonna love the sound of the Tambaqui 😊
It's been great fun following this thread, and who knows, I might stick around for longer time 😁
Literally 10x the cost of an R26 lol, quite the leap! Enjoy!
 
Feb 9, 2024 at 8:57 AM Post #8,383 of 8,835
Has any R26 dac owner checked how much the transfomers secondaries do output dc? It would be nice to know the exact values (both voltage and amperage) for each wire. I am kind of considering upgrading the both power transformers to occ EI-type custom wound ones. Very easy to change those if you know the exact output values.
 
Last edited:
Feb 9, 2024 at 8:59 AM Post #8,384 of 8,835
Gustard R26 forever ))
 

Attachments

  • IMG_20240206_112932.jpg
    IMG_20240206_112932.jpg
    3.4 MB · Views: 0
  • IMG_20240206_112649.jpg
    IMG_20240206_112649.jpg
    3.6 MB · Views: 0
  • IMG_20240206_112533.jpg
    IMG_20240206_112533.jpg
    3.8 MB · Views: 0

Users who are viewing this thread

Back
Top