[Jake2]

This is a subjective audio forum. I doubt Currawong has no idea how to listen to different inputs correctly. Plus my main point is my opinion - I’m saying optical with CDs sounds great to me. I bet this dac sounds great with lots of things so that’s probably the ultimate truth.

+1 for the R26's Toslink sounding good. I use the optical output from AV sources like my TV and Apple TV cleaned up by an IFI S/pdif iPurifier2 (powered by an IFI Power X + iFi DC iPurifier2) and it sounds ridiculously good even with compressed /lower bitrate content. I highly recommend the iFi S/pdif iPurifier2 and as clean power you can provide it with as an affordable way of very effectively regenerating and reclocking a less than perfect s/pdif source. Without the iFi combo those AV sources were thin and brittle. The only other DDCs I'm aware of that take and reclock s/pdif inputs are the Denafrips and Audio-GD DDCs which cost a fair bit more, but also have a broader set of input and output capabilities.

I was also surprised at how close the s/pdif outputs of my U18 DDC came to its I2S out, all feeding the R26. It's all very cable dependent too of course. I have good but not top flight Toslink and i2S cables.

 

[sajunky]

This is a subjective audio forum. I doubt Currawong has no idea how to listen to different inputs correctly. Plus my main point is my opinion - I’m saying optical with CDs sounds great to me. I bet this dac sounds great with lots of things so that’s probably the ultimate truth.

He was wrong about R2R-11, there is no question in this case. The same happens with Gustard. So what? There are two options. Is it Gustard USB implementation defective or Currawong was wrong again?

Remember that Gustard has no galvanic isolator on the USB port, same as my $350 R2R-11. For the last 10 years USB receivers use asynchronous transfers, there is no reason to suspect that Gustard use outdated libraries. Besides, R26 is UAC 2.0 compatible and asynchronous transfer mode is a requirement. Jitter-free transfer. That is.

It is leaving only one possibility — due to a missing galvanic isolation a DAC picks up noise from the USB source, typical ground loops scenario. And according to my experience in all cases where optical sounds better than coax or USB, it can be reversed after taking care of ground loops. The same will happen to you if you do it properly.

 

[DeckHiFi]

He was wrong about R2R-11, there is no question in this case. The same happens with Gustard. So what? There are two options. Is it Gustard USB implementation defective or Currawong was wrong again?

Remember that Gustard has no galvanic isolator on the USB port, same as my $350 R2R-11. For the last 10 years USB receivers use asynchronous transfers, there is no reason to suspect that Gustard use outdated libraries. Besides, R26 is UAC 2.0 compatible and asynchronous transfer mode is a requirement. Jitter-free transfer. That is.

It is leaving only one possibility — due to a missing galvanic isolation a DAC picks up noise from the USB source, typical ground loops scenario. And according to my experience in all cases where optical sounds better than coax or USB, it can be reversed after taking care of ground loops. The same will happen to you if you do it properly.

Thanks for your thoughts. I think USB sounds great with my USB cable that isolates the power. But yeah I get what you're saying. I guess what I'm saying is I loved both. I might just be preferring the sound of CDS and not streaming in general. Just seems like I can hear the little things more in a track now, but I also could just be more intently listening. I think a lot of times with audio we love a new sound, and then get bored and search for something else. Then we find a new great thing, and automatically assume it's better. When 2 things can be true, but not saying you're wrong about anything you said. I hear no hum or problems with my USB though.

But perhaps it's also just better for audio to have a cd player or streamer instead of a MacBook. Not saying it is actually better, I'm too lazy to constantly A/B test things. But it makes sense to me why the straight up device just feeding audio could be better. Plus if it helps me listen a whole album at a time more, that's a positive.

 

[sajunky]

Thanks for your thoughts. I think USB sounds great with my USB cable that isolates the power. But yeah I get what you're saying. I guess what I'm saying is I loved both. I might just be preferring the sound of CDS and not streaming in general. Just seems like I can hear the little things more in a track now, but I also could just be more intently listening. I think a lot of times with audio we love a new sound, and then get bored and search for something else. Then we find a new great thing, and automatically assume it's better. When 2 things can be true, but not saying you're wrong about anything you said. I hear no hum or problems with my USB though.

But perhaps it's also just better for audio to have a cd player or streamer instead of a MacBook. Not saying it is actually better, I'm too lazy to constantly A/B test things. But it makes sense to me why the straight up device just feeding audio could be better. Plus if it helps me listen a whole album at a time more, that's a positive.

You will love a sound more and more. You will rediscover all your old CD library. Damaged NOS mode by opamps (while Gustard lied about discrete output stage), but it is still R2R. I am R2R guy.

A comment on a lack of hum coming from USB port. You will never hear hum, unless ground loops affect the analogue output stage. A rare case, in fact I never experienced it.

A noise that affects DA conversion is not heard as a hum. It corupts a sound, but dissappears during silence. It is jitter. You can not hear jitter, but you notice when it goes away

 

[Fluty]

I'm interested to see mention of CD again. My opinion has for a long time been that CD is superior to streaming, using an excellent CD transport from Audiolab. Earlier this year I upgraded my Audiolab CDT 6000, which I thought was fantastic, to their latest and even more stunning CDT 9000. I thought I was in heaven. Only a few weeks ago I managed to acquire a used HALCRO CD / SACD player. No longer available new, it has a Swiss made high quality metal transport and using the R26 as dac I can't believe how much it has improved on an already superb presentation. It will only play CD (not SACD) through the external dac of course, but it emphasises what a phenominal dac the R26 is when it performs even more brilliantly with this high level transport! (My Audiolab CDT 9000 is for sale, by the way, as new for half the new price at $1,100 with one year manufacturer's warranty remaining).

 

[LucianoBadoe]

Someone here have tested the Denafrips Arce Streamer on R26? I believe that the i2s scheme isn't compatible between both but there is some way to connect they in this interface? With a custom HDMI cabe for example?

 

[dacwizard]

This is a subjective audio forum. I doubt Currawong has no idea how to listen to different inputs correctly. Plus my main point is my opinion - I’m saying optical with CDs sounds great to me. I bet this dac sounds great with lots of things so that’s probably the ultimate truth.

If you think optical input sounds great then it does, simple. I too use optical input (from a streamer in my case), I tried Lan input using the streamer, there was no noticeable improvement, so I use optical as it’s easier and it sounds simply fantastic. Remember everyone’s setup is different, what seems ‘night and day’ better for one system may be worse in another.

 

[dacwizard]

But I can't notice any tonal difference between different settings of the R26. Therefore I was wondering if my R26 might be malfunctioning (got it used). Or whether these settings do only apply to other input types.

I suspect there’s nothing wrong with your system, these changes are certainly not night and day and rapid switching back and forth may not result in any obvious change. I find that if I change (say filters) I might notice no differences on some material, but then some element of another album is just slightly different, or you may find music fatiguing long term with the fast filter in a way you don’t experience with the slow. I found NOS mode just to lose a bit of resolution compared to OS (supplied with a PCM stream) but I think NOS is probably an obvious improvement primarily for if you use an external player with its own filters (hqplayer etc) …which I don’t.

 

[DeckHiFi]

Yea ^ great thoughts! I think the easiest way to tell difference between mid and fast filters (for me) is if you have an album with great vocals. Seems like at same volume mid filter allows the mid range to breathe more where fast sounds more full or brings up the other instruments a bit. Not sure which is better really. Might depend on the album if you want to hear the vocals more or not. Today I’m loving mid filter but I might change tomorrow.

 

[ohcrapgorillas]

A question for the HQPlayer users: what do you use for DAC bits? As I recall, designing an R2R DAC with 24 bits of depth is impossible, so R2R designs usually have something like 18-21 bit depth.

Jussi recommends 18 or 20:

Yes, I would get started with DAC bits set to 20 for the Gustard. Or maybe 18. You can try if you notice any difference between the two.

What are other people using? Does anyone know the true bit depth of this DAC?

 

[camrector]

A question for the HQPlayer users: what do you use for DAC bits? As I recall, designing an R2R DAC with 24 bits of depth is impossible, so R2R designs usually have something like 18-21 bit depth.

Jussi recommends 18 or 20:


What are other people using? Does anyone know the true bit depth of this DAC?

15

 

[Jake2]

A question for the HQPlayer users: what do you use for DAC bits? As I recall, designing an R2R DAC with 24 bits of depth is impossible, so R2R designs usually have something like 18-21 bit depth.

Jussi recommends 18 or 20:


What are other people using? Does anyone know the true bit depth of this DAC?

Strictly speaking I understand HQP's setting relates to a DAC's linearity measured in bits, not bit depth per se, also from my reading sub-20 bits is typical for R2R with the Holo May an outlier at 20 bits IIRC.

But to answer your question, from my initial trial and error of careful AB listening at different HQP bit settings I found resolution dropped off on my R26 playing PCM content with HQP set to greater than 15-16 bits. After a number of tweaks to my R26 and system (incl grounding, cleaner power, vibration damping) over the last 12 months I revisited this and found I could get to 18 bits without resolution drop-off.

YMMV, try this yourself and see what setting sounds best in your system

 

[camrector]

A question for the HQPlayer users: what do you use for DAC bits? As I recall, designing an R2R DAC with 24 bits of depth is impossible, so R2R designs usually have something like 18-21 bit depth.

Jussi recommends 18 or 20:


What are other people using? Does anyone know the true bit depth of this DAC?

Strictly speaking I understand HQP's setting relates to a DAC's linearity measured in bits, not bit depth per se, also from my reading sub-20 bits is typical for R2R with the Holo May an outlier at 20 bits IIRC

Grab any online ENOB calculator.
Then enter the SINAD measurement provided by Wolf here :
https://www.l7audiolab.com/f/gustard-r26/
You will see 15.7 to be the effective number of bits. Set it to 15 and you’re in business.

Not doubting all of @Jake2 amazing tweeks but this DAC will never ever reach the 112db SINAD that would allow for a bit depth setting of 18.

 

[Jake2]

Good historical, if technical, discussion and explanation of benefit of optimising HQP bit setting for a given DAC by Jussi here:

https://community.roonlabs.com/t/which-hqp-filter-are-you-using-2015-2023/6061/545

https://community.roonlabs.com/t/which-hqp-filter-are-you-using-2015-2023/6061/566

 

[sajunky]

Does anyone know the true bit depth of this DAC?

Short answer: It has little relation to the true ladder accuracy. The optimum value (selected by tests) indicate how much dither is applied in a DAC to the sound.

Long:
All ladders have similar accuracy, not exceeding 16-bits unless resistors are laser trimmed. Dither can be added in a DAC to linearize ladder errors. More dither applied, it will look better in the linearity test, as results are averaged. However excess of dither is detrimental for sound quality, it kills details.

A number in HQPlayer indicate how many bits are 'untouched'. If listening tests give the optimum value 20, it means that a DAC itself is inserting extraordinary amount of dither, so don't do it again on more significant bits than 20, as it would be overdose.

If a number is below or close to the theoretical ladder accuracy (16), it means that a DAC do not apply dither to a sound so adding more dither in the player can be beneficial for some users who prefer a smooth - less detailed sound.

In practice bits above 16 in the ladder create natural dithering, this all HQPlayer 'optimalisation' is not needed. But it depends on your preference.