[565321]

I'm puzzled why you waste time on I2S, non-standard, no error correction. It's an interface for short runs on the board. Just asking for trouble. Use the streamer and you're fine and spend more time with funny herbs and good music.

 

[JackJohnson316]

I got my DI-20HE back from loaner.

Can confirm I2S does not work in any configuration between DI-20HE and R26.
Also, I've mentioned this much earlier in the thread, but as a reminder neither did my Hydra-Z I2S work with R26.

They both worked with X26. I know Gustard changed the I2S configuration on the R26, but they had to of changed more than just how the pins are.
It has something to do with one of the NC lines (NC = no connect).

Basically, the R26 is stuck on DSD mode in any configuration settings with DI-20HE or Hydra-Z (and yes both those DDC have configuration for I2S).
So again it's not just a simple config issue, R26 has something funky going on with it.

Of course their U18 (also pictured) works perfect with R26 using I2S. But I find the built in ethernet on R26 is better than U18+I2S, so it's meaningless.

Kinda dumb, but whatever.

I just ordered a Holo RED DDC today, but I won't get it for 5-6 weeks. I'll update this thread then, on if Holo Red I2S works with R26.

EDIT:



Huh, I didn't read this before I posted. This is interesting, is this your direct experience, or did you read this somewhere else?
And if you did read it somewhere else, can you link me that original source?

I'm having quite the opposite issue. DI-20HE is stuck on DSD, not PCM.
If it was stuck on PCM that would be amazing, because I only play PCM music, and I have no DSD, or any desire to ever play DSD.

Bummer for me.

EDIT2:

Upon further reading in this thread, Jake2 was mistaken on the flip. According to @JackJohnson316 it being stuck on DSD is normal.
Pins 15 and 17 need to be shorted for proper PCM flag.

Well, I'll be monitoring this thread for any manufacturers making customer I2S cables to fix this.

From my findings if you short pin 15 to ground (pin 17 in my case) it forces the R26 to play everything as pcm. That means no dsd, although let me test to see if it will still play dsd but display as pcm. Also, I completely took out the pin 15 conductor from my I2S cable and it still registered as dsd and would not play. So from my experiment it seems that no pin 15 conductor/signal triggers dsd mode. And shorted pin 15 to ground triggers pcm mode. And probably the dsd direct mode doesn’t work, but I’ll check tonight when I get home.

 

[Jake2]

Don't want to start a debate here, just want to chime in with my own experiences.
My DDC rankings are (I directly own all of these, and have extensively ABX'd them).

DI-20HE > Hydra-Z+ZPM > U18 > SU6 > Lynx AES16e (tied with Pi2AES) > Eitr > (I've owned these in the past, DI-2014/DI-U8/F1)

The R26 ethernet for me is better than U18 I2S, which stated above, I already think is better than SU6.
I'm surprised by your results, but I wan't argue them, I just want to give my own data point.

The sound signature of SU6 to me (compared to my other DDC) is darker, thicker, a touch of haze.
I've also loaned out the SU6 and had people tell me they thought the same.

I will also say all ethernet testing is with Roon, using their own (and IME best in the market) RAAT network protocol. Which again, ime, is better than UPNP or airplay.

As I've posted a few times up thread I reached the same conclusion - that R26 NAA with ROON/HQP > U18 & SU6 + Zen. Using OCK-1 with R26, and U18 too. A few others prefer R26 NAA over other inputs too.

For me if anything my expectation bias was the U18 (which brought a great improvement over USB direct from several DACs) would be superior to R26 ethernet. I caveated my results by the fact my upstream ethernet/network chain and cabling was basic, and there's better streamers than the Zen, so maybe this audibly favours the R26. (Certainly it favours the R26 from a $ spent /bang per buck)

 

[Jake2]

From my findings if you short pin 15 to ground (pin 17 in my case) it forces the R26 to play everything as pcm. That means no dsd, although let me test to see if it will still play dsd but display as pcm. Also, I completely took out the pin 15 conductor from my I2S cable and it still registered as dsd and would not play. So from my experiment it seems that no pin 15 conductor/signal triggers dsd mode. And shorted pin 15 to ground triggers pcm mode. And probably the dsd direct mode doesn’t work, but I’ll check tonight when I get home.

That's great investigative work @JackJohnson316 . Really interesting that shorting Pin 15 to ground at source end is seen differently by R26 than no signal at all to Pin 15. I'd wrongly inferred that your fix was working because your source was by default sending a constant signal/voltage triggering R26 Pun 15 Flag 1/ DSD so shorting to ground meant no signal = Flag 0/ PCM default. But as per Comzee's experience it's clearly the other way round. R26's default of no signal on Pin 15 = R26 flag '1' /DSD. Not an ideal default by Gustard but then there is no I2S standard to be fair.

 

[OCC7N]

I got my DI-20HE back from loaner.

Can confirm I2S does not work in any configuration between DI-20HE and R26.
Also, I've mentioned this much earlier in the thread, but as a reminder neither did my Hydra-Z I2S work with R26.

They both worked with X26. I know Gustard changed the I2S configuration on the R26, but they had to of changed more than just how the pins are.
It has something to do with one of the NC lines (NC = no connect).

Basically, the R26 is stuck on DSD mode in any configuration settings with DI-20HE or Hydra-Z (and yes both those DDC have configuration for I2S).
So again it's not just a simple config issue, R26 has something funky going on with it.

Of course their U18 (also pictured) works perfect with R26 using I2S. But I find the built in ethernet on R26 is better than U18+I2S, so it's meaningless.

Kinda dumb, but whatever.

I just ordered a Holo RED DDC today, but I won't get it for 5-6 weeks. I'll update this thread then, on if Holo Red I2S works with R26.

EDIT:



Huh, I didn't read this before I posted. This is interesting, is this your direct experience, or did you read this somewhere else?
And if you did read it somewhere else, can you link me that original source?

I'm having quite the opposite issue. DI-20HE is stuck on DSD, not PCM.
If it was stuck on PCM that would be amazing, because I only play PCM music, and I have no DSD, or any desire to ever play DSD.

Bummer for me.

EDIT2:

Upon further reading in this thread, Jake2 was mistaken on the flip. According to @JackJohnson316 it being stuck on DSD is normal.
Pins 15 and 17 need to be shorted for proper PCM flag.

Well, I'll be monitoring this thread for any manufacturers making customer I2S cables to fix this.

I believe the DI20HE has 4 modes for HDMI. Did you use the right one?

I have heard the DI20HE should work without problem, even without shorting pin...Pin 15

Your DI20HE has for hdmi pin modes, did you check it?

Try mode 2 or 4 on DI20HE
http://www.audio-gd.com/R2R/DI20HE/DI20HEEN_Use.htm

I just discovered Matrix x-spdif 2 has a dip switch for this. If not playing any dsd(turning on switch) it will force the pin 15 pcm. I have yet not tried since I did not order the R26.

 

[comzee]

I believe the DI20HE has 4 modes for HDMI. Did you use the right one?

I have heard the DI20HE should work without problem, even without shorting pin...Pin 15

Your DI20HE has for hdmi pin modes, did you check it?

Try mode 2 or 4 on DI20HE
http://www.audio-gd.com/R2R/DI20HE/DI20HEEN_Use.htm

I have tried all 4 modes, per the manual linked, (and which I reference when I tested earlier today).
Of course any spdif output from DI-20HE works with R26, that's a given. But no, with any of the 4 I2S configurations available on the DI-20HE, none of them worked for PCM. R26 was always stuck on DSD.

 

[pankar0]

Can Someone please measured the height with the feet's

 

[camrector]

Can Someone please measured the height with the feet's

3in

 

[Quinto]

Can Someone please measured the height with the feet's

7cm

 

[Nada]

Q is also better with Singser than over Ethernet.

The R26 ethernet for me is better than U18 I2S,

Its hard to know what going on with Ethernet vs I2S sound quality, as the modes with I2S are diverse and haven't been clarified

With I2S the options include running on the 10MHz based synthesised clock versus dedicated I2S clocks. Then there is R26 OS vs non-OS options.

If you want to find what a DDC I2S can do try:

1. over sampling to the max. up stream eg HQP or Foobar
2. on the R26 flip between "OS" and "NOS" modes to test the clocking vs R26 OS choices

Once you have the best sound with I2S then flip to the ethernet to see which sounds best. Cheers.

 

[comzee]

Its hard to know what going on with Ethernet vs I2S sound quality, as the modes with I2S are diverse and haven't been clarified

Di-20He / HydraZ+ZPM via spdif actually edge out R26 Ethernet.
The U18 and SU6, I prefer built in ethernet.

I find for a well designed DDC, the difference between I2S and spdif isn't has big as people make it.
The DDC itself lends more performance as an overall unit, then the specific output you use... (except for edge cases like Pi2aes where the spdif is gimped compared to it's I2S)

The U18 actually has better transparency than R26 ethernet, but it's got narrow soundstage, funky tonality, doesn't sound the most natural.

In the end, it's about synergy, personal preference, etc...

 

[JackJohnson316]

That's great investigative work @JackJohnson316 . Really interesting that shorting Pin 15 to ground at source end is seen differently by R26 than no signal at all to Pin 15. I'd wrongly inferred that your fix was working because your source was by default sending a constant signal/voltage triggering R26 Pun 15 Flag 1/ DSD so shorting to ground meant no signal = Flag 0/ PCM default. But as per Comzee's experience it's clearly the other way round. R26's default of no signal on Pin 15 = R26 flag '1' /DSD. Not an ideal default by Gustard but then there is no I2S standard to be fair.

I'm not able to test DSD files when shorting pin 15 only because my I2S bridge pcie card does not play DSD. I will say that I tested my cheap ($15) DIY I2S cable against a $150 I2S cable and a $500 I2S cable, and after 30 minutes going back and forth I couldn't tell a difference in the sound. The cheap DIY I2S cable has cheap 22 gauge copper conductors from Amazon. The $150 has single crystal copper coated with silver, larger gauge. The $500 I2S cable has 22 gauge solid silver cryo treated conductors with tons of shielding. They all sounded exactly the same to me coming from a high end music server to the R26, but my DIY crackles with 176/24 files and above. That's probably due to the small gauge traces in the actual connector, that was only $5 from Amazon. I ordered a better I2S connector with higher gauge traces to see if that's what causing the crackling sound in higher res files. All that to say that I wonder if we really need to pay so much for I2S cables. I'm not a cable denier, but my little test has caused me to ask myself if we really need a $500 I2S cable? I'm attaching a photo of the pin 15 short I did on the receiving end of the I2S cable. It's only a temp solution.

 

[Jake2]

I'm not able to test DSD files when shorting pin 15 only because my I2S bridge pcie card does not play DSD. I will say that I tested my cheap ($15) DIY I2S cable against a $150 I2S cable and a $500 I2S cable, and after 30 minutes going back and forth I couldn't tell a difference in the sound. The cheap DIY I2S cable has cheap 22 gauge copper conductors from Amazon. The $150 has single crystal copper coated with silver, larger gauge. The $500 I2S cable has 22 gauge solid silver cryo treated conductors with tons of shielding. They all sounded exactly the same to me coming from a high end music server to the R26, but my DIY crackles with 176/24 files and above. That's probably due to the small gauge traces in the actual connector, that was only $5 from Amazon. I ordered a better I2S connector with higher gauge traces to see if that's what causing the crackling sound in higher res files. All that to say that I wonder if we really need to pay so much for I2S cables. I'm not a cable denier, but my little test has caused me to ask myself if we really need a $500 I2S cable? I'm attaching a photo of the pin 15 short I did on the receiving end of the I2S cable. It's only a temp solution.

Interesting observations and nice work on the cable. Keep em coming.

I've only ever used the $15 BJC FE I2S cable, recommended by many who've used more expensive cables as 'as good' or better in some cases. So based on those anecdotes, and the good sound I get in my system, I'd agree they don't need to be expensive to be 'good'/OK. A counter data point - well, anecdote - I came across over in the U18 thread was a long time and content user of the BJC FE in his otherwise pretty high end and resolving system (U18 + X26 were some of the least expensive components). He switched up to the Tubulus Argentus and was pretty surprised at the scale of the improvement it brought to his already dialled in system, as I'm not sure he was expecting much.

That has me I2S curious, I won't lie, as to how much untapped potential there is in the DDC>R26 chain rather than my currently preferred R26 NAA. But then there's observations like yours. So for now I'm trying to just sit back and read with interest as I2S cable observations/experimentation like yours and his continue to roll in.

 

[SupCompracer]

I've been DIY'ing DAC's for ~ ten years before purchasing an R26. In some cases, I2S offered the best sound quality, if implemented properly. That depended on DAC chip as well. ESS Sabre did a great job with SPDIF so you may get only a slight improvement with I2S. Twisted Pear offered Teleporters years ago, or LVDS modules. I could snag PCM/DSD from a CD player and send it to the DAC over a network cable. If you used a one-meter or longer cable without LVDS, you would run into sound quality issues.

I just added an SU-6 and will compare it to the R26 USB and LAN. I'll stay with whatever sounds best.

An example of a well-regarded USB to I2S module with isolation and reclocking. A Sonore Audiobyte. It's in my current DAC, a Twisted Pear BIII9038 Pro with TP Mercury I/V stage.

 

[NoaHimesaka]

Of course

Or you could just always overwrite with Gustard provided firmware... The streamer "update" file is just a full firmware file.