New Audio-gd R-7, R-7HE R-8, R-27, R-27HE, R-28 Flagship Resistor Ladder DACs and DAC/amps

[PLGA]

I have the DI-20 non-HE and the HE-7 mkii and I do prefer the SQ with the DI-20 on the chain.

I was expecting to save money and sell the DI-20 after getting my HE-7, but I couldnt.

 

[Ninja Theory]

I understand from @sajunky that the DI-20/HE does not reclock USB. So I don't think clocks have any bearing on the sound. You should be getting benefit from the isolation. Are you sure there is no galvanic isolation on the DI-20 USB? Also, have you tried other interfaces for example AES/EBU from the DI-20 to your R7HE and does it sound better or worse?

updated: I missed that you're using IIS. If so, I think galvanic isolation is not relevant, so isolation cannot be the reason for any sound quality difference.

I see that Kingwa suggests not to use external USB isolation with the DI-20. So it does appear to have some sort of USB isolation. Well I'm going to test it some more this week and see (hear) if it is a noticeable improvement in SQ. I don't want to keep it in my chain just because of its good reputation if my ears don't tell me the story of the goodness it should be bringing to the party.

 

[dougms3]

I see that Kingwa suggests not to use external USB isolation with the DI-20. So it does appear to have some sort of USB isolation. Well I'm going to test it some more this week and see (hear) if it is a noticeable improvement in SQ. I don't want to keep it in my chain just because of its good reputation if my ears don't tell me the story of the goodness it should be bringing to the party.

He suggests not to use anything in between the pc and di20 except the cable.

 

[FredA]

The DI-20 does not have galvanic isolation on the USB. I was using it to reclock over I2S to the R7HE. The R7HE I2S port has the 2021 I2S isolation upgrade. This leads me to believe (perhaps erroneously?) that the clocks in my DI-20 are inferior to the clocks in my R7HE? The DI-20 is a 2016 model I believe (might be 2017 or even 2018 I am not sure).

Your dac has galvanic isolation on usb. Your di20 could have the 45/49 mhz clock pair, which is inferior to the 90/98. Your dac usb input could be upgraded to the latest with some soldering if i am not mistaken. Something to consider. The latest usb input is really really good.

Which hdmi cable do you use? This makes a difference.

 

[FredA]

I understand from @sajunky that the DI-20/HE does not reclock USB. So I don't think clocks have any bearing on the sound. You should be getting benefit from the isolation. Are you sure there is no galvanic isolation on the DI-20 USB? Also, have you tried other interfaces for example AES/EBU from the DI-20 to your R7HE and does it sound better or worse?

updated: I missed that you're using IIS. If so, I think galvanic isolation is not relevant, so isolation cannot be the reason for any sound quality difference.

The di20/he does not indeed reclock the usb signal. Asynch usb lets the receiver (di20/he) regulate the data flow. The signal out of the di20/he has very little jitter. The level of jitter is similar to the perf. of the internal clocks (90/98 version) . Therefore it is not reclocking but clocking.

Using an external clock on the di20/he, the sound could be better if an older fw that uses the incoming word clock signal (adaptive) is installed on the dac. Something to try for those who do not have a clock input on the dac.

 

[FredA]

Not sure such a fw exists for this dac. At some point, Kingwa decided not to use the incoming word clock. I just don't know when.

 

[Ninja Theory]

Your dac has galvanic isolation on usb. Your di20 could have the 45/49 mhz clock pair, which is inferior to the 90/98. Your dac usb input could be upgraded to the latest with some soldering if i am not mistaken. Something to consider. The latest usb input is really really good.

Which hdmi cable do you use? This makes a difference.

My DI-20 has the 90/98 clocks. I’m using an AudioQuest Carbon HDMI cable.

 

[Ninja Theory]

I tested more after the comments here. I AB'd the song from B-52's, Deadbeat Club. It has a nuanced bass line with a subtle mid bass twang that needs a fairly resolving system to present this texture so it can be appreciated. I2S from the DI-20 does indeed seem to have the edge in SQ. It is difficult to pick up unless you know a song well and can zone in on a specific sound. The bass guitar on this track is presented with more texture over I2S (it's a subtle difference). The sound is slightly rounded off over USB by close comparison. This is after many switching back and forth as there is a few seconds delay between the switch on my system (the Lumin requires the DI-20 to be switched off in order to switch to the other USB output). Like I said though, the difference is subtle and without this focus on back and forth, locking into a sound, I'm not sure I would notice it.

This makes me think the DI-20 is doing something in my chain. Perhaps reducing jitter? Perhaps the isolation on the upgraded I2S input on my R7HE is playing a part? I'm not sure. I'm not even sure if my DI-20 has the same clocks as my R7HE? It was purchased in 2016 with the 90/98 clocks I believe. Regardless, there does seem to be a subtle improvement in SQ so I will probably leave it in my chain again. Perhaps this improvement will become more noticed over some albums of listening time.

 

[gimmeheadroom]

Again, the DI-20 clocks don't have any impact since you're using the USB input, and you get isolation by using IIS (I2S). I guess you won't find a better USB->IIS interface for use with Audio-gd gear than a DI-20...

Might be worth trying a few better HDMI cables that @FredA and @Jandu and @DACLadder and other people can recommend.

 

[Ninja Theory]

Again, the DI-20 clocks don't have any impact since you're using the USB input, and you get isolation by using IIS (I2S). I guess you won't find a better USB->IIS interface for use with Audio-gd gear than a DI-20...

Might be worth trying a few better HDMI cables that @FredA and @Jandu and @DACLadder and other people can recommend.

Thanks. I'm still a little confused about when the DI-20 clocks are/are not used. I was under the impression that the Di-20 clocks are used when going out over I2S. It seems this is not the case. So just for my knowledge, when in fact would the DI-20 clocks come into play?

 

[gimmeheadroom]

My comments are based on trying to put together and understanding based on comments @sajunky has made in this thread. I understand USB input is not reclocked but sometimes S/PDIF is.

 

[Articnoise]

Thanks. I'm still a little confused about when the DI-20 clocks are/are not used. I was under the impression that the Di-20 clocks are used when going out over I2S. It seems this is not the case. So just for my knowledge, when in fact would the DI-20 clocks come into play?

DI-20 clocks are used for sure. The DI-20 or any other DDC wouldn't function without its clocks. Asynchronous USB means that the DAC's clock is pulling data from your source, eg. dictates the timing for the packets to be sent. The opposite is called Synchronous USB and means that the source pushes data to the DAC. and its the source clock that dictates the timing of the packets sent.

 

[FredA]

Of course, they are used. The point is the r7he does not use the timing of the incoming signal.

 

[PLGA]

I tested more after the comments here. I AB'd the song from B-52's, Deadbeat Club. It has a nuanced bass line with a subtle mid bass twang that needs a fairly resolving system to present this texture so it can be appreciated. I2S from the DI-20 does indeed seem to have the edge in SQ. It is difficult to pick up unless you know a song well and can zone in on a specific sound. The bass guitar on this track is presented with more texture over I2S (it's a subtle difference). The sound is slightly rounded off over USB by close comparison. This is after many switching back and forth as there is a few seconds delay between the switch on my system (the Lumin requires the DI-20 to be switched off in order to switch to the other USB output). Like I said though, the difference is subtle and without this focus on back and forth, locking into a sound, I'm not sure I would notice it.

This makes me think the DI-20 is doing something in my chain. Perhaps reducing jitter? Perhaps the isolation on the upgraded I2S input on my R7HE is playing a part? I'm not sure. I'm not even sure if my DI-20 has the same clocks as my R7HE? It was purchased in 2016 with the 90/98 clocks I believe. Regardless, there does seem to be a subtle improvement in SQ so I will probably leave it in my chain again. Perhaps this improvement will become more noticed over some albums of listening time.

In my case the difference was small, but noticeably with the DI-20 (non HE) on the chain. And I believe my DAC has a newer USB input version than yours. I will try to make some more test in the next days to confirm my impressions, but it was pretty obvious.

If it's within your budget and you think you could use it with future upgrades, maybe the pure silver Aliexpress HDMI cable we recommended on the thread could be an option. It really made a difference in my case.

On the other hand, if I have to struggle to detect a difference in SQ adding a piece of gear or a tweak, it's not really worth it. Even more if you are doing AB tests with short delay.

 

[Jandu]

. This is after many switching back and forth as there is a few seconds delay between the switch on my system (the Lumin requires the DI-20 to be switched off in order to switch to the other USB output). Like I said though, the difference is subtle and without this focus on back and forth, locking into a sound, I'm not sure I would notice it.
......

If you have to switch off DI20 to do A/B test, you are not listening to the potential of a warm up DI20. It does make a difference in my system (DI20He) whether it just get turned on or it has been on for 30 min.