I2S compatibility issue between W4S dac2 & Singxer SU-1, two beatiful pieces of hi-end products

[Franatic]

Patiently waiting 

[Wyred 4 Sound]

There are many issues swirling in this thread and the other. We’ll be addressing some general but relevant issues here that hopefully serve as background and explanation of the points brought up by others.

 

Regarding I2S, we were/are very close with PS Audio who was the first company to use this type of signal transmission, and we were one of the first to adopt this technology as our own. We'll get the unfortunate out of the way — admittedly, there is a third-party spreadsheet of pin configurations for ours and other devices that we thought was accurate, but in closer inspection, we realized it was not. We’ve taken this chart down from our website and we apologize for not recognizing the discrepancy and for any inconvenience caused. However, we think it’s important to understand in the context of this discussion and in general— there is no ‘standardized’ implementation of I2S, nor was DSD even on anyone’s radar (not to mention even supported on our DACs) back 10 years ago when we first introduced I2S on our products. Our version has been the same ever since, and open for any company who wishes to implement their I2S device with ours to ensure things function as intended. All of our products that use I2S our compatible with each other, which was our sole intent with this connection path. 

 

The other point brought up was firmware updates. Firmware is not in control of the phase issue discussed here. We have the capability of updating our USB firmware via USB port, however, this is not the input that needs to be addressed. Our product firmware requires a different method, but it also cannot alter these settings. USB firmware would need to be adjusted on the source device as it controls the pins the signal is sent on and can likely be controlled with firmware. Our side of this signal path is purely hardware, not software.

 

We appreciate that there are always those users who are constantly asking for and pushing the envelope on technology past what is mainstream. That same spirit drives us as well. It’s great that companies like Singxer develop new, bleeding-edge products which advance the larger cause, but there also needs to be vetting and evaluation of existing implementations to ensure things work smoothly. We are always willing to collaborate with companies to address issues that arise and explore possible solutions.

 

Regards,

Tony and EJ

[Franatic]

Really????? I am stunned by the ineptitude of this response.

 

So you are what / were what / PS Audio what????? So you "just" realized that a pin configuration that you maintained you went by was wrong. You just now inspected it and took it down. So what is your pin configuration....or is that top secret?? Apparently it is so secret you don't even know.

 

I told Tony 3 months ago that I was going to use the Singxer and i2s. This has been a real headache for me and it just got worse. You need to at least publish your pin out configuration. What exactly are your "variations" from PS Audio??

 

I hope Singxer will fix this for me. But you have to help them out with your data.

[Ienag]

Tony, EJ 

 

Why dont'you contact directly Singxer to give them support to solve this problem with your product so that you can make happy your customers?

I think it is right and fare, solve the problem with Singxer dcc is the most important thing to appreciate conpletely you Dac.

 

You should also publish your I2S configuration to avoid future problem...

 

Last point... for firmware upgrade,  we are on 2017, you could offer the possibility to change the hardware for who lives in Europe without sending the Dac everytime in USA.

[Franatic]

After a bit of chilling, I realize the only important thing is that we Singxer SU-1/ W4s DAC2 customers need to find a path forward. In the end I am more practical than anything else.

Tony, Ej. please work with Leter from Singxer to help him get the right settings to make DSD work properly. I hope Leter is willing to do this. I also hope for future customers there is a unified solution to the PCM issue that was fixed via different bypass capacitor locations.

This i2s path with the SU-1/DAC2 combo sounds amazing. If the issues get ironed out, I can heartily recommend it to others.

[GUTB]

"Bleeding edge", lol.

 

Look, Singxer just wants to get paid to fix your problem -- so pay them. They can release a firmware addressed to DAC2 owners.

[Franatic]

16 minutes ago, GUTB said:

Look, Singxer just wants to get paid to fix your problem -- so pay them. They can release a firmware addressed to DAC2 owners.

W4S still needs to publish their configuration. Even if there is no standard, they have to say what they are. They can make a W4S standard.

At that point, they can insist that DDC manufacturers configure accordingly. However, to maintain they are PS Audio standard and then not to be that, they have no equitable ground to stand on. They should feel an obligation to make this right for their customers.

I bought the Singxer SU-1 knowing it worked well with the PS Audio DirectStream DAC. W4S maintained they were the same configuration as PS Audio. That was not true.

[jojeong8338]

43 minutes ago, Franatic said:

W4S still needs to publish their configuration. Even if there is no standard, they have to say what they are. They can make a W4S standard.

 

I agree on it as well. If HW pin is different with PS AUdio standard, we can test the function with DIY HDMI cable according to the PIN configuration. I think the best solution to satisfy the customers is upto W4S but the configuration should be shared in advance.

[MINORISUKE]

4 hours ago, Wyred 4 Sound said:

.....

Our version has been the same ever since, and open for any company who wishes to implement their I2S device with ours to ensure things function as intended. All of our products that use I2S our compatible with each other, which was our sole intent with this connection path. 

.....

I am relieved that I could get rid of a false accusation on Leter's excellent product.

It is very easy for W4S just to disclose this open-for-any-company pin assignment also to end users.

There still remains a mystery of the different behaviour of the new and old units under the same model mentioned above in this thread.

I wonder if there are variations among the "ever since same" version.

[Ienag]

I think both Manufactures could agree a technical agreement to solve at least the I2S problem with dsd, eventually with a firmware that only Leter can provide and I hope he will provide to permit other W4S dac owners buy the Singxer SU 1 and for us to be satisfied with both the equipments.

In my opinionion it is  interest of both solve the specific problem clarifying all the misunderstandings now and for the future W4S should permit to know its I2S pin out to make future compatibility assured with third parties interfaces.

We are still waiting for a proposal of solution...

[Franatic]

As someone who has gone through a 12 step program, W4S is just beginning their i2s recovery. They are starting to admit they have a problem. They are starting to become honest, at least as honest as they can be at this point. More is going to be required for a solution. I'm rooting for them. Not just because I want my problem fixed, but because I like their company. Before this issue, they had handled other problems very professionally. Once this problem appeared, big time dysfunctionality followed.

 

Together we can get this i2s thing on track. It is worth the effort. As I have stated earlier, it sounds amazing. It is the best audio path for this DAC. But W4S is going to have to be completely upfront about their "variations". Are there 2 different configurations.....or more? It seems that most DACs have the PCM issue and the DSD inversion.

 

 I have over $1000 in my Su-1/SU-1 power supply upgrade/Wireworld Starlight silver HDMI combo. I have another $1000 in my DAC2 DSDse 9038 upgrade. $2000 in upgrades that can't properly play DSD. I will do what I can from my side to help straighten out this problem, and I already have much time and effort invested here.

 

I want to personally thank Minorisuke for his expertise in helping clarify the issues here.

 

C'mon W4S, we are all on the same side.  It is in everybody's interest to get this fixed. 

[ksalno]

For every problem, there is always someone, even if it isn't W4S or Singxer who is willing to provide a fix - for a price.

 

First let's confirm the actual number of problems, are there two or three? It would be helpful if someone could summarize and then everyone agree. So far, I have experienced only two issues with the Singxer and DAC2.

 

1) Noise when trying to play PCM. It has been shown by both manufacturers that the solution is a capacitor to clean up the signal between the Singxer and the DAC2. Each has shown this by putting a capacitor in the others device. Personally, I don't this the location matters, what matters is there is clearly some noise on the interface that when cleaned up, fixes the problem. We can argue all day whether it is the Singxer that is unstable or the DAC2 that is filtering properly. Since neither is willing to admit the issue is theirs, it is left to the consumer to correct the problem. Taking matters into my own hands, I have delivered my Singxer complete with pictures and cap values to a local repair shop that I trust and they are making the mod. it will probably cost me $100 but in the grand scheme between the cost of the DAC2 and the Singxer, this is nothing.

 

2) There is some kind of pin assignment mismatch between the Singxer and the DAC2. Since we know what the pin assignments options are on the Singxer, all we need is for W4S to publish their's and I'm sure we can find some enterprising cable manufacturer who will make up a cross-over HDMI cable for everyone who needs one. Since W4S has already admitted that they had the wrong info on the website and that it is clear that some DAC2 customers used this bad information in purchasing a Singxer, I think it would be very much appreciated by those customers if W4S would publish the pin-out specs for their input, so that we can begin the process of finding someone to make the cross-over cable. I believe this would eliminate the need for any firmware updates to either device, so we can stop arguing about that, too.

 

What am I missing or where am I going wrong, guys?

 

We can be practical and get this problem behind us so we can start listening to music again or we can wait until hell freezes over for these two to stop pointing fingers at each other and work together to come up with a fix that neither one seems particularly interested in providing.

 

Karl

[Ienag]

Karl, I don't know if the phase inversion on one channel during reproduction of dsd files through I2S can be solved by simply customizing an hdmi cable, I think it is more complicated and the new firmware is needed. Only Leter can do and provide it and I know he is working on it even if I don't know what he really will done when finished.

I understand your point of view but the customers have to be supported,  we can't do everything by ourself.

Pcm problem was solved but the solution was left in our field, now for the dsd I would want a clean solution.

[Franatic]

I have emails from January 31st and February 2nd where I asked Tony Holt about DSD via the i2s path. He assured me DSD is supported. I told him I would only go to the i2s path if it supports DSD. This is because I was disappointed that there is no DSD64 via AES or Coax.

I did not blindly go and buy an SU-1 "assuming" it would work. I was given the information that they are formatted to the PS Audio standard AND it supports DSD. I knew PS Audio DACs were working with the SU-1. It would be reasonable to project from that information that the SU-1 and DAC2 would work. W4S has an obligation to figure out a way to make DSD work correctly here. They are the ones who misrepresented their DSD protocol and communicated (at least to me) that they were supporting DSD.

(email chain available upon request)

[Wyred 4 Sound]

Attached please find our I2S pin configuration. Based on PS Audio’s model, pins 1 and 3 need to be addressed from Singxer to fix the phase issue. We have reached out to Leter of Singxer about this and are awaiting his reply. FYI, there are no other variations of our pin configuration; we’ve been doing it this way since the beginning.

 

Regarding the PCM noise issue, putting the cap fix on either side accomplishes the same result. The reason we fixed the Singxer instead of our DAC is that our other I2S sources do not exhibit any noise issues. In addition, we had Fran’s Singxer here, not his DAC. There is a chance that if Singxer is able to update their I2S, the noise issue could also go away due to the improved clocking. Our theory is that it’s caused by isolation jitter on the Singxer output.

 

We’ll post more once we hear back from Singxer.

[MINORISUKE]

1 hour ago, Wyred 4 Sound said:

Attached please find our I2S pin configuration. Based on PS Audio’s model,

.....

DSD is not indicated.  I assume it is the same as PS Audio except

No.1 DSDR+   No.3 DSDR-

It seems that "W4S standards" is a swap of No.1 and 3 from PS Audio both for PCM and DSD.  If so, however, absolute polarity swap should happen also in PCM with PS Audio compatible inputs, which nobody has ever reported (or simply noticed yet?).

[jojeong8338]

I have just tested PIN1 and PIN3 as mentioned.

 

Audirvana 3.0.3 --- Singxer SU-1 --- DIY HDMI cable --- W4S DSDv2se

 

I made DIY HDMI cable : crossed PIN1 and PIN3.

 

The test result:

1. DSD64 file : Play OK, Phase OK

2. FLAC(16-44.1k, 24-44.1k, 24-96k, 24-192k) PCM : Play OK, Phase OK

3. FLAC(16-44.1k, 24-44.1k, 24-96k, 24-192k) Upsampling to DSD128 : Play OK, Phase OK.

 

Before interchanging PIN1 and PIN3,

- PCM play OK, and Phase OK

- DSD play OK and Phase NOK

 

W4S and Singxer, thanks for your more investigation and solution in advance,

Juno

[jojeong8338]

36 minutes ago, jojeong8338 said:

 

Audirvana 3.0.3 --- Singxer SU-1 --- DIY HDMI cable --- W4S DSDv2se

 

I made DIY HDMI cable : crossed PIN1 and PIN3.

 

The test result:

1. DSD64 file : Play OK, Phase OK

2. FLAC(16-44.1k, 24-44.1k, 24-96k, 24-192k) PCM : Play OK, Phase OK

3. FLAC(16-44.1k, 24-44.1k, 24-96k, 24-192k) Upsampling to DSD128 : Play OK, Phase OK.

 

In case of Amarra4.x, the output sound is mixed like noise.

I'm so confused by the difference between Audirvana and Amarra.

[MINORISUKE]

The "PCM Phase" result in the above should be different.

W4S's statement that their PCM playback with PS Audio PWT was correct is now also doubtful.  Swapping DATA+ and DATA- results in absolute polarity swap, but they have been saying in these 10 years that there is no swap according to their post.

[jojeong8338]

54 minutes ago, MINORISUKE said:

 

The "PCM Phase" result in the above should be different.

 

Noisy issue was caused by contact problem.

After power recycling and tighten plug-in cable, I can listen all kinds of audio sources including PCM and DSD.

Thanks a lot.

Juno

[Franatic]

3 hours ago, MINORISUKE said:

The "PCM Phase" result in the above should be different.

W4S's statement that their PCM playback with PS Audio PWT was correct is now also doubtful.  Swapping DATA+ and DATA- results in absolute polarity swap, but they have been saying in these 10 years that there is no swap according to their post.

I can say without a doubt, there is phase inversion with DSD. There also is NO phase inversion with PCM in playback. Minorisuke, if you say this is not possible with data lines swapped, perhaps W4S has fixed PCM phase inversion in their DAC in some other way (like inverting PCM line again in DAC).

Anyway, if Leter can make SU-1 firmware that inverts one channel of DSD, we should be good for in-phase playback of both. The one thing I've learned about phasing is that any even number of inversions (0,2,4,etc) will come out correct. Any odd number of inversions (1,3,5,etc) will come out inverted. 

[Franatic]

On 5/1/2017 at 3:43 PM, Wyred 4 Sound said:

Regarding the PCM noise issue, putting the cap fix on either side accomplishes the same result. The reason we fixed the Singxer instead of our DAC is that our other I2S sources do not exhibit any noise issues. In addition, we had Fran’s Singxer here, not his DAC. There is a chance that if Singxer is able to update their I2S, the noise issue could also go away due to the improved clocking. Our theory is that it’s caused by isolation jitter on the Singxer output.

*******************************************

How convenient to offer the theory/opinion that the PCM noise issue is caused by Singxer. This problem only manifests in the W4S DAC and not in any of the other DACs that Singxer is working well with. I would hope that at this point you would be more open to the possibility the issue is in your i2s application.

Leter bought a used DAC2 to use for testing. I'm sure it cost more than the $399 an SU-1 would cost. Please be more careful about pointing fingers. Please thoroughly test your i2s application. It is becoming popular and more people will be using it in an application (DSD) you have not tested. It is wishful thinking that the PCM noise issue will just "go away". Evidence points that W4S likely has the problem here, not Singxer.

 

[Ienag]

The pcm problem is a W4S firmware issue.

 

Two Dac with different firmware version, one has the issue and one no.. the one with the newer firmware.

 

What phisically do the cap?  Something that can be done with a firmware update but with W4S update the firmware i impossible by ourself.

 

It's so clear...

[JohnDonaldson]

2 hours ago, Ienag said:

The pcm problem is a W4S firmware issue.

 

Two Dac with different firmware version, one has the issue and one no.. the one with the newer firmware.

 

 

@ Wyred 4 Sound, please comment on the availability of a firmware update to the DAC2-V2 that corrects the PCM noise with the SU-1 in I2S.

[Ienag]

On 28/4/2017 at 9:06 PM, Wyred 4 Sound said:

 

The other point brought up was firmware updates. Firmware is not in control of the phase issue discussed here. We have the capability of updating our USB firmware via USB port, however, this is not the input that needs to be addressed. Our product firmware requires a different method, but it also cannot alter these settings. USB firmware would need to be adjusted on the source device as it controls the pins the signal is sent on and can likely be controlled with firmware. Our side of this signal path is purely hardware, not software.

 

 

Regards,

Tony and EJ

 

They have already Answer that the firmware update is impossible via USB port, you have to live with the problem.