exaU2I - Multi-Channel Asynchronous USB to I2S Interface

[oyvine]

Originally Posted by rsdio

... As far as DIY is concerned, the exaU2I is offered via the model of "buy one and see if you can modify it to meet your needs." Thus, we have a hammer. When questions are asked, it seems that no details are forthcoming, so I made the snarky remark that every problem must be a "nail." ...

Hi,
I just wanted to congratulate you with your new thread in the vendor area. It is also nice to see that you still allow critical comments. You are much more likely to get me as a customer with this approach. That being said, I think it would be nice to see some sort of jitter measurements. If it is good it will for sure sell a few boards

regards,
Øyvin

[exa065]

Originally Posted by oyvine

Hi,
I just wanted to congratulate you with your new thread in the vendor area. It is also nice to see that you still allow critical comments. You are much more likely to get me as a customer with this approach. That being said, I think it would be nice to see some sort of jitter measurements. If it is good it will for sure sell a few boards

regards,
Øyvin

Øyvin, this question was answered long ago. Here is my post 211 from 28th March 2011:

"Here is my perspective on the impact of jitter on the exaU2I output. The clocks are responsible for about 2ps of jitter. The FPGA causes about 40ps and the GMRs another 100ps. This would add up to about 108ps on the exaU2I I2S output. Let's assume that it is 20 times more - 2000ps. Here is the jitter sensitivity for Sabre DAC to 2ns of jitter. I don't see any reason to be concerned even if two stages of GMR are as used as presented by RayCtech on the previous page of this thread. "

This is a worst case scenario. Third parties have attempted to measure the jitter and have come up with readings between 40 and 100 ps. We base our modeling on the worst-case scenario.

[exa065]

qusp, barrows, You've made your point repeatedly on this thread. We know your opinion and you've heard our position. It is time to move on. Going forward I will be deleting posts that are repetitive and have been answered in the past.

[rsdio]

Originally Posted by exa065

This is a worst case scenario. Third parties have attempted to measure the jitter and have come up with readings between 40 and 100 ps. We base our modeling on the worst-case scenario.

Thanks for the graph, but it does not appear to be a measurement of the exaU2I. The red line, with higher jitter, is labeled as "a competitor's part" while the dark line, with consistently low jitter, is labeled as "SABRE DAC" ... where is the graph for the exaU2I I2S output?

I think the key point here is that people can look up the part specifications themselves, but they do not want to base their decision on general data sheets, and they do not want worst case models. It seems that people are asking for actual measurements with test equipment designed to indicate jitter.

Forgive me if such test results have been published earlier in this thread. If they have, then might I suggest putting them on your web site since people seem to be remiss to scan this long thread for such information. If the results are there, I'd appreciate a reminder (I may have missed them the first time through).

[exa065]

The graph indicates the ES9018 tolerance to jitter. Please see the original discussion starting on page 22. The worst case scenario for exaU2I is much lower than the 2ns jitter impact indicated in the graph. Further the measurement of jitter is not straight-forward and easy.

What matters is not the jitter on the exaU2I output, but the jitter level at the receiving end - the DAC. For example some USB to SP/DIF devices claim 3ps jitter. This is possible under lab conditions on the near side of the cable. By the time the signal is delivered and received the jitter reading will be off by nanoseconds.

Instead of relying on measurements that will be irrelevant because the impact of the receiving end of the I2S connection is not known, we use the worst case estimate and we make sure that it is within the jitter-cancelling capability of ES9018. Than we evaluate the jitter-related artefacts in the analogue output. It is more pragmatic to verify the end result rather than speculating on measurements that are partial reflection of the reality.

Our measurements on the analogue end correlate with subjective listening tests done by users. This gives us confidence that the balanced approach that we've taken between minimizing noise by using ground isolation and accepting jitter caused by the ground isolation is paying off.

I am not saying that a certain level of jitter/noise is acceptable. exaU2I implements a balance that we believe is optimal. My advice to users is to believe to their own ears. Consider also the opinions of users that have first-hand experience with the device.

[drspence]

I am a very happy exaU2I owner since June.

The performance of your interface is superb and it is dependable. Equally as important, your superb customer service and patient demeanor results in someone I enjoy working with.

Keep up the good work! I look forward to your continued advancements.

George, frankly you have been overly patient with the posters who do not own and are unfamiliar with your product. In my opinion, there is an unjustified and concerted effort to discredit you and your products. Is a hidden (perhaps not so hidden) agenda is at work here!

[Nikola Krivorov]

Two lines from me too - I think I am very representative of a lot of the USB->I2S buyers - I can talk for hours on jitter(it is bad,bad,bad) and what I read it about on forums and how one solution is supposedly better than the other. But I don't really understand it. So what I did is that I BOUGHT most of the available converters and compared them on my Buffalo II. The ExaU2I is in an absolutely different league than most of them. I used to think that GMRs will kill what's good in the Exa but that was when I was trying to imagine how it will sound,I don't care at the moment - I am listening to music.

P.S. I have one old favorite commercial, it kind of says a lot:
Pub Hakkinen/ Becker Mercedes Classe C video on CastTV Video Search

[TheShaman]

Originally Posted by Nikola Krivorov

So what I did is that I BOUGHT most of the available converters and compared them on my Buffalo II. The ExaU2I is in an absolutely different league than most of them

Interesting Nikola.
I'm sure a lot of members (myself included ) would be interested in reading your comments regarding the comparisons you did. Or have you done so already in a post I missed?

[Nikola Krivorov]

Well,

I haven't posted much because I am no authority of any kind and anyone knows that it is not just one device vs. the other, it would take a full description of the system, the music, how well I have tried to optimize each device and so on. I will only say that I had some of my friends who are not interested in computer audio at all and there was no question about which is best. The exa was described as the only one that did not give you the feeling that the computer is the source, the comaprison was done against a top Naim CD player, my ex Examplar 3910, a heavily modified Young DAC and even a turntable. I want to stress again that I feel that is best to decide between devices that you actually own, for me that is the only way to be objectively subjective. I know that audio is passion and we are not always rational, if you want to hear that the ExaU2I is not good you will hear it and vice versa, what is important for me is that it works flawless and sounds fantastic but I must admit that if there is a mK II with better parts I will get one Hope that helps.

[supra]

I also have had all the interesting dacs and interfaces through my system- Audiophilleos, M2HiTech, Emperical Audio, Metrum Octave, Lavrey, etc and the EXA U2I with a Buff2 and tube output stage is in another leauge. And being able to upsample to 352.8Khz and 384Khz is a big plus imo. Even well recorded MP3 sounds a whole lot better when upsampled properly- I'm using PureMusic and Audirvana Plus and it is as good as I've heard.
I'm building 3 more units for friends as once you hear this it's hard to listen to anything else again.
And it's very refreshing to experience personal service in these days of computer generated replies to customer service. George's accent is a bit tricky to an Aussie, but he probably finds my accent even harder to understand, nevertheless he's been able to help me overcome a few set up niggles with relative ease.
The U2I is a sleeper- watch it become the reference standard when the audio world wakes up.