[svyr]

really more of a Preview

 

[Balmoral]

Great review as usual project86!

 

[TwoEars]

Nice review! Certainly seems like a very impressive product, I'm a bit of a minimalist though and I really wonder why they use two different chips for the line output and headphone section - that seems a bit strange and costly to me.
 
Does anyone know if it uses class A outputs or opamps?
 
Won't be buying one - 4k is way to much for me for what you describe as relatively subtile differances compared to the Anedio D1, still nice that products like this exist though.

 

[macrog]

I would consider it if they had used dual mono ESS9018 DAC chips for a compromise free implementation. For 4K I think it is hard to justify to not max out the potential sound quality.
 
It seems that Twisted Pear is the only game in town to Max out the Ess9018.
 
Regards
 
Macrog

 

[axw]

Quote:

I would consider it if they had used dual mono ESS9018 DAC chips for a compromise free implementation. For 4K I think it is hard to justify to not max out the potential sound quality.
 
It seems that Twisted Pear is the only game in town to Max out the Ess9018.
 

I read somewhere that the upcoming DAC-9 from NuForce is also supposed to have a pair of ESS 9018.
 
btw, it may be that in this specific case the benefits of dual chip mode (improved snr/thd only?) may be diminished by the fact that at least some of Sabre's built-in capabilities may not be utilized in single channel mode? As far as I remember Sabre has plenty of onboard features incl. spdif and dsd receivers, various filters, upsampling, some form of de-jittering, etc etc. In other words, wouldn't dual mono mode reduce its capabilities just to d/a (and thus demand adding more chips)?
Unfortunately, ESS does not provide (not openly) full technical specs so I am just speculating, don't know which features benefit from or require multichannel mode.
 

 

[macrog]

For $4K I expect a pretty compromise free implementation. Yes you are right that the ESS 9018 is harder to implement in Dual Mono mode but at this price shouldn't this be an option.
 
I am sure that ESS are hard to work with but it seems that Resonessence Labs should be well enough informed by their connection with ESS to do this. Lets face it Twisted Pear has managed to offer a no compromise solution with very few apparent resources.
 
Regards
 
Macrog

 

[macrog]

Duplicate post sorry
 
Macrog

 

[axw]

Quote:

I am sure that ESS are hard to work with but it seems that Resonessence Labs should be well enough informed by their connection with ESS to do this. Lets face it Twisted Pear has managed to offer a no compromise solution with very few apparent resources.
 

hmm.. probably I am overlooking something -- but where did you find the information that Twisted Pear has a dual mono Sabre board offering? Looking at their website, I find only Buffalo II, Opus and COD, none of them seems dual mono ...
 

 

[svyr]

Probably on the diyaudio forum. I think people suggest a separate psu+buffalo board for each channel or something of the sort

 

[Currawong]

Quote:

For $4K I expect a pretty compromise free implementation. Yes you are right that the ESS 9018 is harder to implement in Dual Mono mode but at this price shouldn't this be an option.
 
I am sure that ESS are hard to work with but it seems that Resonessence Labs should be well enough informed by their connection with ESS to do this. Lets face it Twisted Pear has managed to offer a no compromise solution with very few apparent resources.
 
Regards
 
Macrog

I think we'd be talking out of our rear-ends to be suggesting what would work best to be honest.
 

 

[project86]

Quote:

I think we'd be talking out of our rear-ends to be suggesting what would work best to be honest.
 

Bingo.
 
These guys pretty much made the Sabre DACs including the ES9018 chip. It stands to reason that they know more about its capabilities than anyone else out there. 
 
Also remember that the ES9018 has 8 channels so in a stereo configuration you are already getting 4 per channel. You could use two of them but it wouldn't really be a "dual mono" configuration. Resonessence could certainly afford another chip (roughly $40) in their $4k DAC if they felt it was significant to do so. But they say as it is they are already exceeding some of the published specs.

 

[axw]

 
Quote:

I think we'd be talking out of our rear-ends to be suggesting what would work best to be honest.

 
Exactly.
 
Sabre's clock-free dejittering algorithm seems to be data dependent, so putting it in dual mono may be problematic (common sense reasoning, I am a software person but not a hardware specialist). Check this quote from Sabre:
 
[..] an assessment is made of the width of each pulse based on its relation to recently seen pulse widths and a decision circuit assigns each a width of 1, 2, or 3 units. 
A state machine then operates on the assigned widths in succession; this state machine is searching for the block boundaries and the bit states. The state machine makes no attempt to re-time or otherwise decode the clock  – it simply
“time stamps” the event and passes it to the downstream processor.
 
so in theory at least, it is possible that two different signals (as in left/right channels) result in different time stamping decisions of left and right chip. Or am I wrong?
 
And this quote:
 
the default configuration is optimum and unless some special need is found we recommend use of the default configuration [..] The default configuration for Stereo mode, and the only configuration that can use the SPDIF input, is to wire four output channels in parallel.
 
this looks like balanced stereo output. If I read correctly, in any other mode SPDIF input does not work. Giving up on integrated SPDIF input means adding another chip and throwing in more jitter (eg. Wolfsons 880* are rated 50ps).
 
Anyhow, the quoted doc is here, maybe you can get out of it more than me..

 

[project86]

I've been busy and somehow missed that last comment.
 
I think AXW is spot on with his reasoning. There is a danger to armchair engineering, especially when it comes to devices like this that are clearly very well thought out already. More is not necessarily always better.
 
In any case, Resonessence will be exhibiting the Invicta at the RMAF show in October. Hopefully it will receive some more coverage based on that, because they really do deserve it.

 

[macrog]

Quote:

I think we'd be talking out of our rear-ends to be suggesting what would work best to be honest.
 

 
Are you suggesting the Twisted Pear Buffalo 2 using the same ESS9018 doesn't work well in dual mono mode.  I suggest you research more the results people are getting using Buffalo 2's in dual mono mode before suggesting I am" talking out of my rear end".
 
Regards
 
Macrog

 

[project86]

Macrog - I'm not interested in an argument here. You are obviously starting from the premise that 2 is always better than 1 and making your assumptions from there. Thus far I've not seen any evidence of that being the case. Someone a few posts ago actually gave a good reason why it would NOT be the case.
 
A project made by forum members on a DIY enthusiast site is not good evidence of your theory. It's an interesting project to be sure, but that's about all we can say for now. Twisted Pear Audio is an excellent company and I highly respect their designs, so that isn't even the issue.