Grimm MU1 vs Aurender N30SA vs Taiko Audio

[GSOphile]

It might be apropos to remind everyone that the Internet runs on millions of miles of generic cable. The TCP/IP protocol that transports terabytes of data along the web every second of every day is built to be remarkably fault tolerant, originally designed so that the network would still function even if large parts were taken down by a nuclear strike (the basic architecture of the Internet was designed in the 1960s).

Fault tolerant for computer data is a bit different from high SQ digital audio. It's not just "bits is bits."

 

[kennyb123]

Fault tolerant for computer data is a bit different from high SQ digital audio. It's not just "bits is bits."

Exactly. With high end audio, what matters is that the right bits arrived at exactly the right time and nothing else came along for the ride.

TCP/IP only guarantees that the right bits get transmitted. It can do nothing to address noise nor can it guarantee that the bits are supplied to the DAC at exactly the right time.

 

[Mark Jones]

Yes! It’s very good. However, it still fails to deliver the punch I desire in soft rock. It’s too polite IMO - that is for my taste!

Yes it's polite for sure, but please tell me you were using the AES output and not USB when you initially evaluated the MU1. The MU1's magic is only realized via the AES output.

 

[PYP]

Yes it's polite for sure, but please tell me you were using the AES output and not USB when you initially evaluated the MU1. The MU1's magic is only realized via the AES output.

and choosing the right oversampling rate (3 available) for your particular DAC (right according to your ears).

 

[Mikeg1]

Hello, I too changed to AES from USB out of my Grimm MU1 and it is better! I am very interested in the N30, waiting on its certification as a Roon endpoint. I want to run Roon Core on my Grimm MU1 in my second system, and then possibly use the N30 as streamer in my best system. I understand Roon may not be best sound quality, but I do enjoy its music selections. I have modded my MU1 with a new ethernet cable inside it, replacing the stock generic 1 foot cable. It makes a big difference. Even so much so, that running roon core on this MU1 and then using the Tambaqui internal streamer (bridge) now sounds much better in the second system, even without a dedicated streamer!

fast.freight: I want to know what internal ethernet cable you upgrade to in the MU1?

Questions for the group: I am looking to upgrade from an AfterDark Rossana Server to the Grimm Audio MU1. In my system, I want to incorporate the Theoretica BACCH4MAC Processor/Roon Endpoint so I plan to maintain the following set-up using Wireworld Platinum ethernet cables: Google Fiber Router-Fidelizer AirStream2 switch-Grimm MU1 to A side of Uptone Audio EtherRegen-Waversa REF LAN filter (connected to B side of EtherRegen)-(all ethernet)BACCH4MAC-USB output to Ayon Audio Kronos Ultimate DAC/Preamp. From the reviews, the use of the AES digital cable from the Grimm to DAC takes full advantage of the FPGA chip processing as compared the ethernet connection. I would be buying a current Grimm MU1 directly from Grimm which has the latest internal upgrade. Any inputs?

It looks like I need to move the Grimm MU1 before the DAC and skip the BACCH4MAC so I connect the Grimm using the AES digital out for optimum performance. If I want to maintain the ethernet chain as originally planned, what Server is recommended (Pink Faun, Taiko, Grimm, Antipodes, or Lumin) using all ethernet connections up until the BACCH4MAC?

 

[Bosrt]

I think fast.freight has made some changes since his last post here, they are interesting.

 

[fast.freight]

Hello, I modified my system signature to reflect the MU1 internal modification.
Well worth it!

 

[analogsa]

TCP/IP only guarantees that the right bits get transmitted. It can do nothing to address noise nor can it guarantee that the bits are supplied to the DAC at exactly the right time.

The timing of the bits is a red herring. It matters at the dac data input generally speaking (in many dacs the data input is not particularly jitter sensitive anyway), but it's really unimportant in the ethernet packets. There are all sort of mechanisms aligning, bufferring and reclocking data between ethernet and dac, especially in well engineered high end dacs.

 

[Doc76]

For what it’s worth, different systems may present different results, but I’ve had the Grimm MU1, Aurender N30SA, and the Taiko Audio Extreme all in my system. Taiko was the winner, with the N30SA being app 10-15% of it and the Grimm not even close to the N30SA. Needless to say I own a Taiko.

 

[kennyb123]

The timing of the bits is a red herring. It matters at the dac data input generally speaking (in many dacs the data input is not particularly jitter sensitive anyway), but it's really unimportant in the ethernet packets. There are all sort of mechanisms aligning, bufferring and reclocking data between ethernet and dac, especially in well engineered high end dacs.

The existence of this thread proves that it is not a red herring. That music servers still bring substantial improvements, despite the fact that DACs feature all these mechanisms, tells us that this is an extremely difficult problem to solve. Reducing noise to a point where it can no longer negatively impact sound quality isn’t yet possible.

 

[fast.freight]

Taiko was the winner, with the N30SA being app 10-15% of it and the Grimm not even close to the N30SA.

Hello, can you clarify please? The N30SA performed at 15 % while the Extreme was 100%? And the Grimm was then below 0%? Thanks.

 

[audio.bill]

Hello, can you clarify please? The N30SA performed at 15 % while the Extreme was 100%? And the Grimm was then below 0%? Thanks.

I interpreted it as rating the Aurender N30SA to be within 10-15% (or 85-90% as good as) Taiko Extreme, with the Grimm performing well behind and not even close to the N30SA.

 

[fast.freight]

Thanks for clarifying.

 

[sbo6]

The rest of the English speaking world uses the word "ecosystem" to describe "any system or network of interconnecting and interacting parts". The number of parts that comprise it doesn't factor into whether that term applies. It's just that the parts are interconnected and interacting. Certainly Taiko has been releasing a set of products that interconnect and interact so "ecosystem" is an apt description of that. Just my two cents.

From a product perspective, an ecosystem is a grouping of compatible accessories that co - benefit. It's also a great way to increase revenue, sell em' the big item(s) and you have them bought in to buy the accessories, adapters doodads and gen on gen sell, rinse repeat. Think Apple. And that's definitely Taiko, they seem to make a vg music server, but the product and "ecosystem" is expensive and continued revisions of parts keeps $$ evergreen. A nice dollop of good free advertising doesn't hurt either ;-)

 

[kennyb123]

adapters doodads and gen on gen sell, rinse repeat. Think Apple. And that's definitely Taiko, they seem to make a vg music server, but the product and "ecosystem" is expensive and continued revisions of parts keeps $$ evergreen.

As many have discovered, the network can have an influence on how well a music server performs. I believe Taiko discovered that the solving the network problem was essential to getting their server to the next level of performance. It appears that their customers agreed, despite it costing them more than what most of us spent on our servers to get there.

 

[godofwealth]

I’m not sure what you mean by “solving the network problem”. The TCP/IP protocol that runs the internet solved the network problem and it was developed 60 years ago as a Cold War project by DARPA. Every day literally millions of terabytes flows through the internet flawlessly. Businesses depend on error-free transmission. Perhaps you means something else?

 

[sbo6]

As many have discovered, the network can have an influence on how well a music server performs. I believe Taiko discovered that the solving the network problem was essential to getting their server to the next level of performance. It appears that their customers agreed, despite it costing them more than what most of us spent on our servers to get there.

If you're referring to their network switch, there are dozens out there for decades that will aid on pretty much any network server / streamer. I'm not sure how this is remotely novel to Taiko. Are they generating some unique network traffic / protocol where their "special" switch aids only their device? Let me guess, it's part of their ecosystem. ;-)

 

[kennyb123]

I'm not sure how this is remotely novel to Taiko. Are they generating some unique network traffic / protocol where their "special" switch aids only their device? Let me guess, it's part of their ecosystem. ;-)

I can understand how being disinterested in what is novel about Taiko’s approach might leave one with that impression - though I think most folks would reserve judgement until they’ve done their homework.

 

[sbo6]

I can understand how being disinterested in what is novel about Taiko’s approach might leave one with that impression - though I think most folks would reserve judgement until they’ve done their homework.

You never answered my question. That speaks volumes.

Happy Listening!

 

[benage]

As many have discovered, the network can have an influence on how well a music server performs. I believe Taiko discovered that the solving the network problem was essential to getting their server to the next level of performance. It appears that their customers agreed, despite it costing them more than what most of us spent on our servers to get there.

This seems like a real reach here based on how Ethernet works. It's not like your DAC is decoding your audio at the exact time it's transmitted across the Ethernet cable.
This is why "special" Ethernet cables are snake oil. The media player componentry will cache/buffer the file locally. In reality the song has probably buffered within a second or less, not "live streaming" across the cable.

This is classic audiophile marketing, just like Aurenders "double isolated LAN" port. It's such a joke, regardless of the quality and design of the amp/server design.

As the old saying goes, "a fool and his money are easily parted".