SGM Taiko Extreme

[Miska]

1 minute ago, Nenon said:

I use I2S now, so it matters even more.

 

I2S is for connecting between chips, maximum 20 mm distances. And it needs very very careful PCB layout to not totally screw up. Sure, you can make things very sensitive to all changes (hard to get functional at all) if that is your goal...

 

I just don't prefer to go asking for trouble.

 

[Nenon]

13 minutes ago, Miska said:

 

I2S is for connecting between chips, maximum 20 mm distances. And it needs very very careful PCB layout to not totally screw up.

It's using LVDS with very very careful PCB layout... And I like it better than any USB I have tried so far. 

 

13 minutes ago, Miska said:

Sure, you can make things very sensitive to all changes (hard to get functional at all) if that is your goal...

Not my goal obviously. My point was that changes to upstream clocks with USB have always made a difference with any DAC I have tried, no matter how much DAC manufacturers / engineers keep saying that ONLY the DAC clock matters. I am yet to hear a DAC that is completely immune to crappy clocks in front of it. 

 

[Miska]

3 minutes ago, Nenon said:

It's using LVDS with very very careful PCB layout... And I like it better than any USB I have tried so far. 

 

I have not yet seen such... But I favor Ethernet over USB.

 

4 minutes ago, Nenon said:

Not my goal obviously. My point was that changes to upstream clocks with USB have always made a difference with any DAC I have tried, no matter how much DAC manufacturers / engineers keep saying that ONLY the DAC clock matters. I am yet to hear a DAC that is immune to crappy clocks in front of it. 

 

Can you provide some measurement results? Yes, I've seen using I2S over cable making difference, but towards worse.

 

[Nenon]

1 hour ago, Miska said:

 

I have not yet seen such... But I favor Ethernet over USB.

 

 

Can you provide some measurement results? Yes, I've seen using I2S over cable making difference, but towards worse.

 

 

Sorry, if it is not clear I am an end consumer / DIY-er in some cases , not an industry manufacturer/designer/engineer. I don't own precise measurement equipment, just some basic ones.

 

What I said is that upgraded upstream clocks sound better in my system. I don't believe that the only clock that matters with asynchronous signal is the one on the DAC, because that contradicts my experience. I have not heard a DAC that is completely immune to upstream clock changes. Rajiv's review on the DCS DAC was interesting. I have never tried a DCS DAC or the top of the line MSB DACs. But with the other dozen of DACs I have tried, I can clearly hear audible difference when making changes to the upstream clocks. 

 

When it comes to digital sources, I have tested every product that has piqued my interest. That includes streamers like the microRendu, ultraRendu, sms-200, Allo DigiOne Signature Player. It also includes PCIe cards like the JCAT USB card, PinkFaun USB Bridge, PinkFaun I2S Bridge, SOTM USB card, etc. It includes quite a few different motherboards, CPUs, RAM, power supplies, etc. 

 

After testing all that, I like the sound of the PinkFaun I2S card with ultraOCXO clock and Sean Jacobs DC3 LPS the best of all in my system. However, it is clear that this is DAC dependent. Other implementations (i.e. USB, upsampling, etc.) work better on some DACs. So I always look at the entire chain from the file you play all the way to the analog conversion as one complex system and how software, hardware, and everything in between interracts together. 

 

But back to asynchronous tests with USB cards - a PinkFaun USB bridge with ultra OCXO clock installed sounds a lot better in my system (with all the DACs I have tried) than the same PinkFaun USB bridge without the ultra OCXO clock. Also their lower class OCXO clock installed on the same card sounds different. It's literally a matter of removing the ultra OCXO clock and putting the lower class OCXO clock, so that test eliminates a lot of questionable variables and leaves the clock as the only significant change in the system. That to me is a proof that the clocks upstream of the DAC make some kind of a difference. I have no respect of the manufacturers (at least of the DACs I have tried) who claim the contrary. To me, it is clear that they have no idea what they are talking about. 

 

Lastly, I have no desire to argue with anyone here, neither have the time for this. And I especially try to avoid arguments with people with 5-digit posts. Sorry, I have more important things to do in my life. Those discussions have been had many times in these forums and are not useful to anyone. No matter what you measure and what you say, you will not change my opinion. And I will not try to change yours. So let's move on and talk about the Extreme server, which has a state of the art engineering for this time. I would really like to have the opportunity to listen to one myself. 

 

[Blackmorec]

I have a modem (LPS)—-> router (LPS) ~~Wi-fi~~>access point(LPS)/ethernet cable——>AQVoxSE(LPS)—->ethernet cable —> server —>USB cable ——> DAC & Amps

 

It is totally illogical that anything noise-wise or clock-wise before the server can be heard....yet you can clearly hear the effects of all the power supplies, cables, switches and even anti-vibration measures anywhere within the described network stream.  The WAN cable between modem and router, even the Atacama stand under the router make small but positive contributions. I was as surprised as anyone so please don’t claim its expectation bias. I only made the changes because I kept hearing the differences...not because they were expected. Quite the opposite....i can’t figure why they make a difference given the network layout and thorough isolation. 

What is clear however is that the results of a well screened, vibration optimised, low jitter and noise  data stream are simply stunning when they’re all added together, so I’m in complete agreement with Nenon’s findings

[Miska]

57 minutes ago, Blackmorec said:

i can’t figure why they make a difference given the network layout and thorough isolation. 

What is clear however is that the results of a well screened, vibration optimised, low jitter and noise  data stream are simply stunning when they’re all added together, so I’m in complete agreement with Nenon’s findings

 

If you are using shielded ethernet cables, you are destroying the isolation ethernet would otherwise provide. For audio, only use UTP cables with plastic (non-metallic) connector bodies to make sure that you are no inadvertently creating ground connections between those devices...

 

That's unfortunately quite common mistake.

 

[EuroDriver]

20 hours ago, Nenon said:

 

Sorry, if it is not clear I am an end consumer / DIY-er in some cases , not an industry manufacturer/designer/engineer. I don't own precise measurement equipment, just some basic ones.

 

What I said is that upgraded upstream clocks sound better in my system. I don't believe that the only clock that matters with asynchronous signal is the one on the DAC, because that contradicts my experience. I have not heard a DAC that is completely immune to upstream clock changes. Rajiv's review on the DCS DAC was interesting. I have never tried a DCS DAC or the top of the line MSB DACs. But with the other dozen of DACs I have tried, I can clearly hear audible difference when making changes to the upstream clocks. 

 

When it comes to digital sources, I have tested every product that has piqued my interest. That includes streamers like the microRendu, ultraRendu, sms-200, Allo DigiOne Signature Player. It also includes PCIe cards like the JCAT USB card, PinkFaun USB Bridge, PinkFaun I2S Bridge, SOTM USB card, etc. It includes quite a few different motherboards, CPUs, RAM, power supplies, etc. 

 

After testing all that, I like the sound of the PinkFaun I2S card with ultraOCXO clock and Sean Jacobs DC3 LPS the best of all in my system. However, it is clear that this is DAC dependent. Other implementations (i.e. USB, upsampling, etc.) work better on some DACs. So I always look at the entire chain from the file you play all the way to the analog conversion as one complex system and how software, hardware, and everything in between interracts together. 

 

But back to asynchronous tests with USB cards - a PinkFaun USB bridge with ultra OCXO clock installed sounds a lot better in my system (with all the DACs I have tried) than the same PinkFaun USB bridge without the ultra OCXO clock. Also their lower class OCXO clock installed on the same card sounds different. It's literally a matter of removing the ultra OCXO clock and putting the lower class OCXO clock, so that test eliminates a lot of questionable variables and leaves the clock as the only significant change in the system. That to me is a proof that the clocks upstream of the DAC make some kind of a difference. I have no respect of the manufacturers (at least of the DACs I have tried) who claim the contrary. To me, it is clear that they have no idea what they are talking about. 

 

Lastly, I have no desire to argue with anyone here, neither have the time for this. And I especially try to avoid arguments with people with 5-digit posts. Sorry, I have more important things to do in my life. Those discussions have been had many times in these forums and are not useful to anyone. No matter what you measure and what you say, you will not change my opinion. And I will not try to change yours. So let's move on and talk about the Extreme server, which has a state of the art engineering for this time. I would really like to have the opportunity to listen to one myself. 

 

Hi Nenon,

 

Think of each clock like a radio transmitter with a carrier wave at the clock frequency.  Vibration causes frequency modulation, power supply fluctuations result in amplitude modulation.  RF emissions in the MHz to GHz range are very hard to stop, they get everywhere

[Darryl R]

On 11/3/2019 at 4:24 AM, isquirrel said:

 

If you use USB your DAC receives the data asynchronously, that is to say that your DAC re-clocks the data, which voids any benefits of using a Word Clock sync, I have experimented with different outputs from the W20 and only the AES 110ohm & the SPDIF outputs are affected by the word clock sync option. As I want to be able to take advantage of high res files it means using the USB output anyway. So the word clock sync is of no value. MSB offer a Pro USB module http://www.msbtechnology.com/accessories/prousb/ which effectively reduces any noise out of the USB output by converting the data to fibre. 

 

There is a sense in which I wish I had not seen this thread.  I've been thinking about this server for a while, and like the OP, I've been in denial about what might be going on here.  Emile has been very kind and patient with some of my questions, but he clearly implies this is the server that best complements the MSB Select II DAC.  I'm starting to seriously consider it.  Maybe I need an intervention ;), at least to wait until Emile comes out with the trickle-down version sometime next year.

[isquirrel]

7 hours ago, Darryl R said:

 

There is a sense in which I wish I had not seen this thread.  I've been thinking about this server for a while, and like the OP, I've been in denial about what might be going on here.  Emile has been very kind and patient with some of my questions, but he clearly implies this is the server that best complements the MSB Select II DAC.  I'm starting to seriously consider it.  Maybe I need an intervention ;), at least to wait until Emile comes out with the trickle-down version sometime next year.

I have had the Taiko Extreme for a month or so now, I have over 600 hours on it. It has sounded better and better and I still don't think it has hit its peak. If you have a MSB Select II DAC, you owe it to yourself to bite the bullet and get this beautifully built server. 

 

The major test for any pice of equipment is how much time I spend listening to music through it. I find myself login from listening occasionally once maybe twice a week on rare occasions to listening every day. I can't give the Extreme a better compliment than that. 

 

I have a few friends who are Audiophiles, they seem to be coming around more and more often to listen my system. That never happened before. I have even compared it to 1/2 inch tape masters, which when I download the same file from HDTT it sounds at least as good. 

 

Do you have the ProUSB Module? 

[Darryl R]

Hello,

 

I've ordered it, along with the same cable you have (I use all Shunyata Sigma throughout my system).  This is such a difficult decision, as I would seriously deplete my very limited financial resources.  I discussed this briefly today with Jonathan at MSB via email, and even they said all you need is the galvanic isolation of the ProUSB.  I then sent him a link to this thread and he thought your review was credible, and advised hearing it for myself.

[Darryl R]

Before I check out for the evening, I wanted to leave you with another question.  I looked at some of the early posts in the long WBF thread, where Taiko say their optimizations for the Extreme really took off when Emile applied things he'd learned from modifying the Herzan devices.  Are you using any special vibration control measures with your Extreme?

[isquirrel]

19 minutes ago, Darryl R said:

Before I check out for the evening, I wanted to leave you with another question.  I looked at some of the early posts in the long WBF thread, where Taiko say their optimizations for the Extreme really took off when Emile applied things he'd learned from modifying the Herzan devices.  Are you using any special vibration control measures with your Extreme?

Congratulations on your purchase. You will not regret it. I use mine on an HRS VXR stand with TOTL M3X2 platforms, I have also found that adding a Setchi D3 to the unused USB port on the Extreme has over time (they take time for the full effect to come into play) I have also re-cabled my Network which is quire simple, Modem/Router/Switch with Shunyata Sigma Ethernet Cables.

 

I have a few sets of the HRS Vortex footers, I am interested to see what if any difference they make under the Extreme. The Extreme is so well. engineered that adding this and that accessory to it may change the sound however the real question is has it changed the sound for the better. 

 

I have heard that people have tried the Daiza platforms under the Extreme and DAC with great success. YMMV.

[isquirrel]

On 11/11/2019 at 7:50 AM, Nenon said:

I have never tried a DCS DAC or the top of the line MSB DACs. But with the other dozen of DACs I have tried, I can clearly hear audible difference when making changes to the upstream clocks. 

 

I too thought that ultra precise clocks are important inn a music server, the transition from the Aurender W20 to the Taiko Extreme and the use of MSB Pro USB module has debunked that idea in my system. The MSB Select II DAC has an option of a Femto 33 clock which is the most accurate clock you can buy. The MSB Select II re-clocks all data and I have found removing upstream clocks have improved the sound. Until you have heard it, it's difficult to accept however myself and a bunch of MSB Select II owners have all come to the same realisation.

 

If you get the opportunity try and have an extended listen to the latest MSB DAC's. They really are a revelation. I have always loved Analog front ends, that's changed with my current setup.  

[Darryl R]

1 hour ago, isquirrel said:

Congratulations on your purchase. You will not regret it. I use mine on an HRS VXR stand with TOTL M3X2 platforms, I have also found that adding a Setchi D3 to the unused USB port on the Extreme has over time (they take time for the full effect to come into play) I have also re-cabled my Network which is quire simple, Modem/Router/Switch with Shunyata Sigma Ethernet Cables.

 

I have a few sets of the HRS Vortex footers, I am interested to see what if any difference they make under the Extreme. The Extreme is so well. engineered that adding this and that accessory to it may change the sound however the real question is has it changed the sound for the better. 

 

I have heard that people have tried the Daiza platforms under the Extreme and DAC with great success. YMMV.

 

To clarify, I've ordered the ProUSB, not the SGM server, at least not yet.

 

With respect to your comments about clocks above, John Swenson says something similar about his new EtherREGEN device which removes clock noise.

[Nenon]

1 hour ago, isquirrel said:

I too thought that ultra precise clocks are important inn a music server, the transition from the Aurender W20 to the Taiko Extreme and the use of MSB Pro USB module has debunked that idea in my system. The MSB Select II DAC has an option of a Femto 33 clock which is the most accurate clock you can buy. The MSB Select II re-clocks all data and I have found removing upstream clocks have improved the sound. Until you have heard it, it's difficult to accept however myself and a bunch of MSB Select II owners have all come to the same realisation.

 

2 hours ago, isquirrel said:

I have had the Taiko Extreme for a month or so now, I have over 600 hours on it. It has sounded better and better and I still don't think it has hit its peak. If you have a MSB Select II DAC, you owe it to yourself to bite the bullet and get this beautifully built server. 

 

I am a little bit confused what are you trying to say here. I was trying to say that many manufacturers make claims that upstream devices don't make a difference, but I have never heard a DAC yet that is completely immune to upstream devices (never heard any MSB DACs, or the DCS DAC Rajiv reviewed, in my system, though). I have heard the opposite many times - changing upstream devices makes a bigger difference that changing DACs. By improved upstream clocks I did not mean daisy chaining a bunch of USB de-crapifiers between the server and the DAC. I am pretty sure if you try to do that with your Taiko Extreme, you would end up with worse quality after all. But your Taiko, according to Emile, has something that works better than OCXO clocks, whatever that is... There are obviously clocks in every computer. What I actually meant was generating a clean, low-jitter, low-noise digital signal that goes directly to your DAC. That's what Taiko Extreme (and any other good quality source IMO) does. 

 

So, are you saying that devices upstream from the MSB Select II DAC with Femto 33 clock don't make a difference? If that's the case, hats off to the MSB engineers. But does that mean you can replace your Taiko Extreme with a MacBook (or a NUC, or another genetic computer) and you won't hear any audible difference? I thought you were saying that the Taiko Extreme does make a big difference. So I am a little confused exactly what are you trying to say? 

 

[isquirrel]

My apologies, when I re-read my post I can see how its confusing. So let me start from the beginning:

 

I already had the Aurender W20 when I purchased the MSB Select II DAC (+dual power supplies, 33 Femto clock) I decided to use the AES 110 ohm output from the Aurender as I could then use the additional clock sync option on the W20 & on the DAC. That produced a sound that at the time was probably at or close to Start of the Art digital playback. Not using the USB option mean't that I was missing out on MQA streaming and ultra high res files. Then Jonathan from MSB told me they had engineered a USB solution that should have made music servers all but redundant. This was interesting and I subsequently acquired the ProUSB module and external USB to Fibre module.

 

More info here: http://www.msbtechnology.com/accessories/prousb/

 

Still using the W20 I moved to the Pro USB solution. It was clearly superior although I still found that different USB cables sounded well, different.

 

Fast forward to the Taiko Extreme, a unit was lent to me and much I as tried not to enjoy it, the facts were obvious it sounded considerably better.

 

So yes I 100% agree that even with the ProUSB module and the fact that the DAC re-clocks all data, the source as it has forever been (I am thinking of the Linn Sondek LP12 days :-)) is all important. It doesn't matter how good your DAC is if it doesn't have the best data fed to it, its simply not going to be there. 

 

I have tested the Taiko Extreme with MSB's Bit Perfect test files, unfortunately they only work with MSB's latest DAC's. The files all played Bit Perfect the same as they did on the Aurender W20. So if they were playing bit perfect what was going on to generate such a difference in sound quality? 

 

The answer I have come to believe is noise and isolation. Emile has taken such "extreme" length's to lower the noise floor and or shift it to frequencies which do not cause major issues in the digital playback realm.

 

I trust this explains my findings with greater clarity.

 

MSB offers a number of input options: I have found the following to be the case in order of preference when using either the W20 or the taiko Extreme.

 

1.   ProUSB

2.   USB

3.   Network Renderer V2 

 

 

 

    

[Darryl R]

6 hours ago, isquirrel said:

The answer I have come to believe is noise and isolation. Emile has taken such "extreme" length's to lower the noise floor and or shift it to frequencies which do not cause major issues in the digital playback realm.

 

I trust this explains my findings with greater clarity.   

 

I love that fact that Emile and company have used their God-given freedoms to create something remarkable, and at the moment seemingly inscrutable.  Even if I come to the conclusion I can't afford it, I'm delighted it's there for some to enjoy.

 

That said, this doesn't provide greater clarity for me, not that I insist on having it right now.  I still come back to MSB's statements about the isolation of the ProUSB module.  Jonathan tells me he uses a Roon NUC at home and a Mac at work, and that's all he needs.  John Swenson also minimizes the role of server tweaks in certain scenarios with his groundbreaking EtherREGEN.  But there's something more, and we come back to the the basic question of "what is going on?"

 

For now, I suppose we can refer to the seminal thread and read about manipulation of "RF soup" and "DRAM refresh", all a certain amount of which is going to be the undisclosed intellectual property of Taiko Audio:

 

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/

 

EDIT: Emile told me he plans to produce a "2/3" version in a few months.  That will be interesting to see.  I'm reminded of all the Wilson Audio videos I've recently watched about so many of their products which benefited from pricier offerings.

 

 

 

 

 

 

 

[Nenon]

7 hours ago, isquirrel said:

My apologies, when I re-read my post I can see how its confusing. So let me start from the beginning:

 

I already had the Aurender W20 when I purchased the MSB Select II DAC (+dual power supplies, 33 Femto clock) I decided to use the AES 110 ohm output from the Aurender as I could then use the additional clock sync option on the W20 & on the DAC. That produced a sound that at the time was probably at or close to Start of the Art digital playback. Not using the USB option mean't that I was missing out on MQA streaming and ultra high res files. Then Jonathan from MSB told me they had engineered a USB solution that should have made music servers all but redundant. This was interesting and I subsequently acquired the ProUSB module and external USB to Fibre module.

 

More info here: http://www.msbtechnology.com/accessories/prousb/

 

Still using the W20 I moved to the Pro USB solution. It was clearly superior although I still found that different USB cables sounded well, different.

 

Fast forward to the Taiko Extreme, a unit was lent to me and much I as tried not to enjoy it, the facts were obvious it sounded considerably better.

 

So yes I 100% agree that even with the ProUSB module and the fact that the DAC re-clocks all data, the source as it has forever been (I am thinking of the Linn Sondek LP12 days :-)) is all important. It doesn't matter how good your DAC is if it doesn't have the best data fed to it, its simply not going to be there. 

 

I have tested the Taiko Extreme with MSB's Bit Perfect test files, unfortunately they only work with MSB's latest DAC's. The files all played Bit Perfect the same as they did on the Aurender W20. So if they were playing bit perfect what was going on to generate such a difference in sound quality? 

 

The answer I have come to believe is noise and isolation. Emile has taken such "extreme" length's to lower the noise floor and or shift it to frequencies which do not cause major issues in the digital playback realm.

 

I trust this explains my findings with greater clarity.

 

MSB offers a number of input options: I have found the following to be the case in order of preference when using either the W20 or the taiko Extreme.

 

1.   ProUSB

2.   USB

3.   Network Renderer V2 

 

 

 

    

 

That makes sense and is inline with what I was thinking. The fiber-isolated USB module and the ultra low noise clock in the MSB DAC must be reducing the influence of the upstream components quite significantly, but like with other products, they are still there. Otherwise a different USB cable or upstream server would not make a difference. I am sure that even a MacBook would sound really good, but if you want those last bits of performance and refinement, the Taiko Extreme would do it.

BTW, it's the same with my EtherREGEN switch - while it almost annulates upstream effects, the word "almost" plays a key role here.

Thank you for the feedback. 

[RickyV]

 

1 hour ago, Darryl R said:

Even if I come to the conclusion I can't afford it,

 

Maybe you should check out the streamer from Grimm audio the MU1. They have also built a great streamer.

[Darryl R]

24 minutes ago, RickyV said:

 

 

Maybe you should check out the streamer from Grimm audio the MU1. They have also built a great streamer.

 

Thanks Ricky, I'll check it out later.  But I've really got my eye on these guys.  They invested in highly-sensitive audio and RF detection gear, and spent at least a year of R&D on this (not to mention the TDY to Mike Lavigne's room in Washington).  With that kind of risk taking and persistence, they deserve a rich reward.

[bmichels]

Few years ago I had demo of the Aurender W20 and the SGM2018.

Because I do not like the idea of upsampling and because I was afraid by the complexity of the optimisation needed by the myriad of HQ Player filters.... I decided to go for the W20.

Now it is time for upgrade. I am therefore hesitating between the new Aurender W20SE,  the Playback Design MPT-8 (with CD/SACD reader & the optional Roon core module) and of course, the first choice being the Taiko Audio SGM Extreme... now that it do not upsample anymore with HQ player.

I just wonder: Now that Taiko Audio SGM Extreme do NOT uses HQ Palyer, I hope that there is no more optimisation needed ! will it be as "plug&play" as an Aurender or a Playback Design transport or... are there still optimisation needed to match your system ?

[isquirrel]

8 hours ago, bmichels said:

I just wonder: Now that Taiko Audio SGM Extreme do NOT uses HQ Palyer, I hope that there is no more optimisation needed ! will it be as "plug&play" as an Aurender or a Playback Design transport or... are there still optimisation needed to match your system ?

 

No there are no such optimisations needed to match your system other than installing your DAC's driver which Taiko Audio do for you. Which DAC are you using now? 

[joeling]

Do u think this Taiko extreme can beat the Vivaldi transport ?

[CKKeung]

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/page-88

 👆🏻#1755

 

Romaz has ordered an Extreme.  😃

[BMCG]

Joe - given that Francisco (Microstrip on WBF) has both the full stack and an Extreme...be worth pinging him. 

 

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/post-610079