Audio Grade Router and Switches in RMAF with SOtM

[rickca]

51 minutes ago, ismewor said:

Just got my TCXO router from the linear solution tonight.

What model D-Link router is that based on?  So thelinearsolution offers both routers and switches?  On their website, they are showing 8-port switches but the item in your picture is clearly a router.

[ismewor]

Over the weekend, I was able to try this new audio router from TLS. I currently setup are ASUS AC-3100 Dual band + DLink Switch + Custom LPS 12V -> to Custom Server with ATX LPS -> NAS -> Ethernet to SoTM Trifecta -> Or Chord DAVE -> MC275 Mono -> 800D2

With this new TCXO router I really couldn't believe the different that made. My wow factor was way up high. I would try to say the detail is more into the instrument, I can hear the echo from the guitar box itself. it feel more live to the whole picture. Vocal is the biggest improvement. I can feel the singer feeling within the song i feel the sadness and the up down. this is something once need to experience for. 

I spent 6 hrs on Sat night till 4am in the morning. and 4 hrs on sunday to just try out all my reference. My wife is so picky to music she even ask what have i changed it sound much better. and she said hope you don't cost too much for it. i told her i just replaced a new router. she is totally cool with that.

Some might say what can it improve in sound, i got perfect data and no data loss. I really don't know but this router is the missing link for streaming audio or the final gate for the castle. 

 

Happy listening. 

 

[rickca]

3 hours ago, ismewor said:

With this new TCXO router I really couldn't believe the different that made.

Thanks for your review.  So did you simply replace your Asus RT-AC3100 with this TLS router?

[ismewor]

Yes, I’m using it together now. I just found out it improve tidal streaming by far. Now tidal sound like tidal suppose to be. 

[marce]

The wife in the kitchen.

So the data dosen't change, we just have a better clock on an asynchronous data transfer, that is going to be buffered anyway...

Then fed into the DAC, so the same data will be present at the DAC anyway...

[plissken]

Just let me know when you want to do this blind

[ismewor]

10 hours ago, marce said:

The wife in the kitchen.

So the data dosen't change, we just have a better clock on an asynchronous data transfer, that is going to be buffered anyway...

Then fed into the DAC, so the same data will be present at the DAC anyway...

Buffer is just a storage. the accurate timing, jitters correction is the key.

[marce]

You don't have a clue do you, once buffered the data is re-clocked so any previous jitter is nulled...Usually a FIFO buffer. Another clue is asynchronous.The only jitter of real concern is the final buffer to the DAC's input pins.

[Special Snowflake]

On 2017-10-30 at 3:30 PM, ismewor said:

Over the weekend, I was able to try this new audio router from TLS. I currently setup are ASUS AC-3100 Dual band + DLink Switch + Custom LPS 12V -> to Custom Server with ATX LPS -> NAS -> Ethernet to SoTM Trifecta -> Or Chord DAVE -> MC275 Mono -> 800D2

With this new TCXO router I really couldn't believe the different that made. My wow factor was way up high. I would try to say the detail is more into the instrument, I can hear the echo from the guitar box itself. it feel more live to the whole picture. Vocal is the biggest improvement. I can feel the singer feeling within the song i feel the sadness and the up down. this is something once need to experience for. 

I spent 6 hrs on Sat night till 4am in the morning. and 4 hrs on sunday to just try out all my reference. My wife is so picky to music she even ask what have i changed it sound much better. and she said hope you don't cost too much for it. i told her i just replaced a new router. she is totally cool with that.

Some might say what can it improve in sound, i got perfect data and no data loss. I really don't know but this router is the missing link for streaming audio or the final gate for the castle. 

 

Happy listening. 

 

Wow fascinating results! Bargain for the price!

[Special Snowflake]

15 hours ago, ismewor said:

Buffer is just a storage. the accurate timing, jitters correction is the key.

Well said. This is probably why Uptone uses a low jitter clock in the isoregen. Thousands of audiophiles can’t be wrong. 

[marce]

LOL

The only clock that matters is the clock at the data pins of the DAC. So the critical path is from the last buffer to the DAC. Buffered data has no jitter... jitter is not stored with the data.

If the facts don't fit the physics then thousands of audiophiles can be wrong.

We are also talking about Ethernet data transmission absolutely nothing to do with the regen or the iso regen, try and focus on the topic at hand.

[louisxiawei]

3 hours ago, marce said:

LOL

The only clock that matters is the clock at the data pins of the DAC. So the critical path is from the last buffer to the DAC. Buffered data has no jitter... jitter is not stored with the data.

If the facts don't fit the physics then thousands of audiophiles can be wrong.

We are also talking about Ethernet data transmission absolutely nothing to do with the regen or the iso regen, try and focus on the topic at hand.

When people bought something and shared their experience on the forum, even some products might be nonsense. They expect compliment and want to be validated for their choice. 

 

If a know-how person pointed out the potential truth, they prefer to ignore and believe their ears or use their friends or family member as another example to further validate it. Otherwise, either their choice or ears will be invalidated. So, quite tough choice for them tbh.

 

I know some of these guys here, they are very active on some TAIWAN hifi audio forum, talking BS and get hyped all the time. Fortunately, this is CA. People always come out and dare to point out.

 

Who benefit the most out of this? We all know the answer. 

[Summit]

 

The truth is out there and not always what we assume it to be.

[marce]

No the truth is often based on physics... Surprisingly signal integrity and transmitting data has been studied for quite some time, Oliver Heaviside started it off with the telegraphers equations... circa 1878. We have moved on since then as you may have noticed.

[Summit]

6 minutes ago, marce said:

No the truth is often based on physics... Surprisingly signal integrity and transmitting data has been studied for quite some time, Oliver Heaviside started it off with the telegraphers equations... circa 1878. We have moved on since then as you may have noticed.

 

Yes the truth is often based on physics, but the capacity and tech to store and transmitting data has and are still evolving.

[Superdad]

9 hours ago, marce said:

The only clock that matters is the clock at the data pins of the DAC. So the critical path is from the last buffer to the DAC. Buffered data has no jitter... jitter is not stored with the data.

If the facts don't fit the physics then thousands of audiophiles can be wrong.

 

Maybe assumptions are made about the physics when there is much more happening on a small scale.  I liken some of audio to where medicine and physics was before they proved molecules existed.  Earlier assumptions turned out to not telling the whole story.

While these issues may not matter much for standard Ethernet data transmission, for sensitive audio clocks there are indeed modes of incursion.  Here is the beginnings of a hypothesis based first on proven facts--which will eventually lead to more full measurement and understanding:

 

 

[marce]

Yes but domestic audio is not cutting edge these days, the infra structure is there, look at Ethernet, the first Ethernet cards I did had a BNC connector, now I can type this and send it error free across the world in seconds, a few feet in a music room is not hard...

[marce]

Just now, Superdad said:

 

Maybe assumptions are made about the physics when there is much more happening on a small scale.  I liken some of audio to where medicine and physics was before they proved molecules existed.  Earlier assumptions turned out to not telling the whole story.

While these issues may not matter much for standard Ethernet data transmission, for sensitive audio clocks there are indeed modes of incursion.  Here is the beginnings of a hypothesis based first on proven facts--which will eventually lead to more full measurement and understanding:

 

 

Physics is physics...

As to scale, look at some of the stuff I have posted on both signal integrity, simulation for layouts and systems, including modelling scope probes so you can add them to the simulation to load the circuit so you will see the same waveform on screen as you will on your scope. Power integrity integrity, both for DC current and AC impedance. Then there is the other side of the signal integrity coin, EMC compatibility engineering, well covered these days, as well as the well known and regarded sources of information, automotive EMC guides are a good read, often more practical based. Automotive is the new mill spec both for design and components, working on a couple of military project, where we are using automotive grade components, because of there temp range, reliability etc, it is also a necessity as there are not many real mil spec SMD components these days and the ones that are available are getting a bit long in the tooth and expensive.

I must admit there is a difference in how far down the scale you go depending whether the design is commercial, industrial, automotive, mil/aerospace, cost and what damage failure will cause being the main drivers.

The problem with digital and optical isolators are known, but they have their uses... But without the correct layout and filtering they will also pass noise as they all have some capacitive coupling, parasitic capacitive coupling is also one of the problems with some PCB designs and noise getting through, the main one being overlaps of digital and analogue ground planes when they are on different layers. (one trick is a chassis layer, very similar to the grounding of a connector), this forms not only a capacitive screen, but also a low impedance path to some single point chassis connection to stear the noise away from sensitive components. This is more often used on automotive and other "earth free designs", using PE can often add problems such as loops and depending on your location for mains supply noise. There is no easy fix to any of this just persistent attention to detail.

That said once buffered data is brand spanking new, and its next journey will only be affected by the pertinent clock for that particular transmission...  

[Special Snowflake]

24 minutes ago, marce said:

Physics is physics...

As to scale, look at some of the stuff I have posted on both signal integrity, simulation for layouts and systems, including modelling scope probes so you can add them to the simulation to load the circuit so you will see the same waveform on screen as you will on your scope. Power integrity integrity, both for DC current and AC impedance. Then there is the other side of the signal integrity coin, EMC compatibility engineering, well covered these days, as well as the well known and regarded sources of information, automotive EMC guides are a good read, often more practical based. Automotive is the new mill spec both for design and components, working on a couple of military project, where we are using automotive grade components, because of there temp range, reliability etc, it is also a necessity as there are not many real mil spec SMD components these days and the ones that are available are getting a bit long in the tooth and expensive.

I must admit there is a difference in how far down the scale you go depending whether the design is commercial, industrial, automotive, mil/aerospace, cost and what damage failure will cause being the main drivers.

The problem with digital and optical isolators are known, but they have their uses... But without the correct layout and filtering they will also pass noise as they all have some capacitive coupling, parasitic capacitive coupling is also one of the problems with some PCB designs and noise getting through, the main one being overlaps of digital and analogue ground planes when they are on different layers. (one trick is a chassis layer, very similar to the grounding of a connector), this forms not only a capacitive screen, but also a low impedance path to some single point chassis connection to stear the noise away from sensitive components. This is more often used on automotive and other "earth free designs", using PE can often add problems such as loops and depending on your location for mains supply noise. There is no easy fix to any of this just persistent attention to detail.

That said once buffered data is brand spanking new, and its next journey will only be affected by the pertinent clock for that particular transmission...  

What about the theory that jitter gets “baked” into the audio? Even when the Ethernet clocks run at a different frequency than the audio clocks. And try not to burst any bubbles around here because when a grown man hears a change, that means there was a change. Audiophiles are the only species known to be immune to the power of suggestion. 

[plissken]

12 hours ago, Special Snowflake said:

Well said. This is probably why Uptone uses a low jitter clock in the isoregen. Thousands of audiophiles can’t be wrong. 

 

BenchMark Audio showed this to be false. They wired up a 100 foot cable to their DAC, it was loaded with errors and the audio it was producing was pristine. In a hobby were a $2000 DAC is low cost for many it shows that this isn't hard to accomplish.

 

I can drive two DAC's concurrently. As long as they are a properly designed DAC I would like to to pick the one with the ISOR vs one with just a nice, short (my USB cable is 1.5 foot) cable. 

 

Yes, thousands of audiophiles can indeed be wrong. 

[plissken]

57 minutes ago, Special Snowflake said:

What about the theory that jitter gets “baked” into the audio? Even when the Ethernet clocks run at a different frequency than the audio clocks. And try not to burst any bubbles around here because when a grown man hears a change, that means there was a change. Audiophiles are the only species known to be immune to the power of suggestion. 

 

That theory is quite easily debunked. Let's see if people can figure out where the jitter in this video is introduced. 

 

 

 

[Special Snowflake]

4 minutes ago, plissken said:

 

BenchMark Audio showed this to be false. They wired up a 100 foot cable to their DAC, it was loaded with errors and the audio it was producing was pristine. In a hobby were a $2000 DAC is low cost for many it shows that this isn't hard to accomplish.

 

I can drive two DAC's concurrently. As long as they are a properly designed DAC I would like to to pick the one with the ISOR vs one with just a nice, short (my USB cable is 1.5 foot) cable. 

 

Yes, thousands of audiophiles can indeed be wrong. 

From what I hear the audio that comes out of any Benchmark products isn’t pristine under any circumstances. 

[plissken]

1 minute ago, Special Snowflake said:

From what I hear 

 

Yep that's always suspect now. 

[Special Snowflake]

1 minute ago, plissken said:

 

Yep that's always suspect now. 

I don’t mean with my ears. I’m over 40 so my ears can’t be trusted. I only trust the ears from the under 25 crowd. 

[plissken]

Just now, Special Snowflake said:

I don’t mean with my ears. I’m over 40 so my ears can’t be trusted. I only trust the ears from the under 20 crowd. 

 

You have my condolences. On which part I'll leave to you.