ASR Audio Science Review forum YouTube Channel

[plissken]

19 minutes ago, PeterSt said:

 

 

 

And this is my basement. It will sound like sh*t because it wasn't made to play audio. For that we build different machines.

 

I am shouting for over 2 years now that the PC is more important to the sound than the DAC. And it better be believed because without exception each customer agrees fully. You just never tried it, I suppose. Or you think you know better. Or you think you are an Amir. Hahaha.

But seriously. If you think you are an audiophile you must seriously invest in the playback hardware.

 

I don't do it exactly the same as Taiko.

 

 

So a Server board in there, that 28 or 32 core Xeon processor, running at 500Mhz (settable) and of course with a Linear PSU inside. Oh all passive. No watercooling either. wait, not even heatsinks (except for the CPU heatsink). CPU 60 degrees C max. Weight 13Kg.

No room heater at all.

 

So is this better than the Taiko ? ... -> Unimportant. Both ate infinitely better sounding than your laptop. 

 

You don't need a 32 core Xeon connected directly to a DAC. You need it sitting in a rack somewhere doing all the heavy DSP and/or ASRC and sending that out to an endpoint. The above directly contradicts whole mantra's of 'all that noise' crowd.

 

You'd be better off with a Pi4 as a DNLA client.

 

I've no issue what so ever sitting down with someone and doing this blind.

 

@Miska has the correct approach, IMO

 

[PeterSt]

23 minutes ago, plissken said:

You don't need a 32 core Xeon connected directly to a DAC. You need it sitting in a rack somewhere doing all the heavy DSP and/or ASRC and sending that out to an endpoint.

 

You appear to have no clue. The DSP stuff doesn't happen there.

You also seem to think I work from theory only.

 

How many of your systems have you sold for playing audio ?

You should join AmriM. He does not listen either. 🤪

[The Computer Audiophile]

28 minutes ago, plissken said:

You don't need a 32 core Xeon connected directly to a DAC. You need it sitting in a rack somewhere doing all the heavy DSP and/or ASRC and sending that out to an endpoint.

 

28 minutes ago, plissken said:

You'd be better off with a Pi4 as a DNLA client.

 

I thought bits were bits? Why would this approach be any better than another approach?

[plissken]

38 minutes ago, PeterSt said:

 

You appear to have no clue. The DSP stuff doesn't happen there.

You also seem to think I work from theory only.

 

Network Audio Adapter

Network audio is especially useful to give freedom from cables when player is run on a tablet or other wireless device.

Processing is performed by the player application and the processed data is then asynchronously streamed over network to a very lightweight network audio adapter interfacing to the DAC. Asynchronous FIFO provides maximum isolation between processing and audio reproduction.

 

And from the NAA thread, per Miska:

 

Yes, sort of. NAA is a way to expose remote audio devices to HQPlayer over asynchronous network link. It is just an audio "renderer" with support for number of different audio endpoint devices. Playback by NAA is completely clocked by the audio endpoint device and NAA performs only buffering, no decoding or processing at all. (A bit like asynchronous USB DAC, but having ethernet interface instead.) These endpoints are shown in the HQPlayer device selection just like local devices would, so you could have for example following kind of list:

livingroom: Mytek Stereo192-DSD DAC

livingroom: Musical Fidelity V-Link192

bedroom: Fostex HP-A8C

 

So in a local network there can be any number of NAA devices and each can have any number of audio endpoint devices.

 

All processing is performed by HQPlayer, either the Desktop-version which combines GUI and processing engine, or the Embedded-version which further splits the processing engine and GUI to separate entities.

 

Now who sounds like the fool?

[plissken]

20 minutes ago, The Computer Audiophile said:

 

 

I thought bits were bits? Why would this approach be any better than another approach?

 

Kind of my point Chris...

 

Again any are welcome to do this blind.

[plissken]

37 minutes ago, PeterSt said:

How many of your systems have you sold for playing audio ?

 

 

Huh, what's your excuse then?

[semente]

1 hour ago, PeterSt said:

So a Server board in there, that 28 or 32 core Xeon processor, running at 500Mhz (settable) and of course with a Linear PSU inside. Oh all passive. No watercooling either. wait, not even heatsinks (except for the CPU heatsink). CPU 60 degrees C max. Weight 13Kg.

No room heater at all.

 

Are you opting for that many cores and low speed to keep the heat down?

[plissken]

1 hour ago, PeterSt said:

You also seem to think I work from theory only.

 

No, from stupidity also.

[lucretius]

4 hours ago, charlesphoto said:

ASRanon

 

That made me laugh.  😄

[ASRMichael]

2 hours ago, plissken said:

 

 

I've no issue what so ever sitting down with someone and doing this blind.

 

 

Can you advise what server have you compared yours with? 

 

There might be a reason why no one ever takes you up on your blind test offers. 

[plissken]

14 minutes ago, ASRMichael said:

Can you advise what server have you compared yours with?

Used to own a Cary Audio Streamer. Several custom build PC's, Pi with RoipeeeXL.

 

Far as various A/V servers it's all been broadcast industry and mainly AVB protocol with some AES67 interspersed. So nothing you would see in the home.

 

But let's not try and shift focus away. We just experienced someone with a 100% incorrect understanding of audio distribution and DSP/ASCR etc that is a vendor making claims and recommendations.

It's why it's been CRICKETS since I made my point.

[Superdad]

50 minutes ago, plissken said:

No, from stupidity also.

 

You are calling @PeterSt stupid? 

I can think of some choice words but I'll simply suggest that you take your meds today Mark.

[plissken]

13 minutes ago, Superdad said:

 

You are calling @PeterSt stupid? 

I can think of some choice words but I'll simply suggest that you take your meds today Mark.

 

Stupidity is ignorance left willfully uncorrected Alex.

 

PeterSt maintained (he should now be disabused of that notion) that all the DSP etc happened at the end point. WHY would you take a 32 core CPU and underclock it to 500MHz? You might as well get a Lambo, put in a 4 cylinder Toyota Corolla engine. You are defeating the entire purpose.

 

I'll tell you what, You, Chris, PeterSt, myself can all get on a Hangout's or Duo call that I'll put up on YouTube. I've no problem with those odds. I can more than hold my own. You'll find out what kind of 'meds' I'm on then.

[The Computer Audiophile]

2 hours ago, plissken said:

 

Kind of my point Chris...

 

Again any are welcome to do this blind.

I misunderstood then, why you advocated for the network endpoint style of playback and having more power in another location. All the power can be placed one USB cable away from an audio system, thus saving the extra box, etc...

[ASRMichael]

27 minutes ago, plissken said:

Used to own a Cary Audio Streamer. Several custom build PC's, Pi with RoipeeeXL.

 

Far as various A/V servers it's all been broadcast industry and mainly AVB protocol with some AES67 interspersed. So nothing you would see in the home.

 

But let's not try and shift focus away. We just experienced someone with a 100% incorrect understanding of audio distribution and DSP/ASCR etc that is a vendor making claims and recommendations.

It's why it's been CRICKETS since I made my point.

I’m now 100% confident and agree with your findings. It’s conclusive! Well based on the equipment you have compared against. Which is fine. What you can’t say is your server is equally as good as Taiko or others without actually having compared them. 
 

Yes back the subject on Emile, I think most people who know Emile will know how credible he is. There’s hundreds and hundreds of people who talk about him. Can you send me a link to your credentials? Forums etc that talk about how credible you are? Please don’t say Fox news!! Maybe tongue & cheek, but that’s what we are dealing with here. 
 

😂

[plissken]

24 minutes ago, The Computer Audiophile said:

I misunderstood then, why you advocated for the network endpoint style of playback and having more power in another location. All the power can be placed one USB cable away from an audio system, thus saving the extra box, etc...

 

Let's not get off point here. I specifically laid out that you just need a low power end point and you can do all the heavy lifting at a server elsewhere. PeterSt, in an ill-advised attempt to discredit what I was saying, said I don't understand how DSP etc works that that it all happens at the end point.

 

Now I don't know about anyone else but... I have a garage setup, I have a desk setup, my wife has a setup in her office, and the dedicated room next door has a setup.

 

I'm I going to put in four Xeon based computers? Or put in some low power draw DNLA or NAA end points and put in one server, like I listed, and let it apply everything I want applied?

 

PeterSt compromise is to knee-cap a CPU by running the FSB at 500Mhz to fit it to both acoustical and thermal envelope that a listening room requires. Now I'm going to give out one caveat here: Unless there is some instruction set specifics with the Xeon in question for a player application, then why not find out with Sysmark what the Xeon offers, and look down market at a less expensive solution that will give the same score and you can keep that just as quiet and thermally managed?

[plissken]

54 minutes ago, ASRMichael said:

Yes back the subject on Emile, I think most people who know Emile will know how credible he is. There’s hundreds and hundreds of people who talk about him. Can you send me a link to your credentials? Forums etc that talk about how credible you are? Please don’t say Fox news!! Maybe tongue & cheek, but that’s what we are dealing with here. 
 

😂

 

We don't have to even go out of this very thread. 

 

When PeterSt says: "You appear to have no clue. The DSP stuff doesn't happen there."

 

And we know this is 100% in error on his part. I do not purchase or obtain services from people that have fundamentally flawed understanding about how all this works. You buy what ever you want to buy from Peter. 

[plissken]

Again, I welcome a realtime, face to face Google Duo on this topic for posting to YouTube. 

[ASRMichael]

16 minutes ago, plissken said:

Again, I welcome a realtime, face to face Google Duo on this topic for posting to YouTube. 

😂😂😂! Nobody has taken you up on $4000 offer, now you’re reverted to offering out YouTube battle! Getting close to WWF event! 

[ASRMichael]

18 minutes ago, plissken said:

Again, I welcome a realtime, face to face Google Duo on this topic for posting to YouTube. 

Joking aside! What would be the subject? Audio SQ or your knowledge in IT Networking? 

[plissken]

1 hour ago, ASRMichael said:

What you can’t say is your server is equally as good as Taiko or others without actually having compared them. 

 

I also can't say that Tidal's, Qobuz, Naxos, Amazons servers are equally as good either. But when a streaming service starts offering lossless Redbook or better there's much fanfare at AS and other sites.

[charlesphoto]

18 minutes ago, plissken said:

 

I also can't say that Tidal's, Qobuz, Naxos, Amazons servers are equally as good either. But when a streaming service starts offering lossless Redbook or better there's much fanfare at AS and other sites.

 

??????????????????????????????

[plissken]

9 minutes ago, charlesphoto said:

 

??????????????????????????????

 

I'm operating from the POV that servers are servers and not end point's with direct connection to a DAC. I'll both circle back and double down on Miska's NAA approach. You do all the heavy computational stuff on a server and then send out PCM, DoP, DSD to an endpoint supporting whatever protocol it is. Be it DNLA or NAA or RAAT or RTP or.... The list goes on.

 

Why would you have the device plugged in directly to your DAC doing all the computational work? Off load that somewhere else and just spit out the end result to a nice low power endpoint. @Miska , @jabbr am I missing something here?

[acg]

16 minutes ago, plissken said:

 am I missing something here?

 

Certainly...

[RickyV]

For SQ purpose.