ackoDAC based on ES9018 - Page 68 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 28th January 2013, 01:48 AM   #671
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
some people believe all sorts of ridiculous things improve their audio. its a cheap unit that gives good numbers for the money, its as simple as that, if you have more money for the higher performance part, then buy/use the high performance part. if not then it would seem to be good value for money, but still nothing that stands out.

there really isnt that much a clock has to do, it has to produce a regular waveform with low harmonic distortion. most importantly for audio, is accuracy in the short term, rather than ppm over years.

the SAW are higher jitter both RMS and significantly higher for peak to peak than the crystek (and even worse compared to NDK and some other parts). measuring jitter to a high degree of accuracy is not very accessible to the DIYer, but its not magic, the company itself quotes jitter/phase distortion many times higher than the crystek.

now i've heard someone say that Epson claim its cleaner because its produced with a fundamental frequency multiple, sure OK, but cleaner than what? cleaner than some DSPLL based generators used in an unrelated field? in a similarly low price bracket?

any harmonics produced are measurable and it is not better, have you discovered another parameter that matters more that rigorous testing for all manner of fields of scientific endeavor that deal with frequency have missed, that is only important for audio frequencies?

how about you send one to iancanada? he has built a reasonably accurate test rig for measuring low jitter devices.

Last edited by qusp; 28th January 2013 at 01:51 AM.
  Reply With Quote
Old 28th January 2013, 02:02 AM   #672
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
it could be for instance that the SAW clocks have higher PSRR, or are less affected by improper line termination, but provided both are well taken care of I can see no room for there to be an area it excels at vs higher performance parts.
  Reply With Quote
Old 28th January 2013, 02:34 AM   #673
diyAudio Member
 
Joe Rasmussen's Avatar
 
Join Date: Jan 2003
Location: Sydney, Australia
Send a message via MSN to Joe Rasmussen
1. I am not prone to the ridiculous. Thanks, I really needed that.

2. I know what a clock needs to do, thank you for the lesson.

3. I am aware of the specs of the SAW, didn't know that was to become an issue. If it was, we could have had a discussion about it. Past tense now.

4. I don't believe in Magic (maybe sleight-of-hand, but that is a different kind of Magic).

5. I have spoken to Epson in Singapore, on the phone, and they have made no claims whatsoever, other than being surprised that I want to use it for Audio. They thought it, as you seem to, that it was the wrong type and tried to get my attention to a different product.

6. Price sometimes relates to "goodness" - sometimes does not. A pursuit of understanding is never under-valued.

7. Discoveries? Perhaps, that long-term stability is more important for audio frequencies and that it is difficult to convince certain people that it might just be the case. A mystery that needs to be solved? Why can blind testing hear ULF filtering of an oscillator supply when we are talking about Allan Variance which has been disregarded as "fanciful" and surely can have no influence at audio frequencies - do I know why? Not yet, but I can do a convincing demonstration - BLIND!

8. Can everything be quantified? Einstein was once asked if Beethoven's 9th could be quantified? "Maybe, but it would be useless." Has anybody ever explained conception? Or how a single human thought is formed - where does that little voice in my head come from? Trust me, I can be as much of a septic as anybody, but I would like to think that my scepticism is healthy. But the Universe is still full of Mysteries, and I say "Thank God".

Thank you for your ridicule, I know when I am not wanted, so I shall no longer make any comment and take this thread off my list.

I just tried to be helpful, and I thought being in the same town I was being neighbourly.

Best Regards.

Joe R.
__________________
The "Elsinore Project" DIY Speaker System
Custom Analogue Audio - we also support and promote non-profit DIY

Last edited by Joe Rasmussen; 28th January 2013 at 02:38 AM.
  Reply With Quote
Old 28th January 2013, 03:13 AM   #674
diyAudio Member
 
Joe Rasmussen's Avatar
 
Join Date: Jan 2003
Location: Sydney, Australia
Send a message via MSN to Joe Rasmussen
Quote:
Originally Posted by qusp View Post
it could be for instance that the SAW clocks have higher PSRR, or are less affected by improper line termination, but provided both are well taken care of I can see no room for there to be an area it excels at vs higher performance parts.
BINGO!

That is a suggestion I am definitely open to. In fact PSRR is the likely key and that is what I would put my money on.

How do we react when your "reason" says one thing and when "experience" tells you something different. Here "reason" is a current set of known numbers, to relate numbers to an experience is a fraught activity.

Very early on I sat down Brad Serhan of Orpheus Loudspeakers as my Guinea Pig, and I demonstrated what happened when I lowered an ultra-low noise ultra low frequency servo-controlled power supply, the frequency of the servo was already in the thousandths of a Hertz, and then lowered it by a factor of 10, and he went Holy Catfish on me. I played him Track 1 of one of his favourite CD - and he just wanted to listen to the whole album while constantly saying "I shouldn't be here, I have an appointment" - which he didn't make. LOL

It wasn't even hard to tell. I wish it was as easy to explain.

On paper it should not have made an iota of a difference.

But what do you then do? Ignore it?

I tell you, I just made hundreds of them and saw the smile on people's faces - and about half of them overseas. And guess what, we never claimed "low jitter" - because it wasn't the point.

Something called Allen Variance seems to be at play here, but few want to know about, and the louder I shout the more crazy people will think you are. So I am playing a long game here, just maybe one day it will get taken seriously.

Bill Hobba, a retired academic in QLD, who now uses his time pursuing his passions, and audio is at the top of the list. Bill was talking to some guy in the US who was clearly involved in some serious research. Bill said to him "I have a friend in Sydney who is looking at jitter effects from a sub-1 Hertz perspective." Bill later told me there was a little pause on the line and the voice came back "tell your friend in Sydney he is on the right track."

There are others out there looking at Allan Variance - I am sure in time it will come out. A hundred plus users' ears can't be wrong.

Cheers, Joe R.

PS: I don't know if you have heard of Bill Thalmann - in the US is a stalwart of the audio industry and would be on first name with any significant person you can think us. For 12 years the Designer for Conrad-Johnson, so you can imagine. Well, Bill has not only heard it but in fact his own personal player uses it and more, he actually installs them for his personal clients, some also in the audio industry. One of them is Arnie Nudell of Genesis and Infinity fame. They do actually just listen - and that is enough.
__________________
The "Elsinore Project" DIY Speaker System
Custom Analogue Audio - we also support and promote non-profit DIY

Last edited by Joe Rasmussen; 28th January 2013 at 03:18 AM.
  Reply With Quote
Old 28th January 2013, 04:08 AM   #675
glt is offline glt  United States
diyAudio Member
 
Join Date: Oct 2004
Let the challenge begin... (of the clocks)
__________________
www.hifiduino.wordpress.com
  Reply With Quote
Old 28th January 2013, 07:06 AM   #676
acko is offline acko  Australia
diyAudio Member
 
Join Date: Jan 2008
Location: Sydney, Australia
Quote:
Originally Posted by Joe Rasmussen View Post
Hi Acko

Re: "Surface Acoustic Wave" 100MHz Oscillator.

I don't know if you are aware, but some of the Buffalo guys are experimenting with the 100MHz "SAW" 3.3V oscillator - comparing it against the Crystek 957 version. A number of them are coming around to the view that the "SAW" version is superior, definitely different and very honest to a fault, shades of more detail, more air around instruments, at bit less romantic but more lively pace rhythms etc and one commented that the bass was more solid and deeper.

I myself have not compared it against the Crystek yet (one of the Buffalo guys in Malaysia kindly sent me one, just got it but not listened yet), but it is easily better than any other I have tried so far. It needs a good 3.3V supply - as you would imagine - and I am sure you have that. The form factor is the standard 5x7mm.

Give me a call on 9607-4650 or 0412-203382 and if you like, I can post one to you, no charge (they are not expensive). It would be interesting to see what you make of it, as they do sound rather different.

These oscillators were never designed for audio use (military as they are very sturdy and microwave and networking) and don't come in frequencies for the usual audio use and the lowest is 50MHz. But of course the ES9018 can indeed take it, so the "SAW" is only really compatible with the Sabre DAC.

Cheers, Joe R.
Thanks Joe, all noted

I have got SAW 100/125MHz already since your very early post on this matter but have not yet tested them. All in good time. If you can arrange samples for 90.3136Mhz and 98.304Mhz versons from EPSON that would be great for synchronous clocking!

Most likely will also be used on my new Re-Clocker board. This can take even higher frequencies (500MHz) so may fall into the sweet spot of the SAW technology....
__________________
AckoDAC Project
Now Open:Sync Clock GB
  Reply With Quote
Old 28th January 2013, 07:12 AM   #677
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
haha Joe, ridicule? why is it when someone (in particular me it seems) asks for those making outlandish claims to make some effort to back them up is held to be being unreasonably negative? I made a suggestion for you to put your ideas to the test and asked for some sort of explanation for something that to me is irrational, that is all.

you personalized a sentence that was not even directed at you or anyone in particular. lately there have been many claims made by people for superior performance in situations that quite simply and obviously make matters worse, but I was not even talking about that. some people in audio believe that placing crystals near their system focuses the sound, so when anecdotal claims are made, it doesnt hold any water for me.

you are the one/s making claims, the onus is on you to back them up; I should not be held in contempt for requesting that back-up.

This is a technical forum and in relation to an area fairly well understood area with very little room for variables and those variables there are, are not unable to be tested. Claims that a device that does the exact same thing with only one job to do, yet has specifications many times worse is held to be superior and without any attempt to explain it, are IMO more noise than information.

People and hobbiests are already confused enough about digital audio and what problems are of real concern, without people inventing new ones. Its like these claims of 'low level digital information' being permanently effected by things in bit perfect systems, systems where you cannot have a partial/subtle error. These claims are usually claimed to affect 'soundstage' or 'musicality' or this new kind of 'transparency' that means something else.

the post you quoted of mine re PSRR is a guess that is unsupported and regardless only explains a possible difference for those who have not set up their systems correctly in the first place... for either clock. it is a real possibility, but it does not speak in favor of the SAW, it only suggests the power supply and layout/connection for the Crystek needs work, when fixed it would still be the superior clock.

I have absolutely no issue with people claiming a preference, what I do have a problem with are anecdotal claims for undocumented and seemingly impossible technical differences. the jitter between these 2 devices is not even close, the SAW is 600% worse than the Crystek and more with peak to peak. Measured performance by their numbers puts it equal to, or worse than the generic clock in the fifo and as you mention it does not come in very useful frequencies.
  Reply With Quote
Old 28th January 2013, 08:46 AM   #678
diyAudio Member
 
Joe Rasmussen's Avatar
 
Join Date: Jan 2003
Location: Sydney, Australia
Send a message via MSN to Joe Rasmussen
Quote:
Originally Posted by qusp View Post
making outlandish claims
You just don't get it, do you?

People who actually know me, and clearly you don't, would never say that about me.

But that is not what this is about.

Did you hear about two scientists who made "outlandish claims" and who was pilloried and ridiculed for years on end by the establishment? They had to work in basements and their work "lacked validity" and their names became bywords for jokes.

But gradually things changed, you see, the results spoke for themselves. You see, they had patients who were very happy. And the number grew... it was like a magic cure.

Their names were Barry Marshall and Robin Warren. They had the temerity to challenge prevailing theories about stomach disease/ulcers, finally making an irrefutable case that bacteria caused up to 90% of gastric ulcers. That ulcers were largely bacterial infections. Their discovery meant stomach ulcers were no longer a chronic, disabling condition, but a disease that could be cured. Yes, CURED. Millions of people who had been suffering needlessly would be treated with... antibiotics.

They got a Nobel Prize in 2005.

Right now, in my line of work, we are working on something that has totally stumped several trusted loudspeaker designers. It has to do with the dominant damping in loudspeaker drive units, the Electrical Q or Qes, which is related to the electro-mechanical structure of the driver. This is in parallel with the Mechanical Q or Qms. which is mostly losses in the suspension and surround of the cone. Typically Qes dominates Qms about a ratio of 10:1 - so it is clear that the Total Q, or Qts, will be slightly lower than Qes. Total erosion of Qe would mean only Qms would remain (act as a ceiling) as damping - totally inadequate.

If Qes was eroded by adding external series resistance (or high output impedance of the amplifier), then the Qe damping will be eroded by the corresponding amount, hence doubling the series resistance will double (or actually halve the damping) of the Qes value. Hence the driver needs to see a zero Ohm impedance, or way below 1 Ohm at least.

Take a suitable 8" inch driver, measure its Thiele-Small Parameters, then works out a sealed box volume that will give a 2nd Order Butterworth alignment, say with a sealed box resonance of 40 Hertz. This alignment will be maximally flat (have no rise/peak before roll-off) and only 3dB down at 40 Hertz, the box tuning frequency.

So far so good. But I can add a specific parallel network across the Voice Coil that acts as a Current Mirror (again I was attacked by one person for coming up with that phrase, but it describes what it does), and using Thevenin's Theorum I was able to predict the deviation of the response of that 2nd order Butterworth and by tuning the Current Mirror I was able to tune away any deviation to within a few percent. Now I could add any series resistance I liked, and because the 2nd order Butterworth alignment was retained even when 1000 Ohm was added to an 8 Ohm driver (120:1), the fact that a Butterworth has an in the box Q, or Qb, of 0.707 - then it is clear that the damping is locked in, even though the Voice Coil never sees a zero Ohm impedance. The determining factor is only the Vb, the volume of the box, although if Vb was changed then the tuning of the Current Mirror would lock in the different alignment, say increase Vb and aim for a Bessel alignment, and so on.

It will always be a 2nd order Butterworth (or Bessel), no matter the external connection.

What has happened to the Qes inside that driver? Only Qms is supposed to be left unchanged. There is no rational explanation.

It should not be behaving this way !!!

And yet it is.

At this point everybody been made aware of this has been stumped, and we are speaking about loudspeaker designers with decades of experience.

This is no thought experiment - it is actually happening - it has been done on repeated basis. It works

It also interesting that we now have a 2nd order Butterworth, even when using a current amplifier - I built one that has an output impedance of 200 Ohm and zero so-called mythical 'Damping Factor' just doesn't exist, and the bass is tight as a drum.

Somebody needs to find an explanation.

Yes, it is going to be cracked. Why? Because serious people who know I am a serious person and that these people take me seriously and I have done enough to deserve their respect, and they know my record of getting results. So we are going to find out, because I am humble enough to realise I need help.

Did you get that?

I need help!

Now let's get back to whether Allan Variance has an affect on audio, even though we are talking about a freequency band that makes no sense how this ought to affect audio frequencies.

I admit it, it makes no sense!

Yet real world experience tells us that something is going on. I am prepared to be ridiculed - at least somebody has to be brave.

And I am humble enough to realise I need help.

Did you get that?

I need help!

So don't give the "outlandish claims" line. That is just a recipe' for laziness. Why not take this as a challenge that may just have some validity, and then we can make some progress. The 'Butterworth' challenge is going to be solved, because people are taking it seriously.

So a critical mass will be reached, then it will no longer be ignored. Right now I do believe that there are some who are taking it seriously, that sub-1 Hertz jitter will be accepted as not only audible, but in fact is the cause of the kind of jitter that causes that 'digital' sound we all loathe.

Then we shall all be smiling.

So this not about an outlandish claim, but a question: WHY?

BTW, re the speaker example above, it get crazier than that. Since a Butterworh is a mechanical high-pass filter, then does Current Mirrors work with electrical high-pass filters? Yes! And even low-pass filters. It works both in the electro-acoustic as well as equivalent electrical filters. yes... Crossovers. And again, not just a thought experiment, it actually works in the real world.

Cheers, Joe R.

PS: It just occurred to me, that the real reason for the "outlandish claim" bit is that somehow this is about my ego? Then you really don't know me at all.
__________________
The "Elsinore Project" DIY Speaker System
Custom Analogue Audio - we also support and promote non-profit DIY

Last edited by Joe Rasmussen; 28th January 2013 at 09:06 AM.
  Reply With Quote
Old 29th January 2013, 01:17 AM   #679
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
sorry but yes, claiming a part that has little room to do anything different, but measures significantly worse is better; that IS outlandish and the onus is on you to back it up, not me to prove you wrong.

I offered real suggestions for the reason, I offered suggestions of people who may help to measure the device. that you wrote an essay without really saying a great deal... I simply dont have time to read it, let alone reply to it and it changes nothing.

it may really turn out to be another factor that was not adjusted for with controls in testing. ie providing a properly designed clock circuit and PSU for both devices.

ego? not sure where specifically you think I called you out on ego, but since you mention it, what else is there but ego (the broad meaning, not the derogatory one) in the search for best DIY audio technology? what else is there is the way that stops us all agreeing on what that is?

Last edited by qusp; 29th January 2013 at 01:27 AM.
  Reply With Quote
Old 29th January 2013, 01:46 AM   #680
diyAudio Member
 
Joe Rasmussen's Avatar
 
Join Date: Jan 2003
Location: Sydney, Australia
Send a message via MSN to Joe Rasmussen
Quote:
Originally Posted by qusp View Post
and the onus is on you to back it up, not me to prove you wrong.
Huh?

> "not prove you wrong"

I am at a loss, what do you mean by that? WHY are you making this so personal... "prove you wrong"?

Why would you want to prove me wrong?

Then come around then, I will demonstrate it.

Now the ONUS IS ON YOU!

Others have doubted, but they had an open mind. When something completes transforms the music, then saying "the numbers don't add up" just means we have work to do, it doesn't mean we ignore the event we have just experienced? That makes no sense. Can't you see that?

Are a hundred people wrong. Did I hoodwink them? Are you suggesting they only thought they heard some difference.

Are you saying there was no difference?

You are hardly in a position to say that.

I honestly don't understand your attitude? If everybody had that attitude we would still be in the Stone Age.

Wouldn't you rather be part of solving a seeming mystery? Something is going on, should we not try to figure it out. Conventional jitter specs is not explaining this, so there must be some other explanation.

I mean, are you not risking ending up with egg on your face?

Or maybe you are a Troll, know what I mean?

Maybe you should go across to YouTube, that is where banality of this sort belongs. They like calling each other idiots there. Seems they get some kind of debased enjoyment out of it.
__________________
The "Elsinore Project" DIY Speaker System
Custom Analogue Audio - we also support and promote non-profit DIY

Last edited by Joe Rasmussen; 29th January 2013 at 01:50 AM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
WM8816 based pre-amp panson_hk Chip Amps 13 22nd May 2009 02:16 PM
DIY ESL III (based on an ER Audio Kit) jasonlky Planars & Exotics 7 23rd January 2008 07:09 AM
UCD based amp oldmaelstrom2 Class D 3 27th June 2007 10:10 AM


New To Site? Need Help?

All times are GMT. The time now is 01:51 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2