again, if anybody who don't believe what I'm telling, just ignore or leave a message to mind other ppl, I'm not going to reply or will
against the rules here
get FREE info from others
try to sell anything related with this topic, like cables, whatever I mentioned ...
screw up people's mind by crap
...........
against the rules here
get FREE info from others
try to sell anything related with this topic, like cables, whatever I mentioned ...
screw up people's mind by crap
...........
SY you are certainly right with your comment, but when I read your Isaac Asimov quote,
I think we should not tolerate such anti-logical claims without contradiction. Many people
find such threads by Google (that´s how I found diyaudio.com years ago) and what they
read should contain some (usable) information.
Think I play "The War Against Intelligence" by The Fall now ;-)
I think we should not tolerate such anti-logical claims without contradiction. Many people
find such threads by Google (that´s how I found diyaudio.com years ago) and what they
read should contain some (usable) information.
Think I play "The War Against Intelligence" by The Fall now ;-)
As I told you, Ethernet (whatever physical layer is used) has no timing therefore can't disrupt timing. Timing comes from the receiver. Of course, if the design is so daft that it assumes timing from the Ethernet then it is bound to fail. Same for TCP/IP or UDP; all they do is deliver you a bag of bytes.powerpan said:
Are you once again announcing to your (potential) customers that you don't understand the technology used by your products?
gk7 you can have a free
Ceramic Quantum Silence Enhanser
from post #434, broken in half you get two free stress busters.
Wow, many thanks. For the stress buster function I might take them
to lower my hypertension ;-)
Too late, he already has. Now it's just a matter of trolling to get free design help.
Maybe we need to draw some nice Venn diagrams to straighten up some thinking here ...
OK, circle A is digital functioning, circuitry that worries about logic processing, and passing around of pure information. Everything is beautifully ordered in this world, this is a place where things can be made perfect, we can throw around the term 100% performance without people having a hissy fit.
Then we get circle B, that unpleasant, messy analogue stuff: cough at the wrong time and your measurement, and behaviour can go up the spout. Everything is relative, and the name of the game is S/N, and rejection ratios -- this is the world of the butterfly flapping its wings on the other side of the world.
Now comes the really nasty bit: is there any overlap of circle A and circle B? Most people here are fervently sure there isn't; ie. the overlap is 0%, that is, 0.0000000000000....%. Now, I come from the angle that there is an overlap, it may be, say, 0.01%, or 0.1%, or 0.001% depending upon everything -- just enough to cause an effect ...
Now I want some data: those who believe in 0.0000000000000....%, show me the figures .. .
Frank
OK, circle A is digital functioning, circuitry that worries about logic processing, and passing around of pure information. Everything is beautifully ordered in this world, this is a place where things can be made perfect, we can throw around the term 100% performance without people having a hissy fit.
Then we get circle B, that unpleasant, messy analogue stuff: cough at the wrong time and your measurement, and behaviour can go up the spout. Everything is relative, and the name of the game is S/N, and rejection ratios -- this is the world of the butterfly flapping its wings on the other side of the world.
Now comes the really nasty bit: is there any overlap of circle A and circle B? Most people here are fervently sure there isn't; ie. the overlap is 0%, that is, 0.0000000000000....%. Now, I come from the angle that there is an overlap, it may be, say, 0.01%, or 0.1%, or 0.001% depending upon everything -- just enough to cause an effect ...
Now I want some data: those who believe in 0.0000000000000....%, show me the figures .. .
Frank
The world's weather systems may be chaotic. An Ethernet based DAC should not be, so butterflies should not have any significant effect.
All we are talking about is assembling some data into a buffer, then reading it out at a constant rate. If the readout depends on the cable used to place the data there, then there are serious design flaws.
All we are talking about is assembling some data into a buffer, then reading it out at a constant rate. If the readout depends on the cable used to place the data there, then there are serious design flaws.
Again, we are not worried about whether the readout is correct, we're worrying about the electrical behaviour that was needed to ensure that that readout was correct: is it the case that in every instance, that any analogue circuitry which is the last link in the chain, is 100% isolated from that electrical behaviour ...?
Frank
Frank
I spend my days laying out often very complex PCB both analogue, digital and mixed signal, as do many others, there are engineers doing the schematics, test guys testing what we have designed production guys looking at the assembly side, all liasing with each other and working towards a common goal, creating somthing that works and works well, your simplefictaion is a bit silly as we have pointed out hundreds of time, that no not everything is perfect, but we engineer a product that works. We have never said there is 0% overlap (you dont need the .000 they are pointless, zero is an absolute), we have said that if noise is a problem (and you wont know the scope of the problem using your ears) there are ways of engineering solutions. You dont have to chase demons and eldrich problems that are not there, but concentrate on the real problems and you will find solutions without having to resort to surreal science.
Got ot ask a question, Frank, as you are concerned about high frequency digital noise, do you use shielded interconnects for low level analogue signals and is the shield connected at both ends through 360 degrees?
Au contraire, that's the whole point: if the noise, or distortion issue is so insignificant that your ears can't pick it, then there is no problem: the 'overlap' is sufficiently low that everyone can forget about it. However, there are enough ears out there that are sensitive enough to the interaction, mine included ...
Agreed. Of course there are engineering solutions, but first one has to acknowledge there is an issue, at least for some listeners. No good abusing them for having the ability to hear things, even if oneself doesn't ...
I still have problems with interference, probably because of those reasons. However, creating a near 100% interference free environment is probably something I'll address down the track, when I have a bit more energy for doing so ...
Frank
Last edited:
You are wrong again, read my full statement, "if noise is a problem etc. you are mis quoting me aqnd not reading what I say, If noise is a problem, if you hear distortions that you think are down to noise then you cannot determine the scope of the problem with your ears, you need to determine the frequency of the noise and where it comes from, and for that you need to measure.
Please read what I say and dont twist it to fulfil some audiophile belief I am trying to give answers to problems percieved or real, and how they can be solved by engineering as that what creating a music repreduction system is, its engineering.
I often work on projects where it has to be 100% free of noise and interefrence even for the very sensitive analogue, so I see and have to work to solve these problems every day, but it needs both meassurements and listening, and as Eric Bogatin says "No myths allowed"
Marc
You did extend your explanation somewhat here, but it now makes better sense what you were saying.You are wrong again, read my full statement, "if noise is a problem etc. you are mis quoting me aqnd not reading what I say, If noise is a problem, if you hear distortions that you think are down to noise then you cannot determine the scope of the problem with your ears, you need to determine the frequency of the noise and where it comes from, and for that you need to measure.
Please read what I say and dont twist it to fulfil some audiophile belief I am trying to give answers to problems percieved or real, and how they can be solved by engineering as that what creating a music repreduction system is, its engineering.
I often work on projects where it has to be 100% free of noise and interefrence even for the very sensitive analogue, so I see and have to work to solve these problems every day, but it needs both meassurements and listening, and as Eric Bogatin says "No myths allowed"
Marc
And, I agree. The "scope" of the problem, from an ordinary listener's POV is that the sound is not as good as it could or should be, and the "scope" in the sense that you use it is to fully understand, comprehend the issues. Which, as you say, is something that should be measured, as a good first step. However, to this point in time I see relatively poor levels of discussion and analysis of the situations, with even poorer resolution of them with convincing implementations. If it were otherwise then the audio world would not be the quarrelsome mess that it is.
I haven't the access to the tools which could measure precisely what is occurring, so up to now I've worked with instinct, research, trial and error, and simulation to develop improvements in the areas I worry about. I know what the end goal is, because I've heard it enough times to be certain that it exists, greater understanding of 'everything' would certainly help me get there faster and more consistently, but that may take another lifetime ...
Frank
Last edited:
So let me get this clear: we have an analogue back-end after the DAC which is so poorly designed that it picks up noise and interference from the digital processing before the DAC. This digital processing varies according to the quality of the Ethernet cable, thus demonstrating that the Ethernet receiver circuitry is so poor that it can't cope with normal operation.
I might believe that this is just about possible for a cheap mass-market boombox, where people just throw a few chips on a board and don't bother to do any real design (or even hire a real designer). I could also believe it for 'high-end audio', where people seem to think the usual rules (e.g. Maxwell's equations, circuit theory) don't apply to them so also don't hire a real designer.
Mid-market competent hi-fi audio? No, there they will use people who know how to design mixed-signal circuits and how to test them. Anything which was so poorly designed that it creates these alleged problems would be quite likely to fail the requirements for a CE mark too.
I might believe that this is just about possible for a cheap mass-market boombox, where people just throw a few chips on a board and don't bother to do any real design (or even hire a real designer). I could also believe it for 'high-end audio', where people seem to think the usual rules (e.g. Maxwell's equations, circuit theory) don't apply to them so also don't hire a real designer.
Mid-market competent hi-fi audio? No, there they will use people who know how to design mixed-signal circuits and how to test them. Anything which was so poorly designed that it creates these alleged problems would be quite likely to fail the requirements for a CE mark too.
Yes, they will be tested to ensure they match good, conventional electronic standards of quality. Say, accurate to 1 part in 1000. 0.1% is pretty damn good for most electrical parameters -- should do the job. Trouble is, the ears can be annoyingly sensitive to "errors" at even lower levels than this, which is when the bun fights start. And, is the the company going to pay the engineers, and for the parts, to get 0.01% accuracy, or 0.001% accuracy?
Frank
And, my instincts are informed by my hearing system assessing that one audio system sounds highly convincing, and another sounds like a typical hifi setup. Fortunately I am not constrained when being presented with the fact that the latter system is "correct", because conventional measurements have verified this, to reach the conclusion that my hearing mechanism must be faulty ...
Frank
Frank
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.