For posterity it is much better to have the discussion in one place (the original thread). It is also against Forum Rules to start multiple threads on the same subject.
@Markw4 IIUC, the board in this thread is just the reclocker part? The I2SoverUSB will take audio from USB and output I2S signal. Does the reclocker board then reclocks the signal and then outputs I2S ? The reason i ask this question is that, is it a general purpose board that can feed other DACs like DDDAC that take i2s input?
For example in the pic you posted it says DSDR, BCLK and DSDL
Is it same as BCK, LRCK and DATA
I2SoverUSB board itself can also take separate input power (dirty side) and output power (clean side). Do we use I2SoverUSB board like that?
For example in the pic you posted it says DSDR, BCLK and DSDL
Is it same as BCK, LRCK and DATA
I2SoverUSB board itself can also take separate input power (dirty side) and output power (clean side). Do we use I2SoverUSB board like that?
Last edited:
Not really. The thread started with an open source Clock Board. I was hoping other people would step up to contribute other things needed to make what I consider to be a good dac. But nobody did. If you read through this thread, I reluctantly designed a reclocker to share because nobody else would. Same for the interconnect board, although I complained less about having to do that.
Also, the reason I showed pics in the first post of a couple of ways this stuff could be used was to illustrate there is a lot more than just plugging together boards. Power supplies, ground loops, all sorts of other things matter too. They are all appropriate for discussion if someone wants to know my approach to making a good diy dac.
In that regard, it doesn't help when some guy with a different audio religion comes along and tries to negate and cancel out everything I say. Its hard enough as it is to try to be helpful for free without constantly being trolled. I am already fielding multiple PM conversations at once to try to help people who need clarity and don't want to be in the middle of trolling distractions.
It can. The general purpose boards I designed to illustrate some possible approaches could be used with various dacs, although in some cases some modifications might be needed. Depends what a particular dac needs in the way of input signals .
I expect it could used or adapted for use with that.
I2SoverUSB from JL Sounds is galvanically isolated with clean and dirty sides. That's what I have been using. The dirty side can run on USB power or from an clean isolated +5v supply, which is what I do. That's the blue USB board you see on the lower left side of Cestrian's clock and reclcocking board.
Last edited:
Okay, now that you added the pics, no. Its called a "Clock and Reclocker Board." My open source reclocker board is linked to from the first post of this thread; it doesn't include clocks.
What we have been discussing recently is a simplified, compact clock and reclocker board by Cestrian. That's the pic you showed. Regarding that part of the board, its true the D-flip flops are used for the actual reclocking. However, if you don't have a good MCLK signal to clock the flip flops then you will have garbage in, garbage out. The rest of the whole right side of the board is to produce that good quality MCLK signal, and to synchronize it in time with the incoming I2S data so the the reclocking will have the desired effect. Therefore the right side of the board also has to provide an external MCLK signal to clock the USB board synchronously with the reclocker flip flops.
Those are the I2S bus signal Marcel's RTZ dac needs. That's what Cestrian's board is designed to work with. The signals work fine and are aligned in time to work fine with Marcel's RTZ dac. If you take some other dac, such as an ESS dac, it has different timing requirements. So, you may need to delay one of the signals a little, or you may want to only reclock the MCLK signal. You have to read the data sheet for the dac you want to use and condition the signals as necessary for that particular dac. Now, it isn't that hard to do in most cases. In some cases it can work as-is, or it might be as trivial as making one of the D-flip flop series damping resistors a higher resistance than we otherwise would. Or maybe we would have to add an inverter or two to make a low jitter delay. Depends.
Yes.
Last edited:
Posting measurements that indicate issues with Marcel's RTZ dac and PCM2DSD has nothing to do with audio religion or trolling. Your unsubstantiated claims about IanCanada's products are trolling and rude as other members pointed out.
@MarcelvdG ,
There is a bit more to it than reclocking of it. If people are to consider building an RTZ dac project they need to have some comfort they understand what the dac is and how it basically works. Some of these people hardly have the technical background to understand how to plug together Iancanada RPi stack modules. If you think it better to talk them through a basic understanding of how your RTZ dac works where the introductory/tutorial stuff is all at once place in the thread, maybe it should be in your thread. The question is do you want to do the low level tutoring? Do you want to keep distractions involving audibility versus FFT measurements, versus random noise jitter measurements versus random noise Vref measurements versus audibility, etc., under your control? Can there be a simple tutorial without arguments and distractions?
And then there is the question of support circuitry. Its not just about reclocking. Its about the fact that the dac can range from sounding awful to sounding great depending on a number of details involving the surrounding circuitry. For one example, for that to make sense it helps to connect power supply schemes to dac and power rail loads as they are presented to external power supplies, including noise effects produced by the dac circuitry. Why shouldn't the output stage be powered from the same +-15v rails as the dac board? Does it make no difference? What is the underlying problem if there is one and how does it arise? What's the best solution? Is there any benefit to using shunts? Etc.
The problem with some of that stuff in the last paragraph is I am going to say if the effects are clearly audible on my system. I don't care how they measure if I can hear them. I also don't care too much about measurements if I can't hear them (I don't necessarily need 130dB SINAD even if somebody can measure it, it may not be the most important technical and or audibility problem).
Mark
There is a bit more to it than reclocking of it. If people are to consider building an RTZ dac project they need to have some comfort they understand what the dac is and how it basically works. Some of these people hardly have the technical background to understand how to plug together Iancanada RPi stack modules. If you think it better to talk them through a basic understanding of how your RTZ dac works where the introductory/tutorial stuff is all at once place in the thread, maybe it should be in your thread. The question is do you want to do the low level tutoring? Do you want to keep distractions involving audibility versus FFT measurements, versus random noise jitter measurements versus random noise Vref measurements versus audibility, etc., under your control? Can there be a simple tutorial without arguments and distractions?
And then there is the question of support circuitry. Its not just about reclocking. Its about the fact that the dac can range from sounding awful to sounding great depending on a number of details involving the surrounding circuitry. For one example, for that to make sense it helps to connect power supply schemes to dac and power rail loads as they are presented to external power supplies, including noise effects produced by the dac circuitry. Why shouldn't the output stage be powered from the same +-15v rails as the dac board? Does it make no difference? What is the underlying problem if there is one and how does it arise? What's the best solution? Is there any benefit to using shunts? Etc.
The problem with some of that stuff in the last paragraph is I am going to say if the effects are clearly audible on my system. I don't care how they measure if I can hear them. I also don't care too much about measurements if I can't hear them (I don't necessarily need 130dB SINAD even if somebody can measure it, it may not be the most important technical and or audibility problem).
Mark
Last edited:
Well, if I don't hear anything back I will probably just proceed as I see fit. If Marcel feels its a problem at some point then I will take it very seriously. Also, if I happen to say some thing that's not quite right, I would sure hope that Marcel would be willing to take a moment to correct me.
What you refer to as distractions were in this case started by you yourself as you saw it necessary to bash IanCanada and his devices in his thread without any actual evidence.
https://www.diyaudio.com/community/...eapon-to-fight-the-jitter.192465/post-7974740
https://www.diyaudio.com/community/...eapon-to-fight-the-jitter.192465/post-7974833
This is a diy forum and there is no need make it a competition of who's best especially if you don't have anything concrete but just subjective opinions. Also if I or anybody else prefers something else than you we are allowed to express our opinions just as well without having to receive accusations of trolling or ad hominems.
Simple tutorial about how RTZ dac works does not contain any subjective opinions about audibility or your preferences which actually are the source for these alleged distractions. If you leave your subjective arguments out and just focus on the tutorial side of it there is no need for any distractions.
You are making up stuff out of your imagination. I own a number of Iancanada devices, some of which are good and some of which are not equally as good. Do you? A long time ago I wrote about experiments showing noise coupling between RPi and FIFI_Pi, and between FIFO_Pi and Ian's ES9038Q2M dac. It was only after that he brought out station pi and shield pi to help with those very same noise coupling issues. I still have that dac and he still sells that same one. It didn't sound great, but FIFO_Pi was pretty good compared to what other people were doing. I know probably better than you do. I didn't want to bring up all that old baggage, but you forcing me to do it. You are the one forcing more to come out about Iancanada and thus the one most responsible for causing him damage, more than anything I said up until now. I hope you're happy.
And you do troll, that's your specialty. What you do is try to shut up and or attack people who beliefs about audio are different from your own.
Your audio religion is based around your belief in your measurements as being all there is that matter. The fact is you can't and nobody else can measure the combined effects of dac and clock phase noise together while disentangling it from amplitude noise. But that doesn't mean what you happen to be able to measure is more important under all conditions. What you can easily measure probably isn't more important because its likely once FFT measurements are good enough they matter less and less relative to other, harder to measure problems. Thus focusing everything around your particular sacred belief can be a mistake. It is your sacred beliefs that make yours an audio religion. Its about holding things sacred, like your precious measurements. It is well known they do not exactly correlate with how people hear.
There is much more this that can be covered in a few posts. No point in going into all over again because you will ignore the information and go right back on the attack.
And you do troll, that's your specialty. What you do is try to shut up and or attack people who beliefs about audio are different from your own.
Your audio religion is based around your belief in your measurements as being all there is that matter. The fact is you can't and nobody else can measure the combined effects of dac and clock phase noise together while disentangling it from amplitude noise. But that doesn't mean what you happen to be able to measure is more important under all conditions. What you can easily measure probably isn't more important because its likely once FFT measurements are good enough they matter less and less relative to other, harder to measure problems. Thus focusing everything around your particular sacred belief can be a mistake. It is your sacred beliefs that make yours an audio religion. Its about holding things sacred, like your precious measurements. It is well known they do not exactly correlate with how people hear.
There is much more this that can be covered in a few posts. No point in going into all over again because you will ignore the information and go right back on the attack.
Last edited:
Can you point to a single instance where I have tried to shut you up? Posting opposing views to yours is not attacking or trolling.
It is quite easy to find examples where you attack other members without any real evidence or cause:
https://www.diyaudio.com/community/threads/kaamos-tech.408842/post-7592031
Again, where have I said that? Posting measurements is not an audio religion. I happen to possess equipment that allows me to take reasonably good measurements so why not use them? Even Marcel has found them useful.
Last edited:
Tutoring requires indisputable, objective facts to be conveyed. One don't possess these if one can't back them up with scientific rigour, and thus, the tutoring must be downgraded to fable telling. One should be careful if one are a self appointed expert with a not so small ego - such a person should try to take this into consideration when one word ones gifts to the masses. But I'm sure most on this distinguished forum can master such a balance when trying to inject insight into e.g. fellow "people" that "hardly have the technical background to understand".
//
Nobody can be allowed to tutor astronomy because we don't know for sure if the universe is expanding forever or ultimately contracting? What you seem to be saying is totally at odds with the very nature of science. Science is always a work in progress where there is no absolute certainty, only reduction in doubt over time (often with missteps along the way).
BTW, this has all been discussed in the forum before. Did you ever read about it and or read about the philosophy of science?
Last edited:
No. Tutoring requires the tutor to have something of value to convey and the ability to convey it in ways the students can absorb.
But where someone uses just experience, not indisputable fact, it needs to stated. Except, of course every "fact" has been and likely will be again, disputed.
I can teach you to prepare an amazing fillet mignon. I learned from someone with experience. No indisputable science fact was involved. All experience.
I can make a Wiener Schnitzel as good as Borchart in Berlin, I learned it in their kitchen, actually, in '92. Again, what was involved was experience, not indisputable facts.
I learned how to zero a Dragunov rifle, how to estimate the lead on a target without indisputable facts. In fact, without intuiting the complex integral/differial math in muscle memory, it would take far too long to calculate a firing solution and the target would long have disappeared.
You might say "but this is audio, not cooking or shooting a running target at 500m."
The issue is that at this point in time, "signal fidelity" as tested using instruments such as audio precision et al and "audible fidelity" have no demonstrated link. More, a lot of work on perceptual coding for data compression of audio signals actually demonstrated the opposite.
As such "signal fidelity" and "audible fidelity" are best treated as NOMA.
Which of course is a profound insult to all those cargo cult scientists who mistakenly think that signal fidelity and audible fidelity are the same.
The only people who categorically sure they are absolutely right, are actually those who are wrong.
Thor
Last edited:
I would call people who mistakenly think that there is "the audible fidelity" the real cargo cult disciples (not scientists). You have your audible fidelity, I have mine and ASR members have theirs. For the latter folk signal fidelity and audible fidelity may quite well be the same. And they are as right in their belief as you.
Exactly... and most of time this is pertinantly expressed. So one should only convey what is the current accepted facts. And state clearly when there is a doubt that it is something else expressed.
So no "feel-facts"!
I wish all in the forum good luck with this challenge - including myself.
//
Tell that to JJ (old friend who did all the work on the perceptual side of MP3, using proper science and properly controlled listing tests).
Nope, "my" audible fidelity is actually science, proper science. And it's not mine, it's a matter of public record and extensive publication in peer reviewed journals (not the JAES though) on perceptual science.
Now the rest, I don't know about yours, but over at ASR cargo cult science goes into overdrive.
And I'm not saying that because I got banned, I got banned for saying it and giving arguments that could not be falsified, while being 100% falsifyable.
There is no "belief" in "my" (again, it's not mine) audible fidelity. Just lengthy empirical research, that started by getting the basics right.
That's why 128kVBR MP3 is distinguishable from CD AND for a vast majority of listeners is actually delivering superior AUDIBLE FIDELITY.
Thor
Last edited:
Now we are getting way, way, way off topic on the subject of a general clock board, let me throw in some science that interrests me:
https://tu-dresden.de/tu-dresden/ne...as-was-sie-zu-hoeren-erwarten?set_language=en
Thor
You have linked to that study before in Marcel's RTZ dac thread and it was not anymore convincing back then. What exactly in that makes your audible fidelity more scientific or more correct than that of ASR members?
Quote from the article:
Dr Alejandro Tabas, first author of the publication, states on the findings: "Our subjective beliefs on the physical world have a decisive role on how we perceive reality. Decades of research in neuroscience had already shown that the cerebral cortex, the part of the brain that is most developed in humans and apes, scans the sensory world by testing these beliefs against the actual sensory information. We have now shown that this process also dominates the most primitive and evolutionary conserved parts of the brain. All that we perceive might be deeply contaminated by our subjective beliefs on the physical world."
So IOW our perception is subjective. Which also means ASR members claiming that signal fidelity and audible fidelity are the same base their claim on their subjective belief. Just as you do with yours or I do with mine.
Quote from the article:
Dr Alejandro Tabas, first author of the publication, states on the findings: "Our subjective beliefs on the physical world have a decisive role on how we perceive reality. Decades of research in neuroscience had already shown that the cerebral cortex, the part of the brain that is most developed in humans and apes, scans the sensory world by testing these beliefs against the actual sensory information. We have now shown that this process also dominates the most primitive and evolutionary conserved parts of the brain. All that we perceive might be deeply contaminated by our subjective beliefs on the physical world."
So IOW our perception is subjective. Which also means ASR members claiming that signal fidelity and audible fidelity are the same base their claim on their subjective belief. Just as you do with yours or I do with mine.
Last edited:
Convincing of what? What is the point I am trying to make with quoting such research?
Do you understand the implications for listening tests? Anecdotally we have this issue raised for decades, but there never was a study like this before, which explains what happens, how and why.
It clearly illustrates that listening tests as up to now are not meaningful and their results are makulatur.
It illustrates that we need a fundamentally different approach to the science of listening, with scientific evidence, that essentially could technically be falsified, but is in practice unassailable.
If you find that unconvincing, again, I ask unconvincing of what?
Thor