Okay. But calling people crooks or the equivalent, can constitute a libel if it isn't true (which is illegal). If you say it as a matter of fact, that is.
If you say it as a matter of personal opinion, then its a different matter entirely. But you have make very clear its an opinion.
But if someone solicits funds to run a for profit LLC that gives advise to people he himself knows (and has publicly admitted to) to be based on things irrelevant to the advise and does not provide "health warnings" that he is neither qualified to give such advise nor is what he presents based on any proven evidence on the subject that what is claimed is related to the subject, I call that a fact in a matter of fact way.
You do not?
I did not use the word crook. Crooks steal and rob.
I used a different word that means taking money for something known to be worthless or of much less value than presented and demanded, but still providing something to the punter. The bottle of "Dr Good". Like what you get, for example at the Carnival. Or buying expensive watches from known Felons.
You might even enjoy being taken advantage of. That does not however change the nature of what takes place.
Thor
Asking for and accepting donations by itself does not constitute a contract. A contract requires consideration (e.g. if you do this for me then I will do that for you). Both people have to give something to get something. A gift is different. It is a giving without expectation of anything in return. I believe an untethered donation is a usually gift.
There is no dishonest transaction in that case, unless perhaps a promise was made that the donation was for a particular purpose, say, to get a new file server for the website (which could then be enjoyed by the giver), then that could constitute a contract. If the recipient of the donation instead spent the money on a personal vacation then there could be a breach of contract or even a criminal act. Depends on the specifics.
Last edited:
The phase noise measurement Markw4 has often promoted (https://www.diyaudio.com/community/threads/phase-noise-in-ds-dacs.387862/post-7063038) is quite useful for assessing the implementation quality of Vref and clocking. Objective audibility implications of such measurements are harder to assess as overall there is little real evidence that supports audible benefits of ultra-low phase noise.
I have to disagree with this. There are many very knowledgeable members at ASR and some of those are also members here. Some have even posted on the forementioned Markw4's thread.
What resembles cargo cult at ASR is that unfortunately there are plenty of wannabe experts who will parrot whatever the real experts have to say probably in an attempt to "climb the ladder". So typically every controversial posts will receive tens of responses with similar content. But on the other hand disagreeing with ASR members or getting banned does not make your opinion more correct.
Regarding mountebanks the focus on measurements at ASR has actually exposed plenty of high-end audio manufacturers as the real mountebanks. Besides ASR is not a Cult of SINAD as there are few members there that claim SINAD is all there is. However the rankings based on SINAD has had an unfortunate side effect as some manufacturers (mostly chifi) have clearly made willful questionable design decisions targeted for high ranking at ASR SINAD.
We are in a new domain where we are still the law catching up. On line systems on many levels change the way a lot of things work.
We are not debating a breach of contract.
Hence the word mountebank, as in "confidence trickster".
Not Crook, Thief or Fraudster.
But someone who convinces you to do something that in end is not to your benefit, even to your disbenefit but does not commit outright fraud or illegal acts in doing so. Like the people running games at a carnival.
All you wrote to me confirms to me my use of mountebank is precise and applicable to the case, because it is precisely not alleging fraud or criminal activity, just deceptive behaviour.
Are you still insisting my use is out of order?
After all this Cunning Linguism I'm severely tired. I'm not a Master Debater, so I'll leave it here.
How did we make this Mount Everest out of a one ant anthill?
Please do. And I have to disagree with what you presented. We can leave it at that and go our separate ways.
Too many bytes wated on debating this cargo cult on that site already.
Thor
Last edited:
Hi tamra,
Reclocking may happen multiple times as I2S is processed, but only the last reclocking just before the dac is the one that matters most. It is the one that has the most effect on dac sound.
Regarding the reclocking done by I2SoverUSB, it is done in a CPLD chip. CPLD chips are not that good for reducing jitter.
Reclocking may happen multiple times as I2S is processed, but only the last reclocking just before the dac is the one that matters most. It is the one that has the most effect on dac sound.
Regarding the reclocking done by I2SoverUSB, it is done in a CPLD chip. CPLD chips are not that good for reducing jitter.
The clock board work for both reclocker board and I2SoverUSB at same time?
I am ordering RTZ and your PCBs now.
Also may I ask which regulator board did you use for your build shown on your 1st post?
I am ordering RTZ and your PCBs now.
Also may I ask which regulator board did you use for your build shown on your 1st post?
The clock board can drive both I2SoverUSB and the reclocker at the same time. In that case 22/24MHz clocks should probably be used, and the PLL on the clock board can then produce 45/49MHz for I2SoverUSB (please note the PLL chip will go "last time buy" early next year). The clock board should be configured in a particular way for use with Marcel's dac and I2SoverUSB. I described how to do that in post #144 of this thread. https://www.diyaudio.com/community/threads/general-purpose-dac-clock-board.413001/post-7729060
Most of the external pre-regulators I use are based on LT1083 or LT1084. They tend to sound decent enough. Most of those boards came from ebay or aliexpress. One thing that can make them a little better is to replace the trim pot with a good cerment unit such as Bourns 3296W
EDIT: Regarding the configuration described in post #144, it reduces the basic concept of the Clock Board to look more like this (simplified diagram):
Most of the external pre-regulators I use are based on LT1083 or LT1084. They tend to sound decent enough. Most of those boards came from ebay or aliexpress. One thing that can make them a little better is to replace the trim pot with a good cerment unit such as Bourns 3296W
EDIT: Regarding the configuration described in post #144, it reduces the basic concept of the Clock Board to look more like this (simplified diagram):
Last edited:
Thank you so much.
I will keep some PLL chips.
I have 45/49MHz clocks, accusilicon 318B and 45Mhz Ian's scpure is coming soon.
Do they work for it or I shoul buy 22/24mhz 318B?
Better clocks( including Ian's scpure )are too expensive for my budget.
I will keep some PLL chips.
I have 45/49MHz clocks, accusilicon 318B and 45Mhz Ian's scpure is coming soon.
Do they work for it or I shoul buy 22/24mhz 318B?
Better clocks( including Ian's scpure )are too expensive for my budget.
If using 45/49MHz clocks, there are few different things you could try, some of which I have not tried yet.
1. You could try configuring the Clock Board using the frequency dividers but leave out the constant outputs intended for Andrea Mori type use, and not use the PLL. Not sure if you would find a crosstalk problem or not. In that case the clock board would look more like this (you can also configure the frequency dividers to run from AUX power, which may or may not help):
2. You could do like in 1. above, except leave out the first relay, and keep the PLL to generate the 45/49MHz clock signal for I2SoverUSB. It would look more like this:
3. You could use two clock boards, one board for each clock, then do the clock switching after the clock boards.
1. You could try configuring the Clock Board using the frequency dividers but leave out the constant outputs intended for Andrea Mori type use, and not use the PLL. Not sure if you would find a crosstalk problem or not. In that case the clock board would look more like this (you can also configure the frequency dividers to run from AUX power, which may or may not help):
2. You could do like in 1. above, except leave out the first relay, and keep the PLL to generate the 45/49MHz clock signal for I2SoverUSB. It would look more like this:
3. You could use two clock boards, one board for each clock, then do the clock switching after the clock boards.
Last edited:
If using long clock lines and 45/49 MHz clocks I'd probably recommend LVDS or PECL circuit and running the clock lines differential.
Using clocks with differential PECL out would be good. Dividing can be done in PECL. PECL has basically no appreciable delay or jitter.
Eventually we need to convert ECL to CMOS, again very low jitter IC's exist for that.
2. You could try configuring the Clock Board using the frequency dividers but leave out the constant outputs intended for Andrea Mori type use, and not use the PLL. Not sure if you would find a crosstalk problem or not. In that case the clock board would look more like this (you can also configure the frequency dividers to run from AUX power, which may or may not help):
Switching clocks using RF Relays is a good solution, I have used it myself.
One problem is that it causes the clock to glitch, it may cause problems downstream.
An alternative are CMOS switches aimed at switching 480MBPS USB for example. The NXP/Nexperia 74CB3Q3257BQ can switch 5V signals on 3.3V supplies, have extremely low propagation delay etc. No contact glitches, switching in under 10nS.
Also, unless the clocks must be running continuous, I'd probably gate the unused clock off.
Thor
I see it looks complicated using 45/49MHz,
ThenI would better use 22/24Mhz,
Are there any recommendations to use Andrea's older or cheaper clocks?
I will start from accusilicon 22/24 then upgrade later.
ThenI would better use 22/24Mhz,
Are there any recommendations to use Andrea's older or cheaper clocks?
I will start from accusilicon 22/24 then upgrade later.
If you already have a 45MHz Iancanada clock coming then I think you should try it, because once you hear it then you will want to keep it (after it burns in for a week, that is). Its possible to use I2SoverUSB with only one clock, 45MHz, so long as you only send it 44KHz clock family recordings (such as CD rips). With only one clock there will be no crosstalk to worry about.
Also, I wasn't aware of Accusilicon clocks in 22/24MHz. They used to only make them in 45/49MHz.
Also, I wasn't aware of Accusilicon clocks in 22/24MHz. They used to only make them in 45/49MHz.
Not a problem. The important clock lines can be reasonably short. The only device that needs 45/49MHz clocks is the USB board, and its output will be reclocked before the dac anyway.
If you want to use Andrea clocks, then you should only use the sine wave oscillators without doublers and without squaring. You might need my squaring board in that case.
Did some tests a few years ago with off the shelf audio clocks, such as Accusilicon and Crystek. Found that if they are gated off too long, then part of them has to warm up again to fully stabilize. Bringing them up for only one song doesn't give them time to fully settle. Also, I believe Andrea Mori independently arrived at the same conclusion.
Of course, the unused clock could be gated off using a two relay scheme like @PJotr25 suggested, if more isolation is desired. I know Andrea Mori used a two relay design. Probably because he could measure some crosstalk. I used other methods and found that its possible to get very good audible results using only one relay. Thus I went with only one. It wasn't until I designed my reclocker board that I could hear the crosstalk effects. Prior to that the ferrites and nonlinear caps in my old reclocker board masked any clock crosstalk effects. That's one reason I question measurements without practical verification. A non or minimal problem gets fixed while a real problem is too complex to measure easily, so it is ignored.
Last edited:
Sorry that accusilicon has only 45/49 and 90/98Mhz but I saw this.
https://a.aliexpress.com/_optiG2p
If I like scpure,then I will buy 49Mhz,
And 22/24 later.
Thank you for suggestions but I am too noob to arrange.
https://a.aliexpress.com/_optiG2p
If I like scpure,then I will buy 49Mhz,
And 22/24 later.
Thank you for suggestions but I am too noob to arrange.