"lowest" are not an indicator for either "best" fidelity or "most" error free internal data transmission..
Can you please explain why "lowest" are as important as you emphasis ?
Hi Bunpei
Answers:
1. No. I will do it, maybe soon...
2. Yes, I have the data sheet of ESS9018. I had not yet got in to the I2C world. Of course is on the list. In the last time I have been working on completely diffident priorities. Time is not my best friend...
3. NONE! My original DAC chip (the same I`ve used all the time for my experiences) is still working just wonderful overclocked for months now. As I said many times before, the chip did not even get warmer at all than usually. Why? Ask Dustin...
4. I think it is already published a such picture of 125Mhz clock (or another frequency above 100Mhz) somewhere in the threads I`ve wrote in. It is a very nice sinus anyway. I think to do some more and detailed measurements, and I will come with some more infos and waves snap shots in near future...
Oh, I'm sorry that my description was ambiguous. The "lowest" is not for a sound quality but for a "DPLL Bandwidth" parameter setting.
Dear Coris,
I appreciated your honest and immediate answers very much.
I'd like to recommend you try a synchronous master clocking before you claim an overclocking of out of manufacturer's guarantee manner.
To tell the truth, I have broken three ES9018 chips in my experiments for AVCC over voltage and large capacitance decoupling. I felt really bad and spent much time to recover the boards by replacing the broken chips. Can you imagine such situation? This experience made me realized that I should never recommend such a risky experiment to others even if it looks safe initially.
You may say "My case is safe." Are you ready for taking care of any unexpected malfunction that may happen to those who follow your claim?
I hope you will understand well what I wanted to say to you.
By the way, through my need for repairing broken boards, I could establish a very easy way of replacing the DAC chip on board.
Bunpei
I appreciated your honest and immediate answers very much.
I'd like to recommend you try a synchronous master clocking before you claim an overclocking of out of manufacturer's guarantee manner.
To tell the truth, I have broken three ES9018 chips in my experiments for AVCC over voltage and large capacitance decoupling. I felt really bad and spent much time to recover the boards by replacing the broken chips. Can you imagine such situation? This experience made me realized that I should never recommend such a risky experiment to others even if it looks safe initially.
You may say "My case is safe." Are you ready for taking care of any unexpected malfunction that may happen to those who follow your claim?
I hope you will understand well what I wanted to say to you.
By the way, through my need for repairing broken boards, I could establish a very easy way of replacing the DAC chip on board.
Bunpei
Hi Bunpei, what are the limits on AVCC voltage and large capacitor decoupling?
Hi Coris,
That's very interesting!
Did you measure the current consumption for AVCC, VDD (1.2V) and 3.3V when the chip is overclocked?
From my past experience with overclocking electronics the power consumption and dissipation will goes up not down.
IMHO, if the currents are lower than in non overclocked mode than I presume some internal DAC blocks are not functioning correctly or not at all.
Kind regards,
Mihai
Hi Bunpei
Sorry for your problems and lot of work for replacing the destroyed chips.
I think I must clarify one more time something: What I`ve wrote here about those my experiments results (overclocking and so on) were not meant as a recommendation about how have to be done to have the best results with this ESS9018. I have done some experiments, I have found something, I had/have some results, and I just shared all those here in those dedicated threads. There is about informations here. If one may follow and practice with those informations, try to verify it, try to get some or better results, it is of course on ones own risk.
AVCC over voltage is safe when is do it right. TPA sell theirs Buffalo with over voltage on AVCC and they have for sure success with that DAC design. It came out from some informations, discussions (Dustin) on this forum that an over voltage on AVCC it may be safe with good results. That it were not an recommendation about how have to be done, but an information. Everybody take his own risk to try this or not... I took that risk too, and succeed...
I run my DAC with 3,6V on AVCC for quite long time and nothing wrong with that, only better results. The same about large capacities on decoupling. This can definitely not destroy the DAC chip, only maybe the regulators if is done it wrong...
I just remember that you were gratefully to me in one earlier post for the information I came with here about large decoupling capacities on ESS9018, when you got better result practicing with this information. You say now that this it were destroyed your chips... I have a little problem to understand this.
As a conclusion, I think maybe I have to hold for my self what about my experiments and the results I get... Maybe I wrong understood the principle/idea with this forum...
Buna Mihai
I do not say that the power consummation or dissipation get down, only that dissipation do not get up to overheat that chip... So long that this functioning were OK for me, I did not researched very much to find out why is like this. I will do it ...
I understand that you may presume that the DAC is not functioning right, and some blocks inside are stop working, but then why the resulting sound is better?
you claim you do not encourage others to make such foolhardy modifications? I beg to differ, you encourage many across several forums other than this one as well, simply by making fantastic claims of better sound quality and claiming, with absolutely nothing to back it up, that there is no harm. you even illustrate how to do it with photos and wax lyrical about the wonderful sounds.
without going to google, can you explain by what mechanism adding too much decoupling might cause harm and what steps you have undertaken to avoid it? just so we know you have considered the risks; even if you havent objectively tested ANY of them.
people have had their dacs start to malfunction from overclocking on your recommendation, yet you still deny that there is any problem. You cannot even supply a theory by which it would provide an advantage to do it, forget measurements, I know you have an aversion to them, but you cannot even theorize,
here in Bunpei we have someone far more knowledgeable than you with more experience doing more advanced modifications on ESS dacs than you and with something of a reputation for doing so. He is going on record telling you that he has damaged chips this way, then you answer that with the above?
the truth is you have no idea what is going on inside the chip, if you did you would see overclocking is futile, you feel it with your finger, its not burning, so is OK? your finger temperature test does not hold much water....
lets list it out
overclocking lowers temperature of a fully operational es9018 dac chip?
no, it raises the temperature of a fully operational es9018 dac chip
overclocking has lower noise?
no, overclocking causes more noise
overclocking helps maintain better lock?
no, it makes it more difficult
overclocking produces cleaner audio?
no, it creates more glitches, more noise, more RFI
overclocking puts less load on the power supply and decoupling?
no, it puts a higher load on all of these things
overclocking allows higher sample-rates?
no, there is no material released that the ESS can play with higher than recommended speed clocks that it cannot play without them.
overclocking has lower jitter and allows the use of lower jitter clocks?
no, the clocks you are using are lower quality than the standard rate chrystek, NDK etc parts.
overclocking makes it easier to maintain proper digital and analogue signal integrity?
No, it makes it harder
then we have all the issues of regulators, AC and Z termination, PCB designs etc being designed for different speeds etc etc.
To top this off, you also recommend running the AVCC voltage at up to 4VDC. I run 3.65VDC but I would not run 4v personally, let alone recommend it to others while also overclocking
so what is it, what is the benefit to offset the obvious risks?
there must be one?
yet you paint me as the negative one, i'm just trying to offer a voice of reason when you dont seem to care what happens to other peoples equipment. I ask for a reason to take these risks, you cannot provide one. I'm the negative one because it causes you frustration.
DIYAUDIO is a technical forum where we discuss various techniques to improve our audio equipment, both subjectively and objectively. if you are going to make claims, you should expect now and again to be asked to back them up. If you dont like that, there are plenty of other forums where such claims are accepted without question.
Last edited:
Great post! You have stile, I can say...
One may test first the regulators if they work well with large capacities on its outputs, before coupling it to the DAC chip... It is a basic precaution... As usually all the regulators have over current protection.
I can not see that in one of my posts I claim that ESS9018 have to have 4v on AVCC. I use 3,6v.
Else your last post is full of exaggerations. Specially when you claim that I destroy the folks DACs... Sorry, this it may be only stupid.
By the way, it is enough surprising for me that Bunpei with his experience have destroyed so many chips. But it may happen on some unlucky days...
One may test first the regulators if they work well with large capacities on its outputs, before coupling it to the DAC chip... It is a basic precaution... As usually all the regulators have over current protection.
I can not see that in one of my posts I claim that ESS9018 have to have 4v on AVCC. I use 3,6v.
Else your last post is full of exaggerations. Specially when you claim that I destroy the folks DACs... Sorry, this it may be only stupid.
By the way, it is enough surprising for me that Bunpei with his experience have destroyed so many chips. But it may happen on some unlucky days...
its spelled style... it wasnt meant to be stylish, it was meant to be clear, since you seem to have such trouble understanding the very basic intent of my query. Yet you continue to avoid the questions.
There was not a single exaggeration, rather matter of fact actually, please stop it with playing the victim. There is nothing like the exaggeration where you claim that I said you destroy peoples dac chips, I said peoples dacs have (very predictably) started to malfunction on your recommendation to run higher clock speeds. that is a fact, not an exaggeration.
it is also a fact that you say its OK to run at 4v and try to back it up with Dustin Saying its OK, it is true the pins are 4v tolerant, but not at over 100MHz and you would want to be very sure that the regulator you use does not have higher start up voltage and its voltage reference doesnt vary with temperature, resulting in higher than 4v. running the higher voltage at the same time as overclocking will only increase the likelihood of permanent damage.
There was not a single exaggeration, rather matter of fact actually, please stop it with playing the victim. There is nothing like the exaggeration where you claim that I said you destroy peoples dac chips, I said peoples dacs have (very predictably) started to malfunction on your recommendation to run higher clock speeds. that is a fact, not an exaggeration.
it is also a fact that you say its OK to run at 4v and try to back it up with Dustin Saying its OK, it is true the pins are 4v tolerant, but not at over 100MHz and you would want to be very sure that the regulator you use does not have higher start up voltage and its voltage reference doesnt vary with temperature, resulting in higher than 4v. running the higher voltage at the same time as overclocking will only increase the likelihood of permanent damage.
You just do not want to accept, or do not understand that simple fact that this my "damaging" setup it is working very well and stable for months now, with an exceptional sound?
I see that you want to force in your "logic", based on your own fantasy about those 4v AVCC. I`ve wrote already many times that I do not have 4v on that pins, but 3,6v.
Sorry, but I will not continue to answer your posts from now on. I have got enough of your such interventions...
I repeat here once again for the other readers about my functional (in use) hardware setup for ESS9018: 125Mhz clock, connection 2-3mm between oscillator (SAW type now) out and DAC chip clock in. Oscillator powered directly from battery. Ultra low noise regulators for AVCC - 3,6v, 6-7mm long connection to the DAC, with 1200µf decoupling ceramics caps on this rails, and soldered directly to the chip pins. Same type caps 500µF decoupling on all digital power rails. DAC chip completely shielded, oscillator outside. Silver wire for all connections.
As my time will permit, I think to come back with some figures in the near future...
I see that you want to force in your "logic", based on your own fantasy about those 4v AVCC. I`ve wrote already many times that I do not have 4v on that pins, but 3,6v.
Sorry, but I will not continue to answer your posts from now on. I have got enough of your such interventions...
I repeat here once again for the other readers about my functional (in use) hardware setup for ESS9018: 125Mhz clock, connection 2-3mm between oscillator (SAW type now) out and DAC chip clock in. Oscillator powered directly from battery. Ultra low noise regulators for AVCC - 3,6v, 6-7mm long connection to the DAC, with 1200µf decoupling ceramics caps on this rails, and soldered directly to the chip pins. Same type caps 500µF decoupling on all digital power rails. DAC chip completely shielded, oscillator outside. Silver wire for all connections.
As my time will permit, I think to come back with some figures in the near future...
Last edited:
you have reading comprehension issues...
among others....
you stated yours didnt work at 133MHz, so you 'eased back' to 125MHz and 'the problems stopped' KlipschKids stopped working at 125MHz.
but nooo, nothing wrong with the faster clocks, faster must = better.... it just must... because it just does
among others....
you stated yours didnt work at 133MHz, so you 'eased back' to 125MHz and 'the problems stopped' KlipschKids stopped working at 125MHz.
but nooo, nothing wrong with the faster clocks, faster must = better.... it just must... because it just does
Last edited:
Hi, Coris,
(I hope you are still a read-only member on this thread and you read this post.)
I do not intend to blame you because the overvoltage experiment was not inspired by you. I made my risk-taking decision before the experiment and I did not accuse anyone else. I could souce the DAC chips for replacement easily.
However, many DIY users in this world may have their own situations. Some of them will blame you keenly when they encounter severe problems in case of making modifications base on your claim.
Anyway, I could experience a significant influence of quality of AVCC power supply including a choice of types and values for decoupling capacitors through the series of trials.
Bunpei