If you read the 'ES9038Q2M board' thread, I just made some progress with a clocking problem I was working on. Typically, if using 100MHz clocks to allow dac operation at the highest sample rates then we use Crystek (e.g. 575 or 957). They can be good enough especially if on their own dedicated voltage regulator.
Has he shared his criteria? Measuring oscillators is well understood if challenging for the best. And even the best oscillator source can be seriously compromised if the distribution is not carefully managed.
The ESS has ASRC's for the input that may or may not affect the audio. Another place where lots of questions exist with few answers. Not to mention running it at 100 MHz vs 24.576 MHz.
Oh, I think we know quite a bit more about it now than not too long ago. Its just not something that has been discussed in this particular thread.
You do realise that the carbon black is the ingredient LEAST likely to affect the sound in the cocktail of plasticisers and other things they add?
I would not assume that. It was never mentioned to me in my telephone conversation. Mostly the analog IC/opamp. If anyone thinks clocking jitter is analog etc etc.... you should talk to BenchMark and ask them. They think all that is important but not what makes the most audible difference compared to "analog" part of circuitry.
Just ask them and not assume what 'analog' cicuit means to them. We talked about the differences and where most differences are found today and most variable is from the analog part of the design .
THx-RNMarsh
Static electricity and UV and harder, longer wearing = carbon blk.
Last edited:
I do think that the oscillator, etc IS very important, and I don't know if the ESS evaluation board is really top notch. I seriously doubt it, so IF I wanted to make a design for production by Parasound, I would be compromised, I am pretty sure. Benchmark makes a pretty good D-A, but not necessarily the best.
It wan't an assumption. One of their blog posts shows how the PCB is divided into analog, digital, and power supply sections. Clocks were part of the analog section, IIRC.
In fact, in their app note/blog post here: Inside the DAC2 - Part 1 - Analog Processing - Benchmark Media Systems they specifically call out some clock functions as analog.
Last edited:
The board doesn't have to be top notch. Is should be possible to modify one to make it as good as it needs to be. I have done it starting with less in order to have a way to investigate the effects of various modifications. I am not making this stuff up either.
Markw4, I appreciate your input. I am in the strange position of being offered to make a D-A with an improved discrete I-V designed by me, but I think more is necessary to even bother to make such a product. That is where all this secondary stuff comes in. I already have an OPPO105, and it is marginal. Just replacing the IC's in the I-V is not going to give me audio nirvana, I am pretty sure.
From what I have been told by others, both Shiite and MSB have done significant research on how to improve digital playback better than the Sabre products, including oscillators (MSB). I just don't have the background to improve over Sabre at this time, so I am not going forward at this time.
From what I have been told by others, both Shiite and MSB have done significant research on how to improve digital playback better than the Sabre products, including oscillators (MSB). I just don't have the background to improve over Sabre at this time, so I am not going forward at this time.
"Oscillators Are More Analog Than Digital. ... While they may use logic circuits to transform sinusoidal oscillations into symmetrical square waves, the crystal oscillator is a quintessential analog device. The crystal and its electrical excitation circuit form a kind of miniaturized, mechanical servo loop."
Oscillators Are More Analog Than Digital | Electronic Design
Bypass the ESS ASR for best sound by reclocking the I2S signal at input & using the same clock on the DAC - this has been mentioned before
Oscillators Are More Analog Than Digital | Electronic Design
Bypass the ESS ASR for best sound by reclocking the I2S signal at input & using the same clock on the DAC - this has been mentioned before
You are not working with what I said. that is an entirely different answer to some other issues. Of course they solved or reduced many issues. But at this point......
What part of DAC ... analog or digital ... is the most audible ? It was BM designers opinion that the analog portion was the most variable and audible.
THx-Richard
Last edited:
I have done some of that research too. I think I may have just solved the 2nd to the last problem. Won't know for sure for another day or two, but jitter and low level detail may now be a little better than DAC-3. I could fill you in less than half an hour. In fact, Jam, the guy the designed the popular headphone amp for Pass Labs was over this afternoon and we talked dacs for awhile. I showed him what I have been working on. He invited me over to see what he's working on next. Of course, he is still in this to make a living. I am still thoroughly retired and staying that way. Dacs are just something I started on working on to help diy'ers build better dacs for themselves.
Many people find that clocking and power supplies have dramatically audible effects. I think they do. I have an opamp output stage probably pretty similar to what is in DAC-3. Not that much to it. It works like it should so long as power quality and clocking related issues are right. All, IMHO, of course. Others are quite free to differ.
I agree and BM also agree with you. No doubt it is audible and important to reduce.
-Richard
Last edited:
You are right, I got the top OPPO player model and it was a big disappointment, sonically.... even with typical mods.
JC, make a better one and I'll buy it as my last legacy CD player.
-Richard
Last edited:
It has been too long John, but still many people will be very happy once a "John Curl approved" design for digital audio becomes available.
- Status
- Not open for further replies.