Yes, I am a bit Francophone. Living in Auvergne, le Midi et Paris will do that to a guy. 
(mon fils est Francais).
I have seen some DACs built with buffers, but never actually heard one - that I know of. It seems like a good idea. I did design and build from scratch a DAC based on an ESS chip. It is supposed to be immune to jitter. I swear it sounds smoother and cleaner than other DACs I've used. But is that no jitter or something else? I don't know.
(mon fils est Francais).I have seen some DACs built with buffers, but never actually heard one - that I know of. It seems like a good idea. I did design and build from scratch a DAC based on an ESS chip. It is supposed to be immune to jitter. I swear it sounds smoother and cleaner than other DACs I've used. But is that no jitter or something else? I don't know.
I exaggerated by saying everybody to make the point but if we ignore the CDs with errors we are left with the premise that some CDs are "easier" to read than others - the easier to read ones sound better i.e less PS perturbations?
Sure, it's a great system and works well for data. But, as you said, timing is important when dealing with audio data - what if there is something in "the machine" effecting this timing? I know the timing doesn't come into play until the DAC stage but the PS noise inside a computer can still be transmitted to the DAC via the connection to the DAC
Yeah, I'll go with that. There might be other things going on too. The PS noise ought to be easy enough to see on an O'scope. Has anyone ever compared this?
Sure. And if the HDD is creating noise throughout the power supply rails, that could get into the analog section of the DAC. It might even come down the data buss and get into the analog sections.
Oh, I am pretty convinced that a substantial part of the large number of FIYaudio members never touch a soldering iron or whatever tool to do some DIY.
But that is not the point, or? Every forum has its share of people who raise their voice just for the sake of it, but there are also people who are very well informed and can contribute in a positive way. I don't see a problem in this.
Ah, I see now - you're raising your old Boogaboo argument.
Is it not possible that people can actually hear the effects of jitter (which is what Pano was talking about - timing & which is demonstrated to exist). Is it potentially possible that there is something going on in PC audio not yet understood that explains these reports of sonic differences?
Last edited:
If you're talking about jitter at the DAC chip itself, then audible differences are certainly possible.
Timing IS crucial in exactly two places. The sampling clock during the A-D conversion is one. The other obviously is the data supplied to the D-A converter. In all other places it doesn't mean squat. Data starts and stops but gets stored in buffers. The buffer may only be one time period or as million but as long as the next is available ON TIME is makes no difference if it's a microsecond or a millisecond and no data is added or deleted. This is not at all a difficult thing to do. In video it's called a time base corrector and was first done analog in the '60s and digital starting around '76. Nothing new here.
G²
You Sir, take full marks for being obnoxious and ignorant.
It's sarcastic people like you that stop other people from getting involved further, or speaking their mind.
It may surprise you to know that JK is among the many people worldwide who have gone on record in other forums, where such insulting language is not condoned, to having heard the differences between uploaded .wav files with identical checksums. I have already posted a report by the highly qualified lead designer of XXHE software, but unfortunately that post was culled recently.
I fail to see what was wrong about that particular post.
These other DIYAudio members just do not wish to put up with your kind of personal attacks, so rarely post here.
Sorry John, for dumping you further into this, but I know you will understand, as you continually receive sarcastic replies yourself on the subject of your HiFace LiPo mods, and findings about SPDIF attenuators, DESPITE all the enthusiastic positive reports worldwide.
SandyK
P.S.
Moderators can you please do something to discourage such intimidating behaviour towards other members ?
These days, only the more brave members such as Terry O, JK and a few others are game to post subjective findings for fear of being ridiculed.
If you don't want technical discussion, perhaps it's best to not make technical statements on a technical forum that does NOT censor technical discussion and is frequented by quite a few technically competent people. The forum does not guarantee affirmation.
Personal attacks have no place here.
Sy
Technical discussion is one thing, but technical competence is no guarantee that a person is necessarily correct on every pronouncement they make.
Science is not stationary, and new findings and developments are taking place all the time, often overturning previously accepted norms.
You will also be aware that I have previously posted links to technical articles elsewhere that did not necessarily fit in with personal views of some members here, on the subject of the part analogue plays in digital playback.
You would also be well aware that the vast majority of members do not have access to the equipment needed to substantiate their reports, and regularly demanded by the "technically competent" members here, who more often than not do not have access to the same specialised equipment either.
SandyK
Yes I can Michael, but I am almost 72 and getting tired of the continual necessity to post similar replies over and over.
Rather than just make such claims, I have also been making regular comparison .wav files available in a couple of forums for around 2 years now, even sending a compilation CD to a member of the moderating team.
However, back then, those differences that many heard, were not so obvious . You and a couple of the more regular posters are among the few that were unable to hear these differences.Several even refused to listen after seeing that the files appeared identical using conventional PC analysis. As was stated here by another member recently, the listening equipment used by 90% of people may not be revealing enough. Mine HAS to be revealing in order to compensate a little for age.
Alex
A true analog-like level of performance can be achieved even with Red Book CD audio format. To achieve this there are some details to pay attention to:
- use ripped material (so exclude real time reading of the CD)
- use a DAC fed by data from memory buffers
- fill the buffers over the medium that will not interfere with audio (I prefer wired LAN)
- use a microcontroller that will take care of filling the buffers properly
- use separate linear power supplies for microcontroller, DAC, clock and analog parts
- minimize the jitter as far as it gets
I have just described a networked music client.
It may be that different media used over S/PDIF can sound differently but in that case I would tend to credit the S/PDIF or the operating system used for the differences more than the media used for storing the data. I have tried same material on a network storage, with USB (both stick and hard disk) on a client designed as above without any audible differences. It is an embedded Linux machine (ARM9 based).
- use ripped material (so exclude real time reading of the CD)
- use a DAC fed by data from memory buffers
- fill the buffers over the medium that will not interfere with audio (I prefer wired LAN)
- use a microcontroller that will take care of filling the buffers properly
- use separate linear power supplies for microcontroller, DAC, clock and analog parts
- minimize the jitter as far as it gets
I have just described a networked music client.
It may be that different media used over S/PDIF can sound differently but in that case I would tend to credit the S/PDIF or the operating system used for the differences more than the media used for storing the data. I have tried same material on a network storage, with USB (both stick and hard disk) on a client designed as above without any audible differences. It is an embedded Linux machine (ARM9 based).
The heart of the problem is that you have proposed no mechanism but which these files could be different. They just sound different. But by all anyone who understands computer files knows, there is NO difference in the files. None. Not even 1 part in many millions. So how CAN they sound different? They contain exactly the same bits - in the same order.
Can you see this side of the argument? It's the very crux of the thing.
If you have, or can point us to, something that would indicate why identical checksum files would be not identical, please do so. Is there any theory about where the differences may lie? You seem to think that the checksum is missing something. What could that something be?
Can you see this side of the argument? It's the very crux of the thing.
If you have, or can point us to, something that would indicate why identical checksum files would be not identical, please do so. Is there any theory about where the differences may lie? You seem to think that the checksum is missing something. What could that something be?
Michael
I previously suggested elsewhere that janneman, as a respected technical author could ask the Sony engineers who came up with the BluSpec format,
the reasons why they made such claims, and provided comparison sets.
They must have done quite a bit of research in this area before commercialising a new format.
It is not the answer you are asking for, but to solve a little piece of the puzzle at a time could be helpful ?
Alex
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.