|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
208.102.172.211
In Reply to: RE: Is it possible that high system latency is the culprit - usually the main case of dropouts? posted by carcass93 on February 16, 2015 at 06:02:17
Guys,
Look I have a ton of computers here. Isosynchronous frames have the highest priority on the USB buss.
Lets remember how async works first. The computer sends a stream of samples to the DAC. The DAC is setup to send a feedback pipe back to the computer to indicate if it needs more, less or the sample stream is fine. The USB "IN" for the feedback pipe is a select protocol. The computer sends the IN and the DAC sends the feedback frame. The feedback frame is isosynchronous and therefore does not have error correction (big problem).
The number one problem with asynchronous USB is usually one of the following problems:
1) Cable is poor and the computer sends the IN and the overhang of the cable effects the feedback pipe sent from the DAC to the computer and therefore the computer sees this as an error and discards it. This actually happens allot. Especially with longer cables.
2) The software in the DAC does not compensate for the feedback correctly and the buffer is effected by this and an overrun or underrun occurs and pops and clicks.
3) Poor USB on the computer side, there are many devices off this port including the DAC and the computer device interface effects the performance of the product. Just move the damn cable and it will probably sound better.
When you have a number of oscillators or PLL in the DAC circuit #2 has to compensate for the this. I have designed some software for companies (not mine) that do this as they have several (more than 2) or have specialized oscillators for some other aspect of the product. In all cases of these designs we were able to tweak the software in the feedback to optimize the product.
Hey the good thing is most of these have upgradable software so it can be fixed.
Thanks,
Gordon
J. Gordon Rankin
Follow Ups:
Never the only problem is much more correct.
You have just effectively demolished what Tony Lauck says about usb audio here and elsewhere.
You have also acknowledged that it is the hardware/software interaction in a specific system that is at work, and not the simple block diagram that is used to describe usb audio operation to the masses.
Fred,
Totally incorrect, I am in total agreement with Tony.
So you want to draw pictures to help everyone out. I think that I brought the procedure to the lowest level of understanding. If you want more just ask for it.
Thanks,
G.
J. Gordon Rankin
What do you agree with and what do you disagree with what I said about TL's pronouncements on the Auralic Vega's tight acceptance window on the ESS input?
Once again I respectfully ask you not to send emails directly to me on your posts in this forum. As far as I am concerned, you do not have the right to transfer the AA email system to that of your own, and I do not wish to have my inbox used for this purpose.
My setting in AA has been not to receive notifications on posts, and this is respected with you as the only exception.
Fred,
I am not sending emails to you in any form. I don't even know what your email is.
Look I agree with TL on everything he says here. The ESS is capable of almost any input as long as it abides to the rules. So for ASYNC input to I2S that means 192x and for SPDIF that means 384, for sync it's less but who cares it's all close enough.
Also I agree with TL that all USB Asynchronous feedbacks should adjust to what ever the input frequency selected is. This is just the way it's suppose to work.
Fred, your an engineer, you should know this stuff by now. Why are you confusing the subject. There is enough missinformation here in this forum to drive people crazy. Don't add to the static.
Thanks,
Gordon
J. Gordon Rankin
I have just received this email from your address on your latest post in response to me below:
''From waudio@cinti.net
Subject New AA post Computer ******
To my email address''
Who sent it?
So, why do I receive from you alone notifications bearing the address
waudio@cinti.net?
I get no other notifications from AA posters but I get this often from presumably you.
What you said about the ESS was not how TL described Auralic's implementation, and this in no way contradicts my point about the lack of information on how the Exact Mode on the Auralic had been implemented.
I assume that you. as a known designer, would not want your work to be classed as defective by someone who has no knowledge of how you'd done it. As an engineer I know so and I would only criticise on the basis of first hand information and intimate knowledge.
As a vendor. I would also assume that, if you state in your product specification, that some feature may not work reliably with some systems, then this would be accepted by reasonable punters, commentaters, and competitors.
This is quite encouraging and easily understood. Unfortunately, the company in question must first ADMIT that there is a problem. I will not be holding my breath.
Good luck. You will need it. Depending on how the product was designed and assembled it is entirely possible that portions came from subcontractors who have their own reasons for keeping quiet and the manufacturer may not even know what is going on with the firmware in sufficient detail to be able to fix it. In addition, even if there is one engineer who has responsibility for the entire product design he may be out of his depth and fail to understand the issues. Or he may fully understand them but is being told to keep quiet.
Your best hope for getting the product fixed even if it is a SMOP ("small matter of programming") is to find enough other customers who have similar problems and organize some kind of joint action. As you say, don't hold your breath.
It could be worse...At least you didn't buy equipment for multiple thousands of dollars from a company that took your money, never delivered the product and is now in bankruptcy with (maybe) enough assets left to pay the lawyers.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It is what it is. An absolutely remarkable sounding DAC that fails to operate optimally at random and annoying intervals. It is Chinese in origin and is ulikely to receive any more design attention than it already has.
However, this to shall pass. As the level of DAC sonic quality grows (rapidly), then it is only a matter of time, hopefully not much of it, before a sonic match for the Auralic Vega DAC appears.
Until then, well, what is really so bad about missing a half second of the music here and there?
LA
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: