DCAudioDIY.com

DC Area Audio DIYer's Community
It is currently March 29th, 2024, 5:58 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 182 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15, 16, 17 ... 19  Next
Author Message
PostPosted: August 1st, 2017, 2:35 pm 
Offline

Joined: July 17th, 2016, 6:24 am
Posts: 1099
tomp wrote:

The second part of #2 is not correct. If the packet is not received at the receiving end in a USB transmission the sender is asked to re-transmit. See below from Wikipedia:

Tom


That's true for "standard" USB communications. Not for audio streaming services, there are no ACKs and NAKs for streams.


Top
 Profile  
 
PostPosted: August 1st, 2017, 2:37 pm 
Offline

Joined: July 17th, 2016, 6:24 am
Posts: 1099
Roscoe Primrose wrote:
Cogito wrote:
Even if the ground is common with the PC, it really doesn't matter in the digital domain for deciphering 0s and 1s and the there is a huge voltage range to represent 0s and 1s, the switchover point is far away from both the ground voltage (0v) and power supply voltage (+5v).

It only matters if and only if the xmos ground is connected to the DAC ground.


You've never looked at a PC ground on an o-scope, have you ;)


No I have not, how much voltage variation due to noise?


Top
 Profile  
 
PostPosted: August 1st, 2017, 2:37 pm 
Offline

Joined: June 4th, 2013, 2:39 pm
Posts: 487
Cogito wrote:
Jim G wrote:
If you're willing to load MinimServer on your NAS and BubbleUpnp on a tablet, I'll bring (or get to you if I can't make it) a Sonore renderer to compare at the meet.


Excellent. Let me do some research about how to load minimserver on QNAP NAS. My DAC has only USB interface, is there a way to use a windows PC as a uPNP device?


Oh sure. I think it's even built in. Google it, it shouldn't be hard.

ETA: You could even use your AP-Linux rig. Just install upmpdcli and run it. Go to his website and he explains how to get the package.


Top
 Profile  
 
PostPosted: August 1st, 2017, 2:52 pm 
Offline
User avatar

Joined: February 19th, 2017, 9:43 am
Posts: 530
tomp wrote:
Cogito wrote:
Right Roscoe.

I am trying to make a couple of points to Dave.

1. Network bandwidth is not like a faucet. There is no continuous flow of bits from one end to the other . It's all packets with acknowledgment of each packet, re transmissions etc . Granted if the packet doesn't arrive at the DAC in time or is corrupted, DAC simply drops the packet and moves on.

2. What packets are sent over the USB and when is determined by the time division algorithms of the OS. Hence the talk about deterministic OSs.


The second part of #2 is not correct. If the packet is not received at the receiving end in a USB transmission the sender is asked to re-transmit. See below from Wikipedia:

If a USB host does not receive a response (such as an ACK) for data it has transmitted, it does not know if the data was received or not; the data might have been lost in transit, or it might have been received but the handshake response was lost.
To solve this problem, the device keeps track of the type of DATAx packet it last accepted. If it receives another DATAx packet of the same type, it is acknowledged but ignored as a duplicate. Only a DATAx packet of the opposite type is actually received.
If the data is corrupted while transmitted or received, the CRC check fails. When this happens, the receiver does not generate an ACK, which makes the sender resend the packet


If there is only a small problem with the transmission the re-send will usually happen with enough time to prevent a dropout. If the problem is severe enough, the transmission will fail. Remember that USB is asynchronous so you don't have the problems with jitter like you do with SPDIF or AES/EBU that are synchronous. Instead of jitter noise you will have droupouts that if severe enough and repeated can sound like static.

Tom

USB audio spec uses isochronous transfers, see this excerpt of an XMOS paper:

Isochronous transfers are used to transfer data in real-time between host and device. When an isochronous endpoint is set up by the host, the host allocates a specific amount of bandwidth to the isochronous endpoint, and it regularly performs an IN- or OUT-transfer on that endpoint. For example, the host may OUT 1 KByte of data every 125µs to the device. Since a fixed and limited amount of bandwidth has been allocated, there is no time to resend data if anything goes wrong. The data has a CRC as normal, but if the receiving side detects an error there is no resend mechanism.

The whole paper can be seen here:

http://www.xmos.com/news/articles/17888

_________________
I have too much stuff - https://www.pleasebuymystuff.com


Top
 Profile  
 
PostPosted: August 1st, 2017, 2:54 pm 
Offline
User avatar

Joined: February 19th, 2017, 9:43 am
Posts: 530
Cogito wrote:
Roscoe Primrose wrote:

Yeah, actually, he did. Noise can propagate through the common ground. Noise on the USB ground of the PC ends up in the USB ground of the DAC even if the DAC is not USB powered. The ground still needs to be connected for the signal to have a ground reference. Schiidt just came out with an updated USB board for most of their DACs that uses transformer isolation on the USB inputs to help deal with this problem, but this solution isn't widely used at this point...

Roscoe


I am not sure that is correct in all instances.

For example the xmos I am using is electrically isolated from USB. It is powered by DAC including the ground reference.

Even if the ground is common with the PC, it really doesn't matter in the digital domain for deciphering 0s and 1s and the there is a huge voltage range to represent 0s and 1s, the switchover point is far away from both the ground voltage (0v) and power supply voltage (+5v).

It only matters if and only if the xmos ground is connected to the DAC ground.


You are 100% correct in that the noise will not affect decoding ones and zeros. However, noise can very well affect the signal after it's converted back to analog.

_________________
I have too much stuff - https://www.pleasebuymystuff.com


Top
 Profile  
 
PostPosted: August 1st, 2017, 2:58 pm 
Offline
Site Admin
User avatar

Joined: February 28th, 2013, 10:38 am
Posts: 1682
Cogito wrote:
No I have not, how much voltage variation due to noise?


It can actually be well above the +5v supply voltage. Now, you're thinking it can't possibly be that high or the PC wouldn't work at all ;) It's nearly 100% common mode noise, so the same noise appears on the ground & +5v the same time. However, when you connect multiple devices together, noise from the ground on one device can get into the ground on the second device w/o the matching noise on the second device's +5v supply.... This is generally RF noise, and most audio designers aren't the best RF engineers in the world ;)

Same thing can happen to a lesser extent on an SP/DIF connection. Too bad they chose such a poor optical standard for Toslink, optical cables don't have ground connections...

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.


Top
 Profile  
 
PostPosted: August 1st, 2017, 3:36 pm 
Offline

Joined: July 17th, 2016, 6:24 am
Posts: 1099
DaveR wrote:

You are 100% correct in that the noise will not affect decoding ones and zeros. However, noise can very well affect the signal after it's converted back to analog.


That too iff DAC's analog section ground is connected to USB ground, which I presume doesn't happen is modern day quality DACs.

Thank You.

Which brings us back to the question I raised yesterday about the sonic benefits of providing clean, dedicated linear power supply to SOtM card.


Top
 Profile  
 
PostPosted: August 1st, 2017, 9:01 pm 
Offline

Joined: March 1st, 2013, 11:12 am
Posts: 42
Jim told me about this conversation so I'll chip in a bit as I haven't been following it. Currently I'm using a Aqua La Scala DAC being fed by a CAPS Server computer (Atom processor with SoTM USB output card with SATA filter and other assorted goodies - check out the CAPS Server stuff on computer audiophile). Jim gave me a huge step forward when he told me about using Minimserver on a NAS. I orginally used the Sonore as a renderer with Minimserver and Bubble UP as the DNLA program. Then I noticed that using a laptop with linux running MPD with a USB output set up for bit perfect playback sounded better than the Sonore with it's SPDIF output. That was annoying. Then I found that different computers running linux and MPD (set up with mpd.conf for bit perfect output - check out the web for instructions - you must use the ALSA drivers with output directly to the hardware) sounded different. To say that that was more annoying is an understatement. Then Jim and I compared a linear power supply to a switching supply and were stunned by the improvement. WHAT..........???? I have no idea why this is the case but the LPS was an amazing improvement. Especially with a separate DC supply for the USB card. At this time I'm using a CAPS server Carbon type computer with linux and a SoTM USB output card with a separate linear supply with a LPS for the computer using linux to work MPD with bit perfect output to the La Scala DAC. For the first time in my life I think that analog has lost. I never would have believed it could happen. If anybody is interested they can stop by for a listen after I get back from Grover's (I'm heading out there to see him next week). I don't have to much space but one or two at a time could be accommodated. I haven't listened to analog in months -- hi res digital is more than I could ever have imagined.

Charlie.


Top
 Profile  
 
PostPosted: August 1st, 2017, 9:38 pm 
Offline

Joined: July 17th, 2016, 6:24 am
Posts: 1099
Thanks Charlie for the input.

I would love to listen to yours and Jim's systems when it is convienent.


Top
 Profile  
 
PostPosted: August 1st, 2017, 10:37 pm 
Offline

Joined: July 8th, 2016, 4:34 pm
Posts: 570
This website might be of interest. It has a linear pc power supply among other things of interest -

http://www.teradak.com/products/55.html


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 182 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15, 16, 17 ... 19  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 56 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group