"re-clocking" the Ultra digital out ?

[onlyoneme]

The clocking on the Ultra is very stable, as Amir's review on ASR shows us. Not absolutely top of the class, but better than most other devices he's measured.

Where did you find it, exactly?

 

[anonyless]

In my experience... you just shouldn't be using or owning equipment which has an unstable clock in the first place. It's a sign of poor engineering.

The clocking on the Ultra is very stable, as Amir's review on ASR shows us. Not absolutely top of the class, but better than most other devices he's measured.

could it be an issue with the Ultra's USB output only? i never understood that part in cheapaudioman's video . I plan to first hook up the Ultra to my amp, listen for a couple of days and only then try with external DAC (SU1 which I already own)

 

[hgo58]

Talking about re-clocking only make sense for the S/PDIF interfaces.

The USB connection is asynchronous and don't include any audio clock. The receiving device always have to provide its own clock.

 

[EddNog]

Where did you find it, exactly?

-Ed

 

[onlyoneme]

View attachment 21961

-Ed

Just s commonly seen j- test plot. What's spectacular on it?

 

[EddNog]

Just s commonly seen j- test plot. What's spectacular on it?

I believe @574stereo 's point is that the jitter is low enough to be transparent and not an audible level of concern.

-Ed

 

[onlyoneme]

Yet another "transparent, non-audible" aspekt of the sound which reportedly becomes audible, or unsettled when considered in the time domain.

 

[Bobcel]

Have a SMSL PO 100 from my Ultra USB out. Sounds good, not sure if it's better or not, BUT, The SMSL has active outputs so I can run it via Coax and Optical simultaneously. Like that

 

[dtc]

Talking about re-clocking only make sense for the S/PDIF interfaces.

The USB connection is asynchronous and don't include any audio clock. The receiving device always have to provide its own clock.

Just a nit - but the USB is only asynchronous if the device on the other end supports asynchronous. Most modern DACs do support that but it is not clear if the converter here does.

 

[hgo58]

Just a nit - but the USB is only asynchronous if the device on the other end support asynchronous. Most modern DACs do support that but it is not clear if the converter here does.

USB is always a asynchronous package transfer (frames). There are different modes used for audio data (Isochronous, Interrupt, and Control) but it will never include an audio sampling clock.

For real-time audio the Isochronous transfer is used for the data (no error retry).

 

[dtc]

USB is always a asynchronous package transfer (frames). There are different modes used for audio data (Isochronous, Interrupt, and Control) but it will never include an audio sampling clock.

For real-time audio the Isochronous transfer is used for the data (no error retry).

In computer audio terms, Aynchronous USB means that the DAC (or receiving device) controls the timing. Not all DACs can do that. In those cases, the Ultra uses synchronous transfer, which means it controls the timing. Wiim has confirmed that the Ultra uses synchronous or asynchronous transfer depending on the capability of the receiving device.

This concept was important in the early days of computer audio when the computer controlled the timing and was not very good at it. Gordon Rankin of PS Audio implemented asynchronous USB where the DAC controlled the timing. That is how most DACs work these days.

 

[onlyoneme]

USB is always a asynchronous package transfer (frames). There are different modes used for audio data (Isochronous, Interrupt, and Control) but it will never include an audio sampling clock.

For real-time audio the Isochronous transfer is used for the data (no error retry).

In the USB audio only isochronous transfer is used for audio data. For this transfer 3 modes are available - asynchronous, synchronous and adaptive.

 

[adias]

Yes Ultra USB out into a DAC with asynchronous filtered/isolated USB is superior.

 

[dtc]

In the USB audio only isochronous transfer is used for audio data. For this transfer 3 modes are available - asynchronous, synchronous and adaptive.

Thanks for confirming what I said.

 

[hgo58]

In the USB audio only isochronous transfer is used for audio data. For this transfer 3 modes are available - asynchronous, synchronous and adaptive.

Correct when we talk about how audio data transfer is scheduled but this thread is about re-clocking and compared to S/PDIF, USB is asynchronous and the receiving device always definers the sampling clock.

 

[onlyoneme]

Correct when we talk about how audio data transfer is scheduled but this thread is about re-clocking and compared to S/PDIF, USB is asynchronous and the receiving device always definers the sampling clock.

It doesn't matter. Compared to spdif both are isochronous. And both are "clocked" by the host if usb synchronous mode is used. It becomes different when asynchronous mode is used instead - transfer pace is controlled by the receiver.

 

[hgo58]

It doesn't matter. Compared to spdif both are isochronous. And both are "clocked" by the host if usb synchronous mode is used. It becomes different when asynchronous mode is used instead - transfer pace is controlled by the receiver.

I don't agree. You are mixing up sample rate clocking (that can cause the jitter we want to minimize) and usb data frame transfer.

 

[MattWnAZ]

I tested this out by going USB (fixed volume) out to my Emotiva BasX PT1 which has an Analog Devices AD1955 DAC chip. I also had it connected to the PT1 via toslink, coax, and analog. I did a lot of A/B/D/C listening comparisons. Honestly the differences weren't really discernable. However I landed on the analog out as that that was the most pleasing to my ears. I've since ditched the PT1 and am now just going analog out (variable volume) straight to my amp. (Emotiva BasX A2). It's just a cleaner setup for me and sounds fantastic. I know this whole discussion is just a lot of hair splitting. But we all love to tinker , so tinker away and do what sounds best to your ears.