[zoophaugus]

I'll ask my friend to bring the matrix with his MSB dac next week - i'm curious as well

That would not be a possible comparison- The matrix does not have an Spdif input for reclocking etc.. Its USB input only..
I need feedback from some one using DI-20's Spdif input and or comparing that input with its the USB input . Is one input stronger than the other feeding same bit rate song ? When I say stronger I meaner better or worse..

 

[Wynnytsky]

That would not be a possible comparison

1. win10=>matrix=>msb coax
2. win10=>matrix=>di20he coax=>msb coax
3. win10=>di20he=>msb coax

wouldn't you want this ranked?

 

[zoophaugus]

1. win10=>matrix=>msb coax
2. win10=>matrix=>di20he coax=>msb coax
3. win10=>di20he=>msb coax

wouldn't you want this ranked?

It would be a great comparison for sure but my intention is to not use windows 10 or usb as a source... I want to use a streamer with a digital coax out as source...This is what the matrix is not able to do as it is Usb input only.
The instructions are some what vague for DI-20. Does the Spdif input go through the same reclocking and regenerating process as the Usb input does??

 

[FredA]

It would be a great comparison for sure but my intention is to not use windows 10 or usb as a source... I want to use a streamer with a digital coax out as source...This is what the matrix is not able to do as it is Usb input only.
The instructions are some what vague for DI-20. Does the Spdif input go through the same reclocking and regenerating process as the Usb input does??

Yes and no.

The process is not the exact same, the output signal only needs to be clocked with the usb in, because with asynch usb, it is possible to avoid reclocking. So this means lower jitter at the output of the di20 for usb. For this to be possible, the amanero usb board takes its timing from the fpga, therefore the same clock drives the usb module and the output signals. This is the best possible approach IMO for a usb to spdif/i2s converter.

With the spdif input, the signal absolutely needs reclocking. So in theory, usb will sound better. But sound quality still depends on the incoming signal quality, and i can testify that my usbridge sound better then a rpi4b+.

I did an a/b between the output of my oppo dvd drive directly to my r7he, and through the di20he. The sound quality improvement is major through the di20he. It makes home cinema (2ch only) much more enjoyable.

 

[Jackula]

I never even tried Queen's switching supply. I have the Morion on HDPlex's 2amp low noise tap and the uptone ether regen on HDPlex's 8amp tap (replacing the 4.8amp switching unit it came with). It's the swiss army knife of LPSs.

Your advice got me curious and I bought the new HDPlex 300WLPS hoping it would let me consolidate my power supplies and replace my LPS-1.2. It certainly didn't replace my LPS-1.2, but it did elevate it to another level.

There's hoping it just needs more burn in.

 

[Wynnytsky]

Your advice got me curious and I bought the new HDPlex 300WLPS hoping it would let me consolidate my power supplies and replace my LPS-1.2. It certainly didn't replace my LPS-1.2, but it did elevate it to another level.

There's hoping it just needs more burn in.

My PC on the HDPlex (8amp 19v rcore) makes the sound relaxed and effortless, and pushes the performance further back (more distant). If you have a fatiguing element, this will mitigate that issue like dark wires would.

My PC on the Keces P8 (4amp 19v torroid) makes the HDPlex sound reticent in the treble. The sound steps forward, and the edge of notes have better delineation. Among 4 systems I tried, the Keces edged out the HDPlex on 3 of them (the 4th runs six 500watt class D amps). Two of my friends tried the P3 before the P8 and the advice was to skip the P3. I wonder what the single-tap 8amp P8 sounds like.

If your CPU's TDP is 25watt or less then I hear good things about the SBooster, which delivers just under 2amps. My 35watt TDP computer operates at 0.5amps (0.3amps when Fidelizer underclocks it to 1ghz). I've been burnt before trading in headroom for better engineering so I'll put that on my wish list of stuff I'd like to be loaned.

 

[DACLadder]

@DACLadder
From that, can I say differential is a specialization of balanced? (ie: differential is balanced, but not the other way around)
Thus far when CMR is used in digital transmission I'm always seeing the differential method.

XLR, ACSS, USB, LVDS, wired Ethernet are all examples of balanced differential systems. I always associate differential as balanced. CMR would not work unbalanced. A good example is old style analog telephones. Balanced system and works for kilometers. Until bad insulation, water, or shady connection unbalances the line and results in hum, noise, and crosstalk.

And my preference for LVDS I2S is limited to AGD DACs. Other DACs have different technology.

 

[Jackula]

My PC on the HDPlex (8amp 19v rcore) makes the sound relaxed and effortless, and pushes the performance further back (more distant). If you have a fatiguing element, this will mitigate that issue like dark wires would.

My PC on the Keces P8 (4amp 19v torroid) makes the HDPlex sound reticent in the treble. The sound steps forward, and the edge of notes have better delineation. Among 4 systems I tried, the Keces edged out the HDPlex on 3 of them (the 4th runs six 500watt class D amps). Two of my friends tried the P3 before the P8 and the advice was to skip the P3. I wonder what the single-tap 8amp P8 sounds like.

If your CPU's TDP is 25watt or less then I hear good things about the SBooster, which delivers just under 2amps. My 35watt TDP computer operates at 0.5amps (0.3amps when Fidelizer underclocks it to 1ghz). I've been burnt before trading in headroom for better engineering so I'll put that on my wish list of stuff I'd like to be loaned.

The newer 300W version supposed to be a sonic upgrade over the older 200W, my research tells me the 300W is better than SBooster. However it looks like I won't be able to burn in this thing to see what it's really capable of. It died in about 4 hours and also fried my connected devices. Perhaps it's simply bad luck but at this stage I'd rather a refund than a replacement. Hopefully I'll get a return label soon.

 

[Wynnytsky]

The newer 300W version supposed to be a sonic upgrade over the older 200W, my research tells me the 300W is better than SBooster. However it looks like I won't be able to burn in this thing to see what it's really capable of. It died in about 4 hours and also fried my connected devices. Perhaps it's simply bad luck but at this stage I'd rather a refund than a replacement. Hopefully I'll get a return label soon.

jesus that's a horror story -- I'd need therapy if mine did that.

Good thing about Larry is he responds immediately. I've never known him to be chatty or conversational, but once I had issues with a PC build and he loaned/mailed me a CPU that I could use to flash my mobo to the latest firmware that would work with my CPU, then I mailed it back when I was done.

 

[sajunky]

XLR, ACSS, USB, LVDS, wired Ethernet are all examples of balanced differential systems. I always associate differential as balanced. CMR would not work unbalanced. [...]

And my preference for LVDS I2S is limited to AGD DACs. Other DACs have different technology.

Referal to differential as balanced is wrong. True balanced connection require symmetric power supply (or transformers). Current flow is between V+ and V- power rails and a ground connection serve only as a voltage reference, do not carry any load. The last element is critical for low CMRR. While similar scheme can be deployed for digital transmission (esp. on long distances), different parameters are important: compliance to the logical levels for '0' and '1', a shortest possible transition period and absence of signal reflections. Differential implementation is typically limited to a single power source and two gates, while one pulling up to the Vcc the other one pulling down to the ground and in reverse. Still offering sufficient CMRR reduction and also gives immunity against a major problem that do not exists in analog transmission. It is called "ground bounce", where high and low logic levels are associated with different impedance and in result generate current peaks when switching levels. Not a good things on a digital transmission line.

I strongly suggest to use correct terminology: 'balanced' for trasmission of analog signals and 'differential' for digital signals.

I2S connection between audio gear is not yet standarised. It is good that Audio GD decided for LVDS ( TIA/EIA-644) signaling standard. It can be used both on HDMI and RJ45 connectors.

 

[DACLadder]

@sajunky Nope, I stand by what I said. Digital signals are in fact analog on the line. Signals are propagated in the analog domain. What makes them digital is just encoding/ decoding. ‘Balanced’ for a differential line or pair just implies two wires of the same type with equal impedance along their lengths and equal impedance to ground and other circuits. You don’t necessarily need a tranformer for common mode rejection and depends on the application. XLR analog inputs today (99.9%) do not use transformers and have very high CMRR.

 

[marcelnl]

Did nobody yet take the soldering iron to the Di20?

 

[Wynnytsky]

XLR analog inputs today (99.9%) do not use transformers and have very high CMRR.

I know a preamp can have an XLR output and not be true balanced end-to-end (an opamp splitter can be used to convert single ended to balanced)
Given that both HDMI and USB terminals are differential, shouldn't that mean it's possible for a DAC to be balanced from end to end? Among the AGD dacs with XLR outs, does anyone know if they are true balanced from the digital-ins to the analog-outs?

 

[sajunky]

I know a preamp can have an XLR output and not be true balanced end-to-end (an opamp splitter can be used to convert single ended to balanced)
Given that both HDMI and USB terminals are differential, shouldn't that mean it's possible for a DAC to be balanced from end to end? Among the AGD dacs with XLR outs, does anyone know if they are true balanced from the digital-ins to the analog-outs?

Once again, digital differential are not balanced, but analog balanced are always differential. So, if you rephrase your question and replace word blanced with 'differential', the answer is yes.

As for Audio GD, I am sure that everything is right. Balanced are true balanced.

 

[FredA]

Once again, digital differential are not balanced, but analog balanced are always differential. So, if you rephrase your question and replace word blanced with 'differential', the answer is yes.

As for Audio GD, I am sure that everything is right. Balanced are true balanced.

There many possible definitions for the term differential. For me, it originates from the notion of differential input. Because the term implies doing a subtraction. When a single-ended circuit handles a balanced signal, it is best done using a differential input.

Semantic debates are a waste of time. As far as how things are done with the hdmi i2s signal by audio-gd, it becomes single-ended as soon as it exits the hdmi module. Cause i2s is not balanced internally (or differential if you prefer)..

The analog section however is 100% balanced or very close with the Master series gears.