[erin]

Quote:

oh....well AD don't seem too willing to tell us about what happens in ADUM4160, its public datasheet is as vague as can get...and yes, it's a major god bless for USB audio.
 

 
http://www.analog.com/static/imported-files/data_sheets/ADuM4160.pdf
 
Leeperry, the datasheet looked quite comprehensive to me. what part did you find vague?

 

[leeperry]

 

 
http://www.analog.com/static/imported-files/data_sheets/ADuM4160.pdf
 
Leeperry, the datasheet looked quite comprehensive to me. what part did you find vague?

The USB reclocking for instance, coz it's clearly occuring and wouldn't work otherwise(as they filter the data flow to death)...they only say "the signal must be faithfully reconstructed on the output side of the coupler while retaining precise timing".

 

[erin]

Quote:

 

The USB reclocking for instance, coz it's clearly occuring and wouldn't work otherwise(as they filter the data flow to death)...they only say "the signal must be faithfully reconstructed on the output side of the coupler while retaining precise timing".

Ok, well they are not going to give away all their secrets are they?
The comment you refer to above was a discussion of the issues faced by the design team, which they have apparently solved with their "iCoupler technology"
They say there is a propagation delay of 325ns.
http://www.analog.com/en/interface/digital-isolators/products/CU_over_iCoupler_Digital_Isolation/fca.html
 
The above document explains that there are transformers used for the isolation. I don't think the chip reclocks as such but rather just converts the incoming USB (edge) signals  to 1ns pulses on the input and then converts these to regular USB signals on the output.
The block diagram basically shows what is happening.
I don't interpret this as reclocking, rather a process of the analog buffering and pulse conversion, which results in the propagation delay.(do you disagree?)
But hey, what do I know.
As long as it works

 

[leeperry]

interesting! I've read some theories that pulse transformers on coax outputs actually had a positive effect on jitter...but some ppl strongly disagree and think that they do introduce jitter.
 
both the Firestone Bravo and Audio-GD DI don't use black rectangular pulse transformers like on the Hiface and TeraX2, they use ferrites beads: http://www.audio-gd.com/Pro/dac/USBface/DI5.jpg
 
FWIR on google, PT are actually grossly overpriced ferrite beads: http://www.electro-tech-online.com/electronic-projects-design-ideas-reviews/36019-homebrew-pulse-transformer-using-ferrite-bead-core.html
 
I still wonder why USB audio sounds far better through ADUM4160...the galvanic isolation itself? the filtering you just explained(and that still "smoothes" and reclocks the data flow IIUC)? the useless state events filtering?
 
About disclosing secrets, I think the worst is ESS w/ their Sabre DAC...they won't even provide a datasheet w/o signing NDA's, duh.

 

[erin]

Quote:

 
I still wonder why USB audio sounds far better through ADUM4160...the galvanic isolation itself?
 
About disclosing secrets, I think the worst is ESS w/ their Sabre DAC...they won't even provide a datasheet w/o signing NDA's, duh.

I think the main thing improving the sound would be keeping the PC ground away from the DAC ground.
 
I don't get companies who don't provide a datasheet. You would think less people would be likely to use their product.
Imagine if car manufacturers did that.
 

 

[LSR75]

Quote:

Here's the new driver
 
http://cid-31812c3a17b68ec9.office.live.com/self.aspx/.Public/TeraDak%20X2%20ASIO.rar

X2 audio control panel still doesn't load.
"Application is out of date. Terminate itself automatically."

 

[leeperry]

these drivers are worthless on XP SP3...Reclock hiccups constantly in KS, I get loud glitches at the begining of each song in ASIO.
 
I spoke to the coder of those drivers, but he made me understand that I was a mere end-user and that he only supports his direct customers. Ah well, flawless driver-free operations FTW...there's nothing I hate more than crappy drivers.

 

[ROBSCIX]

Considering others are using them, perhaps it is just your system or bad configuration.

 

[JulioCat2]

I have those drivers on windows 7 and vista without problems, however i haven't test them in Win XP.

 

[leeperry]

actually most of the time it works fine in KS in Reclock, but sometimes when I seek, the sound is full of static and requires a reseek to fix itself(the Reclock logs are also full of errors!). These turd party drivers only make sense on XP IMO, as W7 supports KS and WASAPI natively...so there's no really no need for them on anything else than XP.

 

[Dynobot]

Quote:

actually most of the time it works fine in KS in Reclock, but sometimes when I seek, the sound is full of static and requires a reseek to fix itself(the Reclock logs are also full of errors!). These turd party drivers only make sense on XP IMO, as W7 supports KS and WASAPI natively...so there's no really no need for them on anything else than XP.

Aren't you tired of Windows and all of its problems?
 
Time to go Linux!
 

 

[leeperry]

 

Aren't you tired of Windows and all of its problems?
 
Time to go Linux!

 
OMG, you're right \o/
 
 
 
 
 
 
 
 
 
 
 
 
 
hehe, Linux means: no ffdshow, no Avisynth scripts, no Reclock to get butter-smooth movies in 48.000Hz, no uLilith, no VST plugins, no CoreAVC CUDA, no madVR, no PotPlayer...I mean, yeah? w/o all this, I'd be better off buying a playstation.
 
There's no problems whatsoever on my box when I don't depend on turd party drivers that glitch up all over the place.
 
That's Reclock using those crappy drivers(it never happened w/ any of my other soundcards in KS on the very same box):  

    26.45s 000db4         ChangeAudioRate(95904.10,0.0,48000)
   26.45s 000db4 WARNING Repeated 1 sample (or added 1 silence) (121892-153600)
   26.45s 000db4         InitPid() rates=(1.001000,1.000000) resampler=(1.001000,1.000000) stretch=(1.000000) psc=0
   26.45s 000db4         ChangeAudioRate(95904.10,0.0,48000)
   26.47s 000db4 WARNING Repeated 1 sample (or added 1 silence) (120496-153600)
   26.47s 000db4         InitPid() rates=(1.001000,1.000000) resampler=(1.001000,1.000000) stretch=(1.000000) psc=0
   26.47s 000db4         ChangeAudioRate(95904.10,0.0,48000)
   26.48s 000db4 WARNING Repeated 1 sample (or added 1 silence) (119136-153600)
   26.48s 000db4         InitPid() rates=(1.001000,1.000000) resampler=(1.001000,1.000000) stretch=(1.000000) psc=0

 
logs are available here, DS works fine, KS craps out: http://forum.slysoft.com/showpost.php?p=268190&postcount=34

 

[Dynobot]

Quote:

 
hehe, Linux means: no ffdshow, no Avisynth scripts, no Reclock to get butter-smooth movies in 48.000Hz, no uLilith, no VST plugins, no CoreAVC CUDA, no madVR, no PotPlayer...I mean, yeah? w/o all this, I'd be better off buying a playstation.
 
There's no problems whatsoever on my box when I don't depend on turd party drivers that glitch up all over the place.
 
That's Reclock using those crappy drivers(it never happened w/ any of my other soundcards in KS on the very same box):  

With  Linux you don't need all the stuff to get good sound.  Windows is like getting digital out via USB, it requires a whole bunch of band-aids to get great sound.  Look at all the money people need to spend to make USB work.  Converters, cables, batteries, isolation devices, clockers and reclockers, special power supplies.  All that just to get digital to a Dac.  Then once you get all that hooked up you still need VST plugins for EQ etc. to make it sound right. 
 
BTW, I got my ADum thingy today...
 

 

[leeperry]

hehe, I think you're verging on the Linux fanboy side of the force
 
I've got no problem believing that it can sound very good, but the need for USB Isolation is strictly hardware...Linux doesn't kill groundloops.
 
I use EQ to kill my middle ear resonances on headphones, it's a physiological problem: http://www.head-fi.org/forum/thread/413900/how-to-equalize-your-headphones-a-tutorial
 
And reclocking is the same, Linux doesn't reclock USB or S/PDIF
 
I don't only listen to music, I also watch movies...and Linux doesn't have anything like Reclock/CoreAVC CUDA/ffdshow/Avisynth/PotPlayer. I'm not interested in watching blurry hiccuping films.

 

[_Spanky_]

Quote:

With  Linux you don't need all the stuff to get good sound.  Windows is like getting digital out via USB, it requires a whole bunch of band-aids to get great sound.  Look at all the money people need to spend to make USB work.  Converters, cables, batteries, isolation devices, clockers and reclockers, special power supplies.  All that just to get digital to a Dac.

How does Linux change that? It's software... How does Linux make it so that you can plug a DAC directly into your computer without a USB converter? Unless I'm missing something incredibly amazing about Linux, you still need a USB converter and cables. Batteries and a special power supply means that the converter is running on better regulated power than the switching power supply in your computer (how does Linux change this?). Isolation device reduces grounding and noise issues via USB (again, how does Linux change this?). Clockers and reclockers are hardware and software related and I don't know about Linux's communication so you may win this aspect of the argument but what about the rest?
 
Also, I've gotta agree with leeperry. I don't know what I'd do without CoreAVC and ReClock.