Upsampling USB DAC with AES/EBU output?
Feb 14, 2009 at 9:43 PM Post #16 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Sadly Patrick much of what you assert cannot be possible in this universe.


Are you saying there are many Universes? In that case anything is possible: a way to visualise 10 dimensions - Science Forums, The Original

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
What goes through an SPDIF or lightpipe is identical. SPDIF and AES/EBU are for all practical puposes identical too. It's all 0s and 1s, the quality of those 0s and 1s is utterly irrelevant to audio quality. If you can hear a difference there must be a fault with the cables or your system. This isn't an opinion, it's simple fact.


If you build a house using bricks (1s) and holes (0s) the building will collapse if the hole is in the wrong position. Also, if the bricks are vibrating from an earthquake it will be hard to sleep. But to some people it does not matter, they are still 1s and 0s (what does little jitter do?). But to the person lying under the collapsed building it matters.

I have found that Magix levitation feet reduces a lot of jitter which makes it sound smoother. I could not hear a difference between SACD and CD upsampled to 192kHz when I had Magix under both transports.

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
XLR has 3 connectors, 1 ground, 2 & 3 carring the same digital signal but one of them out of phase, purely for common mode rejection. RCA (SPDIF) has two connectors, a ground and the digital signal.


I have tried having only one conductor connected, I got no signal. I tried with AC power and headphones too! I have also tried connecting the wires together while touching the diaphrams of K1000, it gave off sound but I saw smoke on the surface of K1000 diaphrams, it also burned a hole through them, but I fixed it with scotch tape.

l_6e6c56b9e384b15db4142d061f494319.jpg



Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
BTW, daisy chaining cables is never a good idea as the connections can only degrade the signal.


It is true for analog signals, the shorter the better! But for AC power I have found that using a vibration dampening power cable reduces the AC noise and gives better sound after adding it to the path. Other power cables I have tried worsens the sound. Genesis power cable is the only one that improves the sound even after daisy chaining 3 of them together.

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Upsampling to 192kFs/s cannot make a difference. Anything lost during the creation of a 44.1kFs/s product cannot be restored by resampling at a higher rate. It's gone, forever, it isn't there anymore and it can't be put back. This really is basic audio theory, taught to every first year audio student everywhere in the world. This theory has been tried, tested, proven and implimented for more than 25 years.


Have ears been used during the testing?


Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
If you are hearing a positive difference when upsampling there must be a fault with your system or your ears, or you must be living in a universe with a different set of physical laws!!


Incomplete physical laws are not the correct laws, they are the wrong laws.

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Of course, it is entirely possible to hear a negative effect of upsampling, two or four times the data rate to transfer and process is likely to cause considerably more errors.


If it sounds more like real life is it an error?

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
I know you are going to fight against what I've written here because it is a bitter pill to swallow when you've been a victim of marketing hype.


I experiment for myself instead of read. I haven't read a book in my life!

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Let me put it this way, I've worked in Abbey Road Studios, Air Lyndhurst, The Hit Factory and several other highly specified (and respected/famous) recording studios around the world. None of them spend more than a few bucks per meter for cabling. Go look up the cost of fully screened Klotz or Van Damme cable. I'm afraid if you're spending more than this you have been done. Unless of course you feel that spending 1000 times too much on cabling has actually made your bedroom sound better than Abbey Road Studios?


Recorded music sounds veiled and muddy, I stopped listening years ago to classical "Reference Recordings". I only listen to trance music with different layers pasted on top of each other. It demands more from the system to output the whole music than to record each sound individually.

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
I'm not having a go at you personally, I'm just really annoyed that companies are allowed to mis-lead the consumer so much and that it's not against the law. Sure the specs look better but it doesn't make any difference.


Everything makes a difference. If you look at the sky you won't see that the aliens are looking back at you wondering why you are looking at them.
 
Feb 14, 2009 at 10:06 PM Post #17 of 55
Thanks for that Patrick, you've proved that you are indeed living in a different universe. In this cased everything you said could well be correct, my apologies.

"I have tried having only one conductor connected, I got no signal. I tried with AC power and headphones too!" - If I were you, I'd try the same thing without the cans. IE., AC power straight into your ears, a few hundred amps should do the trick, you might notice a little buzzing and then a slight change in the universe you are inhabiting.

BTW, say hi to Elvis when you see him next.

Cheers, G
 
Feb 15, 2009 at 12:54 AM Post #18 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Thanks for that Patrick, you've proved that you are indeed living in a different universe. In this cased everything you said could well be correct, my apologies.

"I have tried having only one conductor connected, I got no signal. I tried with AC power and headphones too!" - If I were you, I'd try the same thing without the cans. IE., AC power straight into your ears, a few hundred amps should do the trick, you might notice a little buzzing and then a slight change in the universe you are inhabiting.

BTW, say hi to Elvis when you see him next.

Cheers, G



you don't need to be an ass just because he doesn't agree with you
beerchug.gif


patrick, here's the thread for the 26/22,580 DAC:
http://www.head-fi.org/forums/f7/whe...enough-407658/

its not a strict digital to digital though, its a full fledged D/A, so your Benchmark or similar will be removed from the signal chain (and it'll need an amplifier on the output side, obviously)
 
Feb 15, 2009 at 1:20 AM Post #19 of 55
Quote:

Originally Posted by obobskivich /img/forum/go_quote.gif
patrick, here's the thread for the 26/22,580 DAC:
http://www.head-fi.org/forums/f7/whe...enough-407658/

its not a strict digital to digital though, its a full fledged D/A, so your Benchmark or similar will be removed from the signal chain (and it'll need an amplifier on the output side, obviously)



Thanks for the link.
I wonder why they have tubes on all of their DACs. Probably the 32 processors are power hungry and add noise into the system which makes it sound edgy, and the tubes compensate for it by adding smoothness.

I would like to have the lowest wattage component, maybe software upsampling in the computer is better then. Also I just remembered, it needs to have 24bit, not 16bit. When I used the digital output from dCS Verdi Encore CD transport it showed 16bit/44.1kHz on the dCS Elgar Plus DAC display, when I used computer with EMU0404 PCI soundcard it showed 24bit/44.1kHz, there was no comparison at all, computer was clearly better. 16bit sounded horribly muddy, 24bit was clear and detailed.
 
Feb 15, 2009 at 1:29 AM Post #20 of 55
your reading on the tubes is probably accurate, from its website:

Quote:

Between each original databit, 511 more databits are calculated and inserted to complete the music signal which serves to bring back an LP smoothness and warmth in music.


and

Quote:

This unit has been equipped with a true balanced tube output stage to ensure the best dynamics. Superb sonic performance is achieved, as well as low output driving impedance.


in other words upsample the hell out of the digital input, and then feed that through tubes to smooth/even it out before outputting it

probably sounds half decent (for that price, it better, I believe its a few thousand US dollars)

as far as input abilities, not seeing information as to if it takes 16-bit or 24-bit
 
Feb 15, 2009 at 3:42 PM Post #21 of 55
"you don't need to be an ass just because he doesn't agree with you"

It's got nothing to do with whether he agrees with me, it's because he is spouting fantasy as fact. The latest bit of genius mis-information is that 24bit is better than 16bit. Pure fantasy! He plays 24bit on one piece of equipment, 16bit on another, the 24bit sounds better so therefore 24bit is better???! No, its the different equipment which sounds better, duh. What makes this so foolish is that there is no way his system can output the resolution of 24bit, in fact not even the best recording studio in the world can. I would doubt that his system is even able to output the resolution of 16bit, as even full resolution 16bit is beyond most recording studios. In fact, there are no commercial recordings on the market which use more than about 10 bits of resolution and most are less than 5. So again, enjoy wasting your money on extra storage for absolutely no percievable difference whatsoever. Again, this isn't opinion, it's fact, it's the way that digital audio works. There is not a single person on the planet that can hear a difference with a commercial recording between 16bit and 24bit, it is way beyond any sound system/listening environment and even further beyond any human pair of ears. If you hear a difference it has to be a fault with some piece of equipment in the your system. There is no other explanation unless you feel that you are right and generations of scientists around the world in a number of different fields are all wrong.

All this stuff is the most basic audio theory. Why do people who are obviously completely ignorant of how audio works insist on coming out with pure fantasy and passing it off as fact. They must enjoy mis-leading people, all seems a bit sick to me.

I thnik maybe I'll leave this forum alone and go back to the professional forums where people are generally helpful to each other rather than trying to cheat, fool or mis-lead them, presumably just to make themselves feel big.

G
 
Feb 15, 2009 at 3:56 PM Post #22 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
"you don't need to be an ass just because he doesn't agree with you"

It's got nothing to do with whether he agrees with me, it's because he is spouting fantasy as fact. The latest bit of genius mis-information is that 24bit is better than 16bit. Pure fantasy! He plays 24bit on one piece of equipment, 16bit on another, the 24bit sounds better so therefore 24bit is better???! No, its the different equipment which sounds better, duh.



dCS Verdi Encore CD transport is $15 500. EMU0404 PCI soundcard is $99. When both are using Toslink to dCS Elgar Plus DAC, the EMU is clearly better.

The same is true with Cary 303/300 CD player with Toslink to dCS DAC. EMU is better.

However, when using Benchmark DAC1, Cary 303/300 is clearly better than computer at 24bit/192kHz upsampling. (I used Nordost Valhalla AES/EBU digital cable)

dCS DACs accept max 96kHz through a single input.
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
"
What makes this so foolish is that there is no way his system can output the resolution of 24bit, in fact not even the best recording studio in the world can. I would doubt that his system is even able to output the resolution of 16bit, as even full resolution 16bit is beyond most recording studios.



The display of dCS DAC shows bit and sample rate. It is a fact.

Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
"
I thnik maybe I'll leave this forum alone and go back to the professional forums where people are generally helpful to each other rather than trying to cheat, fool or mis-lead them, presumably just to make themselves feel big.

G



They will like you at James Randi's skeptic forum. There they listen to their audio systems paused. I don't believe they know what ears are...they are skeptics, they don't believe in ears, they just use their measurement devices instead.
 
Feb 15, 2009 at 4:58 PM Post #23 of 55
Patrick - Is it possible that your experience breaks the known laws of physics or is it more likely that there is a fault with your equipment or your perception of sound?

What is actually in your 8bits of difference between 16 & 24bit? It's noise, that's all, nothing else. If there were usable sound in all those extra 8bits, your sound system could not reproduce it and even if it could, you would at the very least permanently deafen yourself and at the worst kill yourself. No commercial recording uses even half of the 24bits (the other half is noise), if it did it would be considered a dangerous product and removed from the marketplace, hopefully before someone dies from it. Exposure to 85dB is considered dangerous, so why do you think a product with about 10,000 times that dynamic range is going to be better? I take it that you do realise that extra bit resolution in digital audio only manifests itself as a greater dynamic range?

"The display of dCS DAC shows bit and sample rate. It is a fact." - No it isn't, here we go again more fantasy peddled as fact, what is wrong with you? What your DAC is telling you is the file format of the audio, not how many bits have actually be used to encode the audio or what freqencies it contains. If I take an 8bit photograph and save it as a 24bit photo, is it now a 24bit photo or an 8bit photo in 24bit format? Come on, this isn't rocket science, it's basic computing!

And just for your information, using my ears has earnt me more than good living as a music/audio professional for the last 25 years. Rather than questioning whether I've ever used my ears, maybe you should be asking why you spent $15,500 on something which could apparently be done better for $99!

G
 
Feb 15, 2009 at 5:08 PM Post #24 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Patrick - Is it possible that your experience breaks the known laws of physics or is it more likely that there is a fault with your equipment or your perception of sound?

What is actually in your 8bits of difference between 16 & 24bit? It's noise, that's all, nothing else. If there were usable sound in all those extra 8bits, your sound system could not reproduce it and even if it could, you would at the very least permanently deafen yourself and at the worst kill yourself. No commercial recording uses even half of the 24bits (the other half is noise), if it did it would be considered a dangerous product and removed from the marketplace, hopefully before someone dies from it. Exposure to 85dB is considered dangerous, so why do you think a product with about 10,000 times that dynamic range is going to be better? I take it that you do realise that extra bit resolution in digital audio only manifests itself as a greater dynamic range?

G



On EMU website it says this:

Bit Depths: 24-bit I/O, 32-bit processing

E-MU E-DSP™ 32-bit DSP with 67-bit accumulator (double precision w/ 3 headroom bits)


What does it mean? Why don't they use 16bit if it makes no difference? Why does $99 soundcard sound better than $15 500 CD transport?
 
Feb 15, 2009 at 6:01 PM Post #25 of 55
Patrick - I can explain exactly what this means and why it is there, but the answer is quite technical.

For the end user 16bit is more than sufficient. 24bit has be the standard for recording studios for over a decade becuase it allows a great deal of extra headroom while recording (usually 18dB or 22dB above nominal 0dBv). When mixing, each mix process creates overflow errors which are then summed together. In a complex mix there may be a hundred or more of these processes so these truncation or noise errors can build up to a noticable level, so modern processors tend to work at 48bit (or the slightly weaker 32bit) precision to keep these errors way below the noise floor when summed together. Accumulators are used to store the information during final summing so as not to overload the input side of the summing buss (usually this is a 56bit accumulator) the output side of the buss is truncated (or sometimes dithered) back to 24bit. This allows a master fader to be lowered to about -90dB without incurring signal to noise ratio problems. Once a 24bit mix has been generated it is dithered down to the consumer format of 16bit. Even in the very best recording studios in the world, there should not be any percievable difference between this 24bit master and the 16bit product. This is because at least 14 or so of the LSBs in the 24bit master are infact just noise from processing, the audio chain and the noise floor of the live recording room.

24bit, 32bit, 67bit accumulators are all useful to the audio professional but completely useless to the consumer. If you see these sorts of figures being banded about on equipment for consumers then you can be sure they are using them to try and fool the consumer into thinking that more bits is better and to buy their equipment. Is a 500bit DAC better than a 400bit DAC, as far as specs go sure, but in the real world there would be no way of telling because even the best play back system in the world can't output more than about 17bits worth of resolution and no commercial recordings exist with more than about 10bits of resolution. Have a look at the noise specs of your DAC, if the spec is higher than -144dB then it can't output 24bit resolution. Although of course it can playback 24bit format files. Bet they don't put that in their marketing!

I'm sorry, but it is no more than marketing hype.

G
 
Feb 15, 2009 at 6:19 PM Post #26 of 55
Patrick - I've just had a look at the link you posted and the output dynamic range of the 0404PCI is 116dB which is just over 19bits of resolution and on the input side it's 111dB just over 18bit. So when they advertise it as 24bit is that misleading or what? They would argue it's not, because it can play 24bit format files, what they don't tell you is that it will only play a maximum 19 of those 24bits! Bare in mind that with modern dither algorithms 16bit can encode a dynamic range up to 120db, which is greater than the dynamic range offered by this so called 24bit piece of gear!!! In other words, a really good 16bit DAC could give you a higher resolution than this 24bit one. Are you starting to feel mislead yet? I don't know about you, but it makes me angry even though I know enough to see through the bull!!

Bare in mind as well that once output, you are going to loose another load of bits worth of resolution from going through cabling, connectors and speakers.

G
 
Feb 15, 2009 at 6:36 PM Post #27 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Patrick - I've just had a look at the link you posted and the output dynamic range of the 0404PCI is 116dB which is just over 19bits of resolution and on the input side it's 111dB just over 18bit. So when they advertise it as 24bit is that misleading or what? They would argue it's not, because it can play 24bit format files, what they don't tell you is that it will only play a maximum 19 of those 24bits! Bare in mind that with modern dither algorithms 16bit can encode a dynamic range up to 120db, which is greater than the dynamic range offered by this so called 24bit piece of gear!!! In other words, a really good 16bit DAC could give you a higher resolution than this 24bit one. Are you starting to feel mislead yet? I don't know about you, but it makes me angry even though I know enough to see through the bull!!

Bare in mind as well that once output, you are going to loose another load of bits worth of resolution from going through cabling, connectors and speakers.

G



I'm confused. I don't use analog output of EMU. I use digital output from EMU into my DAC. My DAC said 24bit/44.1kHz input on the display.
 
Feb 15, 2009 at 6:38 PM Post #28 of 55
Quote:

Originally Posted by gregorio /img/forum/go_quote.gif
Bare in mind as well that once output, you are going to loose another load of bits worth of resolution from going through cabling, connectors and speakers.


I thought you didn't believe in cables.
 
Feb 15, 2009 at 7:26 PM Post #29 of 55
"I'm confused. I don't use analog output of EMU. I use digital output from EMU into my DAC. My DAC said 24bit/44.1kHz input on the display."

OK, so the dynamic range of your card is not relevant, have a look at the dynamic range of your DAC. Your DAC is not lying, it is recieving a 24bit file but guaranteed, 10bits or less of that 24bit file actually contains the music, the other 14bits are just noise.

"I thought you didn't believe in cables. "

Then you mis-read my post. How else am I going to connect my studio together? The cabling to my main monitors probably cost less than $20 in total and is the same cable used by world famous studios like Abbey Road. It's the rip-off priced cabling I don't believe in.

G
 
Feb 15, 2009 at 9:28 PM Post #30 of 55
You said that the signal is losing loads of bits of information through the cables, and then you said you are using $20 cabling. I think you should try air or vacuum dielectric cables.
wink.gif
 

Users who are viewing this thread

Back
Top