[Sovkiller]

Any transport nowadays has to accomplish with the minimal requirements that the modern DACs needs to perform satisfactory, or in other words the DACs have become more and more versatile, and are able to work almost in any condition of jitter, errors, etc....That doesn't mean of course that becasue you are using the receiver you will get the best, you have to know what is inside that receiver, and how the implementation was done after the DAC and even the PSU used etc...

 

[sacd lover]

Quote:

Originally Posted by ssportclay The thing that struck me the hardest when I first received my SACDmods Denon 2900 was the increase in dynamic slam caused by the new discreet analog output stage,(zap filter).I also have to turn the volume down as compaired to before the Mods.A new DAC will not even address this issue.Most CD players sound digital and lame to me now where the SACDmods Denon 2900 sounds like music.

Absolutely! The analog output stage is vitally important. You cant worry only about improvements on the digital side and expect the best sound. I swear by the zap filter out put stages too. I will never have another op amp output stage in my digital sources.

 

[wali]

Analong stage is the most crucial part.


Quote:

dynamic slam

oh dear...

 

[drminky]

Quote:

Originally Posted by Samuli That doesn't really seem to make any sense to me.. The critical thing for the digital processors clock would be to realize that the data coming to receiver is 16/44,1khz.. from there receiver stores the bits to its buffer and feeds it out in time by the DACS own clock. It does not have any usage for the clock signal that came with the signal from there on. It seems to me that Stereophile has never heard of buffers Now why would the DACS clock need another clocks assistance timing the bits from its buffer anymore than the transports clock needed to read and time the bits from the cd in the first place? Also what do they mean by digital processor, maybe it's not the same thing as DAC?

..Digital processor means the same thing..

I'm sure pretty much every cdp/dac has buffers. Problem being that the new clock will ALWAYS have to be somewhat tied to the old one. You can't just completely dump the old clock and create a new one. What you will eventually get in this case is a buffer underrun or overrun. It has to GENERALLY follow the speed variations, doing it's best to smooth out the high-frequency aspects of the jitter, but it cannot afford to ignore the lower frequency aspects of the jitter - because doing this will empty or overflow the cache. This is why the stereophile article says the DAC acts as a low pass filter to the jitter. Obviously, the more buffer, the more lower frequency jitter will be able to be removed, but this will also mean a longer delay in playback.

The only people who you'll see claiming a dac is totally impervious to spdif transport jitter are either people who don't understand this concept, or are trying to sell you their dac!

 

[backdrifter]

edit

 

[Samuli]

Quote:

Originally Posted by drminky ..Digital processor means the same thing.. I'm sure pretty much every cdp/dac has buffers. Problem being that the new clock will ALWAYS have to be somewhat tied to the old one. You can't just completely dump the old clock and create a new one. What you will eventually get in this case is a buffer underrun or overrun. It has to GENERALLY follow the speed variations, doing it's best to smooth out the high-frequency aspects of the jitter, but it cannot afford to ignore the lower frequency aspects of the jitter - because doing this will empty or overflow the cache. This is why the stereophile article says the DAC acts as a low pass filter to the jitter. Obviously, the more buffer, the more lower frequency jitter will be able to be removed, but this will also mean a longer delay in playback. The only people who you'll see claiming a dac is totally impervious to spdif transport jitter are either people who don't understand this concept, or are trying to sell you their dac!

I stand corrected. This seems to be case according to Bob Katz too. (Which maybe the most respected studio engineer and an audiophile aswell!)

PS. He didn't warn about suspicious DAC-sellers though

 

[Ferbose]

Quote:

Originally Posted by drminky The only people who you'll see claiming a dac is totally impervious to spdif transport jitter are either people who don't understand this concept, or are trying to sell you their dac!

Really?
Perhaps you know more about this issue than I do.
I read that Weiss DAC is completely immune to S/PDIF jitter.
It is one of the most highly regarded DACs in the pro audio, and I am under the imprerssion that their claim about jitter rejection is well accepted.
My Benchmark DAC1 also claims to do the same.
In the data sheet Benchmark engineers show that neither the injection of absurd amounts of jitter nor the use of 1000 ft bulk cable generate jitter-induced side band in thier product. The measurement graphs are done with audio presicion 2 analyzers and I am inclined to believe them.
Some other DACs like Mytec and Bel Canto also claim to be jitter-immune, although I don't know how valid these claims are.

 

[swiego]

Quote:

Originally Posted by drminky ..Digital processor means the same thing.. I'm sure pretty much every cdp/dac has buffers. Problem being that the new clock will ALWAYS have to be somewhat tied to the old one. You can't just completely dump the old clock and create a new one. What you will eventually get in this case is a buffer underrun or overrun. It has to GENERALLY follow the speed variations, doing it's best to smooth out the high-frequency aspects of the jitter, but it cannot afford to ignore the lower frequency aspects of the jitter - because doing this will empty or overflow the cache. This is why the stereophile article says the DAC acts as a low pass filter to the jitter. Obviously, the more buffer, the more lower frequency jitter will be able to be removed, but this will also mean a longer delay in playback. The only people who you'll see claiming a dac is totally impervious to spdif transport jitter are either people who don't understand this concept, or are trying to sell you their dac!

From the Benchmark DAC1 site,

"Unfortunately most converters derive their conversion clocks directly from the poorly filtered output of an AES/EBU receiver IC. These converters can achieve their rated performance only when driven from very low jitter AES/EBU sources through no more than 3 or 4 feet of cable. It is highly unlikely that these converters can achieve much over 16-bits of performance in a typical installation. In short, test bench performance is not repeatable in a typical installation.

The DAC-1, DAC-104 and the ADC-104 employ Benchmark’s new UltraLock™ technology to eliminate all jitter-induced performance problems. UltraLock™ technology totally isolates the conversion clock from the digital audio interface clocks. Jitter on a DAC digital audio input, or an ADC reference input can never have any effect on the conversion clock of an UltraLock™ converter. In an UltraLock™ converter, the conversion clock is never phase-locked to a reference clock. Instead the converter oversampling-ratio is varied with extremely high precision to achieve phase-lock to the reference clock. Jitter cannot effect the audio conversion, and test bench performance is repeatable in any installation!"

The second to last sentence implies, well, that you are wrong. DrMinky, would you please explain more specifically why such a device is still susceptable to meaningful jitter issues? Thanks.

 

[Zanth]

Quote:

Originally Posted by sacd lover Absolutely! The analog output stage is vitally important. You cant worry only about improvements on the digital side and expect the best sound. I swear by the zap filter out put stages too. I will never have another op amp output stage in my digital sources.

But you know that once implemented, you are only ever hearing the zap filter sound and not the sound of the original circuit. This was the reason why I declined the installation. Chris over at Parts Connexion was really great in detailing how the installation would go and what would happen. It would have completely eliminated the benefit of my tube buffer stage. That said, the zapfilter is supposed to sound really really nice.

 

[Kurt]

Quote:

Originally Posted by Ferbose My Benchmark DAC1 also claims to do the same.

Quote:

Originally Posted by swiego From the Benchmark DAC1 site, ..... Jitter cannot effect the audio conversion, and test bench performance is repeatable in any installation!" The second to last sentence implies, well, that you are wrong.

My 4 months old Benchmark DAC1 sounds considerable better fed with E-MU Cards then with a Chaintech AV-710.

I also heard the Bel Canto with 2 different sources and while the differences were not that big, bec. the sources were of comparable quality, both transferred their own sound signature through the DAC.

Quote:

Originally Posted by swiego DrMinky, would you please explain more specifically why such a device is still susceptable to meaningful jitter issues? Thanks.

I would like to know that too.

TIA

 

[clarke68]

Quote:

Originally Posted by drminky The only people who you'll see claiming a dac is totally impervious to spdif transport jitter are either people who don't understand this concept, or are trying to sell you their dac!

Quote:

Originally Posted by swiego DrMinky, would you please explain more specifically why such a device is still susceptable to meaningful jitter issues?

You're both right, but you're saying slightly different things.

Jitter immunity, for all practical purposes, means that the sonic artifacts of jitter do not exceed the noise floor. So there's no such thing as a DAC that is "totally impervious" to jitter, but there are a few (and only a few) that are not susceptable to meaningful jitter issues.

Quote:

Originally Posted by swiego This seems to be case according to Bob Katz too. (Which maybe the most respected studio engineer and an audiophile aswell!) PS. He didn't warn about suspicious DAC-sellers though

Yes he did! Page 235 of 'Mastering Audio':

Quote:

Successful converter manufacturers must master the techniques of PC board layout, grounding, internal clock distribution, and immaculate separation of digital and analog signals. Things are looking up. But caveat emptor. (emphasis mine)

 

[ssportclay]

Quote:

Originally Posted by Zanth But you know that once implemented, you are only ever hearing the zap filter sound and not the sound of the original circuit. This was the reason why I declined the installation. Chris over at Parts Connexion was really great in detailing how the installation would go and what would happen. It would have completely eliminated the benefit of my tube buffer stage. That said, the zapfilter is supposed to sound really really nice.

I don't understand what you are saying.The zap filter just amplifies the signal comming into it.You should not really hear the zap filter at all.

 

[drminky]

Quote:

Originally Posted by swiego From the Benchmark DAC1 site, The DAC-1, DAC-104 and the ADC-104 employ Benchmark’s new UltraLock™ technology to eliminate all jitter-induced performance problems. UltraLock™ technology totally isolates the conversion clock from the digital audio interface clocks. Jitter on a DAC digital audio input, or an ADC reference input can never have any effect on the conversion clock of an UltraLock™ converter. In an UltraLock™ converter, the conversion clock is never phase-locked to a reference clock. Instead the converter oversampling-ratio is varied with extremely high precision to achieve phase-lock to the reference clock. Jitter cannot effect the audio conversion, and test bench performance is repeatable in any installation!" The second to last sentence implies, well, that you are wrong. DrMinky, would you please explain more specifically why such a device is still susceptable to meaningful jitter issues? Thanks.

What was I saying about people trying to sell you their dac?

FYI the exclusive 'UltraLock™' technology is just a marketing hype term for asynchronous resampling...yes the same 'upsampling' that is in the majority of products released in the last 5 years at least. The only thing special about benchmark and 'UltraLock™' is that are one of the few companies brash, naive, or downright dishonest enough to claim that they 'eliminate ALL jitter-induced performance problems'.

So, yes, this implementation of jitter attenuatuon is a different case to the one I was describing earlier. So, in this case, they have two master clocks. Instead of slowly adjusting the clock around to make sure the buffer doesn't overrun/underrun, they slowly adjust the sampling rate to prevent the same buffer problem. Yes, this is perhaps the most effective way of dealing with jitter we have so far in a standalone dac (so relax all you DAC1 owners!), but it still will not be perfectly effective. Again, all we are doing is a 'averaging off' or 'spreading out' of the jitter artifacts. You will hear the artifacts less because they can now occur at any point in time, rather than at specific points. Basically, it becomes more like dither. That is not to say the jitter has been removed completely because it hasn't and the jitter will have most definietly impressed it's own character onto the signal.

In a nutshell, many people have reported hearing differences in transports using such upsampling techniques, albiet not as much as the previous generation of dacs. Do you believe the marketing hype™? Or do you believe your ears? I know which one I'll go with thank you