Hi rfbrw,
Some of the later Sony transports were not too bad either.
Essentially, a transport should give a nice, clear, stable eye pattern. It would be nice if it were supported and repairable as well, wouldn't you say?
-Chris
Nah, the NEC units were much better (and repairable!). I was never a fan of the Philips system as you couldn't easily optimze the mechanical settings. You had to take what the factory gave you. Wear type issues pretty much forced you to buy an entire transport assembly.
Some of the later Sony transports were not too bad either.
Essentially, a transport should give a nice, clear, stable eye pattern. It would be nice if it were supported and repairable as well, wouldn't you say?
-Chris
Hi Alexandre,
The location of the master clock in these situations has very little effect on the quality of data you can pull out of the disc. A perfect clock with a noisy eye pattern will still generate garbage.
-Chris
? Very simply put: Jitter in the transport = errors in the data. Hopefully repairable, but add imperfections on the disc or wearing parts in the transport and it can really badly affect the quality of data you get to work with. All "perfect" CD have a certain error rate all by themselves. Add errors from the transport to this and you start getting uncorrectable errors. Throw in a finger print and you may get a mute out of the deal, or a big skip. Sending the signal out will generate another slew of errors on top of all this. Starting with the best data you can get is a huge step in the right direction.
The location of the master clock in these situations has very little effect on the quality of data you can pull out of the disc. A perfect clock with a noisy eye pattern will still generate garbage.
-Chris
anatech said:
NO !
unless jitter is in the range of 20ns AND the corrector can't handle the errors, but I've never seen such CDPs
Hi Guido,
Unfortunately, I have. I have had to deal with DAC warranty through the "cheap transport and good DAC" phase in the Toronto area. Been servicing these since the start and still do.
I'm not trying to go head to head with you, but I have to report what I have seen in the field.
Also consider a noisy eye pattern. The demodulator has trouble getting all the one's and zero's where they should be. This has nothing to do with jitter, but jitter will amplify these problems.
-Chris
Unfortunately, I have. I have had to deal with DAC warranty through the "cheap transport and good DAC" phase in the Toronto area. Been servicing these since the start and still do.
I'm not trying to go head to head with you, but I have to report what I have seen in the field.
Also consider a noisy eye pattern. The demodulator has trouble getting all the one's and zero's where they should be. This has nothing to do with jitter, but jitter will amplify these problems.
-Chris
anatech said:
Hi Chris
I agree on the fact that some drives supply horrible eyepatterns, indeed not due to clock jitter, it is due to pick-up quality and maybe some other electrical errors.
I doubt if clock jitter will amplify these problems. Pit jitter usually is 10 to 20ns, bad readout makes that worse, so how could say 100ps (the average value in commercial CDPs) make that even worse ?
best
anatech said:
And would you be able to recommend any 'good' transports then? You can PM or email me if you'd prefer not to derail this thread any further. 🙂
On a side note, does anyone know what transport is used in the Denon DCD-S10/3000? A search here and on google has brought up nought. The player-tpt-dac list floating around here also doesn't list the transport, though it mentions the PCM1702 (2x in 'co-linear' configuration, whatever that means). The PCM1702 is quite a polarising DAC - I've seen both bouquets and brickbats for it over here. Ah - the vagaries of massed opinion! 🙁
anatech said:
Not too many NEC based machines, Nakamichi perhaps?, around here so one has to take what one can get. The CD104 seems plentiful enough to get by with for the moment. When they are all gone I will have to put some real effort into a dedicated hard disc player.
How to understand this?
If a digital interface demands particular reclocking and other reclocking is problemous, than interface is problemous in both engineering and economical point of view.
Why are you silent now? Please take part in discussion you've started 😉
best regards
Re: Re: S/PDIF Jitter: Myth or Reality?
As you can see, he is in the SinBin now.
darkfenriz said:
As you can see, he is in the SinBin now.
Ulas disappeared?
Ulas posted at about post:10 I think after Guido Tent ranting as usual about snake oil and if people found out the truth his clock business is over.Shortly after after surfing somewhere esle I returned to this thread Ulas' post dissapeared into thin air.Either he removed it or...you can draw your own conclusions.
Ulas posted at about post:10 I think after Guido Tent ranting as usual about snake oil and if people found out the truth his clock business is over.Shortly after after surfing somewhere esle I returned to this thread Ulas' post dissapeared into thin air.Either he removed it or...you can draw your own conclusions.
As smart as he is Ulas misses a simple point. Yes, a HC49 oscillator and a pair of 22pf caps will work but that is no reason not to use something better. Despite the fact that a Trabant or a 2CV will get you from A to B, no one seems to object too much to the availabilty of the likes of the Audi RS4 or the Subaru Imprezza. I see no reason why audio should be any different.
Re: Ulas disappeared?
For all his intelligence, Ulas is... 🙄
Hasn't he learnt from history and experience that the ideas that eventually get adopted and accepted are (sadly) not the ones that are forced upon people with a belligerent tone, but the ones for which time is taken to explain, cajole and wheedle.
Just because an idea is better doesn't neccessarily make it one that is automatically adopted. I mean, just look at the choice that was made regarding sampling rate for CDs. Or the audio codec choice for Windows.
It's sad, but certainly a fact of life. 🙁
singa said:
For all his intelligence, Ulas is... 🙄
Hasn't he learnt from history and experience that the ideas that eventually get adopted and accepted are (sadly) not the ones that are forced upon people with a belligerent tone, but the ones for which time is taken to explain, cajole and wheedle.
Just because an idea is better doesn't neccessarily make it one that is automatically adopted. I mean, just look at the choice that was made regarding sampling rate for CDs. Or the audio codec choice for Windows.
It's sad, but certainly a fact of life. 🙁
Agree with Adhoc,
Gallileo is another famous case when he said the earth was/is round and revolve around the sun when established/religious "fact" said it was FLAT.
Still I am inclined to give support to Ulas,I know he is reading/observing this.My personal advice to him is that he should resolve his personal issues first (he knows what they are) .Try spirituality it will help alot,believe me.
Gallileo is another famous case when he said the earth was/is round and revolve around the sun when established/religious "fact" said it was FLAT.
Still I am inclined to give support to Ulas,I know he is reading/observing this.My personal advice to him is that he should resolve his personal issues first (he knows what they are) .Try spirituality it will help alot,believe me.
Re: Ulas disappeared?
I am afraid (at least my sales numbers tell me) the opposite is true: People (and OEM customers) ARE finding out how clock performance affects digital playback (and recording) quality.
rfbrw hit the nail spot on: Any clock will get your DA converters to work but only a few will truly expose what is on your CD's.
best
singa said:
I am afraid (at least my sales numbers tell me) the opposite is true: People (and OEM customers) ARE finding out how clock performance affects digital playback (and recording) quality.
rfbrw hit the nail spot on: Any clock will get your DA converters to work but only a few will truly expose what is on your CD's.
best
After reading this thread and many others where Ulas state his opinions, I feel that I mostly agree with him. But there are still some questions that NEED to be settled.
Jitter does create a spectral "skirt" around tones. Even if this is not audiable at large signal levels, jitter-modulated tones still appear as "larger than spectral skirt"-components and are more easily measured. In the long run, jitter raises noise floor and produces small spectral distortion components.
Now, regarding this thread. Ulas plots say nothing of the existance of a filter, but if the plot is AFTER the filter, things are really bad. But how bad? Looking at the difference in magnitude of signal to noise floor this is a couple of hundred mdB?! Is that really 10^-3 dB? Such low input signals are doomed to show really small distortion components at the output, but with say -60 dBFS or -30dBFS signals the plots should look very different?
An explanation of this AND more plots comparing jitter from an CS8416 (with lousy PLL) with say and ASRC like AD1896 (PLL jitter cutoff starting at 2Hz!). The difference in jitter supression should be easily measured and regarding the acclaimed introduced jitter with an ASRC? Well, that would show up aswell =)
The only thing we should leave to the subjective minds are what amounts of p-p jitter that is audiable, since it does not rest on a technical basis.
Jitter does create a spectral "skirt" around tones. Even if this is not audiable at large signal levels, jitter-modulated tones still appear as "larger than spectral skirt"-components and are more easily measured. In the long run, jitter raises noise floor and produces small spectral distortion components.
Now, regarding this thread. Ulas plots say nothing of the existance of a filter, but if the plot is AFTER the filter, things are really bad. But how bad? Looking at the difference in magnitude of signal to noise floor this is a couple of hundred mdB?! Is that really 10^-3 dB? Such low input signals are doomed to show really small distortion components at the output, but with say -60 dBFS or -30dBFS signals the plots should look very different?
An explanation of this AND more plots comparing jitter from an CS8416 (with lousy PLL) with say and ASRC like AD1896 (PLL jitter cutoff starting at 2Hz!). The difference in jitter supression should be easily measured and regarding the acclaimed introduced jitter with an ASRC? Well, that would show up aswell =)
The only thing we should leave to the subjective minds are what amounts of p-p jitter that is audiable, since it does not rest on a technical basis.
Asgard said:
He wrote:
"The problem is not S/PDIF, but the implementation of particular S/PDIF receivers"
The problem is both, as I never heard nor designed an SPDIF driven DAC that is fully transparant for the properties of the transport in front of it.
In theory it should be possible to solve all SPDIF induced problems at the receiving end, but practice shows it is nearly impossible hence I would be happy with an improved interface which doesn't require the amount of attention an SPDIF receiver needs.
best
- Status
- Not open for further replies.