Author |
Topic Search Topic Options
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 08 2005 at 07:19 |
To talk about the jitter issue we evoked before, I’ve read an article about a Wadia digital ensemble where it’s about jitter on DRIVE, which is due to a desynchronisation of the clock, with the data sream.
This clock is located in the servo control of the drive.
So this”jitter” phenomenon happens in the drive.
It explains why my drive Sonic Frontiers is considered to have one of the lowest jitter rate of the market.
This is what is said in the extract i posted before.
There are two kinds of jitter: the converter’s one, and the drive’s one.
Wadia 850, featuring Teac modified mechanic
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 08 2005 at 07:29 |
Digital extraction eliminates the need for extreme synchronisation of clock and drive.
Digital extraction -> ZERO jitter. So the only kind of jitter to worry about on the computer is D/A conversion jitter, and that's as low as that of musical CD players (100 picoseconds) when you use QUALITY sound cards. I'm not talking about the Audigy 2 ZS which I'm using now, it has a higher jitter rate, but the X-Fi which will soon be released.
Edited by MikeEnRegalia
|
|
|
goose
Forum Senior Member
Joined: June 20 2004
Location: United Kingdom
Status: Offline
Points: 4097
|
Posted: September 08 2005 at 08:04 |
oliverstoned wrote:
>>> First, the classic 16bits CD is quite limited in freq. |
That bit at least just isn't true! In terms of frequency, CD represents perfectly up to 22.05kHz, certainly well above what's needed.
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 08 2005 at 08:30 |
goose wrote:
oliverstoned wrote:
>>> First, the classic 16bits CD is quite limited in freq. |
That bit at least just isn't true! In terms of frequency, CD represents perfectly up to 22.05kHz, certainly well above what's needed.
|
oliverstoned may be right in that some analog equipment can handle up to 30khz.
But:
- The human ear can't hear anything above 20khz. Indeed, most grown up persons can't even hear anything above 18khz. This is FACT, no listening test is needed.
- Some turntables or radios can handle more ... but the INPUT is limited to 16khz. Even good vinyls don't offer more than that.
BTW: oliver, 16bit doesn't have anything to do with FREQUENCY, it specifies the RESOLUTION of the signal. I agree that using 16 bits is a limitation. It means that 65.536 steps are used for each sample. Here the SACD is a real advantage, it uses 16 million steps. But in this area listening tests have shown that very few people are able to hear a difference. I would say that although it is undoubtedly better (the more resolution the better of course), it's not a difference like night and day and surely not a required feature.
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 08 2005 at 09:15 |
True that the human hear is limited to 20 khz and less as people loose auditive acuity (espially when you listen loud!). But supersonic noise is hear somehow and adds harshness! the proof is trhat when you record a CD on a tapedeck limited to 17khz, it's much smoother in the highs as the upper freq are filtered.
I think we have not much acuity above 20khz, but enough to suffer from supersonic noise.
Yes, in theory SACD is better as it has higher resolution. it's not the case for the moment but it will come maybe later.
A good thing that works is the HDCD filter:
"High Definition Compatible Digital (HDCD) is a patented encode/decode process for delivering the full richness and detail of the original microphone feed on compact discs (CDs) and DVD-Audio. HDCD has been used in the recording of more than 5,000 CD titles, which include more than 250 Billboard Top 200 recordings and more than 175 GRAMMY nominations, and account for more than 300 million CDs sold.
HDCD-encoded CDs sound better because they are encoded with 20 bits of real musical information, as compared with 16 bits for all other CDs. HDCD overcomes the limitation of the 16-bit CD format by using a sophisticated system to encode the additional 4 bits onto the CD while remaining completely compatible with the existing CD format. HDCD provides more dynamic range, a more-focused 3-D soundstage, and extremely natural vocal and musical timbre. With HDCD, you get the body, depth, and emotion of the original performance—not a flat digital imitation.
HDCD-encoded recordings sound better on all digital player products because HDCD subtractive, dither A/D conversion and dynamic-filter processes yield a higher resolution signal with lower distortion. For HDCD CD releases, Peak Extend restorable soft limiting can increase resolution by allowing signal levels to be raised by up to 6 dB; Low Level Extend can improve resolution of low-level signals; and HDCD high-frequency dither improves resolution by 6 dB (one bit). HDCD-equipped players improve the sound of all digital recordings because HDCD decoding includes the HDCD precision up-sampling digital filter."
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 08 2005 at 09:28 |
oliverstoned wrote:
True that the human hear is limited to 20 khz and less as people loose auditive acuity (espially when you listen loud!). But supersonic noise is hear somehow and adds harshness! the proof is trhat when you record a CD on a tapedeck limited to 17khz, it's much smoother in the highs as the upper freq are filtered. I think we have not much acuity above 20khz, but enough to suffer from supersonic noise.
|
Then why did you praise the 30khz analog equipment, and bash the CD for being "inferior" with its 22.05khz?
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 08 2005 at 09:40 |
Just to say that analog can go further than digital on the performance level.
Aida Loth-X
Microseiki
MikeEnRegalia wrote:
oliverstoned wrote:
True that the human hear is limited to 20 khz and less as people loose auditive acuity (espially when you listen loud!). But supersonic noise is hear somehow and adds harshness! the proof is trhat when you record a CD on a tapedeck limited to 17khz, it's much smoother in the highs as the upper freq are filtered. I think we have not much acuity above 20khz, but enough to suffer from supersonic noise.
|
Then why did you praise the 30khz analog equipment, and bash the CD for being "inferior" with its 22.05khz?
|
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 08 2005 at 09:44 |
oliverstoned wrote:
Just to say that analog can go further than digital on the performance level.
|
Ever heard of 24bit/192khz? Digital can go way beyond analog as far as frequency is concerned. And they could go to 64bit resolution just as easily, if there was any need for it.
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 08 2005 at 09:55 |
OK, but anyway, it goes far but there's still the supersonic noise which ruins the sound. (and the problem stays with SACD/24 bits resolution).
It will always be under analog...
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 08 2005 at 10:01 |
oliverstoned wrote:
OK, but anyway, it goes far but there's still the supersonic noise which ruins the sound. (and the problem stays with SACD/24 bits resolution). It will always be under analog... |
You can always use a low pass filter which cuts off the high frequencies.
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 08 2005 at 10:10 |
Yes! but unfortunatly this is not the only problem of digital.
The big problem is that when you numerize the signal, you simplifies it (the higher the resolution, the less you simplify)but there will be always info missing.
The ear and the human brain are too sensitive and have to "reconstitute" the missing info. What's make digital to involve fatigue.
When you switch from digital to analog on a big system, you say "ahhhh..." (relief whisper)
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 09 2005 at 08:43 |
I've listen to my new tape deck yesterday evening and it was fantastic. I listen to some vintage tapes: the Kraftwerk compilation "excellor8" (1976 tape!!)and TD "Stratosfer" among others...it smokes every CD as it has the warm and soft analog sound!
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 09 2005 at 08:51 |
oliverstoned wrote:
Yes! but unfortunatly this is not the only problem of digital. The big problem is that when you numerize the signal, you simplifies it (the higher the resolution, the less you simplify)but there will be always info missing. The ear and the human brain are too sensitive and have to "reconstitute" the missing info. What's make digital to involve fatigue.
When you switch from digital to analog on a big system, you say "ahhhh..." (relief whisper) |
I think that when using 24bit resolution there's a huge safety margin ... I'm sure that the resolution of the human ear is significantly lower. Why? Well, listen for yourself.
As far as the frequency is concerned, there can be no doubt that 22khz is enough by any definition.
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 09 2005 at 09:02 |
Of course, 22 khz is largely enough.
My new tape deck goes only at 16khz and it goes already very far in the highs, even on old tapes in Dolby mode. (if it's recorded on Dolby as well).
The pb is not to go to 22 khz, but to do it well.
MikeEnRegalia wrote:
oliverstoned wrote:
Yes! but unfortunatly this is not the only problem of digital. The big problem is that when you numerize the signal, you simplifies it (the higher the resolution, the less you simplify)but there will be always info missing. The ear and the human brain are too sensitive and have to "reconstitute" the missing info. What's make digital to involve fatigue. When you switch from digital to analog on a big system, you say "ahhhh..." (relief whisper) |
I think that when using 24bit resolution there's a huge safety margin ... I'm sure that the resolution of the human ear is significantly lower. Why? Well, listen for yourself.
As far as the frequency is concerned, there can be no doubt that 22khz is enough by any definition.
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 09 2005 at 13:00 |
And 24 bits doesn't solve the digital pb !(only in theory)as the pb is inherent to digital.
Reag planar 3
A very cheap turntable, which can beat big digital, if you put a good moving coil on it:
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 09 2005 at 13:02 |
REGA PLANAR 3
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 09 2005 at 13:26 |
oliverstoned wrote:
And 24 bits doesn't solve the digital pb !(only in theory)as the pb is inherent to digital.
|
Yes, a higher resolution than your ear can cope with solves the problem in theory AND real life. The only question is how high the resolution of the human ear really is.
Why are you always saying these things like they were approved scientific concepts?
|
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 10 2005 at 03:01 |
MikeEnRegalia wrote:
oliverstoned wrote:
And 24 bits doesn't solve the digital pb !(only in theory)as the pb is inherent to digital. |
Yes, a higher resolution than your ear can cope with solves the problem in theory AND real life. The only question is how high the resolution of the human ear really is.
Why are you always saying these things like they were approved scientific concepts? |
I explained you that the pb is that with digital, infos are missing.
The higher the resolution, the less infos missing.
Unfortunatly, there are still infos missing with 24 bits.
And the human ear/brain hears it.
|
|
oliverstoned
Special Collaborator
Honorary Collaborator
Joined: March 26 2004
Location: France
Status: Offline
Points: 6308
|
Posted: September 10 2005 at 03:03 |
...That's why i said the problem is inherent to digital, whatever the technology (24 bits) used.
|
|
MikeEnRegalia
Special Collaborator
Honorary Collaborator
Joined: April 22 2005
Location: Sweden
Status: Offline
Points: 21206
|
Posted: September 10 2005 at 05:35 |
oliverstoned wrote:
MikeEnRegalia wrote:
oliverstoned wrote:
And 24 bits doesn't solve the digital pb !(only in theory)as the pb is inherent to digital. |
Yes, a higher resolution than your ear can cope with solves the problem in theory AND real life. The only question is how high the resolution of the human ear really is.
Why are you always saying these things like they were approved scientific concepts?
| I explained you that the pb is that with digital, infos are missing. The higher the resolution, the less infos missing. Unfortunatly, there are still infos missing with 24 bits. And the human ear/brain hears it. |
That's what YOU think. In listening tests (on good HiFi systems) most people aren't able to tell the sources apart.
BTW: Who said that vinyl captures more details than 24bit does? Sorry to break this to you, but I don't think that the groove of the vinyl disc has more than 16.7 million steps in resolution ...
|
|
|
Donate monthly and keep PA fast-loading and ad-free forever.