clock

message for ray

i understand the last message that you left.
i made a very simple clock reclocking & the difference was huge!!!!!!!!!
but i have a question?
in my player i have a 33.868 meg crystal for the dac, & then
is divided to feed the servo dsp but i notice that you can
put a stabler clock but the servo dsp is still jittery.

i placed a 74ac00 on the out of the servo dsp to the dac &
i get rid of a lot jitter & the difference in sound was
seriously huge!!!!!!!!!

my question is if i put a master clock like the xlo2 from
lc audio, im going to hear more difference?

thank you
richt
 

rfbrw

Member
2001-10-26 11:51 pm
.
richt,
Correct me if i'm wrong but from what you say you have single clock that feeds the the dac and the servo through dividers where necessary. While a better clock will
probably improve things you need to clean up the jitter where it really matters which just before the data enters the dac. Some more information about the chips and their
inputs and outputs would help in this matter.
Out of curiosity how are you using the 74AC00 chips ?

ray.
 
clock reshape

hi ray

on my player the dac have a clock of 33 meg & then feeds
the servo dsp (synk of data between dac & dsp).
a lot of people talk about a master clock for the unit
but the most inportant part is whtat the dac receives!!!

you can have an excellent clock but if your dsp or d. filter is noisy ,poor inprovment will be heard.

in my case im going to install the master clock to the dac
(lc audio) & leave the clock reshaper to the dac (from the dsp).
right now i am trying different ways to reshape because i made that experiment one night for curiosity & the inprovment was so huge that i am looking for more info
or daughter card dedicated to clock reshapers or a chip that
can make this task to optimum performance.

i will post any info later because i consider that everybody should try this.

i have a linn lp12 turntable and it was sounding a lot better than my cd.
now my cd is too close from turntable & its hard to belive!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

richt
 
jitter

i consider folks:

that a well made 16 bit player with an excellent servo section & a jittery free incoming signal to dac will
outperforms better than 24 bits dac or dvd.

folks open your eyes !!!!!!!!!!

the electronic industry are making money bit by bit with
customers pocket.

dont get caught!
 

rfbrw

Member
2001-10-26 11:51 pm
.
hi richt
Try connecting the data and lrclk and if present sclk lines from the dsp to the dac through a 74HCT74 d-type flip-flop and connect the clock input of the 74HCT74 to 768fs master clock.
As for well designed 16bits outperforming well designed 24bits,I think the old boxing saying applies; A good big'un will always beat a good little'un.
ray.
 
hi rfbrw

thank you for your information!
can you explain me what exactly this circuit makes ???
i have 3 signals: l/r clock,bit clock, datastream going to
the dac.
this device make some kind of clock reshape to the 3 data streams?

i used a 74ac00 for bit clock reshaper.
i ordered from lcaudio , the masterclock (33meg).
i considered it is well made & will improve any unit
thank you
richt
 
richt,

I agree with your comments, a cheap 24 bit DVD/SACD player will not out perform a good CD Player. I have read reviews of the new Sony SACD players and a few have thought that a well designed CD Player of similar price can perform just as well. Of course others have raved about SACD, I have yet to try it myself.
 

rfbrw

Member
2001-10-26 11:51 pm
.
richt,
I first saw this trick in a very expensive piece of Sony
pro video eqpt. The 74HCT74 was between the NPC5842 and the two PCM63. The circuit serves to align the data with falling edge of the clock. On second thoughts it may actually make matters worse or stop the dac working in your case. Without further information on the relationships between the master clock and the data and lrclk in your system,I would advise you steer well clear of my original suggestion.Sorry about the possibly misleading information, but i've got PCM63 and PCM1702 dacs on the brain, not to mention FPGA based fifo's.
ray.