PCM1704 or newer chips?

Jump to Latest
#181
Well, Im still messing around with this "slow-roll off" leakage issue. Some recordings are as bad as -84 , -82 @ 23khz 🙁 . It is triggered by ANYTHING percussive if its loud enough (high dynamic range with big transient, I dont listen compressed music ) . Kick drum, hi hat, snare drum , you name it :dodgy:

Meridian Apodizing type filter has a lot of validity IMHO, as many recording has nothing but garbage above 18khz.
 
#183
Re: SharpRolloff and SlowRolloff automatic changing

nagaesan said:
Hi! inertial, thank you.

I am not a developper but a planner of ALPHA processing of DA-S1 or DCD-S1. This developper of ALPHA was Dr.Hayashi and Mr.Ohtsuka. And now I moved to the other sections, recently, young DENON people are in charge this renewal and development.
Click to expand...

very honorable Nagaesan ,thanks for specifications


For such a reason I have no plan of inprovement of ALPHA processing, but recently there is the part which I cannot understand new ALPHA processing which young people developed.
[/B]
Click to expand...


Maybe they are "too" young? ..... 🙂


Original ALPHA processing in 1993 has 2 function.
One is Adaptive Line Pattern Harmonized Algorithm that revises 1LSB stepped wave pattern.
Two is Automatic Low Pass filter Harmonic Adjustment that playback no ringing impulse waveform, more beautiful than WADIA's when we played a test CD.[/B]
Click to expand...


Understood. Two is foundamental IMO....


New ALPHA processing from 2004 has 3 function.
One and Two are same as 1993 original.
Three is "Slow Rolloff" and "Sharp Rolloff" automatic changing function. And these new generation model has very long TAP FIR filter made by powerfull Analog Devices's SHARC DSP.
I think it is no problem, when changing rolloff curve with muring .
[/B]
Click to expand...


Perdone me , I do not understand very well.
Is not Three " in contrast" with two?


I think DA-S1 was not perfect, there are many points that we should be improved refer to WADIA's nice circuit, but I do not admire that doing a thing. Still I don't like "Slow Rolloff".
[/B]
Click to expand...


I'm confused Here.
Implulse response in wadia's are worse than the most cheap Denon Alpha equipped.
One more: Why you do not like "slow rolloff" ? Technical matters or
listening tests ? Quite different motivations IMO.

Thanks,
Paolo
 
#184
Re: AttackKing sound --> Slow Rolloff mode

nagaesan said:
This is a catalogue of DENON DCD-SA11 with slow and sharp rolloff automatic changing new ALPHA system.

I read this figure, DENON seems like to choice slow rolloff if play back attackking sound like attack hi-hats.
I don't understand why they don't choice sharp rolloff that good high frequency responce.

And my biggest suffering is, does not a problem occur even if changing rolloff curve of long TAP FIR filter, when playback music.
Click to expand...


Oh, well, maybe I am newbie/ ignorant, but IMHO those AL24 graphs "sucks"!:hot: :angel:
Old Alpha was "perfect" here............incredible :-(

Is this the reason because you are skeptical about new youngs EE people in Denon developing area? 😀
In this case, I tend to agree with you ; Are they good/capable only to make impressive calculations but at the same time, totally "deaf" ? (hironic)😀

Cheers,
Paolo
 
#185
2nd function of ALPHA (changing LPF carrasterics TestCD & MusicCD)

Hi! inertial,

ALPHA has 2 function.
The first one is revises 1LSB stepped wave pattern, this is not joke.
And the second is, we can get no ringing impulse, much more beautiful than Slow Rolloff when we played a test CD.
This is [my] joke for "Slow Rolloff freaks". I say [ ] reason as follows.

I don't like SlowRolloff that because remaining mirror noise up to 22.05kHz @fs=44.1kHz, very bad high frequency waveform.
My opinion is we must cut mirror noise up to 1/2fs by long TAP Sharp Rolloff FIR and analog LPF.

And many people misunderstand, that they think about short ringing when playback impulse in TestCD is better than long ringing.
Long ringing is very important for good interpolation at D/A conversion. People should notice it.

****[ ] reason*****
The first function of ALPHA processing was devised by Dr.Hayashi, and developped by Mr.Ohtsuka.
But second function, I devised, and Mr.Ohtsuka developped.
In addition, the naming of this 2 function "ALPHA", the person is Tetsuya Nagae who likes technical jokes.
 

Attachments

  • changeal70j.jpg
    changeal70j.jpg
    91.1 KB · Views: 778
#187
EUVL said:
Nagaesan,

You are probably one of the few persons who have the first hand experience. Would you be kind enough to share with us your impression of the PCM1704 playing 24 bit data (e.g. after a HDCD decoder) compared to ES9008 / ES9018 with SPDIF input ?


Thx,
Patrick
Click to expand...


Patrick, I deliberately took the plunge and searched for Nagaesan's comment about this type of comparison inside the ESS thread ( I also think its better to talk about that there ) , here it is:
http://www.diyaudio.com/forums/showthread.php?postid=1735103#post1735103

Hope Tetsuya can tell us more about the ALPHA, we have a great talk here and Im very glad!

So far I like the Meridian's apodizing filter approach. I also listened their AES speech (Freely available on www), and right now investigating my own version of the filter which I christened the "appetizing" filter, because its less steep, as I only want to push the "crud" below -96 dB (my last LSB ). This going to take a lot of averaging and more TAP-s than I wanted initially :bawling:
 
#188
PCM1702/PCM1704/ES9008S/ES9018

Hello EUVL,

So long time I am making PCM1704 DAC, but DIR is very difficult.
I tried some IC, CS8412,8414,8416,AES20,21... but all of these conventional PLL with S/PDIF sound was just "fat sound image & not tight low end".
So, I bought DP- S1/DA-S1 with ST-Clock sync system.
This 384fs super high frequency clock, not word sync, clock direct sync system with ST-Link cable is tight low end, but I want much more better image.

When I had such a trouble, I heared ES9008S eva-board with non PLL S/PDIF.

http://www.teddigital.com/ESS_Es9008s_en.htm

------------------ History ----------------------
CXD2500/SM5843/PCM1702/5534/SSM2139/5534/5%RMG_470uFcap_AC-Coupling__(Original DCD-S1)=Bad focus, Foggy sound, baggy low end... etc

CXD2500/SM5843/PCM1704/OPA627/AD797/BUF634/0.1%Z201_DC-Coupling__(Modified DCD-S1)=Well focus, Tight & Rich low end... etc

CXD2500/SM5842/PCM1704/OPA627/AD797/BUF634/0.1%Z201_DC-Coupling__(Modified DP-S1/ST-Sync/DA-S1)=Good focus, Tight & Rich_low end... etc

DCD-S1_spdif-->ES9008S/AD797_DC-Coupling__(Original ES9008S_evabord)=Very nice focus, clear sound, Slim low end, just hissing femail vocal... etc

DCD-S1_spdif-->ES9008S/AD825_DC-Coupling__(Modified I/V ES9008S_evabord)=Very nice focus, clear sound, Slim low end, not so hissing femail vocal... etc

DCD-S1_spdif-->ES9018S/AD797_DC-Coupling__(Original ES9018_evabord)=Very nice focus, Very nice clear sound, Slim low end, no hissing femail vocal... etc

------------My IDEA and request-------------------
I want ESS's DIR+64/32bitDF IC for PCM1704

1) DIR (S/PDIF) = ES9008S or ES9018's Non PLL Type
2) DF = ES9018(64bit acc/32bit OUT)
3) DAC=+/- Vcc, Low offset I-out DAC ( PCM1704 ) or__Avcc virtual GNDed ES9018/9008S I-out DAC
4) I/V=AD825 with DC offset cancellor
5) I/V Feed Back Register=Vishay Z-foil 0.2ppm TCR
6) I/V Feed Back Cap=SE99 Soshin MICA cap
7) Diff AMP=AD797
8) Diff Matching Register=Vishay Z-foil 0.1%
9) Analog LPF=AD825x2+Vishay Z-foil 0.1%/SE99
10) Buffer=BUF634/DC Coupling super low impedance output
 
#191
the ESS asrc chip is canned by marketing department , but you can turn off downsamplign filter in TI asrc + receiver SRC4392 (which is 28bit).

ESS Asrc with 8/16 OS + user coefficients would be tough one to beat .😀

nagaesan , I sent you some data of my OS filter Including sinewave interpolation please let me know what you think!
 
#192
Re: 2nd function of ALPHA (changing LPF carrasterics TestCD & MusicCD)

nagaesan said:
Hi! inertial,

ALPHA has 2 function.
The first one is revises 1LSB stepped wave pattern, this is not joke.
And the second is, we can get no ringing impulse, much more beautiful than Slow Rolloff when we played a test CD.
This is [my] joke for "Slow Rolloff freaks". I say [ ] reason as follows.

I don't like SlowRolloff that because remaining mirror noise up to 22.05kHz @fs=44.1kHz, very bad high frequency waveform.
My opinion is we must cut mirror noise up to 1/2fs by long TAP Sharp Rolloff FIR and analog LPF.

And many people misunderstand, that they think about short ringing when playback impulse in TestCD is better than long ringing.
Long ringing is very important for good interpolation at D/A conversion. People should notice it.

****[ ] reason*****
The first function of ALPHA processing was devised by Dr.Hayashi, and developped by Mr.Ohtsuka.
But second function, I devised, and Mr.Ohtsuka developped.
In addition, the naming of this 2 function "ALPHA", the person is Tetsuya Nagae who likes technical jokes.
Click to expand...

Hi Tetsuya,
I was sure you was a true audio-guru from your first post!😎

Thanks for explanations, but I am so newbie an maybe english do not help, that I have to read and read various times.
I have to reflect a bit, and maybe it's time for me to start to study a bit of basic PCM theory🙄

Am I the only who do not understand here?😀
BTW,IMHO, seems to me something is contradictory from Denon/ALPHA-claims and Tetsuya statements. 😕

Cheers,
Paolo
 
#193
seconda funzione dell'ALFA (& cambiante di TestCD di carrasterics di LPF; MusicCD)

Ciao! inertial, L'ALFA ha funzione 2. Quello primo è revisioni che 1LSB ha fatto un passo modello di onda, questa non è scherzo. Ed il secondo è, noi può non ottenere impulso di squillo, molto più bello del Rolloff lento quando abbiamo giocato un CD della prova. Ciò è [il mio] scherzo per " Freaks" lento di Rolloff;. Dico [] il motivo come segue. I don' la t gradice SlowRolloff che perché rumore restante dello specchio fino a 22.05kHz @fs=44.1kHz, forma d'onda ad alta frequenza molto difettosa. Il mio parere è noi deve tagliare il rumore dello specchio fino a 1/2fs dall'ABETE tagliente di Rolloff del COLPETTO lungo e dall'analogo LPF. E molta gente comprende male, quello che pensano al breve squillo quando l'impulso di playback in TestCD è migliore lungamente dello squillando. Lungamente squillare è molto importante per buona interpolazione alla conversione di D/A. La gente dovrebbe notarla. ***** di motivo del **** [] La prima funzione di elaborazione dell'ALFA è stata inventata da Dr.Hayashi ed è stata sviluppata da Mr.Ohtsuka. Ma seconda funzione, ho inventato e Mr.Ohtsuka si è sviluppato. In più, nomina di questo " di 2 funzioni; ALPHA" , la persona è Tetsuya Nagae che gradice gli scherzi tecnici.
 

Attachments

  • changeal70j.jpg
    changeal70j.jpg
    91.1 KB · Views: 680
#194
Re: 2nd function of ALPHA (changing LPF carrasterics TestCD & MusicCD)

nagaesan said:
Hi! inertial,

ALPHA has 2 function.
The first one is revises 1LSB stepped wave pattern, this is not joke.
Click to expand...

OK, understood.



And the second is, we can get no ringing impulse, much more beautiful than Slow Rolloff when we played a test CD.
This is [my] joke for "Slow Rolloff freaks". I say [ ] reason as follows.
[/B]
Click to expand...


Understood the NO-RINGING aspect. This also IMO was the claims of DENON .
I do not understood [your] joke for "slow rolloff freaks".....


I don't like SlowRolloff that because remaining mirror noise up to 22.05kHz @fs=44.1kHz, very bad high frequency waveform.
My opinion is we must cut mirror noise up to 1/2fs by long TAP Sharp Rolloff FIR and analog LPF.[/B]
Click to expand...


Ok, this is your "directive".


And many people misunderstand, that they think about short ringing when playback impulse in TestCD is better than long ringing.
Long ringing is very important for good interpolation at D/A conversion. People should notice it.
[/B]
Click to expand...


Here again I' am confused.
Are you saying that RINGING is better than NO-RINGING ? I'm very confused here..... :-(
In short this appears me the opposite of the ALPHA !! Am I right?



****[ ] reason*****
The first function of ALPHA processing was devised by Dr.Hayashi, and developped by Mr.Ohtsuka.
But second function, I devised, and Mr.Ohtsuka developped.
In addition, the naming of this 2 function "ALPHA", the person is Tetsuya Nagae who likes technical jokes. [/B]
Click to expand...


Very fun, you have invented the ALPHA ( for DENON) but was it only a marketing-"balls" for newbie audiophiles ? I can't believe it, sure I have understood wrong Tetsuya San, please correct me!
:bawling:

Cheers,
Paolo
 
#196
LPF characteristics when playback Music or Test CD was changed.

I say that RINGING is better than NO-RINGING for Music.
Do you know "interporation"?
Why slow rolloff 2okHz sign wave is bad than sharp rolloff?

For slow rolloff lovers, NO-RINGING when playback TEST CD (impulse or square wave), it is more beautifl than slowrolloff DF's short-RINGING.



And do you know the purpose and the difference of WADIA's slow rolloff FIR filter by one stage interporation and others(like SM5843A or DF1706's "3 stage cascading FIR filter) ?

WADIA's short RINGING impulse have become slow rolloff by short TAP FIR but one stage 16fs oversampling for no analog LPF device.

DF1706's purpose is ....... getting "short RINGING impulse" itself.
 
#198
Re: LPF characteristics when playback Music or Test CD was changed.

nagaesan said:
I say that RINGING is better than NO-RINGING for Music.
Do you know "interporation"?
Why slow rolloff 2okHz sign wave is bad than sharp rolloff?
Click to expand...

Hi Tetsuya ,

Perdone me . My basic english can seems as like "provocation".
It is not so. I'm sorry if I have exposed bad my "thinking".
I am ignorant on digital, this is evident 🙂
When you say "Why slow rolloff 2okHz sign wave is bad than sharp rolloff? " is this statement motivated by math ( objective) or by listening preferences ( subjective) ?

Interpolation ,I have to study it !




For slow rolloff lovers, NO-RINGING when playback TEST CD (impulse or square wave), it is more beautifl than slowrolloff DF's short-RINGING.
[/B]
Click to expand...


Ok, make sense to me.
When you say more beautiful, do you intend because oscillogrammes looks visually better ( to inexpert eyes) OR because the sound is more pleasant ( to ears)?


And do you know the purpose and the difference of WADIA's slow rolloff FIR filter by one stage interporation and others(like SM5843A or DF1706's "3 stage cascading FIR filter) ?
[/B]
Click to expand...


Maybe I have understood jet the basic principle but nothing more


WADIA's short RINGING impulse have become slow rolloff by short TAP FIR but one stage 16fs oversampling for no analog LPF device.
[/B]
Click to expand...


Ok, this is copiright by wadia. I remember the first listen of the 2000 DAC. WOW!


DF1706's purpose is ....... getting "short RINGING impulse" itself. [/B]
Click to expand...


Ok, thanks for refresh.


BTW seems to me there are two school of thinking here:
-1) slow roll-of lovers ( short ringing OR no-ringing even better)
-2) hard roll-of lovers ( long ringing)

I presume I am in the -1) togethar with analogue LP's fans
who is in the -2) ? maybe digital fans ?and why? 😕

perdone me for excessive symplicism and please correct me where I am wrong 🙂

Thanks,
Paolo
 
#200
http://www.audioasylum.com/audio/digital/messages/14/143572.html

argued over last time (1000000 +1. )

the problem with slow-roll-off is alias leakage around 22100-26000 khz on order of -70-90 and its triggered by anything(!!!). OFC its not issue if you are NOSser thats bout 1000* times worse 😀

Ayre type filter (leaks) has better phase shift characteristics than Meridian apodizing filter (no leak and attempts to "cut" previous embedded pre-ringing in favor of its own).

http://www.wolfsonmicro.com/uploads/documents/en/Ultra_High_Performance_DAC_whitepaper.pdf
 
You must log in or register to reply here.
Top Bottom