Author Topic: DDDAC 1794-NOS 24-192  (Read 416824 times)

Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #100 on: December 05, 2012, 04:55:42 PM »
I missed it, and I don't exactly know what you're doing.

So you're feeding the modulator directly?

Do you know what the modulator is doing?   I don't (on that chip) and am curious to know.

I suspect it is taking 16 or 24bit data, and converting it into 8bit?   But it will be doing so much more,,,,,
It's all about the music,, not the equipment.

tuyen

  • Guest
Re: DDDAC 1794-NOS 24-192
« Reply #101 on: December 05, 2012, 05:09:50 PM »
have you been through the website oz?    if not, i'd recommend it. doede seems to explain pretty much step by step his design process. 

http://www.dddac.com/dddac1794_design.html

not sure if it answers all the questions you have though.

Offline zenelectro

  • Hero Member
  • *****
  • Posts: 825
  • Liked: 177
Re: DDDAC 1794-NOS 24-192
« Reply #102 on: December 05, 2012, 06:08:59 PM »
I missed it, and I don't exactly know what you're doing.

So you're feeding the modulator directly?

Do you know what the modulator is doing?   I don't (on that chip) and am curious to know.

I suspect it is taking 16 or 24bit data, and converting it into 8bit?   But it will be doing so much more,,,,,

The oversampling filter is upsampling IP data by 8 x. You are bypassing this and feeding direct
into advanced segment modulator. So looks like the modulator can actually take sample rates up to
1.536MHz (8 x 192kHz). However looking at the distortion it works best around 350kHz (8 x 44.1).
At 192k the distortion gets worse.

FWIW if you want to know more about advanced segment operation there is a white paper somewhere.

Z

Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #103 on: December 05, 2012, 07:55:33 PM »
have you been through the website oz?    if not, i'd recommend it. doede seems to explain pretty much step by step his design process.  

http://www.dddac.com/dddac1794_design.html

not sure if it answers all the questions you have though.
Hey T, yep read it, and previously complimented Doede's work in documenting/sharing what he has done, you don't see this from any other Dac producer.   I take my hat off to Doede for doing this.

No, unfortunately it doesn't answer my question.    What exactly is the modulator step doing on the PCM1794 chip?   We're bypassing the oversampling filter,  and we're going straight into the modulator.    Curiosity has got me by the short and curly's.    Delta sigma chips resample pcm data,  what will the modulator be resampling to (in word length and frequency)?
It's all about the music,, not the equipment.

Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #104 on: December 05, 2012, 08:01:26 PM »
The oversampling filter is upsampling IP data by 8 x. You are bypassing this and feeding direct
into advanced segment modulator. So looks like the modulator can actually take sample rates up to
1.536MHz (8 x 192kHz). However looking at the distortion it works best around 350kHz (8 x 44.1).
At 192k the distortion gets worse.

FWIW if you want to know more about advanced segment operation there is a white paper somewhere.
I would definitely like to read it, if you can find it easily?

What is driving my question?

If we feed the dddac-1794 a RBCD source (16-44.1), what happens within the chip when we bypass the oversampling step?    What actually happens in that modulator, before the data is given to the Dac conversion step?     It would be nice to know.
It's all about the music,, not the equipment.

Offline zenelectro

  • Hero Member
  • *****
  • Posts: 825
  • Liked: 177
Re: DDDAC 1794-NOS 24-192
« Reply #105 on: December 06, 2012, 02:32:51 PM »
The oversampling filter is upsampling IP data by 8 x. You are bypassing this and feeding direct
into advanced segment modulator. So looks like the modulator can actually take sample rates up to
1.536MHz (8 x 192kHz). However looking at the distortion it works best around 350kHz (8 x 44.1).
At 192k the distortion gets worse.

FWIW if you want to know more about advanced segment operation there is a white paper somewhere.
I would definitely like to read it, if you can find it easily?

What is driving my question?

If we feed the dddac-1794 a RBCD source (16-44.1), what happens within the chip when we bypass the oversampling step?    What actually happens in that modulator, before the data is given to the Dac conversion step?     It would be nice to know.

There is some info on various SD architectures in this one also has segmented DAC included.

Also explains the unity weighted multi bit OP's with various DEM approaches.

If you can get even a little through this you'll realize these guys put a -lot- of work and research into this.


http://www.cscamm.umd.edu/programs/ocq05/adams/adams_ocq05.pdf

Pretty full on :)






Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #106 on: December 06, 2012, 07:31:32 PM »
[sighs] I wish I never read it.   :P

For those who like simplicity, sigma delta is *not* it.  :-\

It's pretty clear the 1794 modulator step is manipulating the bit rate down to probably 4 or 6 bit words.   But if i'm reading that slide pack correctly, the approach seems to rely on the incoming sampling rate being oversampled *before* the modulator.

I don't want to get too deep into this, because I'll end up running in circles until I understand it.

Key slide,,,,,,
Quote
∆Σ Basic Facts
* ∆Σ works by oversampling, coarse quantization and noise-shaping
* High SNR is possible, if OSR and modulator order are high enough
* Low-order modulators (i.e.1st-order and 2nd-order) are susceptible to in-band tones and DC-input deadbands
* Single-bit modulators are inherently linear, but multi-bit modulators have much higher performance
     o Single-bit modulators typically overload for inputs > –3dBFS
* ∆Σ modulators come in many flavors: single-bit/multi-bit, single-loop/multi-loop, lowpass/bandpass and real/quadrature (complex)


So my question remains.  What is this particular 1794 chip doing with a 16/44.1 signal, when the oversampling filter is bypassed?   I have a sinking feeling it is decimating samples, hope I'm wrong.......  It could be dividing the input down into multiple even streams, but if the sampling rate was too low (in each sub stream), you'd imagine distortion would higher.
« Last Edit: December 06, 2012, 08:05:48 PM by ozmillsy »
It's all about the music,, not the equipment.

tuyen

  • Guest
Re: DDDAC 1794-NOS 24-192
« Reply #107 on: December 07, 2012, 04:55:30 PM »
I'm not too sure either what it is doing with 16/44.1 signals, but had one guy post this re the subject:

I only have 44.1/16 music and now upsample it all to 176.4 (rather than 192 as 176.4 is a whole number multiplier of 44.1) using Secret Rabbit Code Level 0 (the highest quality) all performed in software within mpdPup to a WaveIO. It is excellent and gives a major boost to SQ without IMHO any unpleasant artifacts.


another guy's post:

I think that might have to wait.  That I haven't switched off your DAC after a few hours listening and fired up my TDA1541A comfort blanket, says it all really. Don't quote me on this, and of course I am allowed to change my mind if I ever do implement a 1704 daughter-board, but my gut feeling is that 1704 will sound more "digital" than what you have now. There is a naturalness, a kind of organic nature, which reminds me more of old-school Burr Brown and Philips chips, more than anything else I have heard which has a sigma-delta badge, or even 1704 for that matter.

I will post some more later. But I had a most interesting time last night with mpd upsampling 44.1/16 CD rips to 176.4. (I don't have a great deal of >44.1k material, so was just playing with software upsampling redbook to 88.2 and 176.4 on the laptop.) Since then I have installed Triode's EDO plugin and kernel on my Logitech Touch (for USB DAC support and > 96k sample rate support) and currently have the DAC hooked up to that. The idea being that I could after editing a config file on the server, use sox or ssrc, to up/over sample on-the-fly. Only problem with that is that I can't go above 96k at the moment due to the Touch using wireless network. Doesn't have the bandwidth to support 176k sample rates. I'll hook up via ethernet later. Reading the John Swenson post, (linked above), I think he is onto something. If you are going to OS and DF, you can get better results doing it in software, not the filters built into the chips.

Offline zenelectro

  • Hero Member
  • *****
  • Posts: 825
  • Liked: 177
Re: DDDAC 1794-NOS 24-192
« Reply #108 on: December 07, 2012, 06:14:21 PM »
I'm not too sure either what it is doing with 16/44.1 signals, but had one guy post this re the subject:

I only have 44.1/16 music and now upsample it all to 176.4 (rather than 192 as 176.4 is a whole number multiplier of 44.1) using Secret Rabbit Code Level 0 (the highest quality) all performed in software within mpdPup to a WaveIO. It is excellent and gives a major boost to SQ without IMHO any unpleasant artifacts.


another guy's post:

I think that might have to wait.  That I haven't switched off your DAC after a few hours listening and fired up my TDA1541A comfort blanket, says it all really. Don't quote me on this, and of course I am allowed to change my mind if I ever do implement a 1704 daughter-board, but my gut feeling is that 1704 will sound more "digital" than what you have now. There is a naturalness, a kind of organic nature, which reminds me more of old-school Burr Brown and Philips chips, more than anything else I have heard which has a sigma-delta badge, or even 1704 for that matter.

I will post some more later. But I had a most interesting time last night with mpd upsampling 44.1/16 CD rips to 176.4. (I don't have a great deal of >44.1k material, so was just playing with software upsampling redbook to 88.2 and 176.4 on the laptop.) Since then I have installed Triode's EDO plugin and kernel on my Logitech Touch (for USB DAC support and > 96k sample rate support) and currently have the DAC hooked up to that. The idea being that I could after editing a config file on the server, use sox or ssrc, to up/over sample on-the-fly. Only problem with that is that I can't go above 96k at the moment due to the Touch using wireless network. Doesn't have the bandwidth to support 176k sample rates. I'll hook up via ethernet later. Reading the John Swenson post, (linked above), I think he is onto something. If you are going to OS and DF, you can get better results doing it in software, not the filters built into the chips.


These are great reviews.

However what are they really saying?

The DAC designer went to heroic lengths to run a SD DAC without digital filter, implying that was the problem.
The user adds the digital filter in his PC by upsampling to 176.4 and gets great results.
 
1) The 1794 DAC is a great DAC -but-
2) The 1794 standard digital filter is crap 
3) If the digital filter is done right there is no advantage to 0 x oversampling?

More questions

Z



 





tuyen

  • Guest
Re: DDDAC 1794-NOS 24-192
« Reply #109 on: December 07, 2012, 07:00:25 PM »
Unsure Z.   Maybe the DF in the 1794 is poo and 'ruins' the sound.  In in saying that, some people might actually prefer this 'ruined' sound?    Maybe software DF works better than the built in hardware DF on the 1794?        Is it similar story to why people bypass OS on the marantz cdp and prefer the sound when running NOS mode? 

What I personally take from it all is, try, experiment, have fun with it.    all is possible... and there really never is an absolute right or wrong way to do things, as the final sound to each and every person is different and subjective.  Right? :)

For reference, I have not been upsampling any files, just running them natively from foobar2000 to the usb module.  16/44.1 material sounds on the same level (but slightly diff) to the 120chip dddac1543mk2.   the dddac1794 nos  has slightly more open and refined sound. treble and bass definition/extension feels improved.  The older dddac has bit more energy/rawness.  But could be because was using poor power supply to power the older dddac  while with the new dddac, i'm running decent linear 'audiophile' psu.    Few too many factors at play to be able to give any concrete conclusions regarding the actual dac chip  and   sd vs r2r.     

One thing I am quite confident in concluding is that the hi-res recordings (24/88, 24/96, 24/174 & 24/192) can (not all) have the ability to deliver noticeably more fidelity in the sound over standard 16/44.1 cd rips.  The music seems to be less 'forced' out and is more open/effortless. Subtle details are easier identified.  Bass and treble notes seem to have improved finer clarity and definition.

Overall, happy enough with the performance to just stick to the one dac. Paired with a nice valve preamp, makes a really exceptional hi-res supported digital solution. Probably best I have experienced in my system so far (subjective).  All from a $400 dac kit :)

Offline kajak12

  • Hero Member
  • *****
  • Posts: 2468
  • Liked: 78
    • http://killerdac.com/forum/index.php
Re: DDDAC 1794-NOS 24-192
« Reply #110 on: December 07, 2012, 09:33:36 PM »
Unsure Z.   Maybe the DF in the 1794 is poo and 'ruins' the sound.  In in saying that, some people might actually prefer this 'ruined' sound?    Maybe software DF works better than the built in hardware DF on the 1794?        Is it similar story to why people bypass OS on the marantz cdp and prefer the sound when running NOS mode? 

What I personally take from it all is, try, experiment, have fun with it.    all is possible... and there really never is an absolute right or wrong way to do things, as the final sound to each and every person is different and subjective.  Right? :)

For reference, I have not been upsampling any files, just running them natively from foobar2000 to the usb module.  16/44.1 material sounds on the same level (but slightly diff) to the 120chip dddac1543mk2.   the dddac1794 nos  has slightly more open and refined sound. treble and bass definition/extension feels improved.  The older dddac has bit more energy/rawness.  But could be because was using poor power supply to power the older dddac  while with the new dddac, i'm running decent linear 'audiophile' psu.    Few too many factors at play to be able to give any concrete conclusions regarding the actual dac chip  and   sd vs r2r.     

One thing I am quite confident in concluding is that the hi-res recordings (24/88, 24/96, 24/174 & 24/192) can (not all) have the ability to deliver noticeably more fidelity in the sound over standard 16/44.1 cd rips.  The music seems to be less 'forced' out and is more open/effortless. Subtle details are easier identified.  Bass and treble notes seem to have improved finer clarity and definition.

Overall, happy enough with the performance to just stick to the one dac. Paired with a nice valve preamp, makes a really exceptional hi-res supported digital solution. Probably best I have experienced in my system so far (subjective).  All from a $400 dac kit :)
Maybe the DF in the 1794 is poo and



Thats the bottom line tuyen only $400 bargain if stays in your system longer then a month it must be good ;) ;) ;)
still discovering the link between electronics and audio reproduction.so much to learn and so little time

Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #111 on: December 07, 2012, 10:25:34 PM »
I still wanna know what exactly that freaking 1794 modulator is doing,,,,   ???   :P  :P
It's all about the music,, not the equipment.

Offline kajak12

  • Hero Member
  • *****
  • Posts: 2468
  • Liked: 78
    • http://killerdac.com/forum/index.php
Re: DDDAC 1794-NOS 24-192
« Reply #112 on: December 08, 2012, 12:55:42 AM »
I still wanna know what exactly that freaking 1794 modulator is doing,,,,   ???   :P  :P
modulating numbers (heavy duty accountant ) :P
still discovering the link between electronics and audio reproduction.so much to learn and so little time

Offline zenelectro

  • Hero Member
  • *****
  • Posts: 825
  • Liked: 177
Re: DDDAC 1794-NOS 24-192
« Reply #113 on: December 08, 2012, 11:16:11 AM »
I still wanna know what exactly that freaking 1794 modulator is doing,,,,   ???   :P  :P

Do a search for the advanced segment patent. That may shine some more light.

Then there is always the DSD1794 - same DAC but also does DSD.

And I know you will want to know how it handles DSD signals.

I think better off to build - lifes too short.




Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #114 on: December 08, 2012, 08:53:10 PM »
I think I figured out that the multibit DSD chips are doing multiple 1 bit sub streams in parallel,  what I couldn't figure out is precisely how the sampling is handled, and how many sub streams are occurring.   I'm assuming it is just divided down.

24/192k, divided into 4x6bit/48k (then into further 1bit sub streams). Something like that,,,, (16/44.1 is a real worry).

For the DSD1794,   I'm guessing the 1bit incoming stream is converted something like....

1/2.8224m, into 6x1bit/470.4k sub streams, that are multiplexed.

These are purely guesses, based on loose reading.  I could be way off in my thinking.

« Last Edit: December 08, 2012, 08:56:10 PM by ozmillsy »
It's all about the music,, not the equipment.

Offline rhlauranna

  • Hero Member
  • *****
  • Posts: 1386
  • Liked: 198
Re: DDDAC 1794-NOS 24-192
« Reply #115 on: December 10, 2012, 06:49:31 AM »
Quote from: tuyen link=topic=842.msg12339#msg12339 date=
Oh, please tell Mr Schick that I am still waiting for my tonearm!!$#@

Me too.. 15+ months and counting :-X

Good news! Just yesterday I talked personally to Mr. Schick, and he promised me that you will receive your tonearm till the end of this year !!!

Offline zenelectro

  • Hero Member
  • *****
  • Posts: 825
  • Liked: 177
Re: DDDAC 1794-NOS 24-192
« Reply #116 on: December 10, 2012, 09:17:08 AM »
I think I figured out that the multibit DSD chips are doing multiple 1 bit sub streams in parallel,  what I couldn't figure out is precisely how the sampling is handled, and how many sub streams are occurring.   I'm assuming it is just divided down.

24/192k, divided into 4x6bit/48k (then into further 1bit sub streams). Something like that,,,, (16/44.1 is a real worry).

For the DSD1794,   I'm guessing the 1bit incoming stream is converted something like....

1/2.8224m, into 6x1bit/470.4k sub streams, that are multiplexed.

These are purely guesses, based on loose reading.  I could be way off in my thinking.


Oz,

I think you are a bit off but I don't really have time atm to dig up the resources to post here.
I'll see if I can address it this week sonme time

Clocks to build, stuff to fix yada yada

Z


 



Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #117 on: December 10, 2012, 09:32:34 AM »
Yeah, won't be the first time.  ;)

When I get some time, I'll do some further reading/googling on it all.   1 thing that seems pretty clear, that modulator step is dramatically manipulating the data in some way, before presenting it to the dac step.   

Understanding what the modulator does and how it does it, may tell us a fair amount about the effects of skipping the oversampling step?
It's all about the music,, not the equipment.

Offline ozmillsy

  • Hero Member
  • *****
  • Posts: 2249
  • Liked: 277
Re: DDDAC 1794-NOS 24-192
« Reply #118 on: December 10, 2012, 09:43:54 AM »
The DAC designer went to heroic lengths to run a SD DAC without digital filter, implying that was the problem.
The user adds the digital filter in his PC by upsampling to 176.4 and gets great results.
PC based up sampling is becoming quite common.  I've read about a lot of Metrum Octave (nos dac) users doing the same thing.

Some users swearing that 16bit up sampled on the pc to 24/176.4 sounds virtually as good as native 24bit music.    I find these experiences very interesting,  it tells me that the actual resolution of the music isn't the bottleneck,  it's how the data is being handled.
It's all about the music,, not the equipment.

tuyen

  • Guest
Re: DDDAC 1794-NOS 24-192
« Reply #119 on: December 16, 2012, 08:38:19 PM »
Quote from: tuyen link=topic=842.msg12339#msg12339 date=
Oh, please tell Mr Schick that I am still waiting for my tonearm!!$#@

Me too.. 15+ months and counting :-X

Good news! Just yesterday I talked personally to Mr. Schick, and he promised me that you will receive your tonearm till the end of this year !!!

How was the meet, Reinhard?

Stefano has posted some photos and info on his blog.  Looked really enjoyable.  Who was the person from Australia?!

http://twogoodears.blogspot.com.au/2012/12/eau-de-cologne-winter-2012.html