Can an 1 bit ADC can mask a DAC sound signature ? - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 5th March 2014, 05:05 PM   #1
Eldam is offline Eldam  France
diyAudio Member
 
Join Date: Sep 2012
Location: At home, sweet sound home...
Default Can an 1 bit ADC can mask a DAC sound signature ?

Hi,

As we can have very stable clocks nowadays :

Why it is not possible to stack few TDA 1541 to improve the resolution (let say 24 bits) then convert the analog output with an one bit delta Sigma ADC to have the opportunity to go out with a small low pass filter (one or second order) and take advantage of the details with non masked sub harmonics but without the glitchs ?

Do I mix up with theories or do an ADC in front od a DAC can mask its sound signature ?
  Reply With Quote
Old 5th March 2014, 05:32 PM   #2
diyAudio Member
 
Join Date: Apr 2003
Location: Eastern Pennsylvania
Quote:
Originally Posted by Eldam View Post
Hi,

As we can have very stable clocks nowadays :

Why it is not possible to stack few TDA 1541 to improve the resolution (let say 24 bits) then convert the analog output with an one bit delta Sigma ADC to have the opportunity to go out with a small low pass filter (one or second order) and take advantage of the details with non masked sub harmonics but without the glitchs ?

Do I mix up with theories or do an ADC in front od a DAC can mask its sound signature ?
Yes, it sounds like you are mixing up theories. Simply paralleling DACs does not increase bit resolution (lower the quantization noise-floor), it lowers the thermal noise floor, which is a different thing. If I correctly understand, you are proposing to convert digital to analog via TDA-1541s, then convert analog back to digital via 1-bit (DSD?), and finally convert DSD back to analog, all under the notion that this will somehow unmask hidden detail. Can you explain how you came to such a convoluted notion?
__________________
Ken

Last edited by Ken Newton; 5th March 2014 at 05:38 PM.
  Reply With Quote
Old 5th March 2014, 06:18 PM   #3
Eldam is offline Eldam  France
diyAudio Member
 
Join Date: Sep 2012
Location: At home, sweet sound home...
Thank you Ken to answer me, I believe my question do be very funny for a personn with a perferct understandings of theses technologies...

Ok I thought increasing the bit rate could increase resolution because more sequence in a same time length recording. And then stacking DACS of lower bits could give one of higher bits... feel ashamed (but will survive of this day because no Sepuku adept and clichés proof !)

I try to explain to you how i had this bad idea, let's continue for the fun, amusment (do I have to return my fellow card ater ? :

Well, because my bad understanding come from :

- some old ECL multi bits DAC chips seem to sound gound but have harchness in highs because their low wide band (44 khz, 96 khz and Sha/Nyq low = divide by 2....). with 44 khz you heard maxi 22 khz but the brutal low pass filter give the known harchness with these DACs.

- To get low with quantification noise avoiding cuting all details and their sub harmonics in higher range you must apply a strong slope.... 20 db/octave ore more

- believed the to have One bit permit to apply a very low slope (first or second order) in analog output to supress quantification noise but without shorting the width of the band in analog band (if you have for example a 96 khz Dac chip :the analog band is 43 khz, then you get the little sub harmonics of the higher notes but because "magic one bit": quantification noises can be stopped with a low slope LP without affect the width in the analog domain !

- My understanding is Delta Sigma needs more precise and stable crystal XO, and it's easier nowadays ?!

- But Cmos technologie of suchs ADC could mask the benefits of this old good ECL technologie (urban legend?) used in DACs like TDA 1541 if ADC in front of the DAC (in the last hope to profite of the holy one bit to apply the slow slope in analog domain after the analog conversion (need one more 1 bit DAC after the 1 bit ADC seemed to be the Achile's heel of my bad theory but believed there was one chip device with ADC to DAC in one Bit !)

Voilà you know all !

The beginning of these big paella was :

- the SACD with its big and smooth details in highs have its DATA in one bit in the disc itself. Why not try to emulate to have the advantages with our old 16/44 or 20 bits discs

- Why the R2R, ECL, Multibits are famous to have more acurate tones than moderns Cmos chips try to match with difficulties. Is it just because with old chip you can manage the I/V conversion and analog output design yourself and the modern one have already operationel amplifier inside ?

- If the problem is the lake of bandwidth because the quantification noise need a high slope early (which can be heard, ears prefer a slow attenuation with multiple subharmonics and not only a second harmonics then cut with strong slope)

- Saw some fellows like Abraxalito work with multiple low bits DACS (10 bits, 14 bits)

- then return to the beginning with funny but wrong theory question because lake of knowledge !

Hé you know french cooking I mix all ! And when cooking is a second nature you don't know where to begin when you attack an another knowledge domain !

Last edited by Eldam; 5th March 2014 at 06:27 PM.
  Reply With Quote
Old 5th March 2014, 11:40 PM   #4
diyAudio Member
 
Join Date: Apr 2003
Location: Eastern Pennsylvania
Please permit me to say, that I did not intend to cause you embarrassment regarding your technical knowledge. In my opinion, there are few true technical experts here, and I am not among them. Most of us, including myself, are still moving from some level of technical ignorance toward true expertise.

Regarding your reply. While your English is infinitely better than my French, I must still admit to having difficulty following the wording of your explanation. I hope, you'll forgive me if I am unable to craft a response.
__________________
Ken

Last edited by Ken Newton; 5th March 2014 at 11:55 PM.
  Reply With Quote
Old 6th March 2014, 12:17 AM   #5
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 99
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Quote:
Originally Posted by Eldam View Post
- Saw some fellows like Abraxalito work with multiple low bits DACS (10 bits, 14 bits)
Where was that then?

On my blog I've speculated that a future 'ultimate DAC' might be built from say a 14bit chip. That's because that's the highest number of bits that goes really, really fast (communications DACs). I still have to try out my AD768s which are 16bits comms DACs.... Don't think I've ever advocated 10bit DACs though.

Multiple 16bit DACs, sure - TDA1545s galore in my earlier design efforts. That was when I thought linearity mattered a lot. Now I've found power supplies matter more, so paralleling chips has gone on to the back burner for the time being

P.S. Like Ken I can't quite figure out what you're trying to do here but its an interesting conversation anyway....
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler

Last edited by abraxalito; 6th March 2014 at 12:20 AM.
  Reply With Quote
Old 6th March 2014, 12:34 AM   #6
kouiky is offline kouiky  United States
diyAudio Member
 
Join Date: Jun 2009
It seems to me that the methods Eldam is referring to, were used in certain Marantz and Naim CDP's. The venerable CD65II and CD-80 come to mind, using twin instances of the highly regarded and desirable TDA1541. Instead of a typical three stage sequence consisting of a DAC, I/V conversion stage, and a line level output amplifier, there tends to be a 1-Bit conversion module inserted into the mix. I would like to know its purpose, too.
  Reply With Quote
Old 6th March 2014, 12:40 AM   #7
Eldam is offline Eldam  France
diyAudio Member
 
Join Date: Sep 2012
Location: At home, sweet sound home...
I am not embarrassed Ken. I was joking. You know I have a sentence for that : "A walking fool progresses more than a seated inttelectual".

I hopped you can follow my though as you asked but both my english and the confusion with my lake of knowledge let you give up ! I understand...

Sorry... not for the bad question but for my bad English...

@t Braxie : sorry, never said you ever advocated 10 bits DACS, but maybe I read in your blog something close...Anyway often don't understand subtilities because of my English.

Last edited by Eldam; 6th March 2014 at 12:46 AM.
  Reply With Quote
Old 6th March 2014, 01:09 AM   #8
Eldam is offline Eldam  France
diyAudio Member
 
Join Date: Sep 2012
Location: At home, sweet sound home...
Upon my awfull English I think my question can't be understood because I mixed up with concepts and have a bad understanding of what a bit and sampling frequencies are !

Anyway, multiple devices are often a bad ideas in audio.

I was just trying to understand if an "undersampling" in Delta Sigma mode after a good sounding multibit DAC can increase its softness in highs frequencies. Both because the first multi DAC has a known good sound signature and because you can apply a soft slope low pass with 1 bit Delta Sigma. The best of two worlds to do short ! The first technologie has a good sound, the second offer the best way to have better détails with softness especially in the trebles (sacd).

It's a bad focus of me between sound impression and electronic theories. Even if what i wrote could be possible I suppose the sound quality at the end come from the last device chip and One bit ADCs/DACs are not famous to have the best sound.

Mixing up ECL, then after cmos architecture and in the same concept: adding after the first DAC one more level of conversion with an another analog to digital then digital to analog convertion with an another technologie (Cmos, One bit) is a crasy bad idea. Difficult to focuse in a knowledge when you can't master it ! I was hopping for example than taken the current output from a multidac chip like TDA 1541 or AD1862 or taken its voltage after the simplier I/V solution and then re convert it to one bit in digital then after analog with passive one order low pass filter could keep all the qualities of the chain !

Forgett what I wrote and thanks for your kinds answers... Drinking a bad whisky near the Scotish sea can not emulate it in Caol Ila !

Last edited by Eldam; 6th March 2014 at 01:33 AM.
  Reply With Quote
Old 6th March 2014, 03:14 AM   #9
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 99
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Get a fairly sharp LPF on your NOS DAC would be my advice. Tamed the HF roughness for me perfectly. No need to go over to the 'dark side' of S-D to get smooth treble If there's enough demand for filters on NOS DACs then I suppose I could start a thread....
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler
  Reply With Quote
Old 6th March 2014, 04:13 AM   #10
diyAudio Member
 
Join Date: Dec 2008
I rememeber that ECdesigns told in his thread that the human auditory system (the ears) is in fact a LPF filter itself so his theory said it's no need of any filters after NOS dacs; the only thing I don't seem to like on my TDA1541 NOS dac is the small treble roll-off comparing to the OS in the same output scenario; can this be compensated somehow?
__________________
== There are no limitations to the self except those you believe in ==
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
24-bit ADC to 16-bit soundcard input, SB Live 5.1 with PCM1804. elektryk Digital Line Level 0 10th December 2013 09:40 PM
DAC related "sound signature" CENTRAL Digital Source 38 30th June 2013 10:32 AM
FS: Arda AT1201 ADC / ESS ES9018 Sabre 32-bit DAC chips / THAT 4301 promixe Swap Meet 8 25th March 2012 03:09 AM
24 bit DAC and ADC questions - What are you using and liking? 777funk Digital Line Level 4 16th May 2009 01:07 PM


New To Site? Need Help?

All times are GMT. The time now is 05:50 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2