1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)

Discussion in 'Dedicated Source Components' started by purrin, Dec 5, 2013.
First
 
Back
380 381 382 383 384 385 386 387 388 389
391 392 393 394 395 396 397 398 399 400
Next
 
Last
  1. Sapientiam
     
    Cool, thanks for the clarification. I've done several reconstruction filter designs for NOS DACs, they're all freely available schematics on my blog and anyone with questions for clarification can post them on the blog. It was something of a challenge but 'very difficult' it wasn't. Its certainly possible not to impact the HF significantly but becomes more difficult (primarily inductor tolerances get very tricky), in the end I've gone for flatness to within half a dB up to 18kHz or so. I don't see why a custom oversampling filter would need to be designed - off the shelf ones are available. Some glue logic to adapt to I2S might be necessary though - which is a whole lot easier than starting from scratch on a filter. I have developed custom filter code for an ARM (M0, M3) processor myself which if anyone's interested I'd be happy to have a look for on my various disks in differing states of repair... When I did this I found that simply running the DAC faster made the sound worse so I've abandoned oversampling and stuck with NOS. The filter characteristics weren't the cause of the SQ degradation which was a loss of dynamics, a 'greying out' of timbres.
     
  2. Sapientiam
     
    There's no requirement for rolled off highs, there are a few ways to correct for that. I agree plenty of NOS devotees don't mind the 'NOS droop'. However NOS DACs using the TDA1387 are jolly uncommon - over on TaoBao a few designs have shown up from one vendor in the past six months or so though, at very reasonable prices. None of those designs though implement NOS droop correction - if you'd like to hear how a TDA1387 NOS DAC sounds when this is implemented you could do worse than follow the mods suggested over on this thread - http://www.diyaudio.com/forums/digital-line-level/269199-tda1387-x8-dac-lets-check-its-design-mod-not-play-music-not.html
     
  3. purrin
    Cool. SATCH DAC all over again, but like 2.0. Glad to know it sounds better than the DACMagic (which isn't a good sounding DAC BTW).
     
  4. Sapientiam
     
    Timing issues are far more important for a certain kind of D-S DAC, the kind that does not use on-chip switched capacitor filters. SCFs have been employed by many vendors to get around the timing issues inherent with having vast quantitites (exceeding the signal amplitude for lower-level signals) of ultrasonic noise.
     
    If you study treatments of jitter one thing you might take away is that the higher the frequency a DAC produces, the more susceptible it is to jitter. In other words, the faster the signal is moving the more important timing errors become. On a D-S DAC the output ultrasonic noise is one to two orders of magnitude in frequency above the audio, hence one to two orders of magnitude more susceptible to jitter. Add in the fact that the ultrasonics are much higher amplitude relative to the wanted signal when the DAC's playing back quiet stuff (say < -40dBfs) and you do need to be very careful with output sample timing not to 'fold back' ultrasonic noise into the audio band.
     
    Multibit DACs have no such susceptibility - even unfiltered NOS ones - as any ultrasonics produced track the signal level, they don't go beyond it.
     
  5. evillamer
    Interesting read for those interested in history of digital audio:
    http://www.soundfountain.com/amb/cd25years.html
     
  6. wink
    Quote:Tusco 1965
    1. This is SO wrong. Until the equipment becomes perfect, the journey continues.  [​IMG]
     
    2. Sacrilege.  Hand in your 'Headphoneus Supremus'  title immediately.   [​IMG]   [​IMG]     Jose R has only 25 posts in 10 years. He needs to be encouraged to seek musical heaven[​IMG]
     
     
    I believe a severe castigation by way of a PM by a moderator is in order here.  sarcasm_on-1.gif
     
    johnjen likes this.
  7. BassDigger
     
    Firstly, thanks for taking the time to write such a comprehensive and informative explanation.
     
    I guess my terminology was incorrect; "prone to" does suggest that the dac is the cause of the jitter/errors. You're saying DS is more sensitive to them. TBH, I wasn't sure which.
     
    But your explanation does seem to suggest that maybe my severely limited understanding, of the basics, wasn't totally incorrect.
     
    I don't know that anybody thinks that you can actually hear the jitter frequencies. But, I guess, what's in question are the effects of jitter that are audible.
    This understanding seems to be changing as time goes by. Originally, it was totally mis-understood, unknown even! But over the years it seems that engineers are discovering that it's effects are ever more influential to the sound reproduced.
    You seem to believe that jitter can relatively easily be kept so low as have no influence on the sound, even for the sensitive DS designs. I'm not so sure that other learned people would agree, today or in the future.
     
  8. vhsownsbeta

    Ha. Over my head :p
     
  9. Articnoise
     

    Isn’t it better to double the oversample rate like 44.1K -> (88.2K ->) 176.4K instead of go 44.1K -> 192K? My somehow limited understanding and experience is that you should always double or triple the oversample rate in even numbers, no matter if the oversample takes place in the DAC or in a PC. Some DACs are even using two different clocks; one for 44.1, 88.2, 176.4, 352.8 and another for 48, 96, 192, 384.

     

    Btw I really appreciates your posts here. 

     
  10. BassDigger
     
    Firstly, I wonder if anyone can direct me in the direction of a good explanation of oversampling (and digital to analogue conversion, in general); I've been trying to get my head around this for years!
     
    I'm sorry if I'm nitpicking. But do you mean 'upsample'? And if so, does this mean using a program that 'adds' improvised data to the real music data?
    Please tell more; my music collection is redbook; I'm very keen to know of ways that I can get the most out of it.
     
  11. RoundRound
    I have a general question about bit depth and digital volume.
     
    I was looking into it as my Yggy is plugged directly to a power amp and I'm using Jriver to attenuate the signal around 50dB.
     
    I understand that lowering the volume reduced the ENOB (effective number of bits) but then I realized that if I'm listening at around 70dB (via speakers) than in any case I won't hear any sounds quieter than 70dB which is around 13Bit...? Is this correct? If so why bother with 24 or 20 bit Dacs?  Am I missing something?
     
    Thank you,
     
  12. Bibo
    Actually the dynamic range you hear in this case is only  40-50 db max as the noise floor of a very silent room ist a least 20db.
     
  13. RoundRound

    Thanks,
    So what's the point about true 20 bit DACs if we barely hear 8 bit on the real world?
     
  14. BassDigger
     
    Read back a few pages; this topic has already been discussed, at length. 
     
  15. RoundRound
    Thanks mate,
    That's a 380 pages thread, can you perhaps point me to the relevant page?
     
First
 
Back
380 381 382 383 384 385 386 387 388 389
391 392 393 394 395 396 397 398 399 400
Next
 
Last

Share This Page