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.
180 181 182 183 184 185 186 187 188 189
191 192 193 194 195 196 197 198 199 200
  1. prot
    that is mostly a preferences comparison.
    In terms of numbers vinyl is about 10bits. See http://wiki.audacityteam.org/wiki/Bit_Depth#sources. Some HA threads put it at 12 bits. That's not 100% clear math but I would say it's good enough.
    Also 1bit depth ~= 6.02db SNR ... or dynamic range which is approx the same thing. How much is audible above 12bit depth depends on lots of factors: ears, gear, recording quality, environment. E.g. you'll prolly never hear any improvement above 12bit in a car.
    Anyway, I wont care much about those bits. If it sounds good to you, enjoy and nevermind the numbers.
  2. Articnoise


    Not really that I was asking about, but thanks anyway.

    I asked about the more resolving equals’ better accuracy theory. I have never heard of this correlation before.

    To be clear I’m not saying I know much about dac designs, but to me a god vinyl spinner is superb on accuracy, especially if we talking about tonal accuracy. To me accuracy means true to the source - less artificial. But as you said vinyl is less resolving than a good CD, so what does it mean to the more resolving equals’ better accuracy theory?

  3. purrin
    Indeed, digital sounds so different from analog that it makes comparisons difficult. An entirely different language needs to be used with vinyl. First of all, an LP12 (unmodded) sounds nothing like an unsuspended VPI. Second of all, what cartridge and phono-pre? Also what arm? Cart, phono-pre, arm, and table must be thought of in one unit.
    All that side, I do feel that a properly setup TT system for about $1k, assuming equivalent recordings will easily beat something like a Gungnir in microdetail, microdynamics, immediacy, fluidity, but fall short of something like a Theta Gen V. That's were my Project TT setup stands right now. Somewhere between the Gungnir and the Theta Gen V. It's really too hard to compare because digital does some things really really well.
    While vinyl does have much more limited dynamic range than 16-bit digital, the steps within that range are not quantized into discrete steps. 16-bit digital has 65536 discrete steps of volume. The grooves on vinyl, the rises and falls, the peaks and troughs, do not take discrete steps. How this translates into DAC accuracy I cannot answer.
    It's funny because as I was putting the v0.99 iteration of Yggdrasil through its paces last night, I was saying to myself wow, this DAC really gives me that sense of pure immediacy and enjoyment when I was a kid (we only had records back then.)
  4. purrin
    Wrong way around. I didn't mean resolving = better accuracy.
    I mean better numerical accuracy of D-A conversion -> more resolving.
  5. prot
    sorry that was all I had :)
    Guess it depends on how you define accuracy. In general more resolving (detailed) should be same as better accuracy. But if you talk about tonal accuracy all bets are off. Doubt there is any direct relation between the amount of sound detail and tonality. And I also put tonality first because "crispy" highs do hurt my ears. E.g. a senn 800 produces much more audible detail than hd600 ... but after 30mins or so I feel like someone is piercing my ears with tiny needles. So screw the 800 and all that detail, I wanna enjoy music.
    Judgung by the available iggy descriptions, it's one of those rare components that gets both detail and tonality right. Curious to hear that.
  6. Articnoise


    Ok, when it make better sense to m, lol. I thought you there meaning another type of accuracy.

  7. Articnoise


    I value good tonality very high as well, but I’m not sure that more resolving will lead to a less good tonality.

    I’m with you on the “I wanna enjoy music”!

  8. Clemmaster
    I'm not sure your ear can resolve a signal at -90dBFS when there's another signal 70dB higher in level at the same time, due to the combination of masking and ATH. In a sense, being able to resolve at both extreme of the DAC's scale, at the same time, is probably not necessary.
    What if the totalDAC didn't use a linear scale? It would be able to both 1) have a great dynamic range (140dB, as reported by someone; couldn't find that on the site, at first glance), 2) have an excellent low level resolution (as shown on the website: the -100dB sinewave looks extremely clean), but not at the same time.
  9. purrin

    What I am talking about is how errors in DACs (yes, they do not perfectly translate digital codes into signals) effectively lower the number of bits. For instance, there is a DAC's resolution and then there is error which lowers the effective number of bits. Let's take the example of a hypothetical 14-bit DAC with +/-16 LSB INL error with output from from -8.192 volts and 8.191 volts.
    This 14-bit DAC has 16384 discrete values from -8192 to 8191. When the code is 0, we get 0V output. When the code is 8191 we get 8.191V output. You get the idea. 
    Now because the error spec is 16 least significant bits either way, the output of the DAC may be off by that much for any given code. For example if the code is 0, where we should get an output of 0V, we could end up with an output as far off as 0.016V or -0.016V, or anywhere in between. For code 4000 where we should get 4.0V output, we could get 4.012V, or 3.997V, 3.984V, etc. Heck, the next code 4.001 could even result in a smaller output value than the one before. This would be considering something that isn't monotonic. Motonicity = when stuff is supposed to get louder, it gets louder, not softer. 
    In essence, we still have 14-bit resolution, but the accuracy is not guaranteed. Because the accuracy is not guaranteed, any time we want an output, the value is going to be off, in this case as much as 0.016V either +/- way which means 0.032V which translates to 32LSB. Since 2^5 = 32. 5 bits. What we have is 14 bit DAC with an effective number of 9 bits.
    Why is this spec important? Because if the guidance computer of a AMRAAM tells the motor of a control surface to rotate 1.232% degrees, it better not be 1.217 degrees. Engineers need to know what the effective number of bits is, or what bits they can count on so they can properly design a weapons system according to specifications. Now does this translate into audio? Mike and Jason seem to believe so. I wasn't sure about a year before, but now I am fairly convinced, especially after hearing craptastic DACs with known bad specs, mediocre R2R DACs with dubious specs, and the Yggdrasil with known unbelievably great specs.
  10. Sorrodje
    Ok , I understand where You come from
    Hope you'll get a chance to give a listen to a totalDAC yourself to see if it change you mind or not. I know I'll do so for the Yggy and compare it against the current dac i'll own when I'll have the money to purchase the Schiit DAC.  Maybe it will be this totalDAC A1 , maybe something else if I'm not convinced but the A1 in the long term.
  11. prot

    Well, the "accurate" transistor radios of my teens never did sound as good as my granpa's tube monster. Never had the slightest trace of fatigue from that thing in spite of the ridiculous amounts of hiss and cracks.
    But every time I think about getting a turntable, I remember how couch-friendly digital is :)
  12. zerodeefex
    I think you will be very surprised at how much information is contained in redbook content. When I listened to the Yggdrasil, it had been after about two years of bugging Marv and demoing DACs in the 15-20k or less range. The Yggdrasil pulled information from familiar recordings that I had NEVER heard before the sense of realism was ridiculous.
    I'm curious, what DACs have you compared to the TotalDAC? What gear and with what recordings?
  13. purrin
    I actually have not discounted the totalDAC at all. As I've already said, I expect it to be fantastic sounding. It's just more fun to argue with Arnaud than anything else. He says closed-form digital filters and DAC accuracy are "blathering". I say 10000euros for the privilege of 14 effective bits is bull testicles. :) 
    I very much appreciate that the totalDAC designer went through with the effort to create a discrete resistor ladder. I hope more people make multi-bit DACs, discrete parts or not. The lower the cost, the better. There's no reason why HFers who are willing to spend up to $1500 or so should be limited to a bunch of Chinese Sabre DACs being shilled by people who get to keep their gifts "evaluation" units.
    In the end, it's important to know the true enemy of good sound.
  14. prot

    You guys are surely doing a great job at promoting the iggy. It already sounds like the second coming of digital. And I gotta admit I'm mighty curious already.

    May I ask what makes you different? Cause it surely sounds like you'll enjoy keeping your evaluation iggy :)
    I'm definitely grateful that you are testing all those dacs for us but lets not pretend that you are doing charity work for the god of sound.
  15. Sorrodje
    I can't agree more[​IMG] .  And I wholeheartly hope that the 2300$ Yggy will really be the statement dac we all wait. 2300$ is a really a big amount of money but it stays affordable for hobbyists who are patient enough to save the money. MSB, totalDAC offers are way out this area.
180 181 182 183 184 185 186 187 188 189
191 192 193 194 195 196 197 198 199 200

Share This Page