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.
397 398 399 400 401 402 403 404 405 406
408 409 410 411 412 413 414 415 416 417
  1. KeithEmo
    There is some truth to what you're claiming, but it depends to a large degree on your requirements and priorities. Back when digital cameras first appeared, a 5 megapixel model cost $10,000 and you couldn't have a 10 megapixel one at ANY price - because they simply didn't exist. Nowadays you get one of those in a $300 cell phone.
    In the case of DACs, as with most topics in audio, it all boils down to your needs and priorities. The "early technology" DAC chips didn't use oversampling, and simply weren't accurate enough to accurately "resolve" details even to 16 bits. In fact, they were just barely able to deliver marginal performance with 16/44 audio - which was the standard chosen for CDs. This means that, by today's standards, they had relatively poor frequency response, and unacceptably high levels of noise and distortion. Today there are people who still insist that those "NOS" DACs sound better than current Delta-Sigma models, but they can only do so because they base that claim on the fact that, even though the noise and distortion performance is awful, they insist that "there is something else that is more important" that sounds better about them.
    So, if you are satisfied with a DAC that has (by today's standards) poor frequency response, and poor noise and distortion performance, and are willing to accept that it will not be able to play a modern 24/192k file at all, then perhaps, according to your requirements and priorities, "those old DACs sounded better"... but, if you find poor noise and distortion performance to interfere with your enjoyment of what you're listening to, then you're wrong.
    That claim is a bit like saying "Transportation has been going down hill for the last century - because in 1915 I could take a luxury train from New York to California, and it would have better service than the best first class airline flight today". That claim is true - as long as being able to arrive in five hours instead of five days doesn't matter to you, and you don't mind sitting in a hot train car without air conditioning.
    The reality is that modern DAC chips still aren't perfect, but a low cost modern DAC chip will still outperform a vintage one in every way that is considered to be important by most people. (The simple fact is that manufacturers really do their best to deliver what their customers ask for. The whole "conspiracy theory" that modern DAC designers are trying to sell us inferior merchandise because it's cheaper to make is total crap. The vast majority of DAC buyers asked for better frequency response, better resolution, lower noise, and lower distortion... and current DAC chips deliver all of those - AND lower cost. Audiophiles have now "decided" that there are other performance parameters that matter - and sit around pouting that, even though modern chips excell in 99 other ways, "they suck" because they missed one or two of them. Therefore, it would behoove us, rather than describing those old chips in glowing but very vague terms, we figure out the actual specifications and performance parameters involved - so they can be added to next year's $2 chips.)
    As for your other example - Windows.... that's a bad example. I'm not at all a fan of Windows 8.1, but Windows 98 was a huge improvement over Windows 95, and Windows XP was a huge improvement over Windows 98, and none of the software I need to use to do my job will run on Windows 95 or Windows 98 any more, so it doesn't matter how good or bad they were - because today they are useless.


    Think of it like travel. If you wanted to go from NY to Los Angeles in 1850,
    evillamer likes this.
  2. evillamer
    Maybe it's because those NOS dac users got used to the sound signature(even though it's rolled off or grainy or less dynamic) to the point where they love it. Give them a modern ESS sabre dac with its laser like dynamics and sharp treble, they won't like it at all. This could also happen in operating systems where the user got used to the older interface and do not like/cannot adapt to the newer interface. If must be said that despite its bipolar interfaces, window 8.1 is much more stable and secure than the previous OS.

    I am not saying that all modern dacs are bad sounding, they can be better than some of the 90s or 2000 dacs. In fact I think in 2015, with the rise in popularity of high quality lossless streaming services, more audio manufacturers are paying more attention to sound quality than ever before.

    But we should never just take the year of design/manufacture or even price as a basis to good sound/good value as there are still plenty of modern subpar products out there.
  3. hans030390
    Hoping I didn't misunderstand you. While NOS does inherently have more of a roll-off with redbook content (not an issue with higher sampling rates - even then, the NOS roll-off is more of a subtle difference), they usually aren't grainy and can often be wonderfully dynamic. Some suck, some don't. Just like I've heard way too many D/S, oversampled DACs that are compressed and flat dynamically, grey sounding, lifeless, grainy, and, well, boring or painful. And some sound awesome (mmm, give me a good Wolfson or AKM design). I actually started with the typical sort of DACs and found I much prefer the NOS sound. Or, if not that, definitely a really good R2R/multibit sound like those older Thetas. And, hey, some of those DACs actually measure excellently even by today's standards.
    slankoe likes this.
  4. BassDigger
    @evillamer To me, you seem to be kind of misunderstanding your own point. I mean that I agree with you, but for a specific reason that you aren't mentioning.
    Although I have no idea how a good modern Delta-Sigma implementation compares (I lost interest in DS years ago), historically, there seems to have been a definite jump backwards (in terms of SQ), and it has all the hallmarks of cost-cutting!
    I'm talking about the switch from R-2R to bitstream.
    Comparing late 80s/early 90s multibit players to their late 90s/early 00s replacements,.....well, there is no comparison; the older, unserviced and unmodified R-2R players sound better in almost every way, to me. And not just by a small margin!
    Today there are many who believe that a properly implemented R-2R design sounds the best, even with a format, sample rate or apparent measurement disadvantage.
    This, to me, is the best example that newer is not necessarily better. And in this case, despite all the advances in measured performance, we're still little closer to understanding why.
    (I like this quote: "If it sounds good, but measures bad, you're measuring the wrong thing.")
    Sure, I believe that D-S has its place, but not in any 'audiophile' (or studio) gear, where sound quality is supposed to be the most important factor.
    Maxx134 likes this.
  5. Xymordos
    What I believe is that natural sound we hear is not as "distortion free" as we think. By the time the sound traveled to our ears it is already quite distorted. As DACs improve, their quality improves and their distortion becomes much lower than before, and detail retrieval improves greatly too. When we hear music like this, it will sound unnatural since normally we would not be able to hear this amount of detail. Thus the old DACs with less resolution and vinyl discs are favored by most people, as less resolution is actually "more" natural in this case.
    For example, you stick a mic into the piano to record it. But when you listen to a piano normally, you don't stick your ear into the strings to listen to it.
    I remember a comment on the Chinese headphone forums on the JH Roxanne. He said that the resolution was excellent, but he felt like it was suffocating as there was so much detail that it felt like a brick wall hitting him from all sides.
    Argo Duck likes this.
  6. drez
    I recently did a level matched comparison between Metrum Hex and Invicta Mirus (using digital attenuation on the Mirus to compensate for the XLR/RCA level difference).  To be honest the two were very close on the store system...  The only consistent difference I noticed was that the Mirus resolved transients a little more sharply.  Both DAC's are admittedly on the smooth side, and the store system setup is fairly basic and may have not been transparent enough to allow me to hear differences in other areas.  I expected these two DAC's to be miles apart based ton the designs but basing my opinion from that comparison they just weren't.
    Both the DAC's were at similar price points, so I could imagine that they should be comparative in performance, so perhaps this is a factor here, with both DAC's converging towards perfect accuracy.  OTOH maybe this is just showing that the differences between DAC's really is in the bee's appendage scale and can easily disappear in non-ideal conditions.
    In the future though I will try to level match for all my DAC comparisons.
  7. haywood
    Sounds like someone is asking for a listening test between a Theta Gen V and a DC-1? Should be a slam dunk for DS, right? :wink:

    The problem isn't DS resolving too much of the original detail, it's that the reproduction method introduces digital noise. They use filters that are supposed to get rid of that noise but there's debate (obviously) about how effective that is and whether it neuters the microdetails in the sound (plankton). There are many posts in the thread dealing with why on both sides. Anyway the resolution of a 16/44 file isn't going to change no matter what you feed it into, a modem dac will still have the same bits to deal with as the hoariest old battleship. Obviously high-res sound files will be better on DS as the old dac won't decode it, whether that high-res audio on DS is preferable to redbook on multibit vintage is an interesting question.
  8. Staxton
    But what particular DACs or species of DACs are actually used in the studio, especially studios that work primarily with classical or jazz?
  9. BassDigger
    Something tells me that you already have some knowledge, or an opinion, regarding this.
    As far as I know, DS is quite the prevalent and ubiquitous technology, recording studios included.
    I think this was discussed somewhere, maybe earlier in this thread; apparently, there are some recordings, using R-2R ADCs, that synergise and really show what R-2R dacs can do.
  10. Staxton
    Actually, I have very little knowledge what DACs studios use, other than what I've tried to learn here on Head Fi (See What DACs do Professionals Use in the Studio or Doesn't it Matter?and by searching the internet. And I have absolutely no opinion as to which DAC or DAC species is better.
    What I am trying to find out is whether knowing what DACs professionals use in the studio to listen to the music they create is relevant to choosing a DAC for a particular purpose--in my case listening to classical music through good headphones--and, if so, what DACs or types of DACS, they actually use.
  11. BassDigger
    Sorry; maybe I was being a bit defensive. I'm a bit weary of ogres (mad scientists) and trolls. [​IMG]
    It looks like an interesting thread, that you've started.
  12. ashutoshp
    Good question. Delta Sigma DACs I believe. Maybe a pro audio gent/gal can back me up or correct me.
    I am pretty sure there are just a handful of stand-alone DACs south of $1000 because the money/demand is of components upstream, i.e., the ADCs rather than the DACs. Makes sense TBH. I always seem to get the notion that DACs are a foregone conclusion in the pro audio business until you hit the high price brackets. Another thing, its not easy to get the exact part number of the DAC used as well. Apparently, most of the cheaper USB interfaces use the same DAC.
    Some of those interfaces are truly divine sounding. I listened and am looking to purchase Focusrite's Forte soon. top notch stuff. Any opinions? 
    FYI, look at some DACs they review in Stereophile. Quite a few are from the pro audio side of things. Off the top of my head, there's Benchmark, of course, but also Antelope Audio.  
  13. KeithEmo
    I've been involved in "audiophile gear" (and specifically digital) for a very long time - and I simply can't agree with you there. I had a relatively early "high end" CD player - back in the days when R2R DAC chips were all there was. Honestly, it was OK (with nothing to compare it to). I replaced it with a Rotel model that was well regarded at the time - which was a definite improvement. From there I went to a separate DAC and a separate little box that did upsampling and jitter removal (Assemblage), and from there to a Benchmark DAC1. And I would have to say that, at least to my ears, each of those steps was a clear and obvious improvement in sound quality. 
    I currently own about a half dozen "high end" DACs, all of which seem to be Delta-Sigma at the moment. I recently purchased a well regarded "NOS/R2R" DAC (in the $800 price range), to see what I was "missing", and I have to say that I personally wasn't especially impressed. It sounded smooth, and seemed to do a very good job of reproducing clean plucked strings on some especially good quality recordings, but it also didn't seem to do all that well on voice. I would have to say that, when I plugged my DC-1 back in, it sounded a tiny bit better on some things, and a tiny bit worse on others - all in all about as much difference as I normally hear when switching between different digital filters on a DAC that offers that option - with a very slight edge to the DC-1.
    Since "better" or "worse" are matters of personal opinion rather than of technical fact, there is no right or wrong to be found in any discussion based on those terms, and we are all entitled to our own opinions. However, I might suggest that it would be more accurate to state that "many people hear a difference - which some people prefer". 
    I do, however, find it interesting that you are quite certain that "the older R2R units sound better to you in every way" even though you also say that you haven't heard modern D-S DACs (I'm assuming this means that you haven't listened to any of them critically, let alone made any sort of direct comparison). If I were being cynical, I might think that you simply don't like D-S DACs "on principle" because "it has (they have) all the hallmarks of cost-cutting" (your words), even though you haven't actually verified for yourself whether they have lowered cost by reducing quality, or actually do deliver better quality for lower cost.
    Are you actually basing your opinion on how modern D-S DACs sound based on your memory of how a few very early ones sounded?
    Don't you wonder if perhaps things have changed?   
    Argo Duck likes this.
  14. KeithEmo
    The VAST majority of DACs sold for professional or consumer use today are Delta-Sigma DACs. In terms of percentages, R2R DACs rank right up there between horse drawn carriages and hydrogen powered cars. (And, if you look at the parts vendors catalogs, virtually EVERY DAC chip "recommended for professional or consumer audio use" is a D-S DAC. You may find a few companies who have brought back old designs to sell to audiophile niche markets, but very few. This means that the only folks using R2R DACs are companies or individual DIYers who are willing to create their own designs, or adapt chips intended for other purposes, or build their own hardware based on old available schematics and designs.)
  15. BassDigger
    As you say, everyone has their opinion. Mine is based on old, but very convincing experiences, and the advice of someone who's opinion I trust, more than anyone's around here. (Also, he doesn't misquote me, even though English is his 2nd (or 3rd, I'm not sure) language.)
    OK, your opinion is totally the opposite and based on your "very long" involvement; it's worth consideration (I've really appreciated many of your explanations.). But, my own finding showed such a gap between the two technologies. This, combined with other testimonies, would make me very surprised if D-S has caught and passed R-2R, in reproducing music in the way that I prefer. Sure, D-S implementation has got better over the years. But so has R-2R.
    I don't understand your reference to cost cutting. I thought that it's well known that D-S dacs are cheaper; hence their prevalence.
    Maxx134 likes this.
397 398 399 400 401 402 403 404 405 406
408 409 410 411 412 413 414 415 416 417

Share This Page