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

Some questions about settings (sample rate, latency, etc.) for an external USB DAC running off of ASIO

Discussion in 'Sound Science' started by goodyfresh, Jul 30, 2015.
2 3 4 5 6 7 8 9 10
  1. goodyfresh
    So, I am fairly new to the audiophile world, and am enthusiastic to learn as much as I can about how to get the most bang for my buck out of my equipment.  I currently listen from my laptop using my Fiio X3 2nd Gen DAP as a USB DAC to drive my V-Moda M-80 headphones.  Sounds great, to be honest.  Im' wondering about some settings available in Windows as well as the TUSB ASIO control-panel.  In Windows, for the output to the DAC, I can set the bit-rate and sample-rate to various settings. . .16-bit/44.1 all the way up to 24/192, of course.  One guy in another thread told me I should ALWAYS have it set to 24/192 "for the best sound."  That sounds like hokey to me. . .IS it hokey?  If not, or if it is, what SHOULD I set the sample-rate and bit-rate to for the output?  Should I perhaps VARY it depending on the file being played, so as to match it to the original sample and bit-rates of the FLAC files I am playing?  or will it make no difference whatsoever?

    Okay, so now my second question is about the "Buffer" settings in the ASIO control panel for the USB audio.  The same guy who told me the thing about sample-rates above, told me that I should change the USB Streaming settings from the default, "extra safe," all the way to its polar opposite, "minimum latency."  What does "safe" mean in this context?  Does better latency come at the cost of some kind of risk to the components???  What SHOULD I set it to, and why?  And, should I set the ASIO buffer size to teh highest possible, 8192 samples, or the lowest, 64 samples?  Or should I set it somewhere in-between?
  2. PurpleAngel Contributor
    Audio CDs are 16-bit/44.1K, so any audio files (mp3 or FLAC or other) ripped from an audio CD would not benefit being ripped (created) any higher then 16-bit/44.1K.
    So setting your bite-rate or sample rate any high then 16-bit/44.1K would not enhance music audio.
    There are a few music audio tracks that are higher then 16-bit/44.1K.
    Like my Pink Floyd Endless River, which came with a special 24-bit/96K disk, along with the normal 16-bit/44.1K audio CD.
    And websites like HDtracks, which sell high sample rate music files (up to 24-bit/192K).
    My newest Blu-ray action movie disks come with 16-bit/48K audio tacks.
    So can't see you having a benefit from setting the audio to anything higher then 16-bit/48K.
    For music audio, try using the free program Foobar2000, with the WASAPI component installed.
  3. goodyfresh

    That's the thing though.  Some of my music IS high-res FLAC originally sampled at 24/96, such as my copies of Beethoven's Symphonies (expertly played orchestral/symphonic music is, of course, ESPECIALLY able to benefit from higher bit-depth and sample-rates), for example, thanks to a buddy of mine a while back who let me get some stuff off of HD Tracks as well as a few other similar sites once he wasn't using his accounts anymore.  And so, I'm wondering if I should be SWITCHING the output to my USB DAC to match the depth and rate of the files currently being played?

    And then, um, what about the ASIO settings?  I'm even MORE confused about those than the bit-depth and sample-rate, because I have less of an understanding of what they are actually doing.

    Also, since I use a USB DAC that is utilizing ASIO for its digital input from the computer to the DAC, I've quit using WASAPI, since the ASIO is giving me bit-perfect output already.  I've actually tested Foobar2000 with WASAPI output being fed out frmo it into my DAC, and then switched back to the normal ASIO output into the DAC, and it made no audible difference whatsover.

    Edit:  Btw that's pretty schweet that you've got some Pink Floyd in hi-res.  Of all Classic Rock bands, they had some of the most detailed and sophisticated music out there, I'm sure that on audiophile-grade equipment it must sound GREAT in hi-res :)
  4. Music Alchemist
    Nope, the only thing that matters is the master. If you convert 24-bit files to lossless 16-bit / 44.1 kHz, they sound the same. In all likelihood, you also wouldn't hear a difference if you converted to 256 kbps AAC.
    Read this: https://xiph.org/~xiphmont/demo/neil-young.html
  5. goodyfresh

    Hmmm, really?  Judging by your stats/records here in the forums (over ten-thousands posts, really???), I think I'll take your word for it on that :)

    So what about the buffer settings in the ASIO control-panel?  I'm honestly WORRIED about those, guys.  Mostly because they use the word "safe" in some of them >_<
  6. Music Alchemist
    Yeah, I've tested this extensively. That article I linked you to goes into great detail on the subject as well.
    (Oh, and about the 10,000 posts thing...I actually made that many in a single year and started a thread about it.)
    Which Windows audio player do you use? If you use foobar2000, the settings I use are here.
  7. goodyfresh
    See that's the thing.  As I was saying above, I don't bother using WASAPI anymore because the particular DAC I am currently using recieves bit-perfect input anyways through an ASIO driver, so when I do use Foobar I just have it output in ASIO to my DAC :)
  8. Music Alchemist
    I linked to that due to your question about buffer size.
    I was reading the JPLAY manual and noticed it said something somewhat relevant to this discussion:
  9. castleofargh Contributor
    you can read a lot of crap about the different bit perfect solutions, and the only rational answer is: use what works.
    if your DAC has specific drivers using asio, go for asio. end of the story.
    many people are using asio4all( buggy stuff if you ask me) only because they have seen some people saying how asio sounds better or whatever. they're bit perfect or they aren't, if 2 bit perfect solutions really sound audibly different then one is not working (I got that secret intel from captain obvious).[​IMG]
    about asio latency or any latency really, it's very simple, small latency forces the computer to work extra hard to do it all faster(only subjectively faster because it's still just waiting for the next sample like an idiot after doing one. it's interesting only if you have several sources playing the same thing. ok that's not clear, let's say you watch a movie and the sound goes through something with latency, the image and sound will be out of sync by that latency(most video player have a latency correction so no biggy even with crazy high values from complex DSPs). same if you're a million dollar maniac gamer, you may believe that saving 5ms will be what makes you the best ever(rarely delivers on stats TBH but sometimes all you need is believing ^_^).
    now if all you do is listen to music, nobody cares if the sound comes to your ear at time X or 20ms later. as then all the sound is delayed by 20ms and nothing is different. on the other hand, your CPU will thank you for a bigger buffer, even more so if you're doing resource taxing stuff at the same time you play music.
    all in all, going too small on the buffer can lead to actual problems(it might not but it could). and using the maximum buffer won't do much, if you were to have settings like 10seconds of buffer, now that would be dumb, but the standard values are very fine and nobody should worry about that unless you're some pro, mixing different sources at the same time or something.
  10. goodyfresh

    Well I think I can probably afford to set it to minimum latency, then, if only because my laptop has a quite beastly processor (it's a HP Envy with a Intel i7 Quad-Core i4720HQ, base-clock per core at 2.6ghz and turbo-boost up to 3.6ghz).  The processor alone is worth almost 400 bucks, which is more than many people pay for an entire laptop :wink:  I'm trying it now and yeah it can handle the lowest latency settings with the largest buffer as smoothly as the exact opposite settings, so. . .

    Quick question though.  If the latency is up to, say, a few-hundred milliseconds, couln't that dramatically (negatively) affect the experience of watching movies and shows????
  11. goodyfresh

    Um, how do I use Kernel Streaming in something like PotPlayer or Foobar2000?
  12. Music Alchemist
    Here's one benefit for me: having a buffer and loading entire audio tracks into memory puts less stress on my hard drive array (which is designed for nonstop operation anyway, but might as well be nice to it) and possibly my computer too. Video players are so screwy! Even with a gaming laptop, sometimes I have to mess with far too many settings to get HD videos to play properly.
  13. Music Alchemist
  14. goodyfresh
    Okay then so guys, in all seriousness, which SHOULD I use in Foobar2000. . .the TUSB Asio driver that comes with my USB DAC, WASAPI Event or WASAPI Push output to the DAC, or Kernel Streaming to the DAC? 

    And, in the ASIO buffer settings for the TUSB Asio driver, should I set it to minimum latency and the largest buffer, or minimum latency and moderate-size buffer, or what, for the smoothest experience with, say, watching movies?  What about the smoothest experience for music in players other than Foobar (PotPlayer, let's say).  And finally, should I go in the Windows Sound Control Panel for the DAC and set the output to the DAC to match the bit-depth and streaming to match those of the file being played?  Of all those questions, I'm only expecting an actual yes/no answer for the last one, the rest are probably more subjective and I'd just like a definitive statement of what YOU guys would do, PERSONALLY, in my situation. . .
  15. Music Alchemist
    My suggestion is to experiment and see if any of them sound different to you.
2 3 4 5 6 7 8 9 10

Share This Page