[mix4fix]

In Windows, when you connect a DAC, does adjusting sample rate/bit depth make any difference beyond a certain point?

[nlitworld]

Yeah I have noticed a difference in the past when that accidentally resets back to 16/44. As soon as it adjusts back to max res then things are right at home. Find out what your dac max res input is, set it at that, then in your media player, try to use exclusive mode audio output so it bypasses all windows tinkering. Another good performance enhancer is if using foobar (or any other program also could benefit) to also go into task manager and adjust the resources priority to real time highest priority. You wouldn't think something that simple could yield an improvement, but it really did for me.

[Vincent Kars]

Bit depth is simply the arithmetic precision of the data path  between the media player and the DAC. Set it to the max.
Set the sample rate to what is common: 44.1 kHz for audio (CD quality) or 48 kHz (Video).
If you use a media player supporting WASAPI/Exclusive, you can simply ignore these settings as the whole win audio is bypassed.
Bit more detail: https://www.thewelltemperedcomputer.com/SW/Windows/SRC.htm

[CherylJosie]

In short, no, with most audio gear you won't hear any difference.

Another forum has extensive A/B/X testing on various bit depth/sample rate and here's what they found experimentally:

Nobody reliably detects any difference in bit rate and sample depth when listening in a well engineered home theater etc.

The ambient noise floor in such an environment already exceeds the noise floor of -96dB for undithered CD audio. For properly dithered CD audio, the noise floor is -120dB and that approaches most people's threshold of hearing while they are still quite young.

The listening environment, even in a soundproofed room, is already contaminated and masks the noise floor of the sampling format in nearly every case. In a non-soundproofed room, the noise floor can be as high as -45dB. Properly dithered 8 bit audio is nearly adequate for such an environment. In a car, the s/n ratio is even worse.

When using professional quality headphones, some people can hear a little bit of pre-echo/ringing when listening to sample rates below 96Khz. Not everyone will notice. This pre-ringing/echo is an artifact of the extremely sharp brick wall anti-aliasing filter. It requires a longer processing chain to realize a very sharp cutoff in the stop band, and that long processing chain increases the duration of the artifacts from the filter to the point that people with very good hearing who know what to listen for can reliably detect these artifacts. For most listening, nobody is going to notice the pre-ringing.

The true advantage of higher sample rate and bit depth is in the mixing and mastering process. Very low quantization noise and very low filter ringing can make an audible improvement when mixing and mastering multitrack audio where all noise sources add up. Also, since dithering is always the last step in mastering a CD or other audio signal, the effect of noise-shaped dithering is to push the quantization noise up above the maximum frequency that humans can hear. The improvement is only noticeable because the noise floor has been shifted out of our bandwidth. It's still there, but we can't hear it as much. Dogs might.

There is at least one person who claims that sample rate clock jitter can be audible, and he's demonstrated that mathematically, although I don't know if he's actually conducted reliable double-blind A/B/X testing to confirm this claim. He has tested many home theater receivers and claims the clock jitter is bad enough on some of them that they aren't actually audiophile quality. The jitter appears as high frequency tones in the output that are almost certainly masked by the original signal because they aren't there unless the original signal is also there, and their magnitude is also dependent on the magnitude of the original signal. I don't think most people would notice, providing that anyone can notice.

For the average audiophile, especially older people who already have substantial hearing loss/tinnitus/distortion from age, medications, and health issues, audiophile quality doesn't mean much. It's more about the bragging rights than it is about the listening experience, unless we are talking multichannel and then there's a lot of difference. The most important thing in the signal chain is always the speakers and then it's also the room acoustics that can be even more important than speakers if handled improperly. After that, we can relax our concerns, unless we're talking about tape or vinyl or analog broadcast, and then that's another audible difference that we fortunately don't have to deal with any longer if we don't want to.

Very long speaker wire can also induce audible deficiencies if the gauge is too light. 12 gauge is good for up to 50'.

That pretty much covers it.

A/B/X listening tests have revealed a lot of what passes for wisdom to be nonsense 'cuz science. It's economical to have a fully functional sanity checker on board when shopping.

Cheryl's Axiom: By the time the average audiophile can finally afford the system of choice, they can't hear the difference anyway.

[WGH]

I agree with everything Cheryl wrote, most times (everytime?) you won't hear a difference. Those expensive 24bit/96kHz hi-res albums don't really sound any different than the same 16bit/44.1kHz album bought from Bandcamp for $10. Adjusting the bit depth for native hi-res recordings may make a difference but I haven't experimented with the Windows mixer in over 10 years.

Once the Windows mixer is bypassed using WASAPI or ASIO drivers then interesting things begin to happen. Then again, as Cheryl wrote "with most audio gear you won't hear any difference."

It's when you get beyond "most audio gear" that differences can be heard. Almost all DACs oversample, it's not a bad thing, noise is moved up where it can be easily filtered out. Music sounds better. But the DAC chip's native, undefeatable upsampling makes it extremely difficult to compare different sample rates, bit depths or types of dither. All music ends up sounding the same.

Ted Smith, PS Audio Lead Designer, talks about common DAC chips at 4:20 in the video
https://www.youtube.com/watch?v=xafLYw6EuZ4


A few DAC's don't oversample. Non-Oversampling DACs, as long as they are able to play hi-res will reveal differences. Even then the DAC will have to be carefully chosen. The highly regarded NOS R2R Denifrips DACs oversample in the background which will obscure differences.

"This shows that the [Denafrips] DAC is unfortunately NOT actually NOS, but instead oversampling in a way that mimics a NOS output.
Similar things are done on the RME ADI-2’s ‘NOS’ filter for example (though they’re quite transparent in the manual about what that filter is.)
https://goldensound.audio/2023/01/08/denafrips-pontus-2-12th-anniversary-edition-measurements/

A DAC's noise floor can be reliability measured, I wouldn't be surprised if the noise floor of all DACs made today is -120dB. By that measurement alone there is no need for hi-res audio. But there is something else going on that is hard to quantify but easy to hear with hi-res or upsampled music: a 3-dimensional image with space around it, the inherent flatness of 16bit/44.1kHz disappears and is replaced by an performer between the speakers. Note this is with naturally recorded music, studio recordings will still be flat.


I use the HoloAudio May KTE DAC and upsample all music to DSD256 using HQPlayer and a custom made ultra low noise music server. Hapa Audio interconnects excel at low level retrieval and a REL sub goes down to below 20Hz. This is no longer "most audio gear" and it took a long time to assemble not to mention afford on a woodworkers wages.

I may be delusional but I'm in good company. Spend a little time in the HQPlayer forum and you will either decide all these people are nuts or they may be onto something. A huge thread, start at the end and read backwards, then jump around.
https://audiophilestyle.com/forums/topic/19715-hq-player/#comments


If you want to learn more about filters, ringing, and apodizing filters, Jussi Laako, HQPlayer developer, has an easy to understand primer.

hqplayer resampling filter setup guide for ordinary person
https://audiophilestyle.com/forums/topic/13071-hqplayer-resampling-filter-setup-guide-for-ordinary-person/#comment-175928


My door is always open, Tucson is a beautiful place to be when it's -10 degrees out. I'll make an espresso and put on some tunes.

[snaimpally]

I am a part-time musician and record my performances so that I can listen to them afterwards so that I can improve. Initially, I used a Sony WMD3 professional recording walkman. After digital recorders came out, I recorded at 44.1kHz/16 bit using an Edirol R09 recorder with Sound Professionals microphones (that used Audio Technica elements). I used 44.1/16 so that I could burn them to CD for playing in my car and for distribution to the other musicians. After having the recorder for a few years, just out of curioisty, I tried recording 44.1kHz and 24 bits. To me, the difference between recording 16bits and 24 bits is very noticeable. 24 bits sounds much closer to analog. The 24 bits gives me 8 extra bits of resolving power so if I set the levels too low, I can boost the signal in post-processing without losing clarity. I did briefly experiment with recording at 88.2kHz, but didn't find the difference that noticeable.

In terms of the question being asked, my experience is that 24 bits is better than 16 bits. After that, the benefits for higher sample rates and bit depth start to diminish.

[WGH]

I did briefly experiment with recording at 88.2kHz, but didn't find the difference that noticeable.

In terms of the question being asked, my experience is that 24 bits is better than 16 bits. After that, the benefits for higher sample rates and bit depth start to diminish.

You will probably have to record at a higher sampling rate to notice any difference. A few recording studios are doing just that and their recordings are available to download, sometimes for free.

David Cheskey has a free 12 track 24bit/192kHz sampler he recorded. Enter your name and address for the free download.
David Cheskey's The Audiophile Society 2023 Music Sampler
More info here: https://www.audiocircle.com/index.php?topic=186325.msg1953186#msg1953186


Sound Liason sells one to one copies of the original master recording of many of their artists. The mastered analogue signal chain and One Mic + setup is recorded to both 352kHz and 768kHz. Downloads are available in many resolutions: PCM-96kHz/24bit; PCM768kHz/32bit; DSD64; DSD256; DXD352/32bit and many more. The latest Carmen Gomes album "Stones in My Passway" is excellent.
https://soundliaison.com/collections/nativedsd-album-of-the-year-2023/products/carmen-gomes-inc-stones-in-my-passway







2L Records has some of the finest recordings available, unusual too - Nordic Classical. They used to have sample tracks available but no more, I'll guess the bandwidth needed killed that idea, the downloads were in DSD256 and huge. I have 7.23GB of their DSD tracks that were available, I should have downloaded the surround sound too.





John Atkinson just added a 2L album to the Stereophile Records 2 Live 4 2024 list
https://www.stereophile.com/content/records-2-live-4-2024-page-2



Henning Sommerro: Borders - Original source DXD (352.8kHz/24bit)
available for download as 7.1.4 48k, Dolby Atmos TrueHD, 7.1.4 96k Auro-3D, 5.1 24/192 DTS:X, discrete 24/88.2 and DXD 7.1.4 immersive, and 24/192 PCM, 24/352.8 MQA and DXD stereo
https://shop.2l.no/en-us/collections/latest-releases/products/sommerro-borders



Positive Feedback has a new article that also fits in with the mention of bit depth. The post is so new I haven't had the chance to listen to the 4 tracks.

What We Hear With DXD 32-bit Files (Free Sample Downloads)
https://positive-feedback.com/reviews/music-reviews/what-we-hear-with-dxd-32-bit-files/



"RCA Living Presence album Gounod Faust/Bizet Carmen Suite with Alexander Gibson conducting the Royal Opera House Covent Garden Orchestra. Recorded by Kenneth Wilkinson in Kingsway Hall, this is one of the legendary recordings Wilkinson made under Decca's contract with RCA. I've known it for decades in both original LP and in a variety reissues including the excellent 4-LP 45rpm release from Classic Records. This new release from HDTT is, by the way, absolutely terrific! It comes from a 2-track 15ips tape and the clarity and dynamics are marvelous."

This release was edited in DXD PCM from a DSD256 Master then the DXD edited master was used to generate the final DSD files using Merging Technologies Album Publishing.
DXD (352.8KHz 24/32 bit PCM) is one of the best and least destructive formats for post-processing DSD-originated digital recordings

Free download: each of the four tracks are 4:35 minutes long, 24-bit and 32-bit with 3 tracks in PCM and 1 in DSD.
352.8kHz/32-bit: Track 8, "Funeral March Of A Marionette"
352.8kHz/24-bit: Track 8, "Funeral March Of A Marionette"
352.8kHz/24-bit TRUNCATED, NO DITHER: Track 8, "Funeral March Of A Marionette"
1-bit DSD256: Track 8, "Funeral March Of A Marionette"

Rushton Paul wrote:
"... I prefer the 32-bit file on our primary system with the MPD-8. But I agree with Ann's preference for the DSD256 file on her office system using a chip-based DAC (a Teac 501); the DSD256 file sounds better than the PCM files with that DAC. Does it sound as good as what we hear on our primary system? No—the primary system is far superior. Different DACs, different outcomes."

 
 

[GeorgeAb]

In my experience yes and no. I try to find the recording that comes with the highest bit rate and size from the master tapes. Yes, in my experience that makes a difference. When I upsample from there I do not hear any benefit. Some swear they do, so YMMV. I normally listen to DSD 64 files or the SACD standard. When they come from the master tapes in DSD 128 or DSD 256 I believe I may hear some refinement, but it more on intuition. My Directstream DAC takes all PCM input and converts to DSD so that also is a factor, so it may also depend on which DAC you use. So for me DSD files just sound better than PCM, but it may be a function of my DAC. Never tried converting red book (44.1 16bit) files to DSD 64; there may be some benefit to that as it may sound more liquid, natural, and analog like. Again it may depend on your DAC and how it handles PCM and DSD files so YMMV. 

[WGH]

Never tried converting red book (44.1 16bit) files to DSD 64; there may be some benefit to that as it may sound more liquid, natural, and analog like.

I don't think it would make any difference in what you hear and it could be worse. The DirectStream upsamples all PCM and native DSD files to 20x DSD.

Different programs use different methods to convert/upsample. dBPowerAmp and JRiver Media Player use simple algorithms that require the least computer power. HQPlayer uses custom filters and one-bit modulators in a 2-pass configuration that can slow down the fastest computer. My Intel i7-9700 based fanless music server can only upsample to DSD256, any higher and the processing power needed exceeds the chips 65w max and it overheats.
Serious horsepower for serious sound.

The DirectStream will upsample more accurately and cleaner than any program you currently use. A couple of guys in our audio club have DirectStream DACs. The sound is accurate and perfect but will never be described as analog like. The ANK (Audio Note Kit) DACs are analog like along with the R2R HoloAudio DACs. Two other guys in our club have ANK DACs, it's a great group and we get to hear (and sometimes borrow) top notch equipment.