[robertsong]

Actually, I do know what an ABX test is. What I absolutely did not know and never even heard of is this feature in foobar you are talking about. If you can explain how it's done or point me in the right direction I'll gladly do the test through my loudspeakers.
 
But what I have to wonder is, why rely on me? Have you tried the test yourself, head injury?

 

[Head Injury]

  Actually, I do know what an ABX test is. What I absolutely did not know and never even heard of is this feature in foobar you are talking about. If you can explain how it's done or point me in the right direction I'll gladly do the test through my loudspeakers.
 
But what I have to wonder is, why rely on me? Have you tried the test yourself, head injury?

Here's the plugin. Just follow the documentation on how to install and use it, perform about 15 trials, and copy-paste the result!
 
As I said, I've never heard a difference between hi-rez and regular Redbook audio (I have a handful of hi-rez albums because FLAC files weren't available in anything else), and see no reason why there would be differences. I can't confirm it with an ABX test because I'm biased against finding differences. Only someone who's actually trying to hear the difference can pass the test with any significance, because no matter what my score is I'll just be guessing.

 

[Strangelove424]

Below is a file I have made available on Google Drive that isolates the exact differences between a high-resolution 24 bit/96khz version of Fleetwood Mac's Rhiannon and a down-converted CD-quality 16-bit/44.1khz version. A null comparison was done in Audacity. The render of the null file is 24bit/96khz so all details from the original high resolution version have been retained. Audacity is free, so you can try this with any HD music you own.  
 
 
I'll let you judge whether or not you think this is worth 2x the money of CD:
 
https://drive.google.com/file/d/0B6GfCo1MWrJgdVM5aGNMZTlndms/view?usp=sharing
 
Nothing personal to anyone in this thread, but I'm a little tired of ABX tests. You need someone who honestly wants to know the truth, and that can be hard to verify. Most people don't. We shouldn't require human ears, and even worse human minds, to differentiate anymore. The technology is in everyone's hands to get to the point already, so let's just get to the point. That null file is dead silent. 

 

[BlindInOneEar]

Strangelove424, have you ever heard of the Audio Diffmaker?  It too is free and is designed to do "audio differencing."  You might find it fun to fiddle with! 
 
http://libinst.com/Audio%20DiffMaker.htm
 
Check out the prerecorded ".dyf" files on the site!  It's amazing how many "golden ears" out there simply refuse to listen to the "listener's challenge!" 
 
Note that you cannot load .dyf files using the "load" buttons, but must instead use the "Load .Dyf Files" tab on the menu bar. 

 

[Strangelove424]

  Strangelove424, have you ever heard of the Audio Diffmaker?  It too is free and is designed to do "audio differencing."  You might find it fun to fiddle with! 
 
http://libinst.com/Audio%20DiffMaker.htm
 
Check out the prerecorded ".dyf" files on the site!  It's amazing how many "golden ears" out there simply refuse to listen to the "listener's challenge!" 
 
Note that you cannot load .dyf files using the "load" buttons, but must instead use the "Load .Dyf Files" tab on the menu bar. 

 
Yep, Diffmaker is awesome. I learned it from Chewy's listening test (available in a thread on Sound Science somewhere), and I've been using it since then, and then got into Audacity recently when researching freeware audio tools for PC. I would have used Diffmaker for this, but it seems to have problems with certain high resolution files, atleast on my system. It also requires a conversion to WAV. The conversion to WAV doesn't change the results, but it adds a few extra steps. On the plus side, Diffmaker automatically matches levels, so that can be helpful. They're both great tools.  

 

[Music Alchemist]

I'm hearing an echo

 

 
Funny thing is, I was pretty much saying the same thing as you at the same time without realizing it, but submitted my reply shortly after.
 

  Actually, I do know what an ABX test is. What I absolutely did not know and never even heard of is this feature in foobar you are talking about. If you can explain how it's done or point me in the right direction I'll gladly do the test through my loudspeakers.
 
But what I have to wonder is, why rely on me? Have you tried the test yourself, head injury?

 
Follow the instructions in this guide. (I linked to it earlier in this thread.)
 
You didn't answer my questions either. I'm curious whether you downloaded the HD files and converted them to 16-bit / 44.1 kHz yourself or if you simply assumed that the CD was from the same master. (You have to at least eliminate that variable.)
 

  As I said, I've never heard a difference between hi-rez and regular Redbook audio (I have a handful of hi-rez albums because FLAC files weren't available in anything else), and see no reason why there would be differences. I can't confirm it with an ABX test because I'm biased against finding differences. Only someone who's actually trying to hear the difference can pass the test with any significance, because no matter what my score is I'll just be guessing.

 
We both know that the only reason for a real difference is if a different master was used, so you should have included that qualifier.

(But I know we're talking about converting the "HD" files to "CD quality" anyway.)

 

[kraken2109]

High bit depth allows a lower noise floor. 16 bit gives a large noise floor of -96dB. With good dither it can be lower. This is good enough for pretty much any music.
A higher sampling rate allows higher frequencies to be captured. According to the Nyquist-Shannon theory 44.1kHz allows frequencies up to 22.05kHz, I don't know of anyone that can hear that high. Personally I am 20 years old with very little hearing damage and I can do ~18kHz on a good day, maybe 19 at a push.
 
In other words, 44.1/16 is enough - that's why it was chosen in the first place.

 

[ubs28]

I tried DSD music for the first time today. My first impressions is that it sounds better than Flac. 

 

[VandyMan]

 High bit depth allows a lower noise floor. 16 bit gives a large noise floor of -96dB. With good dither it can be lower. This is good enough for pretty much any music.

 
Does upsampling 16 bit to 24 bit lower the noise floor (assuming your DAC has greater than 16 bits of resolution)?

 

[SilverEars]

   
Does upsampling 16 bit to 24 bit lower the noise floor (assuming your DAC has greater than 16 bits of resolution)?

Yes, and I don't know about Kraken's statement regarding 16bit DAC having the expected SNR. You would have to see the measurements to see how close it is. I've seen two DACs measured that is 24bit source.  Based on SNR, affectively, one had bit depth of 17, and other had 21.  

 

[esldude]

  Yes, and I don't know about Kraken's statement regarding 16bit DAC having the expected SNR. You would have to see the measurements to see how close it is. I've seen two DACs measured that is 24bit source.  Based on SNR, affectively, one had bit depth of 17, and other had 21.  

 

   
Does upsampling 16 bit to 24 bit lower the noise floor (assuming your DAC has greater than 16 bits of resolution)?

Well just to be clear the answer would be yes and no.  Yes, the new file would show as a 24 bit file, but it will be the 16 bits with padding for bits 17 thru 24 as zero.  You don't push the existing noise floor in the 16 bit file lower.  Let us say your DAC has 24 bit resolution.  The analog output of the 16 bit file versus the 16 bit file converted to 24 bits would not differ.

 

[Music Alchemist]

  I tried DSD music for the first time today. My first impressions is that it sounds better than Flac. 

 
DSD is more complex and should (for all intents and purposes under most circumstances) be treated as a different master.
 
Depending on various factors, it can sound better or worse than PCM.
 
You can read up about the format here.

 

[castleofargh]

eheh I'm testing the sony a15 and as always I fool around with different codecs and a few test tracks of the same songs at different resolutions. yesterday I went to bed and started listening in shuffle mode. then daft punk "lose yourself to dance" starts. and I'm trembling under each bass impact like really trembling, like when I was looking live at michael jackson, and my chest vibrated with the bass to a point where I guess any pacemaker would go wild \o/ what the hell? I never had that when I tried the song on other DAPs there is just such a strong sub bass power What is happening???????

I see hires on the DAP, could it be that moment where I actually understand the point of having hires files? hallelujah brothers!!!!!!
 
turns out I imported some tracks using media go (crappy software, very horrible) to test it too, as it came with the DAP. and I ended up uploading the "studio master" version instead of my usual test track(yeah that's how fine I was using the app). so now I'm up and very far from sleepy, I look at both versions ^_^, there is like +5DB at 40hz compared to the mids and going down we end up with close to +20DB around 20hz ^_^.
just a slight remaster with a tiny touch of bass. just enough to make your hair dance.
 
 
this was my hires listening experience, thank you.

 

[kraken2109]

   
Does upsampling 16 bit to 24 bit lower the noise floor (assuming your DAC has greater than 16 bits of resolution)?

It would lower the noise floor of your DAC, and any processing happening. However the noise floor of 16 bit audio would remain the same. However, there's no reason not to have your DAC running at 24 bit. It is often the case that (at least the implementation of) the DAC cannot actually achieve the -144dB theoretical noise floor of 24 bit, simply because that's very difficult to achieve in electronics.

 

[RRod]

Ch.  0:  Peak    -0.30 (   -0.30)   RMS    -6.96 (   -5.80)   DR =   5.51
Ch.  1:  Peak    -0.30 (   -0.30)   RMS    -6.91 (   -5.67)   DR =   5.37
Overall dynamic range: DR5
 
 
Ch.  0:  Peak    -0.10 (   -0.10)   RMS    -9.04 (   -7.77)   DR =   7.67
Ch.  1:  Peak    -0.12 (   -0.18)   RMS    -9.58 (   -8.14)   DR =   7.96
Overall dynamic range: DR8
 
I'll let you guess which track is the hi-res master…  So far hi-res delivery services have not proven to me that they care about the real problem in music mastering today (loudness), but simply seem concerned about making beaucoup $$ off tricking people into re-buying their albums again to get sounds they can't hear.