• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

I made a sample rate blind-test (192 vs 48)

V

vintologi

Active Member
Joined
May 7, 2023
Messages
105
Likes
18
I remember doing a similar test years ago with this track and i though the 192 version was better (9 times out of 10)

www.mediafire.com

MediaFire

MediaFire is a simple to use free service that lets you put all your photos, documents, music, and video in a single place so you can access them anywhere and share them everywhere.
www.mediafire.com www.mediafire.com

I used audacity to downsample and upsample (for the 48 KS/s version).

I used https://www.random.org/ integer generator to decide if the 192 version was going to be called A (if 0) or B (if 1).

Album: The New Organ of the Philharmonie Mercatorhalle Duisburg

Song: Thalben-Ball: Variations on a theme of Paganini for pedals

Note that you probably want tweeters that reach well above 30Khz for this.

The difference between the files is definitely not audible by itself (yes i did try listening to it):
difference.png
 
Last edited:
JSmith said:
How did you ascertain preference if you could not establish difference?
Click to expand...
I listened to one 192 version and one 48 version and i got the impression that it sounded better with the ultrasonics included even though the ultrasonics by themselves were not audible.

It was with fairly cheap equipment so it could theoretically be that it was some audible intermodulated distortion i liked.

Or the hypersonic effect is true after all.
 
  • Like
Reactions: Dz1
Use Foobar's ABX Comparator for proper, statistically significant blind test results.

I would take a 192kHz track, downsample it to 48kHz to get rid of ultrasonic content, then sample it back up to 192kHz to avoid pop sounds from the DAC and to ensure equal Nyquist filter response.

Or you can just low-pass the 192kHz file.
 
Sadly I don’t think I need tweeters that reach 30kHz.
Keith
 
staticV3 said:
Use Foobar's ABX Comparator for proper, statistically significant blind test results.
Click to expand...
I used blind XY testing 10 times (after listening to the 192 version multiple times).

I test again when i have better equipment.
 
vintologi said:
I used blind XY testing 10 times.

I test again when i have better equipment.
Click to expand...

Maybe a better methodology, or at least explain a bit more thoroughly what it is you think you've shown.
 
BDWoody said:
Maybe a better methodology, or at least explain a bit more thoroughly what it is you think you've shown.
Click to expand...
I used deadbeef to randomly play one of the versions, it was in a playlist where it would loop so if the 192 version would play first the 48 version would follow and so on.

The problem with testing this with typical speakers is that there will always be intermodulated distortion from the ultrasonics interacting with the lower frequencies. Still it can be interesting to listen for yourself to see if what difference if any you can hear.

Examples of studies that used a separate super-tweeter for the ultrasonics:


Frequencies of Inaudible High-Frequency Sounds Differentially Affect Brain Activity: Positive and Negative Hypersonic Effects - PMC

The hypersonic effect is a phenomenon in which sounds containing significant quantities of non-stationary high-frequency components (HFCs) above the human audible range (max. 20 kHz) activate the midbrain and diencephalon and evoke various ...
www.ncbi.nlm.nih.gov www.ncbi.nlm.nih.gov
 
Jim Taylor said:
Citation, please? :)

Jim
Click to expand...
Speakers are never 100% linear so unless you use a completely separate driver for the ultrasonics there will be intermodulated distrortion from the ultrasonics interacting with the lower frequencies.

Whether or not the faint ultrasonics can actually result in any audible IMD in the typical speaker is a different matter. It will obviously be less likely to be audible with very high-end tweeters.
 
VintageFlanker said:
That's a wrong process.
Click to expand...
No it's the correct process since otherwise you might be able to hear a difference from the dac using a slow rolloff or messing up in some other way. It's better to let audacity handle the upsampling since then there will be no audible difference between below 22.4 Khz.
VintageFlanker said:
What's lost in downsampled track is lost forever.
Click to expand...
Of course it's lost, i only did that for one of the versions obviously.
 
vintologi said:
I used deadbeef to randomly play one of the versions, it was in a playlist where it would loop so if the 192 version would play first the 48 version would follow and so on.

The problem with testing this with typical speakers is that there will always be intermodulated distortion from the ultrasonics interacting with the lower frequencies. Still it can be interesting to listen for yourself to see if what difference if any you can hear.

Examples of studies that used a separate super-tweeter for the ultrasonics:


Frequencies of Inaudible High-Frequency Sounds Differentially Affect Brain Activity: Positive and Negative Hypersonic Effects - PMC

The hypersonic effect is a phenomenon in which sounds containing significant quantities of non-stationary high-frequency components (HFCs) above the human audible range (max. 20 kHz) activate the midbrain and diencephalon and evoke various ...
www.ncbi.nlm.nih.gov www.ncbi.nlm.nih.gov
Click to expand...
I'm not sure what finding in this article supports 192Khz sounds better than 48Khz recordings or that humans can even tell the difference. Perhaps you can show the relevant section.
 
How can there be IM distortion from frequencies that are filtered out?
 
JustJones said:
I'm not sure what finding in this article supports 192Khz sounds better than 48Khz recordings or that humans can even tell the difference. Perhaps you can show the relevant section.
Click to expand...
There papers focused on doing measurements on the brain directly instead of asking the participants to guess which version was of higher quality. The study authors claimed that the ultrasonics made a difference but i need to examine them more before arguing about it.
 
ahofer said:
How can there be IM distortion from frequencies that are filtered out?
Click to expand...
If they are fully filtered out there cannot be but even with tweeters that start to foll of after say 20Khz there will still be ultrasonics being produced (if it's fed to the speakers), not with much accuracy obviously.
 
I know what they did in the study and I'm not interested in arguing was just asking a simple question. If you haven't examined the paper then that's fine . I couldn't find anything that supported your claim of being able to tell 192Khz over 48Khz since both of those are ultra sonics and would produce similar result.
 
vintologi said:
Examples of studies that used a separate super-tweeter for the ultrasonics:
Click to expand...

Oohashi?

Those results never been replicated.

Got anything else?
 
BDWoody said:
Oohashi?

Those results never been replicated.
Click to expand...
It has been replicated at least twice (with regard to ultrasonics affecting brain activity when mixed with lower frequency audio):

Frequencies of Inaudible High-Frequency Sounds Differentially Affect Brain Activity: Positive and Negative Hypersonic Effects - PMC

The hypersonic effect is a phenomenon in which sounds containing significant quantities of non-stationary high-frequency components (HFCs) above the human audible range (max. 20 kHz) activate the midbrain and diencephalon and evoke various ...
www.ncbi.nlm.nih.gov www.ncbi.nlm.nih.gov

www.frontiersin.org

Frontiers | High-Resolution Audio with Inaudible High-Frequency Components Induces a Relaxed Attentional State without Conscious Awareness

High-resolution audio has a higher sampling frequency and a greater bit depth than conventional low-resolution audio such as compact disks. The higher sampli...
www.frontiersin.org www.frontiersin.org
 
  • Like
Reactions: Dz1
You must log in or register to reply here.
Back
Top Bottom