Can you perceive and characterize the difference?

[BenB]

I've been playing around with generating DSP tools, and made myself a new one. However, I don't know how others might perceive the effect of the tool. If you would be so kind, it would be great to hear some other perspectives. Even if your perspective is that the difference is too small to be of concern, that's good information.

If you try an ABX test with the files, let me know how it turned out, and what clues you used.

Thanks in advance.

Please choose either flac, or mp3 below.

Cut A flac:

04. CD Track 04_Cut_A.flac | Powered by Box

app.box.com

Cut B flac:

04. CD Track 04_Cut_B.flac | Powered by Box

app.box.com

Cut A mp3:

04. CD Track 04_Cut_A.mp3 | Powered by Box

app.box.com

Cut B mp3:

04. CD Track 04_Cut_B.mp3 | Powered by Box

app.box.com

 

[Slayer]

I've been playing around with generating DSP tools, and made myself a new one. However, I don't know how others might perceive the effect of the tool. If you would be so kind, it would be great to hear some other perspectives. Even if your perspective is that the difference is too small to be of concern, that's good information.

If you try an ABX test with the files, let me know how it turned out, and what clues you used.

Thanks in advance.

Had to listen several times, even then i could not be 100% certain.
Using the flac file option. Track B seems to have slightly better separation and and a larger sound field. But I wouldn't put any money on my observations being correct.

 

[BenB]

Had to listen several times, even then i could not be 100% certain.
Using the flac file option. Track B seems to have slightly better separation and and a larger sound field. But I wouldn't put any money on my observations being correct.

Thanks for your observations. Can I assume you used speakers rather than headphones?

 

[Robin L]

On the one hand, I can't really ABX properly, as I'll always know what track I'm listening to. And I'm pretty sure my hearing only extends as far as 10khz. So feel free to ignore my response. That said, the Cut A Flac seems to have the fewest undesirable artifacts in the brass sound.

 

[Slayer]

Thanks for your observations. Can I assume you used speakers rather than headphones?

Yes, you assumed correctly.

 

[BenB]

Yes, you assumed correctly.

Thanks for confirming. I'd also be interested to know whether you listen in direct stereo, or if you use a surround upmixer. And would you characterize your speakers as wide dispersion or narrow dispersion, or something in between?

 

[BenB]

Track A is a digitally filtered version of track B. I used adaptive signal processing to up-mix Left and Right stereo into 4 chanels: Exclusive_Left, Exclusive_Right, Both_in_phase, and Both_out_of_phase. To produce track A, I only downmixed with three of the 4 channels.... I left out the content that was found to be in both channels, but out-of-phase. At the listening position (equidistant from the speakers), that out-of-phase content should largely cancel out. That energy will more strongly contribute to reverberant sound, rather than direct. In pro-logic upmixing, the out-of-phase content is sent to the surround speakers.

Personally, I couldn't decide if I liked more or less of that out-of-phase content. I listened to tracks with that content amplified, and found the result to be very engaging. However, ultimately I determined that I lost a little bit of clarity, and my preference eventually came around to prioritize clarity over engagement/envelopment. I could post a version C that would have the content missing from A boosted rather than removed, if anyone was interested.