Author Topic: Can you hear a difference? Native high-resolution versus red book  (Read 532 times)

Online petetherock

  • Trade Count: (+8)
  • Full Member
  • Posts: 22273
  • Sidere Mens Eadem Mutato
    • My Music, Movie and Equipment Musings...
Can you hear any difference?
The first page of my AES PaperWith the help of a couple of readers with experience in statistics (thank you Robert and Sergio), the raw data from 428 submissions to the HD-Audio Challenge II have been preliminarily analyzed and graphed. There were 6181 data points (the number of individual selections of hi-res vs. CD spec audio). Almost 23% of the submissions selected NO CHOICE. Determining which version of the track was actually high-resolution was impossible for almost a quarter of the participants. Others didn't listen to all 20 of the tracks. And while some of the tracks produced moderately better results than others, the data supports the conclusion that the probability of successfully selecting a hi-res audio file over a CD downconversion of the same track is no better than flipping a coin — 50%. And remember this was done using real high-resolution audio tracks NOT digitized classic recordings from the golden era of analog tape.Hi-Res Matters ... and It Doesn'tThe results of my survey will be hard or impossible for many to accept. Those with a financial stake or vested interest in "hi-res" audio will continue to perpetuate the myth. I've had a few polite back and forth exchanges with David Solomon, the evangelist for Qobuz, on one of the FB audio groups that focuses on high-resolution audio. His contention — and one that I completely agree with — is that using high-resolution audio to produce new recordings makes a difference. He bolsters his position with "testimonials" from respected audio engineers and producers including Grammy-winning Bill Schnee and Cookie Marenco. Their arguments are well-founded and true but they do not conflict with the findings of my survey. How can hi-res audio absolutely make a difference and according to my own survey be indistinguishable from standard-res audio?
Because we're not talking about the same thing. David and his group are focused on one end of the production chain and I'm dealing with the other end. Making a recording requires a lot of individual steps. There's time spent in the studio or at a live venue capturing performances through microphones, preamps, and preserving them on recording devices — analog or digital. Additional vocal or instrumental parts are then layered on to the "basic tracks" over the course of weeks or even months. And finally, the individual tracks are blended, tweaked, and balanced into a surround or stereo mix – called the mixdown stage. At the end of the process, a mastering engineer assembles the album, applies EQ and compression, and adds the metadata to the final files prior to replication or distribution. All of this should ideally happen in high-resolution. Why?
Because one of the primary goals of the audio engineers working on the project is to record the instruments and vocals in the best quality possible. You can always make a recording with lots of fidelity sound worse but you can't go the other way. When the artist is singing or playing a new part, the engineer can only guess at how loud the artist will sing or play. Having 24-bits affords recording engineers additional "headroom" before excessively loud moments overload some part of the signal path, distort and sound terrible. Using 24-bits provides over 120 dB of potential signal to noise specifications. Increasing the sampling rate helps with filtering and analog to digital conversion. Moving to high-resolution during the production phases of a commercial recording project makes all the sense in the world.
But engineering a new album project in a professional recording studio is very different than playing back a final mastered, released track in your own home on your own system. As the HD-Audio Challenge II clearly demonstrated, playing back music in "hi-res" audio — with sample rates higher than 44.1 kHz and word lengths longer than 16-bits — doesn't provide any perceptible fidelity improvement over using Red Book standards. The industry will continue to try and defend "hi-res" audio and market ever higher specifications because they want us to purchase new "hi-res" equipment, subscribe to pricier music services, and buy our favorite music all over again. But signing up for the highest tier of Spotify or buying Blu-ray music discs won't make any difference in your music enjoyment.
If I were an evangelist for a "hi-res" streaming service or the head of marketing for a digital music download site (wait a minute I am responsible for, the first high-resolution music download site ), I would have a hard time acknowledging and accepting the results of the HD-Audio Challenge II survey. As someone that has championed hi-res audio for over a decade, the results directly contradict many of the articles on this site. But I didn't manipulate the data to conform to my previous position of advocating for hi-res audio. I wanted to know if real world audiophiles could hear the "dramatic" improvement in fidelity claimed by NARAS, the DEG, CTA/CEA, music services, and the music labels. I'm satisfied that hi-res doesn't matter as a final delivery format. However, it does matter while Bill Schnee, Cookie Marenco. or I are making new recordings. We can all be right. And audiophiles should understand the facts.
Please post instead of sending a pm, so more can learn.

My gear:

Offline jerome_the_lang

  • Trade Count: (0)
  • Full Member
  • Posts: 6353
Anyone else game enough to take up the gauntlet should consider downloading each comparative files into separate folders so that as you exit one folder to go into another folder to play the other files, your standalone streamer will erase playback memory to allow the subsequent file to be played with lesser sonic degradation that had arised due to the memory retention problem.

Just remember never to stop or pause playback as you exit one folder to go into the other folder to cue up the next file to play. In fact, to keep the system condition constant, you should always  be playing something else from the current folder before you move to the other folder to play the comparative file there.

And if you are concerned that you might get distracted from hearing something different, try playing a "digital silent" track. I had AndrewC (who seems to have disappeared from the surface of the earth) created for me a 1 minute digital silent track which I would inserted into each of the folders to be played after the first file finish playing. I'll play this silent track before I exit folder to go into the next one. This procedure actually has an additional layer of memory erase.

Always use a standalone streamer for these listening tests.
Memory retention on PCs are several multitudes more severe.
Many thanks to the mods for continuing to tolerate all my expressive vulgarisms and harassment of fellow forumers...


SimplePortal 2.3.6 © 2008-2014, SimplePortal