Is there research that shows HD recordings sound better than Redbook (16 bit / 44.1 kHz)?

[hdomke]

I'm from Missouri (the Show Me State), and I'm a trained skeptic. I also love to listen to music on my high-end stereo.

 

Can anyone point me to good scientific studies which document that HD recordings sound better than Redbook (16 bit / 44.1 kHz) recordings?

 

When I say good scientific studies, I mean they use good methodology (double blind, randomized) and are published in peer-reviewed journals.

 

Thanks for helping me find published articles on this topic.

 

 

 

[The Computer Audiophile]

I don't think there will ever be any scientific research that concludes with a subjective answer such as "This is better than that." There may be research concluding there is or is not a difference between the two. I think the science behind the concepts of high resolution music v. redbook clearly show a technical reason why one could be considered to sound better.

 

It's really hard to judge based on material released on CD and a higher resolution because who know what the native sample rate is or if there was a remaster etc... I have some tape transfers of the same material at 44.1, 88.2, 96, 176.4, and 192. This was done for the Computer Audiophile Symposium to let people here tracks a different resolutions. I clearly hear a difference and I think high resolution is far better than redbook. But to be honest I have yet to hear a difference between 176.4 and 192.

 

[taniane]

Dave Moulton of Moulton Laboratories (who has published papers in AES - albeit not the following):

 

http://www.moultonlabs.com/more/taking_stock/

 

http://www.moultonlabs.com/more/high_resolution_audio_and_standards/

 

http://www.moultonlabs.com/more/little_bit_more_about_dither_and_bits/

 

 

 

 

 

 

[Nicholas.Bedworth]

Even on my modest, casual-listening office system (M-Audio FW 410 driving a pair of KRK Rokit 8 powered monitors), the difference between 44/16 and 96/24 kHz sampling rates is readily apparent when played back over FireWire from an ordinary HP laptop.

 

What you'll notice is that the high-res source sounds more relaxed, smoother and natural. The timbre of the voices and instruments sounds more like the real thing, the soundstage is better defined, dynamic range may be greater, and so on. You'll be more at ease with the high res material.

 

In the case of instruments with overlapping ranges, it's easier to distinguish cellos from violas from violins, even when they're playing notes of the same pitch. This kind of fine detail comes alive with well-done high-res sources. Sometimes it feels like the lights were turned on or the windows were cleaned.

 

How much of this improvement stems from the sampling rate (96 kHz versus 44) or the sampling resolution (24 bits versus 16) is a matter of some discussion, but in any event, 96/24 certainly is a very nice improvement. With resepct to an experiment to tease this apart, it would be fun to try some 44/24 versus 96/16 source.

 

In the image processing world, contrast resolution in the form of pixel depth (bits per visual sample) generally increases human perception of fine detail, compared with cranking up the spatial (analogous to sampling rate) resolution. The contrast resolution is more responsible for preserving the phase transfer function, compared with the modulation/amplitude transfer function. Most textbooks, engineers, etc., tend to fixate on the MFT and ignore the PTF.

 

Anyway, as Chris mentioned, the most explicit differences will be with material that was natively recorded in high resolution from the get-go. See TAS 197, page 20, for an overview by Steven Stone of ten high-resolution source vendors.

 

Not all high-res material is created equally, as others on the forum have noted.

 

Above all, trust your own ears. Never feel that you have to rely upon somebody else's supposedly "technical" arguments that one thing "must be" better than another. Check it out yourself. With a little practice, you'll build your confidence and be able to appreciate, and articulate, the differences between sources.

 

No need to do elaborate experiments when the answer is obvious.

 

 

 

[Nicholas.Bedworth]

BTW, to finally answer your question explicity, check these out. They'll get you started.

 

1. http://www.aes.org/events/111/workshops/W10.cfm

2. http://www.crc.gc.ca/en/html/aas/home/evaluation/evaluation

3. http://www.crc.gc.ca/en/html/aas/home/perceval/perceval

4. http://www.crc.gc.ca/en/html/aas/home/products/products

 

You'll see references to various JAES papers, which, in my opinion, tend to be very well written, meaning even though the material is definitely technical and often somewhat complicated, the authors (and editor) do a great job of explaining things clearly.

 

Some of the CRC audio evaluation products go up to 32 bits!

 

And here are the bits (as it were) where you can get more information. I suspect that these people maintain a database of all the research organizations conducting research on high and low resolution audio formats.

 

Mr. Louis Thibault

Manager, Advanced Audio Systems

Communications Research Centre

3701 Carling Ave., P.O. Box 11490, Station H

Ottawa, Ontario, Canada K2H 8S2

Tel:+1 613 990-4349

Fax:+1 613 993-9950

E-mail: [email protected]

 

[jphone]

compares Redbook to SACD

 

http://www.aes.org/e-lib/browse.cfm?elib=14195

and

http://www.bostonaudiosociety.org/explanation.htm