24bit vs 16bit, the myth exploded!

[jdandy]

Ivan.......My pleasure. I felt this article, and the manner in which Greg was able to explain the complicated theory was worthy of every digital aficionado's attention. Greg was quite generous to allow us to use his article in our forum. I hope he joins the conversation.

[BuffaloBill]

The article/post has been around for some time. I have seen it in several other forums.

[BearCubinNY]

Originally Posted by BuffaloBill

The article/post has been around for some time. I have seen it in several other forums.

Yeah, I thought I read this before. There was a huge discussion about this on a computer forum about a year and a half ago. Some heavy weights from the industry really got deep into the whole process. Above my pay grade, but quite interesting. Veteran recording, mixing, and mastering engineer Barry Diament (Bob Marley, Pete Townshend, Led Zeppelin and on and on and on) articulated a contrarian view that gained quite a bit of traction. He's in the 24/192 camp.

Reg

[BuffaloBill]

Also, the article/post is consistent with the Meyer and Morgan test conducted in Sept. '07 and included in the Journal of The Audio Engineers Society, i.e., Redbook CD and SACD sound the same.

[metaphacts]

Originally Posted by jdandy

A new member (bart) directed my attention to this article, and I found it so interesting and informative that I am reposting it here for all to see.


24bit vs 16bit, the myth exploded!

[B]If you play a 24bit recording and then the same recording in 16bit and notice a difference, it is either because something has been 'done' to the 16bit recording, some inappropriate processing used or you are hearing a difference because you expect a difference.

G

Interesting indeed.

Juxtapose that with : Weiss :: POW-R for starters.

Read the history of what these guys are doing - they're not new to the rodeo. They're respected engineers/designers/professors on the production and reproduction sides.

We're jumping into the deep end of the pool guys!

[piolaxo]

To simplify a little bit more the Nyquist-Shannon Sampling Theorem, but basically in order to reproduce a signal from digital back to analog one must sample the analog signal at least 2X of the highest frequency being sampled. So if we want the audio spectrum (up to 20KHz), then we must sample at 40KHz. In this case 44.1KHz is more than twice that spectrum. Theoretically speaking the limit of the analog frequency being converted to digital is around 22KHz. I guess this specific number (44.1) was chosen by the industry.

So one thing is the conversion rate and another is the resolution. The ADC or DAC converter has width bits, usually 16, 20, 24, 32 and so on. The more bits the conversion packs, the more resolution, granularity or precision the data will contain. So the more the better.

I completely disagree with the OP. In this case the more the better. When he says that an extra bit doubles the resolution is correct, but it halves the errors in the decoding, no noise. The THD that he is talking about is with respect to the original analog signal, so the more "dynamic range" as he calls it represents a signal closer to the original.

The reason it was 16bits for the CD "dynamic range" was due to amount of data. At the time 16bits was believed to be enough resolution for audio. ADC and DAC's have always been expensive, specially the ones with lower noise and they were more expensive in the 80's.

So 24bits that according to him provide 144db is with respect to the original analog signal. That is way better than 16bits.

This statement below in my opinion is completely wrong:

I know that some people are going to say this is all rubbish, and that "I can easily hear the difference between a 16bit commercial recording and a 24bit Hi-Rez version". Unfortunately, you can't, it's not that you don't have the equipment or the ears, it is not humanly possible in theory or in practice under any conditions!! Not unless you can tell the difference between white noise and white noise that is well below the noise floor of your listening environment!! If you play a 24bit recording and then the same recording in 16bit and notice a difference, it is either because something has been 'done' to the 16bit recording, some inappropriate processing used or you are hearing a difference because you expect a difference.

The original CD's were mastered and recorded in 16bits. Of course, if you take the analog tapes and now convert them into 24bits they will sound much better. It's pointless to take a 16bits converted signal and do something with it.

Most of the new remasters and taken from the original tapes and converted back to digital with 20 or 24 bits with *new* equipment.

Another huge factor in the early 16bits CD was that the mixing was for LP's because pretty much all of the record producers were trained to work on them. CD's didn't need all the compensation needed for records.

Salvador

[piolaxo]

One of the reasons why ADC's were expensive (then) was because they required very precise resistors to divide the signal into bits. The more bits, the more resistors required. In those early days, they didn't have laser cutters to trim the resistors like it can be done today.

[bradleyc]

The new MoFi Stevie Ray Vaughan Texas Flood hybrid CD/SACD sounds fairly good on the redbook layer, but the SACD layer sound quality is nothing short of outstanding. According to some they're supposed to sound the same?? Well they don't sound the same, it's obvious to me, so how can this be?

[Masterlu]

Originally Posted by jdandy

Ivan.......My pleasure. I felt this article, and the manner in which Greg was able to explain the complicated theory was worthy of every digital aficionado's attention. Greg was quite generous to allow us to use his article in our forum. I hope he joins the conversation.

Me as well, it would be my pleasure to give him an AA Subscription if he chooses to join us here.

[Tonepub]

There's a lot that this explanation glosses over. It's just not worth starting a long argument over, but suffice to say that if you'd like to hear a diff, it's pretty easy to do, comparing a digital capture of your favorite analog bit of music, captured at 16/44 and again at 24/96 or even 24/192.

It's not a massive diff., but it is there and pretty easily discernible.

That being said, I've heard plenty of 16/44 done with extreme care and fully utilizing the medium that sounds great and I've heard sloppy high res stuff too.

I'm always amused when someone (usually with a pretty modest system) loudly proclaims you can't hear something....

This subject has been argued for years over on Steve Hoffman's forum to no real end.

Can we go drink some scotch now and forget about this?