• Guys, we've spent considerable money converting the Les Paul Forum to this new XenForo platform, and we have ongoing monthly operating expenses. THE "DONATIONS" TAB IS NOW WORKING, AND WE WOULD APPRECIATE ANY DONATIONS YOU CAN MAKE TO KEEP THE LES PAUL FORUM GOING! Thank you!

Hi-Res Audio Hijinx: Why Only Some Albums Truly Rock

Triburst

Well-known member
Joined
Feb 12, 2006
Messages
4,353
The Wall Street Journal has a very good article about my favorite format: High Resolution Digital Audio (I think the "return" of vinyl is basically a reaction to awful CD recordings, and digital Hi-Res has made some good albums truly listenable again -- without all the vinyl-associated hassles).

Anyway, it's a very good, thoughtfully written article that has some excellent tips and info, along with some recommendations for excellent Hi-Res albums and reissues.

Read it here: WSJ - Hi-Res Audio Hijinx: Why Only Some Albums Truly Rock.
 

J T

Well-known member
Joined
Oct 20, 2005
Messages
10,501
Yes people nowadays are so used to hearing the MP3 mish mash through headphones they don't know what a high quality recording really sounds like. It's amazing how much digital information is lost when mp3's are created.
 

Doc Sausage

Well-known member
Joined
Nov 21, 2006
Messages
1,707
I've said it before, after listening to music for decades, LOUD, and other sources of high decibel industrial noise, my hearing has become low res and no amount of digital enhancement can bring back what I heard in 1972. It's a shame but is what it is. I do still get enjoyment from my guitars at about 95 decibels though!
 
Y

yeti

Guest
Yes people nowadays are so used to hearing the MP3 mish mash through headphones they don't know what a high quality recording really sounds like. It's amazing how much digital information is lost when mp3's are created.

Yes and no. yes, many listeners who grew up during the 2000's or even the 90's don't have any experience with truly great recordings.
No, digital information is only lost via mp3 encoding if it was there to begin with. If the mix has a dynamic range of DR5 then you only need maybe 4 bit (or less) to encode it anyway and all that stuff usually captured by extending wordlength is inaudible by default, so it really depends on the source material. An mp3 of a shitty over-compressed master sounds just as good as that shitty master. They don't have to throw anything audible away. Let's face it, the biggest problem isn't encoding to mp3, it's mixing and mastering for maximum loudness. A ultra hi-res version of a crappy master will still sound crappy.
 

Pickdropper

Active member
Joined
Jul 17, 2001
Messages
1,229
The biggest crime in the recording industry in recent years has been the trend towards horrific mastering with zero dynamic range (read up on the "loudness wars"). Perhaps the biggest contribution high-res audio can bring is a renewed approach to quality mastering techniques. If they don't do that, I wouldn't even bother with high res.

The article leaves a bit to be desired. They state that most recording studios use 24-bit recording, which is true, but not for the implied reason. It's done because it gives greater margin for error in regards to the noise floor of the recording and allows for adjustment during mastering. Most times, it's mastered down to 16-bit word depth for the released version without corresponding loss of fidelity. The exception to this could be a recording with an exceptionally wide dynamic range, which generally doesn't apply to rock.

There are also a few unexplained statements that I am curious about. They state you need to listen with speakers and not with headphones, which is interesting. I've got a very expensive stereo and a bunch of high quality headphones and I find that statement interesting. Certainly the presentation is different (particularly the soundstage not being "in your head" but headphones eliminate variables such as room acoustics and normally allow for listeners to just fidelity quite well (assuming good quality headphones of course).

I've got a high-res player on loan from Pono and I've been meaning to setup a formal listening test but I haven't had time to do it yet. It does sound good, but I need to compare recordings from the same mastering at different sample rates/bit depths to see what the actual differences are. It all needs to be on the same player as well. Comparing a Pono player a remastered high res file with an iPod compareing a different mastering at redbook resolution has way too many variables to make it a useful comparison.
 

Wilko

All Access/Backstage Pass
Joined
Mar 11, 2002
Messages
20,854
While it is certainly true that dynamic range has been damn near destroyed by the volume wars, MP3 encoding does indeed dispense with a huge amount of harmonic content as well. Massive amounts of the bandwidth are eliminated.

Another area of study is that audio CDs at the standard 44k still fair quite well in A/B testing compared with vinyl or other so-called hi-res or analog sources.

I find that using all tube audio gear has an amazing "softening effect" on some of the offensive square wave high frequency artifacts.
 
Y

yeti

Guest
The biggest crime in the recording industry in recent years has been the trend towards horrific mastering with zero dynamic range (read up on the "loudness wars"). Perhaps the biggest contribution high-res audio can bring is a renewed approach to quality mastering techniques. If they don't do that, I wouldn't even bother with high res.

The article leaves a bit to be desired. They state that most recording studios use 24-bit recording, which is true, but not for the implied reason. It's done because it gives greater margin for error in regards to the noise floor of the recording and allows for adjustment during mastering. Most times, it's mastered down to 16-bit word depth for the released version without corresponding loss of fidelity. The exception to this could be a recording with an exceptionally wide dynamic range, which generally doesn't apply to rock.

There are also a few unexplained statements that I am curious about. They state you need to listen with speakers and not with headphones, which is interesting. I've got a very expensive stereo and a bunch of high quality headphones and I find that statement interesting. Certainly the presentation is different (particularly the soundstage not being "in your head" but headphones eliminate variables such as room acoustics and normally allow for listeners to just fidelity quite well (assuming good quality headphones of course).

I've got a high-res player on loan from Pono and I've been meaning to setup a formal listening test but I haven't had time to do it yet. It does sound good, but I need to compare recordings from the same mastering at different sample rates/bit depths to see what the actual differences are. It all needs to be on the same player as well. Comparing a Pono player a remastered high res file with an iPod compareing a different mastering at redbook resolution has way too many variables to make it a useful comparison.

Great post
First, the reason to record at 24 bits is simply to be able to record at conservative levels without running into granulation noise/ quantization errors.
Digital usually offers "headroom" of 18 to 20 dB and that's not enough to print hot levels and get away with it so you record conservative levels with peaks of -12-ish. These days analog front ends, even high end ones do not have sufficient headroom, our Euphonix System 5 allows for +24dBu at the input, our old Soundcraft 800B beat that figure by 13 dB for a staggering 33dB of headroom, that way you can print hot and be safe at the same time.
"Recent years" is a bit inaccurate, it's been going on for a very long time but the good news is that the loudness wars are about to end.What drives the loudness wars is the desire to be the loudest among your peers on the radio/ internet dial. There is a seismic shift going on in the industry from peak normalization (by default, 0dBFS is the normalized value) to loudness normalization. Microsoft Windows 8, Spotify, Itunes radio and others are adopting a loudness standard that will ensure playout at -16 LKFS no matter what, allowing for a higher DR if so desired. If you submit your master at -5LKFS with a DR of 5 you will find your volume turned down by 11 dB on those outlets and your mix will sound like ass in comparison to a mix that was mastered with more DR at the correct levels. In other words, it will not pay off to be the loudest much longer. Good news IMO.
 
Y

yeti

Guest
While it is certainly true that dynamic range has been damn near destroyed by the volume wars, MP3 encoding does indeed dispense with a huge amount of harmonic content as well. Massive amounts of the bandwidth are eliminated.

Another area of study is that audio CDs at the standard 44k still fair quite well in A/B testing compared with vinyl or other so-called hi-res or analog sources.

I find that using all tube audio gear has an amazing "softening effect" on some of the offensive square wave high frequency artifacts.

It's simple math, if your DR (softest sounds to loudest sounds) is 6 then you need 1 to 2 bits to encode it, multiply times 2 for good measure and you'll see that 4 X 44.1 = 176.4 Kbps = close to mp3 at 192Kbps. Not very "lossy" because only inaudible components were tossed out. Someone with more expertise want to correct me here?
 
Top