'hi rez' DVD-A? Sometimes.

For those of use less educated on the ripping tools/graphs (me ) can you explain the scales/graphs? To me it looks like the CD version of 'death on two legs' is a better copy than the DVDA.

And those are the reasons why some are good and some are bad. I wonder if the six channel mixes would be better.

Harvest sounds great. Steely Dan sounds great. Fagan is nice.
Some bad ones just sound like the CD only LOUDER.

The good ones are amazing to hear in 6 channel. The best sounding disks are the ones that have quality recording, mixing and mastering. DVD-A should be compared to DTS and Dolby-D not CD's anyway. A well done CD sounds fine, but those are few and far between.

Hi krabapple,

Those graphs are interesting. Can you post info on how you were able to generate them? I'd like to try this on some DVD-As that I have...

Thanks,

- Steve O.

What's the program that generates the graphs?

BTW, despite that horrid graph, Speaking In Tongues still sounds great (to me).

I just want to analyze the compression (or lack of compression) of some of my remastered CDs vs. the original CDs. What program is best to use simply for CDs? Thanks.

Any standard wav editor/recorder. The one used in the screenshots is Adobe Audition, but there are free ones all over the place. Chances are one came with your sound card.
All you do is open the wav file/CD track (may have to rip first depending on the program) and it will show the waveform like that.

Is it possible to reverse the dynamic compression with a tool like the Adobe product mentioned?

I've never tried that. My first instinct is probably no, because you'd be trying to get back parts of the waveform that was simply cut out with the compression - like how a lossy codec simply cuts out information, and it's lost.
You'd have to go back to the original source. There may be some smart methods/algorithms to try and *improve* it, but again I'm no expert there. Either way, I'm sure that a DRC-applied track cannot be fully restored to the original dynamic range.

I tried that - knowing that what was lost could never be regained. I still had to try though, and it was a failed experiment :|

the information is just not there anymore, and all you get is a lower level that may not clip, but theres no dynamic range still, and probably even less satisfying that the first version. at least that one was loud..

You CAN reverse dynamic compression, like that done with a dynamic range compressor or limiter - but NOT "compression" as in lossy data compression such as MP3 and the like.

I still use a dbx-128 processor in my system, which is an all-in-one Dynamic Range Expander/Peak Un-Limiter and dbx Type-II Noise Reduction unit. The, uh, "modern" normalized CD's actually sound quite a bit better when ran through the 128 with an expansion ratio of about 1.4. I set the threshold way up so that, basically, everything but the loudest sound is being expanded downward, and it really seems to nicely restore the impact and overall quality of these nasty normalized CD's.

Also, if you do much FM listening, the dbx-128 is great for peak un-limiting FM signals, which usually have a lot of peak-limiting applied. A gentle expansion ratio of about 1.2 will push down the noise just enough to make the sound much more listenable.

Expansion with peak un-limiting also works well with films that have highly compressed soundtracks meant for that was optimized for optical sound prints - like the HD-DVD audio of the Charlton Heston version of The Omega Man. And if you still watch LaserDisc's, it's great for non-CX encoded FM analog tracks or hissy Dolby-A mastered films (a good example is the original LD of Neil Diamond 'film' The Jazz Singer).

Krabapple,

Do you have the "un-released", non-Brian May mixed/mastered version of the DVD-A of A Night At The Opera? I wonder if it would show different results from the officially-released version that you used? (I assume you used the Brian May mixed version of ANATO)

BTW, for those who haven't heard it (or know the story), the non-Brian May version of ANATO DVD-Audio is truly retched in its 6-track mix... And on "Bohemian Rhapsody", guitars are even missing in certain parts! When the album was originally remixed/mastered for DVD-A by DTS Entertainment, Brian May wasn't involved at all - DTS had the resulting DVD-A pressed, sent to reviewers and ready to ship to stores when Brian May finally heard the disc and said "NO WAY!" - DTS put the release on "hold" while Brian supervised a proper remix - DTS claimed the differences between the two versions were "minor", but that is a vast understatement - besides the actual surround mix differences, which are numerous and obvious, the non-Brian May DVD-A has such heavy NR applied you can hear the hiss duck on and off as voices and instruments come and go. It's just awful.

Anyway, I wonder if it's clipped like the version you used Krabapple?

It goes to show that the record industry cares more about money than giving a good quality product to the consumer.
And it seems some don't even respect the DVD-Audio format. IMO they should take the opportunity for MP3 to be the domain for all the detritus, and keep the good mastering for the hi-res prints.

If this juggernaut trend of crap audio production and consumption cannot be stopped, then we will have to have a definitively separate version of each product for the people who care good sound.
I'm sure that not ONE (good) mastering engineer would ever employ anything like hard limiting of their own volition. They are told what to do by their tyrant executives.

Sometimes the artists also insist on certain "undesirable" techniques (I know I've read about a few such bands, though the specifics elude me at the moment). But the bulk of the "demands" come from marketing departments who don't want their CDs to be 10dB down from the others in the mega changer in the music store.

Well in that case, it is part of their art and if you don't like it it means you don't like that particular track/album that they've applied those techniques to. It's like some films are deliberately grainy or with a certain effect...
I'm guessing genres like heavy rock/metal would be the more likely ones?

I didn't mean to imply it would be 'exactly' the way an uncompressed studio master would sound - just that it can really, really help a lot - whereas with lossy compression systems, there's nothing to be done for them!

Yeah, I have both DVD-A's - during Bohemian Rhapsody, after the middle chorus, when Brian May's guitar is supposed to come in, it's completely missing... VERY bizzare sounding. And as I said, the aggressive gating applied for NR is just awful.

Guys, I don't understand what you're getting at. The subject was "Hi Rez". Nothing in this thread has anything to do with resolution, only dynamic range compression. Every commercial recording ever made (except for classical) uses dynamic range compression; it has to, otherwise a mix would be imposssible. All the way from the classic UREI LA-2a's, 3a's, 1176LNs, and now their digital plug-in counterparts. Usually these days, the peaks just get lobbed off.

It's all up to the discretion and artistry of the mastering engineer. None of this has anything to do with resolution. Can you explain your point?

The first thing that a microphone goes into -- at least in a pop/rock music recording session -- before even being captured on tape or on a digital workstation -- is a dynamic range compressor. It has to. The human voice is too unmanageable in dynamic range to ever record without going into distortion without doing a touch of compression as step #1. Not only that, but it just doesn't sound very good without a touch of compression. It's the same reason a singer in concert pulls the mike away from their mouth when singing a loud note -- that is dynamic range compression performed by the performer.

These days, pop/rock records usually compress vocals down to a no more than 10db dynamic range -- then manually edit out all the breath sounds.

In remixes, compressors are again used just to balance out the mix; otherwise it would be musical chaos. It would just sound like a jam session.

And please don't confuse dynamic range compression with digital encoding compression, i.e. lossy vs. lossless. Neither one has anything to do with the other. Nothing. Except that they both have "compression" in their name. It's like confusing spring the season vs. spring the water vs. spring the twisty metal thing. Neither has anything to do with the other.


-John

Average volume, dynamic range and crest factor are highly correlated with resolution.

No they're not, they have nothing to do with each other. Dude, I'm really sorry, I don't know who you are, and I want you to be a friend, but I have been working as an audio engineer all my life, making my living at it since the mid 1970's. So did my dad and my grandfather. My friend Dave Haynes developed the Studer Dyaxis, the world's first digital workstation. I've been working with digital audio since the early 1980's often traveling to Sony in Teaneck just to evaluate equipment designs and prototypes.

I've been a consultant and an evaluator at Dolby, spent three years developing the country's first media digital audio network at Skywalker Ranch, and have a couple Emmy awards for my work in TV sound. Right now I'm working on several major projects with film studios.

I really, really hate "pulling rank", and I don't want to offend you, but you're spouting words you don't even understand yourself, obviously. Stop it; you're just confusing people.

=John

Irrelevant to the facts of the matter!

http://www.audio-software.com/K-System.html
Narrow Dynamic Range Pop Music
“We can avoid a new loudness race and consequent quality reduction if we unite behind the K-System before we start fresh with high-resolution audio media such as DVD-A and SACD. Similar to the above classical music example, pop music with a crest factor much less than 14 dB should not be mastered to peak to full scale, as it will sound too loud.”

But none of that has anything to do with resolution. 16-bit, 24-bit, and a zillion-bit recordings can still be mastered too loud, clipping the peaks, if that's what you're talking about -- but that has nothing to do with resolution. Resolution is how many bits you're using for each sample.

Are you talking about clipping peaks when mastering too loudly? I, too, hate that, but it has nothing whatsoever to do with resolution, it's simply dynamic range clipping. Bob Katz would be the first to agree with me. But again, it has nothing to do with resolution. Please keep the terminology accurate; I apologize for Bob.

And especially this has nothing to do with whether something is a "high res" recording. A 24/96 recording can still be poorly mastered, with extremely high peak clipping (just too darn loud), but being 96/24, it's still a "high res" recording, by definition. High res is simply anything above 44.1/16, it has nothing to do with how well or poorly mastered it was.

Yeah. I agree. The thread title misleading. Compressed or not, the examples presented as compressed DVD-As are still "hirez".

I DO think the dynamic range compression sounds better in hirez.