How many people are dissatisfied with DVD sound quality?

Sure, DVD standard definition picture quality leaves something to be desired on a 1080p monitor, but AC-3 through an S/PDIF cable into a nice home theater receiver and decent speakers seems perfectly good already.

I doubt a higher bit rate coder would make a perceptible difference to most ordinary listeners. The limit of human hearing is just over 20KHz (analog). For digital sampling, you multiply that by around 2.2. For example, phone transmissions are engineered for 3KHz analog and transported at 8KHz digital. So I think these 96KHz and 192KHz sampling rates are a waste. The standard 48KHz is more than adequate for the limits of human hearing.

And then there's the 24-bit sampling depth. There is no audio recording equipment on earth that can record that much dynamic range. And even if it could, nobody would want it. When the source audio track goes to near silent, many listeners will be annoyed. Most well engineered movie sound tracks always have something going (music, birds chirping, etc.) to keep the listener interested. So the dynamic range of most movies could probably be captured with around 12 bits. 16 bits is already more than adequate.

Then there's all the hoopla over compression algorithms. Many audio purists will dismiss any lossy compression algorithms without even giving it a proper listening test. The concept of audio compression goes like this: If you hear an alarm clock ticking, then the alarm bell rings, you don't need to encode the ticking when the alarm rings since no one will hear it. Double-blind listening tests have shown that many lossy encoded sound tracks can't be picked out even by audio mastering engineers (golden-ears people who do this for a living). Interestingly, it's often people with significant hearing loss that tend to pick out the lossy compression algorithms. If you can't hear the high frequencies of the alarm bell, then you'll notice the ticking has gone away.

I suspect all these new Dolby / DTS standards are mostly aimed at getting people to buy new home theater sound equipment. Remember that Dolby Labs & DTS both make a small royalty for each new receiver unit sold. Given that existing audio equipment can always use the core Dolby Digital track (without the higher bit rate extensions) over S/PDIF, I would welcome a Tom's listening review of these new sound schemes to see if there is any real difference.

Dear Dogman:
If you check out the quality sites that rate DVD audio, you'll see comments from people with very "good ears" that give Dolby TrueHD and DTS-HD Master Audio considerably higher ratings than they give to core Dolby Digital or DTS sound tracks. My own experience on my "less-than-state-of-the-art" but good quality audio gear (5.1 is as high as I go right now, because that's as many speaker as I've cared to purchase, position, and EQ for my home theater rig) confirms these confirms these other claims. I can hear more detail and richer sound when decoding DolbyTrue HD than I can when decoding normal Dolby Digital.
The sites that do a good job of rating Blu-ray DVD audio include Cinema Squid, www.highdefdigest.com (I cited a couple of their reviews in this story), www.blu-ray.com, www.soundandvisionmag.com, and www.hometheaterforum.com. A little digging can turn up loads of opinions around this subject matter, most of which leans in the direction that decoding the lossless codecs indeed requires new (and more expensive) gear, but is worth to those to whom such things matter.
HTH,
--Ed--

Hi Ed,

I really hope that you continue to do home theater articles for Tom's. This is, by far, the best article that I have ever seen Tom's publish on home theater. It sounds like a.) you really know your stuff, or b.) you took the time to thoroughly research the topic or c.) both. The in-depth info on the various HD audio formats is fantastic.

I hope that the part 2 article is just as in-depth, and I am hoping that you will at least partly cover the Asus Xonar AV1.

My personal preference is to have a stand-alone Blu-ray recorder, but in the US, those do not exist so far. I have heard rumors that Toshiba will release one later this year, however, if they do, it remains to be seen if it will record from an HDMI input. Perhaps the combination of the Xonar AV1 in a HTPC would provide an acceptable alternative.

Thanks again for the superb article.

Best Regards,
Matthew

Gee thanks Matthew. As far as your a-b-c list goes, it's mostly (b) with a growing amount of (a), hopefully heading toward (c)! I've had the good fortune to work with more accomplished home theater gurus including Mike Chin of SilentPCReview.com, and also Matt Wright (formerly of HTPCnews.com, now with missingremote.com), and to establish good relationships with home theater/audio experts at AMD, NVidia, and Intel. With a support network like that, I'm glad to dig into these topics for the readership, as much out of a desire to get out of home theater PCs what I think they should deliver, as out of plain orneriness (when it doesn't work that way) and curiousity (when it does).

And you're right: the only Blu-ray recorders I'm aware of that are currently available include $3,000+ models from Panasonic and Sony available only in Japan at present. It would be quite a coup for the HD DVD champion to be first to market in the US (or anywhere) with an affordable Blu-ray burner.

And yes, I do plan to touch on the Asus Xonar as well as some other upcoming products (most notably from Auzentech, a Korean-based outfit that plans to release direct HDMI 1.3a audio support later this year, and upcoming stuff from Nvidia that promises to do likewise). If anything, I hope you'll find Part 2 even *more* interesting than Part 1...

--Ed--

People claiming music of one standard sounding superior to another may not be doing a double blind test. I've read serious posts about how XM radio sounds as good as a CD - anecdotal evidence is an oxymoron. If studio engineers could not tell the difference in a blind test - well, I trust science not opinion. Unless it's my opinion of course

This is an excellent article on the subject!

I have researched the subject and stopped my build of the media pc when the truth became clear....

The audio part is not ready for prime time yet!

I am of the middle-aged plus males group where hearing loss becomes rather common, due to nerve damage. Somehow I can still discriminate between media and hardware that are very good versus not good, although there is much of the dynamic range that is no longer heard. Simply, there is much of what is new each year that is just wasted $$$ for those of us with hearing losses. I like new equipment like the next person, but I'm not sure what is worth my money and what isn't.

One avenue that is open to us for saving some money, currently, is by building our own speakers. There are many places via Internet and elsewhere for tutorials and less than retail priced components.

Awesome article - can't wait for the follow-ups. Quick response to one of the comments above -

Unfortunately there's no way any device that is HDCP compliant will be recording via HDMI as the HDCP license explicity defines what an HDCP "Presentation" device (any device capable of receiving, decrypting and displaying HDCP protected content) is allowed to do.

The HDCP license says:

---------
A Presentation Device shall not make any copies of Decrypted HDCP Content for any purpose, except for such temporary buffers
---------
And
---------
Decrypted HDCP Content may be temporarily buffered in a Presentation Device to enable and perform the Presentation Function, image processing function (e.g., picture-in-picture display, image overlay, image enhancement and brightness adjustment) or ?freeze frame? of a single frame of Audiovisual Content, provided that such buffer shall not persist for more time than is necessary to perform such function.
---------

Dear Jawshoeaw:

You're absolutely correct about the double-blind comparison. That said, if you check the various Web sites I cited in my previous post, they do provide separate ratings and rankings for Dolby Digital versus Dolby TrueHD, and DTS versus DTS-HD Master Audio *for the same movies*. Two separate auditions, two separate rankings from people who know their stuff is good enough for me, though that doesn't mean it needs to be good enough for you. That said, I have personally compared the two sets of tracks on my own equipment and (though I didn't do a double blind comparison because I have only one Blu-ray player to work with) I believe I can hear an audible difference betweeen Dolby Digital and Dolby TrueHD, as do other members of my family, and friends who've come over to watch movies. Again that's only FWIW, which may be more to me than to you.

--Ed--

Dear AVenVY:

Great interpretation of the standards. I didn't mean to suggest that this device would allow copying of decrypted HDCP content, though it should finally provide a superior way to burn HD camcorder content onto DVD. The content providers (aka movie studios) would go ballistic if such devices permitted knockoffs of *gasp!* copyrighted material.
--Ed--

Have you ever even heard a high resolution audio source such as DVD-A, SACD, or high res download from places like MusicGiants SuperHD collection? If you want to understand how practical considerations impact sample and bit rate theory have a read of Dan Lavry's (one of the most respected audio design engineers in the business) white papers on the topics (http://www.lavryengineering.com/supportpage.html).

And there's no recording equipment on earth capable of 24-bit resolution? Umm how about just about any pro-level ADC such as those from Lavry (http://www.lavryengineering.com/), Apogee (http://www.apogeedigital.com/), RME (http://www.rme-audio.de/), or Lynx (http://www.lynxstudio.com/) to name a few off the top of my head?

Not to mention all of the Blu-ray and HD DVD sources also recorded at 24 bit sampling rates. But hey: a rant is a rant, and some rants are better than others!
;-)
--Ed--

Meanwhile a fanless HTPC with the free (as in speech as well as in beer) FLAC or WavPack codecs can deliver the same quality for a lot cheaper. Those codec wars, along with the media wars, exist only to make people spend more money.

Meanwhile a fanless HTPC with the free (as in speech as well as in beer) FLAC or WavPack codecs can deliver the same quality for a lot cheaper. Those codec wars, along with the media wars, exist only to make people spend more money.

Meanwhile a fanless HTPC with the free (as in speech as well as in beer) FLAC or WavPack codecs can deliver the same quality for a lot cheaper. Those codec wars, along with the media wars, exist only to make people spend more money.

Meanwhile a fanless HTPC with the free (as in speech as well as in beer) FLAC or WavPack codecs can deliver the same quality for a lot cheaper. Those codec wars, along with the media wars, exist only to make people spend more money.

Meanwhile a fanless HTPC with the free (as in speech as well as in beer) FLAC or WavPack codecs can deliver the same quality for a lot cheaper. Those codec wars, along with the media wars, exist only to make people spend more money.

This is a very well-written and informative article on what is a potentially confusing and contentious subject. If I ever get around to writing a FAQ section for my website, I'll make sure to link back to this.

First, thank you for the many generous mentions of my site in your article. Cinema Squid is still very young (only about two months old), so the information presented on the site may not be entirely accurate, but I am continually correcting entries as reports are given and personal investigations are conducted. Much of the current content is taken from the "back of the box" so to speak, and the studios are notoriously incapable of correctly representing the contents of their discs. However, I recommend that anyone interested in the subject of collecting accurate information about Blu-ray disc releases participate in the "Unofficial Blu-ray Audio and Video Specifications Thread" at AVS Forum (http://www.avsforum.com/avs-vb/showthread.php?t=760714) which is a community effort to use free and open source tools to examine commercial Blu-ray releases for the real statistical story of their contents.

I should say that my personal audio discrimination is not the greatest and thus the audio rankings on Cinema Squid are currently derived as weighted averages of reviews from Blu-ray.com, High-Def Digest, DVD Talk, DVDTOWN, Home Theater Spot, The Man Room and Upcoming Discs. I also include user rankings from the AVS Forum "Tier Thread for Audio" (http://www.avsforum.com/avs-vb/showthread.php?t=918734) which is a good place to participate in tiering audio tracks if you've been blessed with golden ears and top-end equipment. I am always looking to include good sources for audio rankings, since it is probably the trickiest and most misunderstood subjective area for home video releases.

Several notes on audio codecs:

I have not personally encountered a Dolby Digital Plus audio track on a Blu-ray disc, although it is a supported optional audio codec in the BD spec. Some titles that claim this codec (such as "A View From Space With Heavenly Music") are likely misprints copied from the HD DVD version of the title - I have not yet been motivated to add any of these to my Netflix queue to find out for sure. In any case, DD+ (and DTS-HD High Resolution) are largely useless codecs that do not seem likely to have much of a future, since they can easily be substituted with lossless Dolby TrueHD and DTS-HD Master Audio tracks with only a bit more disc space usage.

The actual bit depth of Dolby TrueHD and DTS-HD Master Audio tracks remains somewhat of a mystery at the moment. The physical audio headers for TrueHD tracks do not report the bit depth (unlike MLP, but similarly to AC3). At the moment, their seems to be an empirical trend for tracks with a bitrate under ~3500 kbps to be 16-bit and those with higher bitrates to be 24-bit. However, to complicate matters, there are reports from industry insiders that some TrueHD tracks, such as those on The Fifth Element and Ghost Rider, were actually encoded at 20-bit.

DTS-HD Master Audio tracks do report their bit depths, but somewhat indirectly. The actual reported bit depth is that of the source uncompressed LPCM audio track. Therefore, this does rule out the possibility that a particular DTS-MA track has not been requantized from a 24-bit source into a 16-bit track or vice versa. Basically all US domestic releases with Master Audio tracks are reporting themselves as 24-bit, with 16-bit tracks relegated to European and Asian releases. Making the situation more impenetrable is the fact that there are no open tools at the moment that can convert DTS-MA tracks into their uncompressed LPCM counterparts (which would allow the audio samples to measured to determine the necessary bit depth to fit the data). There are a number of domestic DTS-MA tracks with suspiciously low bitrates which would seem to imply a 16-bit depth.

I had avoided publishing the bitrates for lossless compressed codecs (TrueHD, Master Audio) since they should be considered deceptive numbers and unrelated to audio quality. The bitrate of these two codecs has no relation to audio fidelity, since they are lossless codecs and thus these bitrates only represent the compressibility of the source. However, due to the unresolved issues of bit depth, I have begun adding these numbers to the listings.

Any comparison between two versions of "the same disc" (i.e Dolby vs. DTS or SACD/DVD-A vs. CD) cannot be taken seriously as you have no confidence that equal mastering techniques were used for those versions.
A SACD version of an album, for expample, might have much less audio compression applied to it (no clipping and overvolumed, distorted sound) compared to a CD version (which is usually overcompressed and distorted these days). This may apply to some extent to Dolby vs. DTS.
You simply can't be sure whether the perceived difference is caused by compression technique itself or by other factors (encoder implementation/settings used, digital audio processing employed etc.).
It has been confirmed that some hybrid SACD audio discs carried an intentionally squashed CD track to make SACD sound "superior" while, pure technically and biologically, vast majority of humans is not able to discern a 44/16 recording from a 96/24 one if proper mastering techniques and proper playback equipment are used for both. So they have to "help them hear the difference". What a BS!