AVS › AVS Forum › Audio › Surround Music Formats › Anyone hearing big differences between DTS-HD MA and TrueHD and regular Dolby 5.1?
New Posts  All Forums:Forum Nav:

Anyone hearing big differences between DTS-HD MA and TrueHD and regular Dolby 5.1?

post #1 of 34
10/12/12 at 9:31am
Thread Starter 

I searched for information on this and couldn't readily find an existing thread. I wonder if anyone has done any measurements of the content of a DTS-HD MA/TrueHD lossless track and of its counterpart DTS/DD 5.1 lossy track and then subtracted one result from the other to arrive at a definitive answer as to what the difference actually is?

 

Despite having a flagship AVP (Onkyo 5509), good amps (Emotiva XPA-3/UPA-2) and very good speakers (M&K S150s/SS150s), I find the difference between the audio formats to be more subtle than I would expect them to be considering one is lossless and the other lossy. I guess it depends on what exactly is 'lost' in the 5.1 encodes. 

 

Has anyone done the measurements?  If so, have they published the results?  If not, is anyone willing to do them?

Reply
Reply

AVS Top Picks

post #2 of 34
10/12/12 at 10:47am
Many people found very little difference between 1.5Mbps DTS / 640kbps DD tracks and their HD variant.
I guess lower bitrate audio track (such as 320kbps) will be more pronounced.
Reply
Reply
post #3 of 34
10/12/12 at 11:25am
Remember that both Dolby and DTS did substantial research to determine what sounds their lossy encoding algorithms could remove without causing any significant audible effects.
Reply
Reply
post #4 of 34
10/12/12 at 11:51am
Sure. Listen to the Dolby Digital 5.1 track for mid-bass as well as sharpness on cymbals. Dolby Digital has a tendancy to distort lower-level (but audible) mid-bass regions as well as "overshooting" on peaks. So cymbals have a kind-of MP3-ish quality to them. Not as bad, but somewhat.

Don't get me wrong, for the small bandwidth that AC3 uses, it is really impressive but not the same as an LPCM (which is what Dolby TrueHD and DTS-MA HD becomes). Of course, a bad mix or mastering of audio will greatly outweigh any difference between the two. And, bad room accoustics will also tend to muddy the waters, as well.
Reply
Reply
post #5 of 34
10/13/12 at 12:58am
The comparison should really be between PCM and the lossy formats. Throwing lossless codecs in the mix just confuses the issues because of possible confounding factors like DialNorm, difference in levels between dts-HD and PCM in some AVRs, etc.

Objective comparison has already been done. Results posted in 2009 from someone who encodes audio for concert and opera DVDs and BDs. Scroll down a bit for spectrum, unfortunately most pictures are lost.
Quote:
Originally Posted by Andrew_HD View Post


I've done my own test and results are quite different. The difference is that I have my source and I can compare results to the source, not just only to each other.

Full DTS is very good, almost no difference up to 20Khz.

Half DTS and DD at 448Kbs are transparent up to about 10Khz, but than you can see clear difference in the spectrum. The region where half DTS wins with 448kbs Dolby is above 14Khz. Dolby at 640kbs is very similar to half DTS, but still not as good as full DTS.

Objective proof is that full DTS is much closer to the source than DD at 640kps.

640kbs DD is close to half DTS, but half DTS is better (mostly in high frequencies) than DD at 448kbs.

There is really no magic- DD and DTS have almost the same efficiency, but DD lacks high bitrate option.
Reply
Reply
post #6 of 34
10/13/12 at 3:58am
Thread Starter 

I remember reading  comment on AVS from FilmMixer who said something along the lines that if you measured the HD codecs and the lossy codecs and subtracted the relevant information to look at the differences, all that was really removed in the lossy codec was what, IIRC, he described as 'mush'. I'll try to find the post.

Reply
Reply
post #7 of 34
10/13/12 at 10:41am
Quote:
Originally Posted by kbarnes701 View Post

I remember reading  comment on AVS from FilmMixer who said something along the lines that if you measured the HD codecs and the lossy codecs and subtracted the relevant information to look at the differences, all that was really removed in the lossy codec was what, IIRC, he described as 'mush'. I'll try to find the post.

Be careful with that. If I take a good singer who double-tracks his vocals, the two vocals will be slightly different. The two vocals are very close but the phasing between the vocal tracks is what is slightly different. If the two vocals are then subtracted from each other, the result will tend to show up as mush - almost soundling like room reverb or differences. So, just subtracting and getting mush can be misleading at times. It doesn't necessarily mean a lack of frequency response but could also indicate a number of other differences.

If you take a Dolby Digital output and compare against the original LPCM, what you'll notice is that on sharp peaks, the peak levels go higher with the signal restored from the Dolby Digital encoding. This is an overshoot and has to do with the way Dolby Digital compresses and then restores dynamic levels. DD/AC3 is not just an encoding scheme where the bits are compressed but it also changes the audio signal in a way that is designed to be mostly-transparent to the listener. Most of the time it works, sometimes it doesn't.
Reply
Reply
post #8 of 34
10/13/12 at 4:27pm
Quote:
Originally Posted by alk3997 View Post

If you take a Dolby Digital output and compare against the original LPCM, what you'll notice is that on sharp peaks, the peak levels go higher with the signal restored from the Dolby Digital encoding.
Could you post the data showing these overshoots? Thanks!
Reply
Reply
post #9 of 34
10/13/12 at 4:51pm
Quote:
Originally Posted by alk3997 View Post

If you take a Dolby Digital output and compare against the original LPCM, what you'll notice is that on sharp peaks, the peak levels go higher with the signal restored from the Dolby Digital encoding.
Quote:
Originally Posted by Roger Dressler View Post

Could you post the data showing these overshoots? Thanks!

I'd be interested in seeing them as well. smile.gif
Reply
Reply
post #10 of 34
10/19/12 at 10:03pm
Couple of the above replies may very well be in recording studio type of setting.

I have done side by side comparisons and heard no discernable differences.

Your ears and your system will vary but most will not be able to hear any difference whatsoever.
Reply
Reply
post #11 of 34
10/20/12 at 12:11am
I did not find much diff for the mid-range but, in my case, bass becomes tighter and highs crisper.
Reply
Reply
post #12 of 34
10/20/12 at 2:49am
Subjectively HD sounds better to me. Dolby Digital sounds harsh to me.
Reply
Reply
post #13 of 34
10/20/12 at 4:25am
Thread Starter 
Quote:
Originally Posted by Jetmeck View Post

Couple of the above replies may very well be in recording studio type of setting.

I have done side by side comparisons and heard no discernable differences.

Your ears and your system will vary but most will not be able to hear any difference whatsoever.

 

Quote:
Originally Posted by macuniverse View Post

I did not find much diff for the mid-range but, in my case, bass becomes tighter and highs crisper.

 

Quote:
Originally Posted by Theresa View Post

Subjectively HD sounds better to me. Dolby Digital sounds harsh to me.

 

I believe I can detect differences too, but they are fairly subtle - nothing like the big difference between SD and HD video. I have some discs where I prefer the SQ of the DD track over the DTS-MA track.  Remember this is on a very high quality system with reference-grade speakers and prepro and high quality, hight power external amps. Someone listening on a typical system or a mid-range system may struggle to hear any real differences - yet almost universally people seem to be saying that the HD codecs are way better.

 

I would love too see some objective measurements of the FR showing what exactly is 'lost' in the lossy formats, but I have searched and cannot find anything suitable. 

Reply
Reply
post #14 of 34
10/20/12 at 5:18am
Yes, I'd like to see some measurements too. As you know, my system is high end too. Of course "high end" is a subjective term with some audiophiles saying all Onkyo products are mid-fi. I am still satisfied with the SQ of my 5508, Emotiva amps, and actively crossed over speakers.
Reply
Reply
post #15 of 34
10/20/12 at 5:26am
Thread Starter 
Quote:
Originally Posted by Theresa View Post

Yes, I'd like to see some measurements too. As you know, my system is high end too. Of course "high end" is a subjective term with some audiophiles saying all Onkyo products are mid-fi. I am still satisfied with the SQ of my 5508, Emotiva amps, and actively crossed over speakers.

Well you know my feelings about modern electronics working within their design parameters - very little influence on final SQ. Room and speakers is where the differences are heard and I know you are very well equipped in that regard. I would have expected a pretty substantial difference between a lossy and a lossless codec but I guess it depends on what is thrown away with the lossy format, which is where some measurements would be useful.

Reply
Reply
post #16 of 34
10/20/12 at 6:21am
Yes I know your views and am 100% in agreement. The biggest subjective audio difference that I notice is the very poor audio qualiity of Netflix using a Windows PC compared to nearly any other format.
Edited by Theresa - 10/20/12 at 6:33am
Reply
Reply
post #17 of 34
10/20/12 at 6:45am
Thread Starter 
Quote:
Originally Posted by Theresa View Post

Yes I know your views and am 100% in agreement. The biggest subjective audio difference that I notice is the very poor audio qualiity of Netflix using a Windows PC compared to nearly any other format.

 

I don't do any streaming of HD content because the quality isn’t there. Seems to me the best way to deliver HD content ATM (and for the foreseeable) is the good old-fashioned Bluray disc. Given the sheer amount of data on a 50Gb BD, I can't see the day coming any time soon when streaming or downloading services stand a chance of competing on a quality basis with like-for-like AQ and SQ. Setting aside the tangible benefits of actually owning a disc (or even renting a disc for that matter), I think it will be a long time for the tech to catch up to enable us to receive top quality content in any other way than tangible formats. Heck, in the countryside where I live, I can't even get 2 meg broadband - try streaming with that! (I had to go to expensive satellite broadband with a huge dish (30 inches diameter) to get 12 meg BB - not an option for many people.

Reply
Reply
post #18 of 34
10/20/12 at 7:05am
Yes, Netflix quality is very poor, especially compared to an actual BD. I also subscribe to Netflix' "disk rental program" where I get three disks at a time. Sometimes I watch them on my main system, other times on my computer with the audio handled by the Asus STX headphone card. Its with the latter, listening through headphones of course, that the difference in SQ of various compression schemes becomes audible to me.
Edited by Theresa - 10/20/12 at 7:37am
Reply
Reply
post #19 of 34
10/20/12 at 4:18pm
Quote:
Originally Posted by Theresa View Post

Yes, Netflix quality is very poor, especially compared to an actual BD. I also subscribe to Netflix' "disk rental program" where I get three disks at a time. Sometimes I watch them on my main system, other times on my computer with the audio handled by the Asus STX headphone card. Its with the latter, listening through headphones of course, that the difference in SQ of various compression schemes becomes audible to me.

I got rid of Netflix streaming only get two Blu Ray at a time. Streaming on a 10 feet wide screen doesn't look good smile.gif
Reply
Reply
post #20 of 34
10/21/12 at 4:55am
Quote:
Originally Posted by wse View Post

I got rid of Netflix streaming only get two Blu Ray at a time. Streaming on a 10 feet wide screen doesn't look good smile.gif

VUDU is the best option for streaming video/audio ((up to) DD plus 7.1) quality.
Reply
Reply
post #21 of 34
10/21/12 at 4:59am
Yes, my TV is "only" 50" and viewed from 7-8 feet away. I'm certain this covers a multitude of "sins." Old tv shows have to be watched uncritically.
Reply
Reply
post #22 of 34
10/27/12 at 11:47am
Subjectively, I find the HD/MA tracks on Bluray noticeably superior to DD or DTS 5.1. Not only is dynamic range better, the detail to be had in the HD tracks is incredible.

Now, that's on a finely tuned rig in a custom room, etc. etc. YMV, but to me, totally worth hunting down the BluRay / DVD-A / SACD of any recording.

Objectively, the delta's between the decoded LPCM of an HD vs lossy-encoded track do show the difference in favor of lossless high-rez.
Reply
Reply
post #23 of 34
10/27/12 at 12:42pm
I agree but why?
Reply
Reply
post #24 of 34
1/6/13 at 12:16pm
Dolby TrueHD and DTS Ma 5.1 have bit rate of 7 to 9 Mbps/sec. Dolbt Digital is only 448 or 640 Kbps/sec. There is difference in SQ
Reply
Reply
post #25 of 34
1/6/13 at 2:29pm
The lossy DD 5.1 and DTS outputs on Blu-ray are done at higher bitrates than the ones encoded on DVD. So, there are really three comparisons to make - lossy DVD, lossy Blu-ray, and lossless. While I have never seen a rigorous study on the subject of audible differences, the anecdotal evidence seems to suggest the differences are subtle at best, especially when it comes to lossless and the higher bitrate DD 5.1 and DTS encodes on Blu-ray. In my personal experience, I rarely find lossless better than the max legacy bitrate lossy encodes on BD.
Reply
Reply
post #26 of 34
1/6/13 at 9:39pm
what is the bit rate of Dolby Digital on Blu-Ray? Know that DTS on DVD is either 700Kbps/sec or 1.5 Mbps/sec.
Reply
Reply
post #27 of 34
1/7/13 at 4:34am
Thread Starter 
Quote:
Originally Posted by mgpt6 View Post

Dolby TrueHD and DTS Ma 5.1 have bit rate of 7 to 9 Mbps/sec. Dolbt Digital is only 448 or 640 Kbps/sec. There is difference in SQ

 

If it's the bitrate that makes the difference then it ought to be fairly easy to measure and graph what we are actually losing. There are thousands of subjective views and opinions on this - what I am looking for is some evidence of what the differences are. If there is an audible difference then it can be measured - surely someone must have done this so we can see exactly what we are 'losing' on a lossy disc?

Reply
Reply
post #28 of 34
1/19/13 at 8:06am
HI, I'm not an audiophile (I actually have some hearing damage on one side due to the army, but it's minor).

I don't know how you could measure the difference, but it definitely is there.

I just used a pc with some in ear headphones (made sure they are the low hz type, down to <6hz, not particularly expensive) and load up a 1080p movie that has both DD AC3 and TrueHD. I loaded up 2 instances of VLC and the movie twice (one for either track), set it to the same scene and turned off the lights and closed my eyes thinking I would really have to concentrate. But to my surprise the difference was almost night and day. There is so much more sound information in the True HD one. It's hard to describe but like in a restaurant scene if someone puts down a fork on a table and it hits a knife in the DD you's only hear a plink, where in the True HD you'd hear the fork hit the metal of the knife and the cloth of the table. The True HD one is like you are there on the set.

The 2 files I compared were A52 Audio AC3 sample rate 48000, Bitrate 640 kb/s and the other track was TrueHD audio, Sample rate 48000hz, Bits per sample:32. The movie that I used to compare was Ghostbusters 1984 1080p.

Maybe the best way to compare is not through a complicated high end processor but simple (low hz) in ear head phones and a pc and compare the pure soundfile.

Come to think of it, perhaps you could just mux out the soundtrack and cut say 5 seconds of it and just subtract both waveforms using Audacity or something, that way you could physically hear the difference.
Reply
Reply
post #29 of 34
1/19/13 at 1:32pm
Quote:
Originally Posted by gold333 View Post

I loaded up 2 instances of VLC and the movie twice (one for either track)
While lossless decoders are expected to be, well, lossless, there is no assurance that the Dolby decoder is working to its fullest extent. Lossy decoders are not "bit for bit" accurate to any reference. For example, certain computations need to be performed with extended precision in order to meet Dolby performance requirements, and that information is not implicit in the A/52 spec. It is part of the information Dolby provides to its chip licensees. Since we do not know the lineage of an open source DD decoder, definitive conclusions cannot be made.
Quote:
The 2 files I compared were A52 Audio AC3 sample rate 48000, Bitrate 640 kb/s and the other track was TrueHD audio, Sample rate 48000hz, Bits per sample:32. The movie that I used to compare was Ghostbusters 1984 1080p.
Is there a way to know that both soundtracks were made from the exact same master? This would more likely if they both came from the same disc, but that was not the case here.
Reply
Reply
post #30 of 34
1/20/13 at 7:08am
To be honest your answer goes over my head I'm afraid, I'm not very technical.

I just have a tendency now for when I have a choice in watching a movie to select the True HD option because of that experience. Anyhow it's not like those extra mbps are empty.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Surround Music Formats

AVS Top Picks

AVS › AVS Forum › Audio › Surround Music Formats › Anyone hearing big differences between DTS-HD MA and TrueHD and regular Dolby 5.1?