AVS Forum banner
1 - 20 of 25 Posts

·
Registered
Joined
·
175 Posts
Discussion Starter · #1 ·
I've read many posts from people deriding others for claiming that they hear a difference between uncompressed and lossless audio codecs.


I recently came across a post on the quadrophic forum that could explain the difference.


It's the last post on this page by Jimby.
http://www.quadraphonicquad.com/foru...ersal+dualdisc

Quote:
I've produced or executive produced every U.S domestic DVD-A and Dual Disc release for UMG, which is about 50 titles at my last count. As you know (or you should know), MLP compression can be quite variable based on the content being compressed. Now there are tricks to make audio compress better with MLP, such as high and low pass filtering, or truncating bit depth, but we don't play those games.

So, even though a track is lossless, it could have been filtered to help it compress down to a smaller file size than it would have otherwise. I assume an uncompressed track would be the same size with or without filtering and their would be no advantage for it in that sense.


Just something to think about next time someone posts about preferring uncompressed tracks.
 

·
Registered
Joined
·
13,326 Posts
I am one of the people that think LPCM, may sound a bit better than TrueHD. But, alas knowing all I know, I am skeptical. It is a difficult thing to A/B in all honesty.


From the highest bit rate lossy / compressed formats to the LPCM, your really starting to split hairs here. Test data should be able to show this clearly.


My real amazement over all this is how seemingly low requard basic DD 5.1 low bit rate is held in these days. I mean to me, even this is still high quality audio to this day. Maybe I am easily pleased.
 

·
Registered
Joined
·
1,003 Posts
Well I havent got a HDMI reciever so I havent hear uncompressed PCM yet but the TrueHD sounds great. I also am very impressed with the core DTS MA at 1.5. I accually prefer the lossy core to the analog TrueHD because my reciever has more options to adjust the mix and I get my pretty blue light. I wouldnt mind loosing another 6 wires from my rats nest behind my reciever either. Between my PS3 and my A-1 I have both of my HDMI inputs filled on my tv and I got my 6412 Phase 2 replaced yesterday with a 3416 Phase 1 and it has an HDMI output that I cant use until I get either a HDMI switch or a HDMI switching reciever.


After a 60A2000 an A-1 and a PS3 in one year my wife has let me know that a new 1000-1500 reciever isnt in the best interest of my health.
 

·
Registered
Joined
·
6,213 Posts
If the studio filters the soundtrack in any way before lossless encoding it, then the cow has already escaped the barn. (Presuming the same filtering wasn't applied to the LPCM version as well.)


Given the identical original soundtrack, the decoded lossless Vs. uncompressed either have identical bitstreams or the processing wasn't lossless.


Speaking for myself I consider DD5.1 to be good quality sound, but definitely not high quality sound. Too much is lost for my ears (despite the much vaunted Dolby perceptual processing compression). Based on my listening, I consider DTS at 1.2Mbps or greater to be the minimum for high quality sound, DD+ at 640Kbps is still too much of a compromise (so is ~768Kbps DTS).


PS. Of course, this raises the issue of filtering beyond the audible range, which leads to arguments regarding what is audible and what is not, and includes filter induced phase shift effects.
 

·
Registered
Joined
·
479 Posts

Quote:
Originally Posted by TrevorS /forum/post/0


If the studio filters the soundtrack in any way before lossless encoding it, then the cow has already escaped the barn. (Presuming the same filtering wasn't applied to the LPCM version as well.)


Given the identical original soundtrack, the decoded lossless Vs. uncompressed either have identical bitstreams or the processing wasn't lossless.

I definitely agree w/ TrevorS. If the compressed track is first filtered to get a better compression ratio out of MLP, while the uncompressed PCM remains unfiltered, then of course the end results will not be the same.


In such a case though, it would be inaccurate to say it means uncompressed sounds better than compressed. It would simply mean unfiltered sounds better than filtered.


You could reverse the roles and have an uncompressed track that was digitized at 8KHz w/ 8-bit samples. In this case this uncompressed track would sound like a telephone, but it would not be accurate to subsequently claim that something like Dolby Digital compression actually improves sound quality.
 

·
Registered
Joined
·
2,184 Posts
Honestly, I am willing to bet if they did scientific comparisons most people would not be able to tell the difference. The problem is that hearing is subjective, and it varies from person to person. The only thing that might be of use is a comparison of original vs lossless vs uncompressed on a digital level. Unfortunately, they would have to be compared against the original track, so this is a useless comparison for laymen.


I would caution people that swear they hear a difference to remember that often times people in studies take a placebo in place of the actual drug and swear that they are healed, too. It is not hard to convince yourself that something sounds better if you think it should! This advice goes for high bitrate DD+ or DTS vs TrueHD tracks as well. I doubt there is as big of a difference there as we might hope.
 

·
Registered
Joined
·
829 Posts
This is one of those questions I'd like to see answered by a credible researcher that would perform a double-blind test with an appropriate sample population that included your J6P's as well as enthusiasts such as those on this forum. Only then could a statistically significant answer be given. Until then, its all speculation.
 

·
Registered
Joined
·
2,646 Posts
I would think that Dolby TrueHD lossless or dts-HD Master Audio lossless is easy for an audio engineer to verify against linear PCM. Just convert/capture to .wav and do a digital compare of the bits. No subjective hearing needs to be involved. Of course this needs to be evenly matched for the bit depth (16, 20, 24) bits and sampling (44.1, 48, 96, 192) kHz.
 

·
Registered
Joined
·
3,265 Posts

Quote:
Originally Posted by MSmith83 /forum/post/0


The stupid thing is that people on this forum are comparing two different movies in order to assess the difference between uncompressed and losslessly compressed soundtracks.

EXACTLY! I'm flabbergasted each time I come across somebody saying that they think LPCM sounds better then TrueHD, considering that no movie has been encoded in each. And from long standing AVS members as well. *cough* Gary *cough*
 

·
Registered
Joined
·
3,265 Posts

Quote:
Originally Posted by bobgpsr /forum/post/0


I would think that Dolby TrueHD lossless or dts-HD Master Audio lossless is easy for an audio engineer to verify against linear PCM. Just convert/capture to .wav and do a digital compare of the bits. No subjective hearing needs to be involved. Of course this needs to be evenly matched for the bit depth (16, 20, 24) bits and sampling (44.1, 48, 96, 192) kHz.

Except that a lossless compression scheme simply "zips" the audio stream, so once it's "unzipped", it will be identical to it's original form. In the same way that if I use winrar to archive a dvd, I will get the exact same data once I unrar it.
 

·
Registered
Joined
·
3,993 Posts
Some of the amusing comments remind of some of the comments on Hygrogen Audio regarding people asking about Wav vs FLAC.


I think some people are stuck on thinking "Lossless" audio is just marketing hype and they do not realize that with most lossless codecs you can return the audio back into a bit for bit copy of the original source.


Another thing is that I've read that the levels on some TrueHD discs are lower for the TrueHD tracks. If this is the case it will most certainly be perceived as a quality loss.
 

·
Registered
Joined
·
7,997 Posts

Quote:
Originally Posted by miata /forum/post/0


This stuff is getting too complicated for me. I thought that by definition lossless meant lossless

Yes, but if the original is already compromised ....


The question I have for OP (and others) is ... do you guys think there is no other processing that goes on ? And these other processing things are not doing anything like filtering or EQ ?


I don't think we get to hear much of unprocessed - accoustic - sound. Its all heavily processed cr*p to start with. One of the reasons I don't care about the audio part in movies - compared to music.
 

·
Registered
Joined
·
6,213 Posts

Quote:
Originally Posted by c.kingsley /forum/post/0


Honestly, I am willing to bet if they did scientific comparisons most people would not be able to tell the difference. The problem is that hearing is subjective, and it varies from person to person. The only thing that might be of use is a comparison of original vs lossless vs uncompressed on a digital level. Unfortunately, they would have to be compared against the original track, so this is a useless comparison for laymen.


I would caution people that swear they hear a difference to remember that often times people in studies take a placebo in place of the actual drug and swear that they are healed, too. It is not hard to convince yourself that something sounds better if you think it should! This advice goes for high bitrate DD+ or DTS vs TrueHD tracks as well. I doubt there is as big of a difference there as we might hope.

I think what is more likely to be audible is 16/48 Vs. 20/48. Take an LPCM track at 16/48 and compare it to TrueHD at 20/48 and I would expect the DTHD track to sound more natural -- implying, of course, they would sound different.
 

·
Registered
Joined
·
6,213 Posts

Quote:
Originally Posted by Capek /forum/post/0


EXACTLY! I'm flabbergasted each time I come across somebody saying that they think LPCM sounds better then TrueHD, considering that no movie has been encoded in each. And from long standing AVS members as well. *cough* Gary *cough*

Hey now, if it's BR, it's better -- besides, HD-DVD has a higher bar to jump!
 

·
Registered
Joined
·
1,545 Posts

Quote:
Now there are tricks to make audio compress better with MLP, such as high and low pass filtering, or truncating bit depth, but we don't play those games.

Sounds like a Red Herring to me. Doing the things described here would make an audio stream more compression friendly with just about any CODEC, especially the obvious "truncating bit depth" step.
 

·
Registered
Joined
·
13,531 Posts

Quote:
Sounds like a Red Herring to me. Doing the things described here would make an audio stream more compression friendly with just about any CODEC, especially the obvious "truncating bit depth" step.

Unfortunately, this is already done on LPCM.


Take a look at the Unofficial Blu-ray Audio and Video Specifications Thread and count the number of disks with 24/48 LPCM tracks and compare it to the number of disks with filtered and truncated 16/48 LPCM.
 

·
Registered
Joined
·
13,326 Posts
I think in the end we just like the LPCM concept.



As I said really hard if not impossible to A/B this.


Test results with analog testing equipment would show nearly identicle results.. 1.5mb lossy -vs- -TrueHD -vs- DTS-MA -vs- LPCM. I would bet on it.
 
1 - 20 of 25 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top