DVD/DTS Poll Chris Squire - FISH OUT OF WATER [DTS DVD]

[Jim the Oldbie]

This is just me, but I'm having a hard time assigning a single rating number to this stuff. I don't mind talking about it (and talking, and talking...), but to summarize the experience with a numbered vote is difficult.

 

[jimfisheye]

This is just me, but I'm having a hard time assigning a single rating number to this stuff. I don't mind talking about it (and talking, and talking...), but to summarize the experience with a numbered vote is difficult.

Me too. And then the discussion usually digresses. I just comment now and skip the voting part.

 

[quicksrt]

It would seem to me that the purpose of not utilizing MLP or even LPCM 96/24 is the majors are playing it safe for use in players which don't have DVD~A capability but advertising a codec as 96/24 which, for the majority of us, decodes as a LOSSY codec is simply downright bizarre!

Many do not realize that DTS MA or HD24/96 or whatever DTS want to call it is simply a brand name and not a rating of bit depth or sampling rates. It's like calling a brand of milk and cheese "Organics" but the product is not organic.

 

[ssully]

DTS-HD Master Audio (DTS-HD MA) is in fact a lossless codec that *supports* any sample rate/bit depth up to 196/24 . Which makes it no different from , say, MLP -- you can losslessly package whatever you put into it. No more than MLP (or 'DVD-A' ) on the label or format info guarantees that the content is 96/24, DTS-HD MA doesn't either. Similarly, SACD (and DSD) don't guarantee that the material on the disc is actually 'hi rez'. No more than .flac at the end of a file name says anything about the sample rate or bit depth. That, ultimately, is a function of the digital recording rate (for an all digital recording) , or transfer rate of the analog tape transfer . Not the type of lossless encoding. As many, or at least some, know by now I hope.

DTS is fanatical about making its products 'back compatible' so even lossless DTS-HD will default to a 'lossy' version if the DTS-HD MA decoder isn't available in your hardware. DTS 96/24, which btw is lossy even if decoded by a proper DTS 96/24 decoder, defaults to plain ('lossier') 48/24 DTS if you don't have a proper DTS 96/24 decoder. You can think of DTS products as always having a reliable, extractable lossy 'core' version (output @ 48/24) hiding in them, in case of emergencies. Unless your AVR is completely bereft of *any* DTS decoder, you will always be able to play a DTS source, whatever its species.

It's a feature, not a bug. The amount of time anyone should worry about it, in terms of audibility, approaches zero. But in audiophile world, it's not fun believing that.

 

[jimfisheye]

Many do not realize that DTS MA or HD24/96 or whatever DTS want to call it is simply a brand name and not a rating of bit depth or sampling rates. It's like calling a brand of milk and cheese "Organics" but the product is not organic.

That's not right quicksrt. DTS may be a brand but DTS HD is in fact a format spec that is a lossless container (for their brand). They don't sell lossy in DTS HD containers as a bait and switch.

 

[quicksrt]

That's not right quicksrt. DTS may be a brand but DTS HD is in fact a format spec that is a lossless container (for their brand). They don't sell lossy in DTS HD containers as a bait and switch.

Oh, it's the other DTS 24/96 which is not necessarily 24/96Khz? The HD (flavor) one is just as great as PCM 24/96kHz?

 

[quicksrt]

Yes, and it sounds absolutely *terrible* because as smart people know, that's what Dolby Digital does. It dulls the sound. You can even see it in EQ analysis, if you look really carefully and wear special glasses and a hat made of thin aluminum.

So make sure you do not buy it. (Even though there is a DTS choice and it says 'DTS 96/24' when the disc is playing, it's all lies)

No, Jimfisheye just corrected me, he is saying it's real lossless 24/96 when that darn disc is playing back and your equipment can handle real 24/96. *I was also suspicious but now I'm better. I thought it was all DTS marketing B.S and not real 24/96khz too.

 

[jimfisheye]

That DTS 2496 is in fact the "ringer". Not bait and switch though, just a really really bad idea! Decoded fully, it's slightly but genuinely transparent lossy 24/96. Decoded on gear that only decodes the "core" dts signal, it's heavily lossy. That it also decodes to 48k (in core lossy mode) in that scenario is the telltale but the lossy is magnitudes worse than merely a sample rate conversion to 48k.

Now this may raise an audiophile's eyebrows but it's still not anywhere near say, the mp3 meat grinder. (For reference. Not to justify using this.) I've heard mastering destruction often enough that crosses the line even further than any of the lossy formats so much to make them a moot point. Some of the complaints on some of these releases have more to do with that.

You have to pick your battles I guess. Considering how transparent and very much HD DTS 2496 can be if decoded properly and some of the titles available, it's worth it to jump through a few hoops and get set up. I bring this up often as an FYI because this is a tricky one to deal with. Hopefully at least a couple people will suddenly be listening to their DTS 2496 titles in fullest quality for the first time.

The treble blast "mastering" where apparently someone wanted to be able to listen with earbuds - with the earbuds sitting out in the next room from them! - is still the biggest offender in many releases. And that's not the fault of any lossy format.

There was an opportunity with that first Tull Aqualung 5.1 edition to compare DTS 2496 with LCPM 24/96. They nearly nulled. Whereas comparing the fully decoded DTS 2496 to the lossy core decoded of the same was striking and obvious immediately.

I still don't like this kind of screwing around but I'll take this over stuff like knowing there's a SW remix of Deep Purple Made in Japan that someone refuses to release. Or a pricey Moody Blues boxed edition that turns out to have the worst upmix you've ever heard for the 5.1 "mix".

I still think this Squire remix sounds really good! Who really mixed this?

 

[quicksrt]

"That DTS 2496 is in fact the "ringer". Not bait and switch though, just a really really bad idea! Decoded fully, it's slightly but genuinely transparent lossy 24/96. Decoded on gear that only decodes the "core" dts signal, it's heavily lossy. That it also decodes to 48k (in core lossy mode) in that scenario is the telltale but the lossy is magnitudes worse than merely a sample rate conversion to 48k. "

This is some fantastic writing, that I am not sure I get it at all. The "ringer" flavor is the brand name plain DTS24/96 that is not 24/96kHz like my LPCM ELP or Grateful Dead DVD-As are? And the top of the line DTS "flavor brand" to look for is the "MA" flavor, like chili cheese flavor Fritos, not plain? DTS MA is the deluxe real lossless 24/96kHz I take it?

 

[jimfisheye]

"That DTS 2496 is in fact the "ringer". Not bait and switch though, just a really really bad idea! Decoded fully, it's slightly but genuinely transparent lossy 24/96. Decoded on gear that only decodes the "core" dts signal, it's heavily lossy. That it also decodes to 48k (in core lossy mode) in that scenario is the telltale but the lossy is magnitudes worse than merely a sample rate conversion to 48k. "

This is some fantastic writing, that I am not sure I get it at all. The "ringer" flavor is the brand name plain DTS24/96 that is not 24/96kHz like my LPCM ELP or Grateful Dead DVD-As are? And the top of the line DTS "flavor brand" to look for is the "MA" flavor, like chili cheese flavor Fritos, not plain? DTS MA is the deluxe real lossless 24/96kHz I take it?

It's not cryptic like that, no.

DTS 2496 is encoded from an LPCM 24/96 master.
If you use a codec that decodes it as fully as possible (per the design), it's a very transparent barely lossy copy of the original 24/96 LPCM. The decoded file is 96k. To the point that it almost nulls with the original even.
If you use a 'legacy' codec (as found in many hardware and even software DVD players) only part of the DTS 2496 data stream is decoded. The decoded file here is 48k.

I say "ringer" because there are so many hardware players in the wild that only have the 'legacy' decoder.

Someone made the decision that it was preferred to have older hardware at least output some sound as opposed to rejecting the disc. Opponents of this decision argue that it would be better to reject the disc (forcing someone to upgrade if they wanted to play it) than to play a degraded version. I call it a ringer because there seem to be a lot of hardware and even software media players that only use the legacy decoder.

DTS is a brand sure. Dolby is another one. (They had the significantly poorer lossy format but they now also offer a lossless version. I don't think they have an "in-between" "ringer" format equivalent to DTS 2496). These aren't meaningless marketing buzzwords. They're companies that make these formats.

Another argument for moving into the 21st century and using a computer and software media players IMHO. When stuff like this comes along, it's just a software hunt. No replacing expensive hardware.

I posted the test results using the Tull files I mentioned a few years ago actually. Not sure if I saved them. When you can take two files and A/B them and hear no difference, and then subtract one from the other (null test) and get nearly zero, that's pretty matter of fact. Likewise when you subtract the core-only dts 2496 decode from a full dts 2496 decode and both hear obvious difference between them and they don't even come close to nulling, that's pretty matter of fact too. (And yes it should go without saying that you need to verify they are at precisely the same volume. Before someone wants to suggest this is an amateur hour volume mismatch between them.)

Someone can feel free to move this tech corner digression to another thread if this is too much.

 

[himey]

It's not cryptic like that, no.

DTS 2496 is encoded from an LPCM 24/96 master.
If you use a codec that decodes it as fully as possible (per the design), it's a very transparent barely lossy copy of the original 24/96 LPCM. The decoded file is 96k. To the point that it almost nulls with the original even.
If you use a 'legacy' codec (as found in many hardware and even software DVD players) only part of the DTS 2496 data stream is decoded. The decoded file here is 48k.

I say "ringer" because there are so many hardware players in the wild that only have the 'legacy' decoder.

Someone made the decision that it was preferred to have older hardware at least output some sound as opposed to rejecting the disc. Opponents of this decision argue that it would be better to reject the disc (forcing someone to upgrade if they wanted to play it) than to play a degraded version. I call it a ringer because there seem to be a lot of hardware and even software media players that only use the legacy decoder.

DTS is a brand sure. Dolby is another one. (They had the significantly poorer lossy format but they now also offer a lossless version. I don't think they have an "in-between" "ringer" format equivalent to DTS 2496). These aren't meaningless marketing buzzwords. They're companies that make these formats.

Another argument for moving into the 21st century and using a computer and software media players IMHO. When stuff like this comes along, it's just a software hunt. No replacing expensive hardware.

I posted the test results using the Tull files I mentioned a few years ago actually. Not sure if I saved them. When you can take two files and A/B them and hear no difference, and then subtract one from the other (null test) and get nearly zero, that's pretty matter of fact. Likewise when you subtract the core-only dts 2496 decode from a full dts 2496 decode and both hear obvious difference between them and they don't even come close to nulling, that's pretty matter of fact too. (And yes it should go without saying that you need to verify they are at precisely the same volume. Before someone wants to suggest this is an amateur hour volume mismatch between them.)

Someone can feel free to move this tech corner digression to another thread if this is too much.

Have you ever encoded your own DTS files from 24/96 PCM sources to both standard DTS and DTS 24.96? You do realize that both will have the same bitrate?

Having done my own DTS encoding in the past, before my equipment did uncompressed PCM, I don't hold DTS 24/96 to any higher standard than standard DTS. Again, they both have the same bitrate and of course they are both lossy. My opinion is they both sound equal and much better than Dulby because the bitrate is more than double.

My question would be, when playing a DTS 24/96 disc on a player that can't handle it and only plays the core, does the bitrate get limited in some way? All my players can handle the DTS 24/96 spec so I can't compare the sound difference.

 

[JediJoker]

"That DTS 2496 is in fact the "ringer". Not bait and switch though, just a really really bad idea! Decoded fully, it's slightly but genuinely transparent lossy 24/96. Decoded on gear that only decodes the "core" dts signal, it's heavily lossy. That it also decodes to 48k (in core lossy mode) in that scenario is the telltale but the lossy is magnitudes worse than merely a sample rate conversion to 48k. "

This is some fantastic writing, that I am not sure I get it at all. The "ringer" flavor is the brand name plain DTS24/96 that is not 24/96kHz like my LPCM ELP or Grateful Dead DVD-As are? And the top of the line DTS "flavor brand" to look for is the "MA" flavor, like chili cheese flavor Fritos, not plain? DTS MA is the deluxe real lossless 24/96kHz I take it?

You seem to be conflating two concepts:

1. File compression: lossy (data discarded to decrease file size) vs. lossless (file size decreased while retaining all data)

2. "High resolution" audio: greater-than-CD (16-bit/44.1kHz) resolution in bit-depth, sample rate, or both

It is possible, as in the case of DTS 96/24, to be both lossy and hi-res. The audio plays back at 24-bit/96kHz, but data has been permanently removed in the encoding process, thereby decreasing the bitrate but not the bit-depth or sample rate.

 

[jimfisheye]

Have you ever encoded your own DTS files from 24/96 PCM sources to both standard DTS and DTS 24.96? You do realize that both will have the same bitrate?

Having done my own DTS encoding in the past, before my equipment did uncompressed PCM, I don't hold DTS 24/96 to any higher standard than standard DTS. Again, they both have the same bitrate and of course they are both lossy. My opinion is they both sound equal and much better than Dulby because the bitrate is more than double.

My question would be, when playing a DTS 24/96 disc on a player that can't handle it and only plays the core, does the bitrate get limited in some way? All my players can handle the DTS 24/96 spec so I can't compare the sound difference.

The test I did with the Tull files is exactly like I said.
You really can't have two lossy files start out the same (entertaining the premise that core dts and fully decoded dts are literally the same data stream - which goes against what the creator of this format tells you and further would mean the decoding software is spoofing for one or the other) and by some fluke, one of these lossy data streams somehow upgrades itself to nearly null with the LPCM file. That's just a ludicrous insane stretch! A null test is a hard reality. Random samples of different sounding audio will not come remotely close to nulling. If you think that's happening... It really isn't and you really have to look for where you made an error. That's why this tool is so useful and such a hard black and white telltale.

If someone has a listening space so dialed in (and speakers to match) that you can clearly hear something like a sample rate conversion (using SOX) between 96k and 48k, or hear an obvious difference between lossy but fully decoded (to the best of what the system was designed to do) dts 2496 and LPCM 24/96... OK. Must suck to only be able to appreciate about 6 albums in the world that make the cut for you mastering/delivery-wise in this age of volume war hash!

All I can tell you is this is my experience. I'm not making anything up and I didn't falsify anything on the testing I did. I don't work for any of these companies. Maybe there are more people out there with 6 figure reference systems in professionally treated rooms with everything professionally calibrated than I realize and I'm in gorilla theater here and easy to please? If so... well, there you go! And lucky me. I get to enjoy dts 2496 fare. While very much noticing mastering flaws that bug me with about 80% of music releases that other people claim to not hear. I don't mean to be flippant about it. I'm trying to be logical and thorough to the best of my ability.

 

[ssully]

'96/24' in DTS 96/24 means that the audio data input into the compression codec in the studio was 96/24 PCM. Of course that doesn't necessarily mean it had any usable content above 22 kHz....or even that it was originally recorded or transferred at 96/24. There's plenty of room for shenanigans on the part of the audio file supplier (the record companies) , as experience with 'HDtracks' has shown.

DTS 96/24 is still lossy for the simple reason that removing >24kHz content is not the *only* thing involved in DTS lossy data compression. So 'restoring' the signal at these (inaudible, btw) frequencies doesn't turn lossy into lossless.

If your AVR or playback chain has a DTS 96/24 decoder, then a DTS 96/24 audio stream fed to it will be decoded as DTS 96/24 (most AVRs IME have some way to indicate this is happening). If yours AVR/chain does not have a DTS 96/24 decoder (in which case it's likely pretty darn old), then it will be decoded as DTS 48/24, (NOT 'downsampled') i.e., without those extended frequencies folded back in...unless your AVR is truly ancient and has no DTS decoder at all.

DTS-HD MA is a true lossless compression algorithm, like FLAC or MLP. It shrinks files but does not lose information.

And as for DTS vs "Dullby" (gah), for the nth time: when the lossy algorithms themselves are different (as AC3 and DTS are) then you can't just go by bitrate specsmanship -- 'this number's bigger therefore better'. AC3 and DTS do things differently. In fact it's quite difficult to compare them fairly. But they both work. As described in one of the best online documents about the case of Dolby vs DTS, well worth reading carefully:

http://www.spannerworks.net/reference/10_1a.asp


Leaving aside the technically difficulties of fair comparison, I have no doubt that proper double blind tests of DTS vs Dolby , DTS vs DTS 96/24 , and DTS or Dolby vs lossless , would make hash of many deeply held beliefs and overwrought claims of audibility, if those who hold them and claim them were to be subjects of such tests. "Nulling" test differences do not correlate linearly to degree of audible difference, when lossy codecs are involved, because the auditory system doesn't work like a simple waveforms comparer, and lossy codecs are modeled on how the auditory system works (psychoacoustics).

 

[watsy_73]

Really glad i got this, great album!
Not just a great bass player but a very competent vocalist to boot.
The mix is very good, a very nice balance to my ears.

 

[haikubass]

I'm enjoying this mix more and more, and the louder is definitely the better, too.
Sorry to say this, but I was completely ignorant of this record, so I'm hearing it as fresh as it gets.

Regarding the debate here, I can imagine that lossless codec would help the punch and overall clarity even more, but actually I have no complaints sonics-wise - except maybe for a bit hairy sounding hi-hat, but that would be a really minor and highly personal quibble.

The best part for me is how well-balanced the whole mix is. And yes, so is the entire album.
If I could, I'd pull out a perfect 8,5. Well, instead of tossing a coin, I'll go with the higher rating, let's say just because I like me some bass.

Fave tracks: Hold Out Your Hand, Lucky Seven.

 

[quicksrt]

The whole lossy VS lossless thing really means nothing to me. Can the human ear even really hear the difference? Probably not. Same with all the bit rate wars like 16bit VS 24bit. It may look better in some stupid spectiral analyis graph but does it really sound better? NOOOOOOOOOO!!!! And it takes up too much storage space.

The one thing I can say is that when comparing the 5.1 DD to the DTS, the DTS always sounds better to me. So that's all I need.

Yes, one can tell a difference if their equipment is up to it and their hearing is all there as well. If one cannot hear true high-resolution's improvement, then that might not be a bad thing the way things are going,

 

[quicksrt]

Is the super deluxe issue of Chris Squire - Fish Out Of Water running low or about to go OOP? I see Amazon has them for around $88 plus shipping and it says a couple left. Ebay prices are $120 to $300, as if the big time flippers have taken what's left and jacked them through the roof.

Should I spring for one while I can at the normal outrageous price on Amazon? Just to get that 5.1 mix?

 

[H.A.L.]

Should I spring for one while I can at the normal outrageous price on Amazon? Just to get that 5.1 mix?

A few cents less than 70 bucks on Discogs:

https://www.discogs.com/Chris-Squire-Fish-Out-Of-Water-Deluxe-Edition/release/11861590

 

[Matt Wilson]

It's fun but as with other Jakko mixes, not as faithful to the source material as Wilson's mixes are. Since I usually give the Yes sets a 9, I'm going to go ahead and give this one an 8. Bit surprising that I'm the only one to do so thus far, makes me think I'm being too harsh or I'm missing something. It's also a tad pricey but that's probably because of the vinyl.