DTS 96/24 Uncovered

[Yayita]

I am creating a DTS CD from a DTS 96/24 DVD source.

I was succesful, however, I am puzzled with the sampling rate of DTS 96/24. It is supposed to be 96 khz, so claims dts.

When I convert the original dts file (decrypted with dvd decrypter) to wav using the method:

azidts -core( -input "<source>.dts" -output "<destination>-" -6chfloat -logfilea "<logfile>.log" )

using besweet.dll 1.5.23.0 and azidts 1 and IVIAudio 6.06.42

I get 6 wav files (I know channel mapping is wrong), but they are 48khz / 32 bit wav's, not 96khz / 24 bit.

Q1: My IVIAudio window just passes through the screen for an instant, is there an option to OK this window before decoding continues.


I am puzzled by this 48 /96 khz question. The ifo for the dvd audio says that the DTS file sample rate is 48 khz and world length (Dynamic Range Control instead of 24) . My receiver with DTS 96/24 says the audio has the DTS 96/24 flag, and treats it as such, with cool dts 96/24 light and all, but it says that the sample rate is 48khz also. In comparison the IFO for the LPCM 96/24 track in the same DVD says unrecognized sampling rate (i.e. 96khz) , and 24 bit for quantization (i.e. word length). Just for comparison my receiver also recognizes as PCM 96 khz.

Q2: So is DTS 96/24 really 96khz or not?

Q3: If it is really 96 khz, is there a way to decode to 96/24 wavs.


Thanks in advance to the board for the input.

Y

[Sycho]

dts 96/24 is so stupid, I can't understand why dts even marketed it. It's a regular dts track @ about 1.1mbps with a extension that only compress the sound from 24-48kHz, with bad results. The point of 96kHz sampling rate is to get the digital sound closer to the analog master, dts does the complete opposite here.

[specise_8472]

Originally posted by Sycho
dts 96/24 is so stupid, I can't understand why dts even marketed it. It's a regular dts track @ about 1.1mbps with a extension that only compress the sound from 24-48kHz, with bad results. The point of 96kHz sampling rate is to get the digital sound closer to the analog master, dts does the complete opposite here.

Get your facts right.
The Extension you talk about are the frequency components beyond 48khz.
If your player can handle 96 then this extension stream is unpacked and added to the 48 stream to bring it up to 96.
If not then you have a usual 48khz stream playing.
As for bad results, try the Queen 96/24 - the results are very good.

THIS FROM DTS
A compatible bit stream is made up of ‘core’ data, representing 48 kHz encoded audio signals,
and ‘extension’ data, representing differential audio data and/or higher frequency components. Existing core-only
decoders make use of the core data to reproduce 48kHz-sampled audio. New 96kHz decoders make use of both
core and extension data to reproduce 96kHz audio.

[Sycho]

Originally posted by specise_8472
Get your facts right.
The Extension you talk about are the frequency components beyond 48khz.
If your player can handle 96 then this extension stream is unpacked and added to the 48 stream to bring it up to 96.
If not then you have a usual 48khz stream playing.
As for bad results, try the Queen 96/24 - the results are very good.

THIS FROM DTS
A compatible bit stream is made up of ‘core’ data, representing 48 kHz encoded audio signals,
and ‘extension’ data, representing differential audio data and/or higher frequency components. Existing core-only
decoders make use of the core data to reproduce 48kHz-sampled audio. New 96kHz decoders make use of both
core and extension data to reproduce 96kHz audio. Only the frquency above 24kHz are sampled at 96kHz the rest is a lower resoultion dts stream ( between full and half rate). dts 96/24 is useless

[Yayita]

Okay here is what DTS says, Specise already quoted part of this:

The DTS coding system has a “core + extension” structure. The “core” represents the DTS data as has been known since the first home decoders. The “extension” can carry data for future applications or enhancements of any sort. All DTS decoders recognize and use the core data. Basic decoders ignore the extension data, while advanced decoders can make use of it. This allows for full backward compatibility for any scheme using the extension. DTS has recently used the extension field for two purposes. In the first case, it has been used to carry an additional channel for 6.1 dis-crete. In the second case, the extension field carries the additional spectral data added by 96-kHz sampling. For a program in DTS 96/24, existing decoders read the core at 48-kHz and reproduce the standard spectrum. DTS 96/24 decoders read both core and extension and reproduce the extended spectrum. The data rate for 96/24 is 1.536Mbit/s, the higher of the two DTS rates presently used. While numerically this might suggest twice as much compression, there is in fact negligible additional compres-sion on the core data. This is because there is relatively little information in the range 24-48kHz, so it can be coded very compactly. The 96/24 stream passes through the S/PDIF just as standard DTS does.


So DTS 96/24 is really encoded at 48 khz, not 96 khz.

[In the case of DTS 96/24]... the extension field carries the additional spectral data added by 96-kHz sampling.

I am not sure then that you can claim that your stream is 96 khz in a scientifically sound way.

I have no information if sycho's claim

Only the frquency above 24kHz are sampled at 96kHz the rest is a lower resoultion dts stream ( between full and half rate). dts 96/24 is useless

is true. However, my limited understanding of digital audio tells me that higher sampling rates allow the reproduction of higher frequency sounds. So it is plausible, but would be really stupid to use data rate to encode supersonic inaudible parts.

I know you audio engineers hang out in this board, illuminate us.

Y

[pieroxy]

AFAIK, dts 96/24 is fully compatible with dts. This means all current dts programs (that don't support 96/24) will see a "regular" dts stream (48KHz). All freq up to 24KHz are present in such a packet. This is probably why the original poster found 48KHz wav files after decoding with a decoder that is not 96/24 aware.

Now (and pardon my poor terminology) there are the "core packets" which all dts capable machine* should be able to read, and the "extended packets", which are designed to be understood only by dts 96/24 capable machines.

Such a machine would read the core packets and append to that the extended packets, containing frequencies between 24KHz and 48KHz. To render these frequencies (0-48KHz), the output would be a 96KHz stream.

So in effect, dts codes 0-24KHz and dts 96/24 stores 0-48KHz. I don't see what is stupid about that. I actually find this pretty smart as it allows anyone with a dts decoder to read a dts 96/24 stream, albeit not at full quality.



*: By "Machine" I mean a piece of software or hardware.

[Sycho]

Originally posted by pieroxy
So in effect, dts codes 0-24KHz and dts 96/24 stores 0-48KHz. I don't see what is stupid about that. I actually find this pretty smart as it allows anyone with a dts decoder to read a dts 96/24 stream, albeit not at full quality. It's stupid because the already bitrate starved dts stream is even more limited, and the 24kHz - 48kHz spectrum is extremly compressed, it is not worth it since it is nothing like the analog wavform.

Another reason is that 96kHz sampling is for better representation of analog waveforms throught out the audilbe spectrume, dts 96/24 does not offer better resoultion in those frequencies.

[pieroxy]

Your argument is that since you compress the strem it is not worth it? I don't get it, that's the entire point of dts and ac3...

Second, sampling at 96KHz is not only for better representation of the "audible spectrum" as you put it. It is also meant to reproduct frequencies up to 48KHz which are audible (or "perceptible') on high end audio equipment. Or so it is said.

As for a better representation of the freqs below 24KHz, what makes you think dts 96/24 doesn't benefit from that? Of course, a mere dts player would not benefit from it as it would convert the dts signal to a 48KHz waveform before sending it to the amp, but a dts 96/24 can.

[Sycho]

Second, sampling at 96KHz is not only for better representation of the "audible spectrum" as you put it. It is also meant to reproduct frequencies up to 48KHz which are audible (or "perceptible') on high end audio equipment. Or so it is said. There are no real results that I know of on perceptible audio frequencies, since most humans can only hear up to around 15kHz, what is the point of going higher then 24kHz, and how would that "area" be monitored, since it can't be heard, how would we know if something there by listening,

As for a better representation of the freqs below 24KHz, what makes you think dts 96/24 doesn't benefit from that? Of course, a mere dts player would not benefit from it as it would convert the dts signal to a 48KHz waveform before sending it to the amp, but a dts 96/24 can. dts 96/24 does not benefit from better reproduction of 0-24kHz range, since it is a lower bitrate standerd stream muxed with a extremly compressed 24khz-48khz stream. Playback in the audible portion of the spectrum would be the same on legacy equipment and on dts 96/24 equpiment. 96/24 will only add on the higher non aubile end.

[scharfis_brain]

ever heard of nyqist?

if one wants to transmit a 24 khz tone through an 48 khz sampled transmission channel, the chances are good to loose the signal!

if you sample (by accident) the maxima of your 24 khz-sine tone you'll be happy: the tranmission was successful!

BUT if the sine gets shifed by 90 degrees all you do sample are zero-points, meaning you are tranmitting NOTHING.

or imagine a signal near to 24 khz. let's say 23.995 khz:
all you get after tranmission is a 5 Hz amplitude modulated 23.995 khz signal.
(alternating between silence and maximum ampliture 5x in one second)

Those are the reasons, why one does the 'oversampling' on transmitter side.

It is not the higher achievable frequency spectrum.
It is the higher phase-quality!