DTS playback with non-DTS hardware

[Mark Gulbrandsen]

For the same reason Dolby is not releasing a multichannel analog input board for CP950 Unfortunately, this is not an economically relevant market segment to justify the investment.

I believe that the only possible funder, at the moment, for this type of preservation work would be FIAF (which should similarly pose the problem for Dolby Digital).

Many but not all of the DTS Patents are at the link below. Also, There were actually DTS processors for Large Format Theaters way before Jurassic Park was a twinkle in Spielberg's eyes. I've had to deal with several of those and that dealing with almost always meant removal and installing a T.C. Generator and DTS 6D in it's place. The Large Format distributers simply sto0pped supplying the disks for the old players. So, in the updated systems, a shaft encoder referenced the time code generator to the projector speed and the Time Code generator referenced the modified player. Note that this is even before the XD-10 appeared.

Patents assigned to DTS (Partial but big list)
https://patents.justia.com/assignee/dts-inc

[Scott Norwood]

Scott, all editing on.film productions was done on 24 fps equipment. Most big editing swstems could do 24, or 30 fps. But it utilized data on.hard drives, not tape.

Interesting. I've read about the processes to match back a work tape to a film negative, but maybe that was intended more for TV shows that wanted to archive on film rather than theatrical features. It does not sound like a fun process.

[Michael Zarits]

Coding a program on a Raspberry or whatever to sync audio files to the DTS timecode and bypass the DTS playback units should be easy to accomplish. As David already mentioned, the DTS tc is not that complicated. Two different block types:
Timestamp block: Carries the reel and block number.
Serial Block: Containes only the serial number and 4 empty bits (8 visual dash lengths) which identify this block as a serial block.
The timecode is 30 blocks per second, 80 perfs ahead of the corresponding picture, starting 4 perfs after the sync pop, the first 58 blocks are muted.

[Gordon McLeod]

There is sitll something missing as there was a timecode change with firmware 1.46 that stopped the use of timecode generators driving a DTS player

[Michael Zarits]

The timecode itself did not really change. Since ~2005 the audio on dts discs is encrypted, meaning that the timecode's serial block had to occur in a specific pattern matching the feature's unique "AUE key". That prevented the playback by just hooking up a tc generator and pirating the soundtrack. Encrypted discs had AUE or APX files instead of AUD files (to be exact: The file's header at position 0x5C is set to 1 if a file is encrypted). For non encrypted discs, the pattern of the serial block does not matter. Btw: The ecryption starts at 2 minutes into a file/reel. The first 2 minutes are always decrypted.

[Marco Giustini]

amazing, thanks Michael for those info! It's so cool to see how that very simple technology worked.
So interesting to know how the "encryption" worked. It was indeed very easy before that to just use the player to play sound even with no TC.

If I understand correctly, with the "encrypted" serial pattern, one could not use a standard TC generator and back in the days I guess it would not have been easy to generate a custom TC with a special pattern on the serial field?

[David Ferguson]

The timecode itself did not really change. Since ~2005 the audio on dts discs is encrypted, meaning that the timecode's serial block had to occur in a specific pattern matching the feature's unique "AUE key". That prevented the playback by just hooking up a tc generator and pirating the soundtrack. Encrypted discs had AUE or APX files instead of AUD files (to be exact: The file's header at position 0x5C is set to 1 if a file is encrypted). For non encrypted discs, the pattern of the serial block does not matter. Btw: The ecryption starts at 2 minutes into a file/reel. The first 2 minutes are always decrypted.

Fascinating Michael, I wasn't aware of how the timecode was different on the later encrypted prints. I'm tempted ton hook my logic analyser up to my DTS reader and see for myself now.

amazing, thanks Michael for those info! It's so cool to see how that very simple technology worked.
So interesting to know how the "encryption" worked. It was indeed very easy before that to just use the player to play sound even with no TC.

If I understand correctly, with the "encrypted" serial pattern, one could not use a standard TC generator and back in the days I guess it would not have been easy to generate a custom TC with a special pattern on the serial field?

Correct, with the DTS timecode generator it will play the first minute or so, and then cut out. This isn't an issue today though, as the freely available software Foobar2000 with apt-x100 plugin will happily convert an encrypted AUE to AUD for you, which doesn't have that limitation.

[Marco Giustini]

ahaha I didn't know the plugin had gone so far! Well I guess nowadays DTS sound files are seen as a massive unsecure asset from the studios. But after all there is no more R&D on DTS players so a better encryption mode is out of the question. The XD10 should be able to deal with something more modern than a scrambled serial number but older players would not as based on 80486 or even 80386sx (basically a 80286 with 386 instructions) intel processors.

Too bad there is no demand for DTS audio on print, I guess that a Raspberry pi would be more than capable of replacing a DTS cinema player

[Mark Gulbrandsen]

This would be very easy to do and could in fact be done for any movie brand new or old without altering the DTS time code on the print at all. All it takes is someone willing to pay to do it and get the players involved to play nice with each other.

That is exactly what was done on the Iwerks and CDC 8/70 Linear Loop, and 15/70 large screen systems. However there existed a first DTS player, yes, the very first model player they made, that I have only seen in the support racks twice now. It was built as more of an industrial rack mount computer, and a separate monitor was hooked to it for doing set up's. Once Jurassic Park was released DTS no longer supported these processors, which also provided audio to High Center. So they had to be replaced with a modified 6D(no more high center) and DTS time code generator. The generator was driven off the shaft encoder in the right eye projector. Can't remember the encoder output frequency, but I'd guess 240 or 360 HZ. Last one I converted was just south of Salt Lake City, and we had a really hard time finding a DTS time code generator. I believe Mike Smith finally dug one up for us.

[Gordon McLeod]

Until the advent of the 1.46 firmware one could install a different eprom in the JSK generator and it would drive a DTS player off a shaft encoder.

[Pete Naples]

The 8/70 iWerks I serviced used a DTS6D SV (Special Venue), that has an extra output IC fitted in (I'm going on long memory here) the audio output board, so it had the high centre output.
That snippet sticks in my head because I had to swap it out once for a standard 6D, which I modified into a 6D SV by swapping that IC from the defective player. It was damned awkward to get out IIRC.

The installation used a shaft encoder and TC generator, but also had a DTS70 head because the SMPTE TC didn't work past 1 hour, it went back and repeated the first 1hr audio. We found that out on a rehearsal of the Polar Express in 8/70. So I broke out the jumper that selected DTS or SMPTE timecode and fitted a toggle switch.
They also had a 8 track DAT player that used 8mm video tapes which ran in sync as backup.

And those awful Peavey processors.

Later we added a Vic 8 for 5/70 and 35mm, CP65 & MPU.

The venue is of course now digital, the audio is handled by a Trinnov Ovation, the first one in the UK, it was a nice easy way to drive bi-amped surrounds, multiple subs and that high centre.

[Michael Zarits]

They also had a 8 track DAT player that used 8mm video tapes which ran in sync as backup.

Tascam DTRS, 8 digital tracks on Hi8 tapes.

[Pete Naples]

Thanks Michael, that's the one. I couldn't remember the name. Ironic as I have one here at home!

[Gordon McLeod]

At cinesphere we often had workprints and one finished one that use the DA-88 Tascam

[Daniel Schulz]

Wow this thread is a blast from the past for me (I was with DTS/Datasat through almost the end, laid off once the deal with Inventure was secure and just prior to ATI acquiring Datasat). Just thinking out loud - rather than try to figure out how to sync an Atmos track to a DTS timecode track, a studio could pre-render the Atmos soundtrack to some fixed-but-high channel count (sort of like IMAX does with their 12 channel system). The XD10 could play 8 channels, seems like with today's computing horsepower you could easily do 12 or 16 channels.