From an audio standpoint, Jurassic Park is my favourite movie. So I went searching for as much information as I could about how this audio mix has been treated over the years. A lot of what I found was conflicting, unsubstantiated, or just plain incorrect. So I got together the audio tracks from every release that was available to me and set about comparing waveforms. This is what I found for the benefit of anyone interested.


Jurassic Park 1993 35mm 1080p Cinema DTS v2.0 - DTS-HD MA 5.1



Let's start with the original Cinema DTS track from the 35mm v2.0 release. You would think, having been sourced from the original DTS CD-ROMs, that this would be the final word in theatrical accuracy. But you’d be wrong.

First of all the surround channels are too loud. Due to the way cinema sound systems are calibrated compared to home systems, any theatrical DTS track intended to be played at home theatre levels needs to have its surround channels lowered by -3dB. It should be noted that the standalone DTSHD release of the Cinema DTS track already has the surrounds attenuated but (the one on MySpleen at least) has a completely empty LFE channel.

The second problem is that the LFE channel is not loud enough. According to schorman who compared the output of various DTS decoder plugins with actual DTS hardware, movies pre-1999 need the surround channels upped by +3dB.

So by applying these two changes we should get a Cinema DTS track that we can assume is accurate to theatrical levels. I'll call this Theatrical DTS (Corrected) and this will be the mix that all others are compared to.


Theatrical DTS (Corrected)



There is one possible problem with this mix in that there are more sounds present in the LFE channel than in any of the home video mixes. This could be because the theatrical DTS simply was this way or it could be due to incorrect decoding of the matrixed LFE channel from the surrounds. In his comparison of DTS decoders schorman noted that different software plugins (even different versions of the same plugin) produced different results in separating the surround and LFE channels. So without knowing exactly what software and method was used to decode this track we can't know for sure that the resulting channels contain the correct information. My personal view is that, since DTS themselves handled the decoding for the home video mixes, then those are likely correct and something is slightly amiss with this one.


1997 DTS LaserDisc – DTS 5.1 1411kbps



The LD DTS is the most divisive home video mix. Some rank it amongst the greatest LD mixes ever produced whereas others call it overcooked with unattenuated surrounds and crazy LFE. Compared to Theatrical DTS (Corrected) the surround channels indeed have not had the required -3dB reduction applied. It’s also true that the LFE channel is much louder resulting in an overpowering mix. In all other respects, this track is practically the same as the theatrical DTS.


2000 DTS DVD (1st release) – DTS 5.1 768kbps



The first DTS DVD release is widely considered to be a defective disc with claims that it had a severely neutered LFE channel or no LFE channel at all. But this is untrue. The LFE channel is there and actually has a slight boost compared to Theatrical DTS (Corrected). The surrounds are also correctly attenuated. The big problem is that the centre channel is too loud to the point of clipping. Perhaps this was an attempt at a near-field mix. Overall though I would say this track is far from defective and is actually a lot closer to the theatrical mix than the AC-3 DVD to which it is frequently compared.


2000 DVD – AC-3 5.1 448kbps



This one was a surprise particularly given that, whenever I’ve seen it compared to the first DTS DVD, this is invariably held as the superior track. In fact, this is a total remix bearing very little resemblance to the theatrical mix. Maybe the reason people thought this sounded better is that many of the LFE effects are way too loud. Also, the left and right channels are mixed a lot lower relative to the others. Comparing this to the first DTS DVD is like comparing apples and oranges which makes it surprising that a second DTS disc was ever issued.


2000 DTS DVD (2nd release) – DTS 5.1 768kbps



The second DTS DVD’s mix is essentially the exact same mix as the LaserDisc but with surrounds correctly attenuated this time. I guess the studio figured that since everybody enjoyed the overcooked LFE of the LaserDisc they would just throw that onto the disc, fix the surrounds and call it a day.


Chinese Region 6 DTS DVD – DTS 5.1 768kbps



The Chinese R6 DVD has a unique DTS mix compared to all other home releases and it’s extremely close to Theatrical DTS (Corrected) with all the relative channel levels looking right. The LFE channel has not been boosted and the surrounds are correctly attenuated. In fact I would say that this appears to be a genuine effort to release the original unaltered audio mix.

It doesn’t have the additional LFE information that the Cinema DTS track has but whether that makes it more or less theatrically accurate is a matter for debate.

The one and only problem with this mix is that it’s from a PAL release and hence needs correcting to play at the correct speed and pitch. Also it's also wrongly reported on the net as being full bitrate DTS but it’s 768kbps just like the other DVDs.


2011 Blu-ray – DTS-HD 7.1



The 7.1 track on the BD release is another complete remix with some added and altered sound effects to boot. It is certainly not, as is reported elsewhere, the theatrical mix with the surrounds split. I almost wasn’t going to include it in this comparison since if the goal is theatrical accuracy then it’s of no use whatsoever.


2013 3D Blu-ray – DTS-HR 7.1



Finally we have the 7.1 track on the 3D BD release. This based on the BD 7.1 with the same altered effects but is perhaps of some historical interest since it did play in cinemas making it a theatrical mix and it was also overseen by original sound designer Gary Rydstrom. In practice it’s just a slightly more refined version of the previous 7.1 mix.


Missing Mixes
Not included in this comparison are the following tracks simply because I don’t have them:

LD AC-3
R2 Superbit DVD

If anybody has these tracks and would care to share then I’ll gladly add them here.


Conclusion (TL;DR)

There is not yet any release of Jurassic Park that has theatrically accurate audio out-of-the-box. The 35mm v2.0 DTS needs the surround and LFE levels adjusted and the LFE content is arguable. The LD DTS needs surround and LFE reductions. The R6 DVD has all the correct channel levels but must be PAL corrected. Any of these adjusted tracks should get you very close to the original 1993 DTS mix.

My thanks to borisanddoris and NeonBible for their invaluable help.

Very nice rundown

Interesting writeup!

The LD DTS heading incorrectly states the bitrate as 768Kbps, the LD flavour of DTS is 44.1kHz 1411Kbps. Also the initial pressing of the DTS DVDs did in fact have an incredibly weak LFE channel (I know as I bought it back in the day and had to return it). The disc was re-authored with a corrected LFE channel.

With regards to the Cinema DTS track, were you able to work with the original files ie the .aud files from the CD-ROMS? It's worth noting that the subwoofer channel of Cinema DTS (APT-X100 codec) is extracted from the surround channels so the LFE channel will also include low frequencies from the surrounds, making it look 'busier' than dedicated LFE channels from discrete 5.1 mixes. Also the LFE tops out at around 90Hz on cinema DTS as opposed to 120Hz on discrete theatrical formats like AC-3 and SDDS.

great write up. thank you

am i right to say that dts cinema mixes were originally 5.0 anyway? I recall seeing a comment in a thread about Mission Impossible Ghost Protocol somewhere and it was mentioned that the LFE was mixed into the speakers instead of discrete

(2021-09-08, 02:08 PM)Bigrob Wrote: great write up. thank you

am i right to say that dts cinema mixes were originally 5.0 anyway? I recall seeing a comment in a thread about Mission Impossible Ghost Protocol somewhere and it was mentioned that the LFE was mixed into the speakers instead of discrete

In the Cinema DTS format, that's the case. It doesn't allow for a discrete LFE afaik. Hence the point about correctly unmixing the LFE from the speakers like a cinema processor does. I don't know if the mixing process was as such also done in 5.0 or in 5.1 and the LFE mixed into the speakers for the DTS CDs.

(2021-09-08, 01:37 PM)zoidberg Wrote: Interesting writeup!

The LD DTS heading incorrectly states the bitrate as 768Kbps, the LD flavour of DTS is 44.1kHz 1411Kbps. Also the initial pressing of the DTS DVDs did in fact have an incredibly weak LFE channel (I know as I bought it back in the day and had to return it). The disc was re-authored with a corrected LFE channel.

With regards to the Cinema DTS track, were you able to work with the original files ie the .aud files from the CD-ROMS? It's worth noting that the subwoofer channel of Cinema DTS (APT-X100 codec) is extracted from the surround channels so the LFE channel will also include low frequencies from the surrounds, making it look 'busier' than dedicated LFE channels from discrete 5.1 mixes. Also the LFE tops out at around 90Hz on cinema DTS as opposed to 120Hz on discrete theatrical formats like AC-3 and SDDS.

Thanks for the correction on the LD bitrate. I've updated the post accordingly.

The initial DTS DVD had the LFE level set more or less correctly which is actually a lot lower than the overcooked LFE of the LD and AC-3 DVD. It also didn't help that the centre channel was set too high so any A/B listening tests done with dialogue level as a basis would make LFE sound almost non-existent.

I didn't use the original DTS CD-ROMs as this was intended to be a comparison of tracks taken from releases that are generally available to the community. Hence I used the 35mm scan v2.0 release.

Jurassic Park is one of my favorite films, fantastic write-up.
My go-to track has lately been the 3D 7.1, as it sounds great and syncs to the 4k easily. Though I'd be interested in "correcting" my version of the cinema DTS and comparing it. A few things I would add:

• There's minor dialogue distortions in all mixes up until the BD release. It's not terrible or off-putting, but present and worth noting.
  
• The JP franchise has always been a shining example of the limits of lossy compression, mostly due to the excellent sound design and emphasis on the Dino roars. Listening in a high end setup and a very loud volume, all lossy mixes (even full bitrate DTS) are brought to their knees compared to the lossless sound on the BD.
  
• The 4k UHD has a DTS:X remix. It's based on the 7.1 BD but heavily filtered, has neutered bass, and little bit of added foley. I'd take the lossy DVD mix over it.
  

How do I know which DTS DVD release I've?

(2021-09-08, 04:07 PM)bendermac Wrote: How do I know which DTS DVD release I've?

From thedigitalbits.com:

Quote:Unfortunately, there is no sure-fire way to identify the new disc from its packaging alone. The original release of the disc had two stickers on the front, one which promoted the Jurassic Park III weblinks the disc provided, and one which promoted the discs PlayStation 2 compatibility. Some people are reporting that the new discs don't have the PS2 sticker. Mine didn't have either. One thing you should look for is the spine label at the top of the keepcase - on the new discs, it should read: Jurassic Park (WS) (DTS). That said, THE ONLY SURE WAY TO IDENTIFY THE FIXED VERSION IS BY THE NUMBERS ON THE INSIDE RING OF THE ACTUAL DISC. The FIXED disc will read: PEMC-D2R2. The earlier pressings - the defective ones - read either PEMC-D2R0 or PEMC-D2R1.

Bigrob,

Cinema DTS is a 6-track system with 5 front channels and a subwoofer channel (with 10db in-band gain). The subwoofer channel is stored in the surround channels and recovered during playback. They did it that way as it saved the need to encode an additional channel which increased the amount of audio each CD-ROM could hold.

With regards to the DTS DVD there is probably an edge code you can check to see which version you have (edit: Turisu made mention of this in previous post). To my ears the LFE was seriously weak on the initial release and much better on the re-issue (which was a stealth re-issue, I don't think universal ever really acknowledged it). The LFE is intended to be insane