Sign In

Post #616898

Author
spoRv
Parent topic
The 1997 Star Wars Special Edition Trilogy Preservation Standards Thread [Work In Progress]
Link to post in topic
https://originaltrilogy.com/post/id/616898/action/topic#616898
Date created
27-Dec-2012, 6:50 PM

CapableMetal said:

I have captured the PAL laserdiscs for ANH and they have 1 frame missing from the 'standardised' figures on here, at the end in the victory celebration, I think. Inserting a null frame would solve it, although its so near the end of the film a 1 frame discrepancy should be unnoticeable.

Just for curiosity, stating that "Star Wars" begins at frame 711, what is the number of the missing frame?

The DVB broadcasts are certainly more detailed than any laserdisc releases, but that means that the DVNR smearing is more apparent. Of course, it can vary from version to version, but only with experimentation can we find a nice compromise ;)

Please follow my thoughts:

all of those DVB were broadcasted years ago, almost surely before 2004/5, because George "changed his mind again" and then after 2004/5 only the third version was available (also for broadcasting use). The compression techniques were not refined as today; also, the experience was shorter.

And, who can be sure that these DVB programs were recorded directly in digital domain? As satellite internal tuners were not so easy to find, probably who recorded them used a digital satellite receiver, and used a PC or a DVD recorder to capture the movies; in both cases, the original signal was digital, then converted in analog and then reconverted again from analog to digital by the PC or DVD recorder... I say so, because I did quite the same; I recorded an aerial DVB broadcast with my DVD recorder in 2005, which was connected by a SCART cable to the DVB receiver!

...add to this also signal interference (in the broadcast, antennas, cables etc.), original compression artifacts, wrong settings... and MPEG2 compression - all of those versions were released on DVD - with all its inherent problems.

So here you are the possible scenarios:

  • the capture was made using an internal PC satellite tuner (or an HD connected to a digital satellite receiver, and then downloaded to PC); the DVD was made without recompression. Result: the quality is the same of the DVB broadcast; very good. Probability: very low.
  • the capture was made using a DVD recorder (or a PC analog capture card) connected to a digital external receiver; the DVD was made without recompression. Result: the quality depends by analog DACs and ADCs, cables, DVD or PC capture settings; from good to poor. Probability: very high
  • the capture was made using a DVD recorder (or a PC analog capture card) connected to a digital external receiver; the DVD was made WITH recompression. Result: the quality depends by analog DACs and ADCs, cables, DVD or PC capture AND compression settings; from medium to poor. Probability: medium

 

In the best case scenario, you have a perfect digital copy of the DVB broadcasting on DVD... still MPEG2 compressed!

 

Let's see together some interesting facts, according to myspleen torrent:

the ANH DVD, from GKar set, which is supposedly the best of the DVB broadcasting, has three soundtracks, one German DD 5.1, one English DD 5.1, and one English 2.0; as I have not the DVD myself, I have to guess how much space those soundtracks take.

ANH PAL GOUT is 116 minutes, 1997 SE should be 120 minutes; so this DVD AC3 soundtracks could be:

120 minutes x 60 seconds = 7200 seconds

  • 448kbit/s soundtrack: 7200s x 448 kbps = 393MB
  • 384kbit/s soundtrack: 7200s x 384 kbps = 337MB
  • 256kbit/s soundtrack: 7200s x 256 kbps = 223MB
  • 224kbit/s soundtrack: 7200s x 224 kbps = 197MB
  • 192kbit/s soundtrack: 7200s x 192 kbps = 169MB

 

  • The max space used by the soundtracks could be 3 x 393MB = 1179MB
  • The space I think it used is 337MB + 337MB + 169MB = 843MB
  • The min space used by the soundtracks could be 3 x 169MB = 507MB

 

A blank DVD+R/-R is 4700MB (I don't count space for menus) at the end, the space left for video, and its bitrate would be:

  • 4700 - 1179 = 3521MB / 7200s = 4006kbps
  • 4700 - 843 = 3857MB / 7200s = 4388kbps
  • 4700 - 507 = 4193MB / 7200s = 4770kbps

 

that is good for a commercial DVD, but, in reality, the GKar ANH DVD iso is only 3800MB, so probably its video bitrate is

3800 - 843 = 2957MB / 7200s = 3366kbps (a bit more, as iso is slightly compressed)

and  think that GKar was recorded in 2001 or 2002, then there is the highest probability that is was NOT a direct digital recording... draw your own conclusion.

 

I don't say the laserdisc, as an analog medium, is better than digital; I state that, following my thought exposed before, and my brief experience with the OUT ruLes project, it *may* be better to start with uncompressed material taken directly from laserdisc captures, using good hardwares and sources, medianed and averaged, than using compressed, low bitrate material recorded who knows how. This is only my personal opinion, that's it.

I must add, just for the sake of completion, that I could use my Pioneer HLD-X9 to capture not only from one NTSC US boxset, but from two different copies of it! And I could use also another MUSE player! ...If only I have learned to do proper IVTC... (xx_)

I'd be interested in seeing some test captures of any SE sets you have, particularly in high-motion scenes so compare to what I have already. Any SE source, of course, needs some colour correction due to that horrendous pink tint on certain shots, particularly the Mos Eisley effects shots.

I have the UK PAL and US NTSC sets.

I have your same UK PAL and US NTSC sets (actually two of the last), plus the german and french PAL boxsets (the last one in english).

I'm quite sure I could squeeze more details from these PAL laserdisc than the ones I used for OUT ruLes project, as they are newer and remastered. For the color correction, I'll try to do my best, but I'm sure I'll be here to ask for help once again!

New year, new captures! Just wait some days.