I wouldn't give up on the US laserdisc based on those samples. It was more a bad capture -- with horrendous R/G/B blowout & crush (which I've isolated from the pictures here) in the worse capture, but still there in the best. Even the letterbox was affected . .





These indicate the obvious: a too-high contrast setting pushed the picture out of bounds in both directions, which induced color-drain and loss-of-detail in those upper/lower areas.

i am not quite sure where you want to got with that? surely the capture hardware and setup could have been bad, though out of experience LDs aren't really the best source. furthermore, japanese releases are mostly better than the respective US release. so i would rather look at the jp LD if possible.

anyway i updated the earlier comparisons with the R2J DVD (first DVD release from 2004(?)), 4k | 1080p. disregarding the difference in technology (higher res, better codec, etc), the DVD transfer is better than the BD one as a refference, not as much black crush. the colours are less teal tinted than the BD and less redish than the HDTV source. i would say the DVD is probably one of the most truthful to the source material representation of GitS.

Not actually taking the laserdisc capture observation anywhere -- just digging for verifiable pieces of the puzzle. Generally, the US LDs have their problems (I don't actually know about the GITS LD as I don't have one) but JP LDs have their fair share of sloppy work, too.

Case-in-point is Slipstream (1989). All US releases (LD & DVD) are cropped fullscreen. Only the JP LD is letterbox widescreen. The problem is ... the picture is crushed and blown right off the disc . .


(The "quick fix" is to demonstrate what it might look like. Of course, a real fix would include repairing the flattened high and low ends with detail from non-flattened RED or GREEN or BLUE  color layers.)

Interesting to note the orange tinge of the "damage" LD looks allot like ... Ghost In The Shell 2.0's new sections! Hmmm ... could this be a JP-thing?  

sry, i didn't make myself completely clear with my comment about JP releases being better than US releases, since i am solely dealing with anime (or related) sources. i specifically meant original JP releases.

i believe the orange tint (or red/pink tint) isn't uncommon to japanese releases. even original japanese releases were affected by it. i read about the technical reasons for this but can't find it anymore. i believe, but don't quote me on that, it had something to do how TVs in japan were 'calibrated' back then and to compensate for that or something.

The "legal range" for US  is 16-235 while JP  is 0-235 (out of the full range of 0-255). Could that be the crucial factor for doing the math without compensation ("US"? what's that?) ... it comes out orange?

So, regarding Slipstream (forgive the digression), if I played this on a JP system, it would be normal? Or would it be permanently damaged but still looking as "normal" as my "quick fix"? The answer would be interesting for a general, best-fix approach to correction.

(2018-07-25, 03:29 PM)der richter Wrote: anyway i updated the earlier comparisons with the R2J DVD (first DVD release from 2004(?)), 4k | 1080p. ... the DVD is probably one of the most truthful to the source material representation of GitS.

Thanks for the new additions! Can you fully describe to what each category-abbreviation refers? Some I get, but others I'm not sure.

Without the proper equipment (and I am without proper 4K equipment), I'm still interested in it's features (resolution, mostly). Would a down-rez'd 4K produce a better HD than their plain, old HD release? I understand that there really isn't any detail up to HD standard to begin with.

Forget about ever having Ghost In The Shell restored by it's re-filming. Another eBay auction with another cel ... and another one bites the dust . .

(2018-07-30, 09:06 PM)BronzeTitan Wrote: Forget about ever having Ghost In The Shell restored by it's re-filming. Another eBay auction with another cel ... and another one bites the dust .

Sad days indeed!
It gives me a weird feeling inside when I see how badly treated original elements of pieces of work like this are treated. Once a film is released and has made its initial run it really is discarded!
Finding Bladerunner all those years back in a warehouse with it's destroy paperwork says it all.

The negative is (was) the thing with film production. Everything involved in obtaining the final negative is a means to an end. It's the same with live action, extremely expensive props/animatronics become 'junk' (as far as the production is concerned) once the photography is complete and the picture is locked.

The real irony is that the original elements were rescanned and recomposited... for 2.0 

The thing to keep in mind about animation cels is that the colours are chosen based on how they will look when projected, due to the inherent nature of positive print film (how it resolves highlights/shadow detail and the way colour shifts). Most productions will meticulously screen test the colours and then ensure enough paint is mixed for each chosen colour to last the entire production.

(2018-07-28, 01:21 AM)BronzeTitan Wrote: The "legal range" for US  is 16-235 while JP  is 0-235 (out of the full range of 0-255). Could that be the crucial factor for doing the math without compensation ("US"? what's that?) ... it comes out orange?

So, regarding Slipstream (forgive the digression), if I played this on a JP system, it would be normal? Or would it be permanently damaged but still looking as "normal" as my "quick fix"? The answer would be interesting for a general, best-fix approach to correction.

i am under the impression that the limited range for JP release is the same as everywhere else, eg 16-235 on luma and 16-240 on chroma planes. or did mean this is something specific to laserdiscs?

from what i remember it's just the latter. permanently damaged but looks 'normal'. basically, it introduced a colour shift to counter a general colour shift on TV equipment. one of those dumb legacy stuff things that was kept sometimes in newer processing for the heck of it. it wasn't even done consistently everywhere.

(2018-07-28, 02:04 AM)BronzeTitan Wrote: Thanks for the new additions! Can you fully describe to what each category-abbreviation refers? Some I get, but others I'm not sure.

Without the proper equipment (and I am without proper 4K equipment), I'm still interested in it's features (resolution, mostly). Would a down-rez'd 4K produce a better HD than their plain, old HD release? I understand that there really isn't any detail up to HD standard to begin with.

for the 4k comparison:
4K UHD BD - the screens i found and weren't done by me from the new 4K UHD BD
BDMV - screens by me straight from the first GitS japanese BD (a bit less bitrate starved than the 25th anniversary BD)
DVD R2J - screens by me from the first japanese region 2 DVD
Judgment new - screens by me from my second filtered release (reencode of the WMV3 raw)
Judgment old - screens by me from my first release (the unfiltered WMV3 one)

for the 1080p comparison:
BD - see BDMV above
DVD R2J - see DVD R2J above
Judgment - see Judgment old above
Judgment filter - see Judgment new above, just a lossless encode
Judgment v2 - see Judgment new above

from what i have seen so far, yeah the new 4k BD downsized to 1080p will produce a better result than the old normal BD, if properly processed. though just on a side note, there is also a new normal BD from the new transfer that might look a bit better depending on how bad the HDR conversation of the 4k BD was.

i feel the 4k BD was tempered a bit with too much to make it look "HDR". like the colours are bit too vibrant especially the reds.