- Post
- #689900
- Topic
- Movies with wrong color grading *** UPDATED ***
- Link
- https://originaltrilogy.com/post/id/689900/action/topic#689900
- Time
Demolition Man
BD
MIX (50% BD, 50% BD with DVD colors)
BD with DVD colors
Demolition Man
BD
MIX (50% BD, 50% BD with DVD colors)
BD with DVD colors
Opinions on "Demolition Man"? I think the best color grading should lie in the middle, towards the BD...
You are right, the MPEG2 is not that bad; and yes, the DNR has scrubbed a lot of details in the UHE; but I wanted to make these tests, I've done them, and here I'd like to hear what you think about them... so I will make good use of your tips, as usual, you know it!
But maybe I derailed the discussion too much, talking about this grain plate technique... Should I start a new thread?
Even if I love MPEG2 - and I'm sure you know it - AFAIK this one is ridden by many compression artifacts, while the UHE is DNR'ed a lot... neither is perfect, and probably I'll end to choose the MPEG2 even if has those artifacts; but, as the UHE has more details and no compression problems, and *IF* a grain plate overlay could make it more watchable, ***MAYBE*** the final result could be preferable by some... that's it!
I don't plan to do a restoration now, but maybe some users who watched the test clips could think if it will be an interesting thing to do, or simply a waste of time...
At the end, is just an "excercise of style" to understand if the grain plate technique could improve a bit a DNR'ed video... I thought also to test it with the MPEG2 version, to see if it could mask a bit the compression artifacts... what do you think?
Actually, An American Werewolf in London is AAWIL... (^^,)
Sure, we all know that it will be impossible to restore the UHE, but at least with some grain slapped onto it, it's a bit more bearable... or not?!?
New grain test: http://www.sendspace.com/file/9zikzt
(between 35mm and "16mm" grain plates)
Sorry, I mistyped "downscaled"... indeed, it was UPscaled, and less than double size... so, now I'm testing a new grain that is between the 35mm and the "16mm" recreated grain; I'll post results ASAP.
CSchmidlapp said:
Out of interest is there any way you can copy detail from on source (MPEG 2 Blu ray) and overlay it on another (UHE Blu) like you colour match scripts?
AFAIK, is not possible... but I will be more than happy to be wrong about it!!!
@dvdmike: what is AAWWIL?
The coarse grain (that I called 16mm just to remember...) is actually the 35mm downscaled...
I'm aware that the "16mm" grain covers better the DNR, but it seems to me a little bit too coarse... maybe I should try something in the middle.
Here you are!
Fine grain test: http://www.sendspace.com/file/d6zq1h
Coarse grain test: http://www.sendspace.com/file/nqhrce
Opinions from anybody are welcome!
In few days I'll re-encode and re-author it, and I hope to test in several standalone players - that will be not my PS3!
I'll post results here.
Indeed, bit-perfect doesn't apply to analog capture; as you sent me some laserdisc capture for my projects, and they were amazing, I'm sure also this time the result will be exceptional! GOOD JOB!
I have not the two versions - only the old DVD and laserdisc - but many wrote that the UHE is not *extremely* bad - not as Patton, for example - so I'd like to make a test, because I'm too curious to compare the results... (^^,)
Obviously that, even if a grain plate will make slightly improvements over the overscrubbed UHE, a new 4K scan will blow it out of the water...
Agree 100%, and again, I LOVE that "...a tragic crime against humanity"... could I steal this?
Attention: a grain plate is NOT equal to fake grain... it is REAL grain from real film!
Said so, of course using a grain plate DO NOT add details; but it's just a trick for our brain to let him *think* it... I must admit that, even if I know it's only a trick, and as I did the tests myself, I didn't think my brain actually could be tricked as well as other's, but it was... (^^,)
Probably the MPEG2 is better than a DNR'ed+grain plate'ed UHE version; still, I'd like to test it anyway.
And yes, the only way to have a perfect version is a new rescan, but will it be done?!?
deho, I don't know if any human being will be able to discern if a soundtrack is 1ms out of sync... maybe one of my cats, but they don't understand english, so... (^^,)
Personally, sometimes I could notice out of sync at about 20ms, often 25ms, quite every time at 30ms, but it depends from many factors, and it's different for every one of us.
So I'm pretty sure your soundtrack will be in sync; but, as actually I have not the DCP (yet), I could not test it; hope someone with the DCP will download your track, and test it ASAP!
Meanwhile, I'd like to know if our french friend here has captured the french laserdisc soundtrack - honestly, I can't remember if he has done it or not... I could capture the italian laserdisc soundtrack for BTTF1 (the only laserdisc of the trilogy available in this language) but I will be really happy if someone else could do the same for the german and spanish soundtracks!
Yes, as you can see UHE has more details, but also a lot of DNR... obviously, it will be impossible to recover the lost grain, but adding a grain plate taken from actual film improves the perceived details and hyde some (but not all) problems caused by DNR... but, as this was the worst offender frame, I think the rest of the movie should be improved a little more than this, and, even it will be not perfect, it *could* be a little bit better than the old MPEG2...
This is the reason why I'd like to receive two clips from both versions; to test grain plate on UHE, and to see if final result could be better than the MPEG2.
Probably is due to the fact that laserdisc has some missing frames; is not that easy to put in sync a laserdisc soundtrack perfectly...
Nope... but I think it will have a so fine texture to render it no discernable...
As this is the pre-project test phase, anything should be tested, and anyone has right to say his opinion.
Said so, I agree, I think the best compromise is with grain 2.
Any other opinions?
test clip 3 with grain 1: http://www.sendspace.com/file/msny52
As the grain size was really similar to grain 2, to show something different I decided to create a grain 2.5 with a grain size that is a bit bigger... take a look at this, and let me know what do you think.
test clip 3 with grain 2.5: http://www.sendspace.com/file/ldi4ls
Well, if the source is IMDB, sometimes it is missing some info; but probably you are right... at least, there is always the gorgeous DTS-HD MA from BD!!!
Plus, I'm pretty sure the laserdisc AC3 soundtrack is a (almost) direct port of the theatrical Dolby Digital soundtrack...
You are right; a 30mpps is perfect for 1920x1080 16:9 29.97fps, but 20/22mbps is good enough for a 2.35:1 1920x816 23.976fps... Escape From New York was quite good, while Indiana Jones trilogy is better than BD for many - me included, based for the moment only on SS comparisons, though...
Cinema DTS.
GREAT!
May I dare to ask you if "The Arrival" is available?
(^^,)