[In progress] Big Trouble In Little China (1986) [Project: Celluloid — Regrade]

[LucasGodzilla]

Hello everyone, just wanted to post that I am not dead but have been merely very busy lately. My free time, simply put, is dwindling so I don't get as many opportunities to work on my projects, however I do want to give a status report of sorts with the creation of this thread.

I've been hard at work for the past few months trying to regrade the European master of Big Trouble In Little China (with many issues arising, forcing me to restart from scratch essentially three times) and I thought it'd be a nice tease for what's to come as well as a means of getting feedback as I am currently pressed with a little issue of sorts.

I'm currently trying to determine what'd be a better approach for this project, attempting to recover more details in the shadows and highlights at the cost of lowering the image contrast and mood to a certain extent or present a regrade more true to the 35mm source I'm working off of (which still holds more details in both ranges I may add but less explicitly).

Higher Contrast | Lower Contrast
 

 

Which would y'all say is better?

I think the higher contrast looks fine on my display, but I seem to find that the rule of thumb is that usually when it looks good to my eyes it looks too crushed to someone else's.

I'm also curious to hear how it looks to those who've seen the movie projected as the saturation has been unintentionally boosted after altering with the regraded footage in Resolve to boost shadows and retain some highlight detail (as the straight LUT crushed and clipped a lot of data given the print source's original contrast), leading to a lot of scenes with richer blues than expected (with a fair amount of gamut clipping in some areas that I'm certain will be pulled back on the encoding stage through some AVS limited palette commands; I've also already applied a bit of saturation compression in Resolve to help).

Both are fine to me, I’d be happier with either choice you make.

But if I had to make a choice, I lean towards the lower contrast version just because it’s a good split between the color and mood of the 35 and seeing more details from the BD.

[bronan]

I prefer the lower contrast, but the highlights seem a bit dim, especially on the skin. I tried to fix one in Photoshop, compare the highlights on her cheeks.

https://i.imgur.com/jzd3f5L.png

[X5gb]

Agree with everyone else, lower contrast. But to be sure, maybe you could post a small sample clip of both so we can see them in motion to be sure.

[LucasGodzilla]

I prefer the lower contrast, but the highlights seem a bit dim, especially on the skin. I tried to fix one in Photoshop, compare the highlights on her cheeks.

Yeah, the highlights is something I'm trying to figure out the best approach in trying to make the most of the Euro master's highlights which appears to hold more information than the Fox master—which clip a lot of highlights outright—but it's rather compressed, forcing me to figure out a way to stretch it down so-to-speak to showcase more of it.

Experimenting with some alterations in the HDR wheel grading settings, here's what I have now.

Untampered Highlights | Dimmed Highlights V2



 

Which of the two is "better", I don't know as it looks fine to me in either case, I just know some people out there prefer the highlights to be pulled back a bit to showcase a smidge more detail in the upper range. I will stick with the lower-contrast edit though for the shadows since that seems to be the consensus.

Both look good to me, but the lower contrast is nice for retaining details, and probably gives the best balance. Regarding the hilights, I don't see much different either and they both look fine to me also, You're not going to have much more hi-light information to pull out with a SDR master anyway. Just me two cents. Thanks for all the hard work.

Lower contrast looks preferable, though I can't tell the difference at all between the untampered highlights and dimmed highlights. So take that with a grain of salt.

Here's to the Army and Navy and the battles they have won; here's to America's colors, the colors that never run.

[LucasGodzilla]

Hi everyone, just wanted to make a little minor update post to say where I'm at with this. I ended up going back and redoing much of the regrade by switching to shot-by-shot LUTs and am currently somewhere past the 45-minute mark (my life has been very busy as of late so I haven't been able to make as much progress with this as I'd have liked).

I'm posting the screenshot thumbnails below for a fun comparison between the grade and the masters.

As you can see, colors will not be 100% accurate to the reference source, but that's a result of me tweaking the resultant Dr. Dre LUT to lift the shadows and pull back the highlights to get more out of the picture which did alter the coloration a smidge.

[LucasGodzilla]

So I was pondering to myself that I wanted to ask the FanRes Hive Mind about to see if there may be any sort of credence to the theory.

After seeing some of the results in the past with the AutoOverlay in The Abyss thread again...

I've made some tests with development version of AutoOverlay plugin. It supports a new color matching algorithm and warp transformations.
Also there is a new overlay filter that extract most complexity (i.e. high quality) parts of all sources and merge them into the one.
There are two OAR and two open matte sources overlaid in that way.

It sort of got me thinking about whether or not it'd be theoretically possible to do try and merge together as many of the blu-ray encodes of the Euro master to try and inch out as much detail from it as possible. I don't know if it's really worth the endeavor or not though since I only have a WebDL and the Italian BD so I can't even do experiment with this at the current moment.

I also wanted to ask what people's thoughts are on the DVD mix since I hear that the current consensus is that it's the 70mm spread mix? Doing minor comparisons between it and the 5.1 mix in Audacity, I was able to determine that the rear tracks on the 5.1 are essentially the same as the 4.1 but at different pitches (I presume the mono-rear is accurate since I believe the movie didn't have split-surround) and that the sub-bass track is is quite... different? I'm not sure how to show this exactly, but the filtering was different than what you'd usually see on typical surround mixes; Less of a low-pass than most typical mixes which is strange to see / hear. Don't suppose anyone has any explanation as to why that is?

Its a lot of effort to possibly get a little more detail. It can't hurt to try a small sample but I suspect its more effort then its worth in this particular case. Just my 2 cents.