854 The Tree of Wooden Clogs

[tenia]

My question is why, then, there are so many other restoration houses and so many other restorations which are then not doing this, if there is such a solid logic behind doing so ? Also, why hasn't this been done earlier ? I don't recall seeing restorations having this generic Ritrovata grading before about 2013.
Furthermore, why colorists using Ritrovata scans don't follow at all this look ? See for instance Criterion's Investigation of a Citizen above Suspicion. Basically, every time the color timing isn't performed by Bologna or Eclair, you don't get a movie with this generic yellow flat grading.

The Tree of Wooden Clogs seems like a different case, since according to the restoration credits Olmi requested the new color timing.

Olmi requests a new color timing for Wooden Clogs : yellowish cold flat grading.
Shoji Ueda supervises the grading for the new restoration of Ran : yellowish cold flat grading.
Hua Hui-Ying supervises the grading for the new restoration of Dragon Inn : yellowish cold flat grading.
Tovoli supervises the grading for the new restoration of Deep Red : yellowish cold flat grading (which Arrow had to get color-corrected because it was just so bad).
Ernio Guarnieri supervises the grading for the new restoration of Marriage Italian Style : yellowish cold flat grading.
Whoever at Eclair does the restorations and the gradings for Gaumont Decouverte's color movies : yellowish cold flat grading.
Whoever did the work on SC's Purple Noon 4K restoration : yellowish cold flat grading.


I get that Bologna might be adjusting the color gradings to look like theatrical release prints, but surely these movies weren't all looking the freaking same in theaters ? If you're adjusting different things, they should still be different in the end, and yet, all these movies all look as if they were shot in the same place, at the same time, with the same material, by the same DoP.

Or did nobody noticed at the time that Ran really looks a lot like La valise ?

[swo17]

My question is why, then, there are so many other restoration houses and so many other restorations which are then not doing this, if there is such a solid logic behind doing so ?

Because there is also a logic for not doing so, and there is not necessarily one right answer. Like that 2-perf post suggested: should a restoration of a movie look like the projection of a pristine Release Print or should it look like color corrected Original Camera Negative?

[movielocke]

My question is why, then, there are so many other restoration houses and so many other restorations which are then not doing this, if there is such a solid logic behind doing so ? Also, why hasn't this been done earlier ? I don't recall seeing restorations having this generic Ritrovata grading before about 2013.
Furthermore, why colorists using Ritrovata scans don't follow at all this look ? See for instance Criterion's Investigation of a Citizen above Suspicion. Basically, every time the color timing isn't performed by Bologna or Eclair, you don't get a movie with this generic yellow flat grading.

The Tree of Wooden Clogs seems like a different case, since according to the restoration credits Olmi requested the new color timing.

Olmi requests a new color timing for Wooden Clogs : yellowish cold flat grading.
Shoji Ueda supervises the grading for the new restoration of Ran : yellowish cold flat grading.
Hua Hui-Ying supervises the grading for the new restoration of Dragon Inn : yellowish cold flat grading.
Tovoli supervises the grading for the new restoration of Deep Red : yellowish cold flat grading (which Arrow had to get color-corrected because it was just so bad).
Ernio Guarnieri supervises the grading for the new restoration of Marriage Italian Style : yellowish cold flat grading.
Whoever at Eclair does the restorations and the gradings for Gaumont Decouverte's color movies : yellowish cold flat grading.
Whoever did the work on SC's Purple Noon 4K restoration : yellowish cold flat grading.


I get that Bologna might be adjusting the color gradings to look like theatrical release prints, but surely these movies weren't all looking the freaking same in theaters ? If you're adjusting different things, they should still be different in the end, and yet, all these movies all look as if they were shot in the same place, at the same time, with the same material, by the same DoP.

Or did nobody noticed at the time that Ran really looks a lot like La valise ?

Well first let us not consider any DVD release authoritative, they didn't theoretically do it right at one point only to malevolently decide to do it wrong later.

So consider the film chain, perhaps the DVD was taken from a tape that was taken from a release print, that doesn't mean it is accurate because the DVD crushed the prints color reproduction and dynamic range into the 1950s limitations of ntsc televisions. Shoot take a photo and change its color profile and the changes are significant and it will print quite differently.

Additionally, the release print was timed for the standard projection bulb color temperature of the market it is being released into, the print may be natively yellow but projected with a 6500K bulb neutralizes the color cast as intended in the grading. Scanning the release print with a neutral scanning light like 5000K and assuming that the yellow color cast that inadvertently results is what was meant to be seen is a very common occurrence of the home video era.

And the kind of light projection bulbs cast has changed over time, arc light projectors had a very different color cast to them than modern projection bulbs, so projecting a vintage release print with a modern bulb or modern biased scanning light, does not render a correct color reproduction (regardless of fading)

[MichaelB]

There's also Paolo & Vittorio Taviani's Kaos, which I don't think is as extreme as the examples mentioned above, but it still shows yellowish tendencies - and yet it was nonetheless approved by both the directors and their cinematographer.

(And because of this, I wasn't inclined to make any changes to the Arrow version: I'd have been arguing with the creative decisions of the three people in the strongest possible position of authority, and who am I to contradict them? Also, I thought it looked pretty terrific: working on that film was an absolute joy.)

[tenia]

Well first let us not consider any DVD release authoritative, they didn't theoretically do it right at one point only to malevolently decide to do it wrong later.

So consider the film chain, perhaps the DVD was taken from a tape that was taken from a release print, that doesn't mean it is accurate because the DVD crushed the prints color reproduction and dynamic range into the 1950s limitations of ntsc televisions. Shoot take a photo and change its color profile and the changes are significant and it will print quite differently.

We have numerous 2K and 4K restorations, made let's say between 2010 and 2013, that don't show at all this specific color grading. So it's not a "DVD were wrong" story. But as swo answered to me : in the end, it might just be Bologna and Eclair adhering to a certain philosophy that nobody else share, so when they started doing more and more work, we started to get more and more movies with this grading.

But again, it doesn't explain why their restorations all look the same. Applying the same type of corrections, sure, OK, but applying adjustments to different things should still output different things. That's my main point against their gradings : I highly doubt all these movies originally shared the same photography but now they do.

[Drucker]

I have no expertise here, so I'm just thinking aloud. I'm wondering if what we are seeing is slightly unique color-timing on a film being made generic. Many people attested that A Touch of Zen did in fact have a yellow-tint in release prints they've seen, and we saw one posted on a blog. But when it comes time to give it that yellow tint, the digital program doing only has one way to "give a yellow tint."

We have seen DNR programs that erase details that shouldn't be erased. I figure the assumption being that a program was run to remove detail, and it made some mistakes. Couldn't the same "apply all" type philosophy associated with these digital programs just be leading to the same outcomes?

[tenia]

That's the theory I tend to have : there is Auto-Clean and now there might also be an automated color-grading (which would mean that the workflow isn't biasing the input but is aiming towards a specific output).

[jsteffe]

My question is why, then, there are so many other restoration houses and so many other restorations which are then not doing this, if there is such a solid logic behind doing so ? Also, why hasn't this been done earlier ? I don't recall seeing restorations having this generic Ritrovata grading before about 2013.
Furthermore, why colorists using Ritrovata scans don't follow at all this look ? See for instance Criterion's Investigation of a Citizen above Suspicion. Basically, every time the color timing isn't performed by Bologna or Eclair, you don't get a movie with this generic yellow flat grading.

I think the *principle* of going by release prints is good. If you doing color timing from scratch working from the OCN, it could introduce significant changes to the look of the film. Also, the OCN is great for revealing detail, but sometimes it can reveal more detail than people were supposed to see in the theaters - it can cause an actor's makeup to look overly artificial or betray special effects (such as the wires in The Wizard of Oz).

I don't necessarily endorse every restoration Bologna or Eclair has done. Besides being very blue, the screen caps of The Tree of Wooden Clogs appear to have an overly compressed dynamic range. I'd really like see the actual Blu-ray in playback.

You're right that if a single facility puts out a large number of yellowish or teal-leaning restorations, it suggests a problem. There are also possible differences between the Blu-ray releases and the DCPs of the same restorations due to changes introduced at the Blu-ray mastering and encoding stage. I thought the DCP of Ran looked very good, for instance, and didn't seem as yellow as the Blu-ray screen caps, though I may be wrong.

I guess what I mean to say is that the story is probably more complex than just a recent spate of randomly yellow restorations, although that may be true to a significant extent.

[jsteffe]

Well first let us not consider any DVD release authoritative, they didn't theoretically do it right at one point only to malevolently decide to do it wrong later.

So consider the film chain, perhaps the DVD was taken from a tape that was taken from a release print, that doesn't mean it is accurate because the DVD crushed the prints color reproduction and dynamic range into the 1950s limitations of ntsc televisions. Shoot take a photo and change its color profile and the changes are significant and it will print quite differently.

Additionally, the release print was timed for the standard projection bulb color temperature of the market it is being released into, the print may be natively yellow but projected with a 6500K bulb neutralizes the color cast as intended in the grading. Scanning the release print with a neutral scanning light like 5000K and assuming that the yellow color cast that inadvertently results is what was meant to be seen is a very common occurrence of the home video era.

And the kind of light projection bulbs cast has changed over time, arc light projectors had a very different color cast to them than modern projection bulbs, so projecting a vintage release print with a modern bulb or modern biased scanning light, does not render a correct color reproduction (regardless of fading)

These are all really good points, and they suggest just how complex all of these problems really are.

[tenia]

randomly yellow restorations

The fact that you can guess where the restoration has been performed just by its color grading suggests it isn't so random. I don't think all these movies have been said to have similar photographies when they were released theatrically, but now they do, as if whatever the movie, the restoration will have this grading anyway.

[Costa]

So, this theory says in essence that every film in the past when projected looked like teal-yellowish? without any pure whites etc.?

I don't know but it doesn't stand so much for me, since I saw both Indochine and La Reine Margot on release, and they definitely weren't like this.
On the other hand, prints in the theaters differ from one theater to the other, right?

by the way, I assume one of those 2 involved studios did the restoration to the Gavras films too, right?

[tenia]

Most Costa-Gravas 2K restorations were done by Eclair, so here you go.

[Costa]

Most Costa-Gravas 2K restorations were done by Eclair, so here you go.

I see.
The thing I don't understand though is why people more knowledgeable and in the restoration game, don't criticize such tactics.
because if they were, maybe those studios thought it better afterwards.

I was thinking of starting a new thread at hometheaterforum, specifically refering to these 2 studios in the subject and make a serious discussion on the matter, but I'm afraid I'll be bashed by members and especially one.

[tenia]

Comparaison available now at caps-a-holic between the Criterion and Arrow BDs. The combination of the yellow dull color-grading and their usual crappy compression makes at times the Criterion disc looking like it's using a much older restoration than what Arrow is using.

[ellipsis7]

Interesting, there's certainly a difference...

However the Criterion edition, released in February 2017, is sourced so...

The Tree of Wooden Clogs was restored (in 4K) by the Cineteca di Bologna at L’Immagine Ritrovata (and supervised by director Ermanno Olmi), with funding provided by The Film Foundation.

New program featuring cast and crew discussing the film at the Cinema Ritrovato film festival in Bologna, Italy, in 2016

So a recent restoration, directly from the original 16mm negative presumably...

The Arrow Academy dual format edition was released in August 2017 and is sourced similarly...

Brand-new 4K restoration by The Film Foundation

So to all intents and purposes they are one and the same restoration, regardless of the varying end results....

[tenia]

Both are from the same restoration, but Arrow seemingly had it color corrected for their release. It wouldn't be the first time : they had Deep Red color-corrected for their BD release.

[mhofmann]

Comparaison available now at caps-a-holic between the Criterion and Arrow BDs. The combination of the yellow dull color-grading and their usual crappy compression makes at times the Criterion disc looking like it's using a much older restoration than what Arrow is using.

I agree, and yes, Arrow purposefully re-graded the restoration, as far as I know to be more close to release prints.
I have neither seen release prints nor have I sat next to Olmi during the restoration grading, so whatever I say is unreliable and subjective by definition.

Even though I am very concerned that the Criterion release is a victim of the usual "Ritrovatarization", I find the more green-blue leaning color scheme sometimes a bit more appealing and mood-transporting than the more naturalistic brownish scheme of the Arrow release which occasionally leans a bit too much into the magenta tones. One could look at it as a painting in distinct stages of restoration/decay.
What is the truth here? No idea. Maybe it's good to have both.

The Criterion transfer (as released) definitely drops the ball on dynamic range and black levels, though -- it's really far too flat, crushed and murky, and I have a very hard time believing many scenes were meant to be this way. This is where the dreaded LUT might have destroyed details after Olmi signed off on a restoration with an admittedly revised color scheme. It's too extreme.

The other thing that is shocking to me is the lack of grain as well as fine details (especially in the quoted caps) on the Criterion disc. The Arrow compression is much more careful here - kudos to a job well done!
Yes, it's well-known that Criterion compression can be quite sub-par in many cases, but I don't understand why they still defend their authoring house. It's not that people haven't ever tried to tell them.