Restoring HIROSHIMA MON AMOUR
A 4K restoration of Alain Resnais's classic 1959 film HIROSHIMA MON AMOUR was undertaken in 2013 by Argos Films, the Technicolor Foundation, the Groupama Gan Foundation and Cineteca di Bologna. The work was carried out at L'Immagine Ritrovata, Cineteca di Bologna's restoration lab, where facility director Davide Pozzi oversaw the delicate task of scanning the original camera negative. He speaks here about how the restoration process was aided by the unique advantages of the ARRISCAN and ARRI Wet Gate.
A 4K restoration of Alain Resnais's classic 1959 film HIROSHIMA MON AMOUR was undertaken in 2013 at the Italian restoration lab L'Immagine Ritrovata, where an ARRISCAN and ARRI archive tools were used for the delicate task of scanning the original camera negative.
When did L'Immagine Ritrovata acquire its first ARRISCAN?
In 2009 we decided that in order to carry out the restorations we wanted to do, we needed a scanner that could digitize any kind of material, including nitrate and non-standard formats, as well as shrunken or damaged films. We went to ARRI Munich to do a test with some unusual Lumière brothers' footage, which had only one round perforation per frame. We scanned it with an ARRISCAN and the Sprocketless Transport, and it was one of the best days of my professional life because I realized we had found a machine that could handle archive work. That was in March 2009 and by August we were finalizing our purchase of an ARRISCAN; I think ours was the first Sprocketless Transport to be delivered.
Did you invest in further ARRI archive tools?
We had already invested in an ARRILASER in 2008, and in 2012 we bought the 16 mm and 35 mm Wet Gates for our ARRISCAN, which was a very important step. Then in 2013 we bought our second ARRISCAN and at that point we decided to build a special, climate-controlled room for our ARRI machines. The new room is now completed and my hope is to buy a third ARRISCAN to go in there. We're doing restorations for clients worldwide and our scanners are running all day, but we'd like to increase our volume still further. That's why my goal is to have three ARRISCANs, two of them with a Wet Gate. We also have the 16 mm and 35 mm Archive Gates, and we hope to test ARRI's Built-in Stabilization soon.
What materials did you have for HIROSHIMA MON AMOUR?
We had the original camera negative and an inter-positive. Mainly we were able to scan the film dry, but there were two reels that were affected by mold and we decided to scan those with the Wet Gate. The miracle of the Wet Gate was that in this case it completely fixed the problem and we didn't have to add extra digital cleaning hours into our schedule. Some people emphasize the Wet Gate's ability to fix scratches and dirt -- and it can -- but for me its true power is in being able to fix mold that might otherwise mean having to use another element, rather than the original camera negative. We are working more and more with film that has been stored in humid countries and is often very affected by mold; since getting the Wet Gate the quality we can achieve for these films has increased dramatically.
There were a few shots with frames that were either missing or badly damaged, and for these we turned to the inter-positive, but having the ARRISCAN Wet Gate allowed us to work almost entirely from the original camera negative. The perforations were in good condition on this film, so we didn't have to use the Sprocketless Transport.
Having the ARRISCAN Wet Gate allowed us to work almost entirely from the original camera negative.
Did Alain Resnais get involved in the restoration?
Yes, we were fortunate to have his input and we were very sad when he passed away in March this year. At the start, Resnais was concerned about the fact that he had created HIROSHIMA MON AMOUR on film and we were going to restore and present it digitally. He wanted to be sure that the grain was going to look right -- that the film would still have a cinema look, not a video look -- so we spent many hours doing tests to ensure that our restoration respected the original grain as much as possible. Those tests were screened for Resnais and in the end he was happy with the result.
How did you get the grain right?
The best way to respect the grain of the original film stock is with a 4K restoration. With a 2K workflow you can have a lot of trouble respecting the look of the original grain, but by following a very linear and simple 4K workflow everything is much more natural. Put simply, with a 2K workflow you see pixels, but with 4K you see the original grain.
What other challenges did you face?
Another important step in the restoration was the color correction. Our invaluable consultant was a cinematographer called Renato Berta, who didn't work on HIROSHIMA MON AMOUR, but he did shoot other films for Alain Resnais and he came to Bologna to be part of the color correction process. He was very helpful, mainly because he was friends with Resnais and could call him to ask anything we needed to know. But he also knew how Resnais worked on set and could tell us what kind of lighting he liked on faces and things like that. It helped us take the right approach.
Why was color correction necessary for a black-and-white film?
The color correction for a black-and-white film is very important because there are a lot of different kinds of black-and-white. You can have more contrast or less contrast, it really depends on the film, so it's vital to find out as much as you can about what film stocks and lab processes were used. Even with black-and-white films, color correction is always the most delicate issue of every restoration. Applying a modern color correction to a 1959 movie would have meant creating a fake. On every single film you have to respect what was originally done and the way it was released at the time; this is very important.