- Joined
- Nov 30, 2002
- Messages
- 115
- Real Name
- Film Restoration & Preservation
Hello everyone,
I hope everyone has and I wish everyone a very merry X-mas. It's been rather a long time since my last visit, and I hope people here will forgive me for "not returning the call" at times. The reason is simply that things were very busy at this end. And from what I could gather during my few peeks here once in a while so were the others, especially Charlie and Robert. But I want to "make up for it" by writing something here with a bit of details of what we did over the year (a sort of an extended blog if you like - I hope noone is discouraged by its length) focusing on one project that we are about to finish and that some of you may even look forward to as it will also yield a Blu-ray Edition (provided you like Fritz Lang's film "[COLOR= rgb(255, 0, 0)]M[/COLOR]"). The details I put at the end of this post.
Looking back on this year the one term that would describe best most projects we worked on is probably unexpectedly complicated. Projects that initially were expected to have "just" a remastering status turned out to become more demanding both in picture and sound, some more than others, taking much more time and effort than expected. The A-Z restoration/preservation of a documentary called VIVA PORTUGAL originally shot on 16mm was especially taxing as it had to be restored from the 3 best surving elements, all having troubles with missing footage, fading, color registration and balance (or rather lack of it), warpage, shrinkage, tears, scratches, severe stains and the list goes on. It literally was assembled shot for shot, with the sound restored from two magnetic masters that were equally problematic.
Alongside working also on some great films such as the wonderfully colorful VELVET GOLDMINE, THE YELLOW ROLLS ROYCE, THE HALLELUJAH TRAIL or THE GREAT ESCAPE our other focus of attention throughout pretty much most of the year was FRITZ LANGS "[COLOR= rgb(255, 0, 0)]M[/COLOR]".
Yes, "[COLOR= rgb(255, 0, 0)]M[/COLOR]" - again. Why? some or even many may ask. Well, the reasons are easily explained.
Now, I myself had pretty much set my focus on the other projects post the release of the Criterion Blu-ray (where Lee Kline and I revisited the color timing on their master in Jan 2010) when yet another grand opportunity came practically out of the blue. The distributor (Germany's Universum Film) was very keen in releasing documentary materials the other releases do not feature, and there were extensive materials we could provide, including our documentary feature THE HUNT FOR M. That alone was interesting enough. However, the focus of the discussion of course also shifted toward the issue of the film and the restoration itself, the key question being "if you were to revisit this, would the results show a noticeable difference from the others ?"
As Robert will no doubt confirm the work on a project is a) never quite finished and b) the only aspects aside from the film elements available that dictate the scope of good you can do in your work are time and budget/money.
With the Criterion Edition, it was a great opportunity to begin with to revisit Fritz Lang's work after our initial work on "[COLOR= rgb(255, 0, 0)]M[/COLOR]" back in 2003. However, Criterion's Lee Kline and I had extremely little time to go over the corrections necessary for the various shots. At the time (as I said Jan 2010) both he and I were happy we could make the corrections at all just in time for the release of the BD. With that in mind, I would have been (and I said as much) very happy with it to "close the book" and was. Then, soon after, that call from Universum Film came, opening the door again, but with an entirely different, exciting set of possibilities.
Most importantly, we could work on the material with a LOAD more time, and invested our efforts to go each individual shot for shot, frame by frame, where we could go into much more detail than we had the chance on the Criterion Edition. Also, what was not to be then(as Lee was not able to come to us here in Berlin) we could do now: work in our 2K DCI enviroment (suite) with native 2K projection on the 6-meter projection screen. This, alonside studio 2K displays and monitoring equipment is essential for such work. This allowed us to be much more precise and accurate in balancing as well as matching gradation and density - in a way far better than we ever could - to the original makers intention (as laid out by Lang and the various reference materials on record as well as nirate reference materials themselves).
This by itself was very welcome, but the time we had also gave us the opportunity to dig much deeper and for the first time takle and fix long standing issues such the very apparent, inherent instabilities in the picture caused by perforation damage[COLOR= rgb(255, 0, 0)]*[/COLOR] (mostly on the duplicate negative materials) and copying errors made in the 1930s as well as thick splices, tears, etc[COLOR= rgb(255, 0, 0)]**[/COLOR]. In many, many sequences that had to be restored from several elements aside from the OCN the densities and gradation had to be matched as closely as technically possible (which is at times extremely difficult given the difference in generation as developmental issues can make that job hard at best, next to impossible at worst). But in most sequences, the positions did not exactly match 100%, either on the photochemical restoration elements, since the precision in the photochemical realm has its limits to what you can do. In very frequent cases the frames would slide downward[COLOR= rgb(255, 0, 0)]**[/COLOR] very visibly at the beginning of a shot or even when the materials would vary.
This was another main focus of our attention, along with the many, many scratches, remnants of tears, wire scratches and stains that were, of course, "also on the menu" along with dust removal (the negative elements had quite a load of all)
)).
We fixed most of the sequences to a degree where they either move much less visibly or did not move/tilt downward at all anymore depending on what the 35mm elements would allow. We tested furthergoing stabilization in general, especially on the duplicate negative material, which suffered some perforation problems ("[COLOR= rgb(255, 0, 0)]M[/COLOR]" is pretty unstable both in X as well as Y axis in those dupe neg sequences) but the results in the end caused artifacts with (too) little gain in return for it to go any further.
Another very important issue was the sound. The problem here is two-fold:
For one, Fritz Lang intended very clearly (as the variable density sound negative clearly shows) this film to be a silent / sound combination with scenes featuring sound (dialogue, ambient sounds etc) while other scenes were silent with merely a few bits of sound added here or there or scenes with no sound whatsoever. In many of these scenes he wanted to focus the viewers attention on the image alone, while in the others he added just bits of sound to "rattle" the audience a bit (loud sound of honking cars, bells, shreaking noises and whistles).
Now, this can be said with absolute certainty since the original variable density negative clearly shows how Langs team worked. In the sections where sound was recorded / intended to be heard the variable density track is visible on the neg; the silent sections are just blank film. No track lines whatsoever. The problem for Lang was, however, that once the negative was copied to a positive source, the intention to have total silence at times was severely limited by the nature of the duplicate "movietone" track: even silent sources had background noise from the emulsion carrier. So you would still hear some noise, and over the years due to decomposition, inclusion of moisture and dust etc the very fragile variable density track system would deteriorate so the noise floor would increase.
That was the status back in 2001, when Martin Sawyer and his team made a beautiful preservation track of that variable density negative in colaboration with photochemical restoration producer Martin Koerber. And this preservation master will also be on the Blu-ray Disc Edtion, together with another sound track option that will reflect Langs original intentions - with a new rebalancing of the dynamic detail in the frequency band and reduction of that noise floor. The preservation track is the basis of that new audio restoration, and after very long tests (and many dissatisfying because of[even slightly] resulting audible artifacts) we are looking at one variation / written program that after having QCd about 1/3 of the films' audio works surprisingly well with regard to the reduction of noise floor without creating distracting artefacts. As for the scenes / parts of shots that Lang intended to be silent they will be exactly so, with no (unintended) background noise from the V/D track. I felt it to be important, however, that on the BD the viewer/buyer/collector gets the option to choose what track he or she wants to listen to - and Universum Film agreed (as they did with so many other things re: Extras and the films restoration efforts). In general, this was and still is a very good collaboration from all sides involved.
For those of you, who are interested, I have listed the various technical data and contents in an attached PDF. With regard to Qs I am all ears)
Happy holidays !
TK
I hope everyone has and I wish everyone a very merry X-mas. It's been rather a long time since my last visit, and I hope people here will forgive me for "not returning the call" at times. The reason is simply that things were very busy at this end. And from what I could gather during my few peeks here once in a while so were the others, especially Charlie and Robert. But I want to "make up for it" by writing something here with a bit of details of what we did over the year (a sort of an extended blog if you like - I hope noone is discouraged by its length) focusing on one project that we are about to finish and that some of you may even look forward to as it will also yield a Blu-ray Edition (provided you like Fritz Lang's film "[COLOR= rgb(255, 0, 0)]M[/COLOR]"). The details I put at the end of this post.
Looking back on this year the one term that would describe best most projects we worked on is probably unexpectedly complicated. Projects that initially were expected to have "just" a remastering status turned out to become more demanding both in picture and sound, some more than others, taking much more time and effort than expected. The A-Z restoration/preservation of a documentary called VIVA PORTUGAL originally shot on 16mm was especially taxing as it had to be restored from the 3 best surving elements, all having troubles with missing footage, fading, color registration and balance (or rather lack of it), warpage, shrinkage, tears, scratches, severe stains and the list goes on. It literally was assembled shot for shot, with the sound restored from two magnetic masters that were equally problematic.
Alongside working also on some great films such as the wonderfully colorful VELVET GOLDMINE, THE YELLOW ROLLS ROYCE, THE HALLELUJAH TRAIL or THE GREAT ESCAPE our other focus of attention throughout pretty much most of the year was FRITZ LANGS "[COLOR= rgb(255, 0, 0)]M[/COLOR]".
Yes, "[COLOR= rgb(255, 0, 0)]M[/COLOR]" - again. Why? some or even many may ask. Well, the reasons are easily explained.
Now, I myself had pretty much set my focus on the other projects post the release of the Criterion Blu-ray (where Lee Kline and I revisited the color timing on their master in Jan 2010) when yet another grand opportunity came practically out of the blue. The distributor (Germany's Universum Film) was very keen in releasing documentary materials the other releases do not feature, and there were extensive materials we could provide, including our documentary feature THE HUNT FOR M. That alone was interesting enough. However, the focus of the discussion of course also shifted toward the issue of the film and the restoration itself, the key question being "if you were to revisit this, would the results show a noticeable difference from the others ?"
As Robert will no doubt confirm the work on a project is a) never quite finished and b) the only aspects aside from the film elements available that dictate the scope of good you can do in your work are time and budget/money.
With the Criterion Edition, it was a great opportunity to begin with to revisit Fritz Lang's work after our initial work on "[COLOR= rgb(255, 0, 0)]M[/COLOR]" back in 2003. However, Criterion's Lee Kline and I had extremely little time to go over the corrections necessary for the various shots. At the time (as I said Jan 2010) both he and I were happy we could make the corrections at all just in time for the release of the BD. With that in mind, I would have been (and I said as much) very happy with it to "close the book" and was. Then, soon after, that call from Universum Film came, opening the door again, but with an entirely different, exciting set of possibilities.
Most importantly, we could work on the material with a LOAD more time, and invested our efforts to go each individual shot for shot, frame by frame, where we could go into much more detail than we had the chance on the Criterion Edition. Also, what was not to be then(as Lee was not able to come to us here in Berlin) we could do now: work in our 2K DCI enviroment (suite) with native 2K projection on the 6-meter projection screen. This, alonside studio 2K displays and monitoring equipment is essential for such work. This allowed us to be much more precise and accurate in balancing as well as matching gradation and density - in a way far better than we ever could - to the original makers intention (as laid out by Lang and the various reference materials on record as well as nirate reference materials themselves).
This by itself was very welcome, but the time we had also gave us the opportunity to dig much deeper and for the first time takle and fix long standing issues such the very apparent, inherent instabilities in the picture caused by perforation damage[COLOR= rgb(255, 0, 0)]*[/COLOR] (mostly on the duplicate negative materials) and copying errors made in the 1930s as well as thick splices, tears, etc[COLOR= rgb(255, 0, 0)]**[/COLOR]. In many, many sequences that had to be restored from several elements aside from the OCN the densities and gradation had to be matched as closely as technically possible (which is at times extremely difficult given the difference in generation as developmental issues can make that job hard at best, next to impossible at worst). But in most sequences, the positions did not exactly match 100%, either on the photochemical restoration elements, since the precision in the photochemical realm has its limits to what you can do. In very frequent cases the frames would slide downward[COLOR= rgb(255, 0, 0)]**[/COLOR] very visibly at the beginning of a shot or even when the materials would vary.
This was another main focus of our attention, along with the many, many scratches, remnants of tears, wire scratches and stains that were, of course, "also on the menu" along with dust removal (the negative elements had quite a load of all)
)). We fixed most of the sequences to a degree where they either move much less visibly or did not move/tilt downward at all anymore depending on what the 35mm elements would allow. We tested furthergoing stabilization in general, especially on the duplicate negative material, which suffered some perforation problems ("[COLOR= rgb(255, 0, 0)]M[/COLOR]" is pretty unstable both in X as well as Y axis in those dupe neg sequences) but the results in the end caused artifacts with (too) little gain in return for it to go any further.
Another very important issue was the sound. The problem here is two-fold:
For one, Fritz Lang intended very clearly (as the variable density sound negative clearly shows) this film to be a silent / sound combination with scenes featuring sound (dialogue, ambient sounds etc) while other scenes were silent with merely a few bits of sound added here or there or scenes with no sound whatsoever. In many of these scenes he wanted to focus the viewers attention on the image alone, while in the others he added just bits of sound to "rattle" the audience a bit (loud sound of honking cars, bells, shreaking noises and whistles).
Now, this can be said with absolute certainty since the original variable density negative clearly shows how Langs team worked. In the sections where sound was recorded / intended to be heard the variable density track is visible on the neg; the silent sections are just blank film. No track lines whatsoever. The problem for Lang was, however, that once the negative was copied to a positive source, the intention to have total silence at times was severely limited by the nature of the duplicate "movietone" track: even silent sources had background noise from the emulsion carrier. So you would still hear some noise, and over the years due to decomposition, inclusion of moisture and dust etc the very fragile variable density track system would deteriorate so the noise floor would increase.
That was the status back in 2001, when Martin Sawyer and his team made a beautiful preservation track of that variable density negative in colaboration with photochemical restoration producer Martin Koerber. And this preservation master will also be on the Blu-ray Disc Edtion, together with another sound track option that will reflect Langs original intentions - with a new rebalancing of the dynamic detail in the frequency band and reduction of that noise floor. The preservation track is the basis of that new audio restoration, and after very long tests (and many dissatisfying because of[even slightly] resulting audible artifacts) we are looking at one variation / written program that after having QCd about 1/3 of the films' audio works surprisingly well with regard to the reduction of noise floor without creating distracting artefacts. As for the scenes / parts of shots that Lang intended to be silent they will be exactly so, with no (unintended) background noise from the V/D track. I felt it to be important, however, that on the BD the viewer/buyer/collector gets the option to choose what track he or she wants to listen to - and Universum Film agreed (as they did with so many other things re: Extras and the films restoration efforts). In general, this was and still is a very good collaboration from all sides involved.
For those of you, who are interested, I have listed the various technical data and contents in an attached PDF. With regard to Qs I am all ears
)Happy holidays !
TK
