Tweet Film Preservation in the Digital Age

An interesting article from the Australian National Film and Sound Archives (the original also has some worthwhile hyperlinks - http://www.nfsa.gov.au/blog/2011/01/02/are-we-losing-picture/)

-------------------------------------

Are we losing the picture?
02 January 2011
by Dominic Case

Nearly a decade after the first digital projectors were installed to screen Star Wars Episode 2, the change to digital is gathering speed very rapidly indeed. Driven by the success of digital 3D movies such as Avatar, cinemas in the US are installing digital systems as fast as manufacturers can supply them. After a hesitant start, the same is happening in Australia, where the switch to digital also means that independent regional cinemas will at last be able to screen features at the same time as the metropolitan chains. Across Europe, the change is almost complete.

And so here is the dilemma. Film archives around the world have developed expertise in preserving film. Many expect their collections to last for 400 years or more, stored in carefully-controlled conditions. The old problems of inflammable nitrate film, shrinking and decomposing acetate film, fading colour dyes, are all understood and controlled.

So how will we screen these prints in the future? And when will that future arrive? Sooner than we expect. A correspondent writes: 'All screens in Belgium will be digital by the end of the month. No more film prints. They have dismantled the film projectors from the booths. They plan to keep perhaps one per multiplex, just in case, for one year or so.’ In Australia, nearly all cinemas will be digital in three to five years, and few of them will still be able to project film.

So why don’t we simply digitise everything?

There are so many answers to this question. For a start, no-one knows how long digital data can be preserved or how much more it will cost. Best estimates for most digital formats are around ten years, while a couple of years ago the Academy of Motion Picture Arts and Sciences (the organisation behind the Academy Awards or Oscars) in its seminal report The Digital Dilemma estimated that long-term, secure digital preservation would cost about 12 times as much as conventional film preservation.

Do we trust the technology? All you need to see a film image is a light source. Once the equipment to retrieve a digital image is lost, the image is lost too.

Then there is the question of security. Digital copies of commercial feature films are encrypted so that they can only be shown on specific projectors at specific times, in an effort to reduce illegal copying. Who knows when or where an archival copy will be required for screening?

And then the moral dilemma. We want to see artworks in the way their makers meant them to be seen. An archive needs to preserve film as film, even if we make digital copies as well.

So what are we to do? Film? Digital? Both? More later.

Interesting.

For me, this is the key:

We want to see artworks in the way their makers meant them to be seen.

For a start, no-one knows how long digital data can be preserved.

I think this is a misunderstanding of the digital medium. When you're talking about film, you need to preserve the physical object because the picture and the substrate (film) it's printed on are inseparable. But digital can be "printed" (stored) on any media device that's large enough to hold the file. So, while DVDs, hard disks, and Flash drives certainly won't last forever, it's not a big issue to move their contents to some new media device before the old device fails or random cosmic rays somehow corrupt the data.

[clip] . . . a couple of years ago the Academy of Motion Picture Arts and Sciences (the organisation behind the Academy Awards or Oscars) in its seminal report, "The Digital Dilemma," estimated that long-term, secure digital preservation would cost about 12 times as much as conventional film preservation . . . [clip]

I'd be interested in knowing how the Academy came up with that number. Aside from the initial one-time expense of digitizing the film, the amount of storage space needed and the environment requirements (temp and humidity) should be minimal. And they'd need to make a digital master anyway if they ever wanted to release the film commercially on DVD (hint, hint).

Conversely, I believe film is supposed to be stored at around 45 degrees Farenheit or colder, and it also needs to be inspected routinely for signs of decomposition. If and when that occurs, the only option is to copy the image onto another newer piece of film, with the resultant loss of image quality -- assuming that the original film hasn't shrunk or warped in the interim, creating even more transfer problems. SETH

[clip] . . . a couple of years ago the Academy of Motion Picture Arts and Sciences (the organisation behind the Academy Awards or Oscars) in its seminal report, "The Digital Dilemma," estimated that long-term, secure digital preservation would cost about 12 times as much as conventional film preservation . . . [clip]

I'd be interested in knowing how the Academy came up with that number.

They are referring to the preservation of the movies that are made today. When we talk about an original film element from fifty years ago, we're talking about the original negative or what have you - a couple of reels in a box. The Academy quotes the following as the amount of memory required for a modern digitally shot movie:

• More than 50 Megabytes per frame (and 24 frames are produced every second)
• More than 8 Terabytes per master version of a two hour movie (and there can be 40 different masters)
• More than 2 Petabytes for an entire digital movie production

(there's a decent summary of what's in it at http://www.moviemaker.com/distribution/page2/facing_the_digital_dilemma_20080919/)

2 Petabytes is a SERIOUSLY large amount of memory (2,000 trillion bytes). It could be that for the film buff of a century from now, a copy that is only available in a poorly-compressed version with irrevocably diminished picture quality is our equivalent of a film that exists only as a poorly-kinescoped copy.

But we are talking about the digitalization of movies that were made 50+ years ago. If I can buy a single $20 Blu-ray disc that contains a 1080p version of "Casablanca" as well as a bunch of extra features, we can't be talking about a zillion megabytes of memory in order to accomplish that.

So I'm still confused about these estimates. SETH

[Danny Burk]

But we are talking about the digitalization of movies that were made 50+ years ago. If I can buy a single $20 Blu-ray disc that contains a 1080p version of "Casablanca" as well as a bunch of extra features, we can't be talking about a zillion megabytes of memory in order to accomplish that.

So I'm still confused about these estimates. SETH

A blu-ray isn't a preservation copy. It may look great on your TV, but its real resolution is much less than an actual 35mm print (regardless of the print's age), and a digital preservation copy is a high-resolution scan that captures all of the detail that's actually present in the print. A blu-ray's resolution at 1080p is far less than 35mm film, even before factoring in compression to fit onto a blu-ray disc.

But you can scan a film at, say, 4k or 6k and use lossless compression or very high bit rate lossy compression and still have something that's essentially higher resolution than a film dupe, so I'm fine with it. You could easily fit a film that to the naked eye, projected on a full-size movie screen would look indistinguishable from film onto a 2 terabyte drive.

And remember that unless you're scanning from the original negative, which for the vast majority of old films doesn't exist, you're not starting with the full resolving power of 35mm.

Thanks, Danny, for the additional info, it's helpful in clarifying the situation. SETH

While digital certainly can be copied losslessly forever, the concern from a preservation perspective is "Will it be? Sure, the big money generating titles will be carefully preserved at 2/4K archival resolution and copied to to new media as needed, but what about the thousands of other titles, that eventually will be copied to digital media and then sit on the shelf for decades. Unlike film, if they pull the only archived 2/4K copy off the self in 50 years and the media has degraded to the point it will no longer play, that is the end. Consumer resolution copies are not really preservation copies, as they are only a quarter of the resolution of the original.

The other, and related issue, is what happens when the media becomes so obsolete that there is no equipment left to play it, or the proprietary codec used to encode it are no longer available. Again, the big titles will have been copied over to new media and codecs though years, but not the thousands of others.

20 years ago, the state of the art for consumer digital video was CD-I. These days, those discs are coasters because there is no equipment that can still play them back outside of a few hobbyists, and certainly no software that can copy them over to new formats in the digital domain. When those few surviving CD-I players go, that is the end. Fortunately, CD-I was a consumer format, but my point is that it's obsolescence was only 20 years.

Does this mean, Mitchell, that we shouldn't make the copies? What do we do in fifty years when we go to the original and it's dust?

Bob

While digital certainly can be copied losslessly forever, the concern from a preservation perspective is "Will it be? Sure, the big money generating titles will be carefully preserved at 2/4K archival resolution and copied to to new media as needed, but what about the thousands of other titles,

Well, that's hardly a digital vs. film thing-- whether a studio thinks a title is worth preserving has little to do with whether we're in a digital world or an analog one. That said though, digital is a bit quicker and easier and cheaper so it just might lead to more preservation. Better still, it's DEFINITELY leading to more access, just look at the Warner Archives and the upcoming Archive collections from other studios.

As for media becoming obsolete, there are two ways to deal with that:
1) Keep the hardware and software around for as long as possible. There are companies already who specialize in being able to read obsolete media.

2) Update the media every few years. This is not as big of an issue as you'd think, either, because media gets faster and bigger each year, so it becomes easier to do with each passing year. I myself have some DVDs of things that went from a 1987 hard drive... to 5 1/4 inch floppy... to 3.5 inch floppy... to 88mb Syquest cartridge... to 125 mb Bernoulli cartridge... to Zip disk... to Jazz disk... to CD... and now to I might even consolidate the DVDs to Blu-ray. I also still keep the "obsolete" media, though I haven't tried most of them in years.

As for codecs, those are easier to hold onto than hardware. You can just store them on the media itself. And I can't think of any relevant codecs that it's no longer possible to use. Doesn't mean it'll never happen, but transcoding to newer codecs could be done while transferring to new media.

The thing that I do love about actual film (and books) is that it's such a low tech media and doesn't require much to get at least some use out of it. For example, if all else fails, you can hold it up to the light and get something out of it. (Books are even better in that regard.)

Bob, you missed my point. The issue is not that we shouldn't copy, but that lesser titles will not be copied do to the copyright owners lack of interest.

fwtep, it is a film vs digital thing. First, properly stored film has a much longer known lifespan than current digital media. Second, film damage from decomposition, shrinkage, etc will only affect that section of the film, and unless totally destroyed, can be restored in either the digital or analog domain. Once digital media fails, the entire copy is lost.

As to storing the codecs with the copy, sounds like a good idea, but what do you do with it in 60 to 70 years, when computer architecture will be vastly different than it is today. If you had the source code to the codec, I suppose you could write software to play these back, but many of the high resolution and audio codecs are proprietary and the source is not available.

Film projectors and printers are fairly straight forward mechanical devices, and analog sound, even Dolby Stereo, is fairly low tech. The specifications for film are widely published. In a 100 or 200 years, someone could still fabricate projector from scratch if the need is there. Digital is processed in IC chips, which can currently be cloned, but in 100 years that is unlikely to be an option.

Remember, we are talking preservation on this thread, not consumer.

To me, the migration to new media every so many years is one of the big problems with considering that digital can be a preservation medium. It is easy to say to do this, but as the years go by, ownerships change, original producers go away and the elements are stored who knows where.
It is not like a major studio, with a department following up on all this.
With film, we know of hundreds of pictures where the negatives were lost, but some later generation was found in some storage facility.
Depending on migration will possibly result in lots of things being lost.

But again, Richard, the point is that whether a studio bothers to do it or not has nothing to do with whether it's film or digital. Both have maintenance requirements. Yet at least with digital it maintains a higher quality product (no further generation of dupes).

And the migration to new media need only really be done once every 10 years or so. So a film as old as Birth of a Nation would have only had to migrate about 9 times so far.

There's also quite a possibility that in the not too distant future there will be no need to migrate as often (or at all) because the media will not become obsolescent as fast, if at all. For example, look at hard drives. You can still use hard drives from 20 years ago. They were too small to be viable for storing films, but a hard drive of today can store a film and it'll still be usable way after CDs and DVDs and Blu-rays readers are gone.

Any film preservation requires effort on the part of the copyright holder. That effort is less for digital and the results are better. It isn't perfect, but neither is film.

You just proved the financial problems with digital, it will have to be transferred every 5-10 years if it is to be preserved, whereas film has only to have maybe been transferred once in a 100 years, from nitrate to safety. Both the previous answers hit the nail on the head, digital requires so much more than film, be it preserving the hard drives, the codes, etc. Ownership of so much product is changing so much now, and much won't be passed over or will be forgotten in the future. At that point, without knowing its source or code, it will be worthless and lost to history. Lesser known titles, and probably most silents, will not make the transfer to new media in the future, because the vast majority of companies won't think it will make them any money, which it barely does now. They will consider it too costly to maintain the material, much less transfer it, to new media. And this media requires the same or even colder temperatures than film to be preserved, and should be dust free, unlike film. One scratch won't ruin a film, but it destroys a disc.

[Mike Gebert]

And the migration to new media need only really be done once every 10 years or so. So a film as old as Birth of a Nation would have only had to migrate about 9 times so far.

So we'd have lost it in 1935-- after its commercial value was up, but before MOMA could have saved it.

You just proved the financial problems with digital, it will have to be transferred every 5-10 years if it is to be preserved, whereas film has only to have maybe been transferred once in a 100 years, from nitrate to safety.

Um, about $200 or less per film every 10 years is hardly a "financial problem."

Lesser known titles, and probably most silents, will not make the transfer to new media in the future

Well then they won't be taken care of on film either.

And this media requires the same or even colder temperatures than film to be preserved, and should be dust free, unlike film. One scratch won't ruin a film, but it destroys a disc.

Colder? No. I don't know where you heard that. Ditto dust free. And one scratch doesn't destroy a disk unless it's one heck of a scratch, which is pretty unlikely to happen since the discs don't need to be handled more than once every 10 or so years. (Scratches can be polished out.) I doubt any archivist will be harder on disks than my kids, and their DVDs are still playable.

And even so, with digital there will be MANY COPIES, rather than just one preserved film print. (And the quality of a film preservation goes down with each succeeding copy.)

So we'd have lost it in 1935-- after its commercial value was up, but before MOMA could have saved it.

Not really. Digital media from decades ago is still readable, so to apply that to my example would mean that it would have lasted plenty long until someone else felt it worth continuing to preserve. And the prints we'd have today would look literally as good as the film did on its first release. I sure wish, for example, that we had digital copies of Chaplin's Keystones from 1914 rather than the sorry state of them that we have on film. And believe me, Chaplin would have gotten and kept a digital copy.

And remember, media/formats that are specifically made for archiving have a long life, both physically and in terms of archivists keeping the formats accessible.

So far, the only advantage to film print preservation is that it basically requires no technology. That IS a big plus, but to me, the advantages to digital far outweigh it: Multiple 100% identical copies... automatically makes it viable for distribution with virtually no effort (digital downloads)... disaster-proof (because of multiple copies at different locations)... cheaper (assuming proper archiving, not merely kept in a box in the attic, though that would work fine with optical disc storage too) etc.

One last note for now: Media storage changes have slowed down drastically since the early/mid 90's. That was early enough in the home computer era that everyone was trying to come up with the standard. So we had lots and lots of options (Zip disks, Jazz disks, Syquest, etc.). But it did eventually standardize around CDs. CDs are still around, and far from obsolete even after the introduction of DVDs and Blu-ray. I have lots of discs from 1997 that still work (I use them regularly) even though they're not stored in a way that a professional archivist would approve (I do take care of them though). I think the average person will have no trouble accessing a CD/DVD/Blu-ray for at least the next 20 years; an archivist for significantly longer. I don't know what home storage (as opposed to online) will come along to displace the current optical disks, but whatever it is, it will hold more data and will last-- as a purchasable product and the media itself-- for a very long time. So for about $5 I'll be able to consolidate all of my current disks to one (plus I'll make back-ups of course). And the speed of duplicating will increase too.

So cheer up, we'll have pristine prints of Jack Black in Gulliver's Travels forever!

The record for hard drives and digital files in the real world is not nearly as sunny as fwtep suggests, as a cursory look at the AMIA listserve would confirm. Hard drive failures occur in all brands and formats and for a wide variety of reasons. Furthermore, the problem of adapting, updating, and simply paying for software is consuming more and more time and money. Format obsolescence remains a very real threat.

That's even before the migration process begins. And yes, $200 a film doesn't look that bad every five years (which is the more likely cycle than ten years). But say you're a small regional archive with 500 titles, and migration costs quickly become impossible.

The fact that some discs from 1997 that "still work" is not a standard an archivist can accept.

And the migration to new media need only really be done once every 10 years or so. So a film as old as Birth of a Nation would have only had to migrate about 9 times so far.

So we'd have lost it in 1935-- after its commercial value was up, but before MOMA could have saved it.

I'm sure the Klan would have kept some copies - for recruitment purposes or for purely sentimental reasons.

The record for hard drives and digital files in the real world is not nearly as sunny as fwtep suggests, as a cursory look at the AMIA listserve would confirm. Hard drive failures occur in all brands and formats and for a wide variety of reasons. Furthermore, the problem of adapting, updating, and simply paying for software is consuming more and more time and money. Format obsolescence remains a very real threat.

That's even before the migration process begins. And yes, $200 a film doesn't look that bad every five years (which is the more likely cycle than ten years). But say you're a small regional archive with 500 titles, and migration costs quickly become impossible.

The fact that some discs from 1997 that "still work" is not a standard an archivist can accept.

I'm not saying to store them on hard drives. But if you want to talk hard drives, what's the failure rate on hardly-used drives that are stored in optimal conditions, as opposed to the failure rate in the average household? Also, remember that there'd be perhaps hundreds of copies of the material, all identical, so losing one or two drives wouldn't be a problem. Lastly, while drives may fail to the point where the average person can't use them, the vast majority of the time the data is recoverable.

As for small regional archives, doing it in bulk would drive the price down. And the price will only come down anyway. It's about $200 per film today, maybe, but that will go way down. In 1993 the company I was working for spent over $1500 for a 1gb drive. Today you can buy a 2000gb drive for $80. But anyway, I think optical, or whatever replaces it will be what's used. Right now 25gb Blu-ray blanks are about $1 or $2. 50gb ones are more, maybe $10 or $20.

The other thing to think about is that archives could collaborate, since it would be so easy and inexpensive to do so. But anyway, it's still less expensive than film even if you add it up. Let's say the price per migration is $100. In a hundred years that means about $1000 at every ten years, or $2000 for every 5 years. Correct me if I'm wrong, and I very well could be, but isn't $2000 is far less than one or two film prints? And the quality would be better. (Realistically though, 10 years is the more likely.)

I think that what it really boils down to is this though: For better or worse, film is going away. I'm just saying it's not necessarily as bad as some are thinking.

fwtep, I still disagree with you. You say Digital media from decades ago is still readable, and even if true, what would you read it with? I still have some 8" (not 5") floppies and some IOMEGA 10" disc cartridges from the early 1980's, but nowhere that I am aware of where I can get the data transferred to a more current format. Even if I had a working disk drive, the bus architecture is so different now that they would be worthless. Do you really believe that any current digital media will be usable, even if the information is intact, in 60 to 100 years?

Because of DVD/Bluray, many films are unlikely to become "lost" films, but again DVD/Bluray is not capable of the resolution necessary for true preservation. Saying it does is like saying that original oil paintings in museums are unnecessary because we have photographs of the paintings.

Motion picture film has a solid track record of surviving in less than ideal storage conditions for decades. Digital has a very poor track record to date.

As too the cost of copying preservation copies, ultimately cost is a red herring. The real issue is desire on the part of the copyright owners and the ability of archives. If past history is any indicator of future performance, the future looks bleak.

Unfortunately, what you say about film going away is probably true. What everyone interested in our motion picture history needs to be concerned about is making sure that what replaces it does not degrade the work for future generations. Remember that it was just 30 years ago that many people felt VHS was good enough.

You say Digital media from decades ago is still readable, and even if true, what would you read it with?

Well, if you're an archivist you'd store a drive, or even a complete system to read it. Or not, since new formats don't disappear entirely within 10 years (not standards, anyway). And as I've said there are companies whose sole purpose is to get data from obsolete or hard-to-get formats. But that's irrelevant since I'm talking about *planned* archiving, not just saying "Oh, I think I have that on an old drive, let me see if I can find it and get it running."

Basically you seemed to have missed (or misread) all of my posts explaining my points. Here's a couple of clarifications to a couple of your comments:

1) I'm not talking about Blu-ray video for archiving. When I spoke of Blu-ray it was as an optical storage format (data). Of course one thing to remember about Blu-ray as a video system is that it's only 68 pixels short of "movie resolution" as it's been for about the last 20 years. Until recently (and even so, not always) the resolution for post production work (visual effects, etc.) was 2048 by Z (where Z is the height and would depend on the aspect ratio). So what you were seeing up on the screen was, say, 2048x1024. HD is 1920x1080 (again, depending on aspect). The bigger issue with Blu-ray video for archiving is compression. Still, I'll take a nice Blu-ray with its compression over a scratchy, multi-generation dupe any day. Even projecting it. (I have a 120 inch projection screen here and it looks fantastic.)

And all that being said, within 5 years we'll have a much higher density storage format, probably in the multi-terabyte range. So I don't think Blu-ray data disks will be the way to go. (Wasn't TCM one of the test companies for holographic storage a few years ago? I seem to remember reading something about that.)

2) Digital may or may not have a great track record so far, but it's only been fairly recently that people (and companies) have woken up to the fact that long-term stability is necessary. So now there are higher quality materials with longer shelf life. And anyway, with 90% or so of silent films being lost, it's hard to go around saying how safe things are on film. (Yes, I'm aware that a lot of films were destroyed intentionally by studios who thought they were worthless, but many more were lost to problems with the media that the images were stored on. At least Blu-rays aren't made of explosives.

You just proved the financial problems with digital, it will have to be transferred every 5-10 years if it is to be preserved, whereas film has only to have maybe been transferred once in a 100 years, from nitrate to safety.

Um, about $200 or less per film every 10 years is hardly a "financial problem."

Even if this dollar amount is correct; the current holding of just the
Library of Congress now holds is:

The collections of the Motion Picture, Broadcasting and Recorded Sound Division include approximately 150,000 motion picture titles, with several thousand titles added each year. Piece count estimates currently run to 490,000 reels of safety and nitrate film and 162,500 videotapes

information of holdings pulled from this site.

http://www.loc.gov/acq/devpol/colloverviews/mopix.html

So the fast math is $30,000,000 for just one copy to digital let alone every 10 years each time costing more as the collection expands.
Since this is not a financial problem. Fwtep you can just make the check
out to the Library of Congress. I am sure they will feel warm and fuzzy
with that kind of donation.

Ok Enough poking fun. I just do not think you understand the true size
of the problem here, and making archival scans of the film frames and
then maintaining these with the constent upgrades in technology. Film
is film it has not changed in over 100 years. Projectors are super simple
machines. This is why it is a good way to store film. Yes Digital is compact,
but when you figure in the cost and come to grasp the size of the task
you will understand the digital storage is not as simple as it seems.

Remember I just gave the figures for ONE Archive. There are about 90
major archrives around the world. The swells that fun dollar amount
to $2,700,000,000 for a complete digital transfer you perposed. This does
not count the smaller archives that in some cases need preservation
much worse than the top 90 archives.

This is a major job that could take several generation just to make the
first copy, let along maintaining the digital copys and upgrading them.

Pookybear

Then fine, just archive the playback system along with the media and deal with it only every 100 years.

As far as that $200 figure I mentioned, actually, the price would come down significantly, both because of the bulk and because of the media costing less per byte. So it could be down to $10 per movie. Even with today's prices and tech that would get you 500gb at a compression rate of less than half that of a Blu-ray video (which looks great already).

So let's say $20 per movie (two copies at $10 each) every 20 years. In 100 years that's $100 (plus the cost of the telecine, which is roughly equivalent to making a film dupe in cost, so it's a wash). Archives would save money though in that the climate doesn't have to be as tightly controlled as it does for film. With the price of heating and air conditioning these days (and likely so in future days), that's a BIG savings. That savings alone would probably pay for the whole shebang. And also keep in mind that at the 100 year point you just keep copying, whereas with film you'd want to make another expensive and image degrading dupe (if the machinery, chemicals and film stock is even available).

By the way, I think that media would last at least 30 years, so I think the cost would be even less per 100 years. And remember too that once its in a digital format it makes it WAY easier to make it available either for screenings or for sale. Streaming would even make distribution practically free to the studios.

Just imagine if all of the films in all of the archives were in a digital format! Copyrights and quirks of the people running the archives aside, it would make access to those films much easier.

Then fine, just archive the playback system along with the media and deal with it only every 100 years.

As far as that $200 figure I mentioned, actually, the price would come down significantly, both because of the bulk and because of the media costing less per byte. So it could be down to $10 per movie. Even with today's prices and tech that would get you 500gb at a compression rate of less than half that of a Blu-ray video (which looks great already).

So let's say $20 per movie (two copies at $10 each) every 20 years. In 100 years that's $100 (plus the cost of the telecine, which is roughly equivalent to making a film dupe in cost, so it's a wash). Archives would save money though in that the climate doesn't have to be as tightly controlled as it does for film. With the price of heating and air conditioning these days (and likely so in future days), that's a BIG savings. That savings alone would probably pay for the whole shebang. And also keep in mind that at the 100 year point you just keep copying, whereas with film you'd want to make another expensive and image degrading dupe (if the machinery, chemicals and film stock is even available).

By the way, I think that media would last at least 30 years, so I think the cost would be even less per 100 years. And remember too that once its in a digital format it makes it WAY easier to make it available either for screenings or for sale. Streaming would even make distribution practically free to the studios.

Just imagine if all of the films in all of the archives were in a digital format! Copyrights and quirks of the people running the archives aside, it would make access to those films much easier.

Well we seem to have forgotten some of the basics of archiving materials
here. The original is never disposed! This means even if we went with
the 100% digital archive with your new cost saving numbers. One would
still have to maintain and store all those old reels of film.

As You have noted new technologies are being made everyday. So it does
go with saying that the digital copies may not be the best they can
be a present, but still the best we can do right now.

So one needs to keep the original material just in case for this reason
alone. Let alone the fact that the digital copy may fail anyways. Again
keeping the original would be most helpful.

Plus the whole word "compression" would never be used in an archive.
That is the whole point of an archival copy. Just think how good your
Bluray will look when projected to say a height of 30 to 50 feet.

I know one thing film still looks good.

On a different note here and to give you another view of why digital is
not use more lets go to music.

As you know sound waves are a smooth curve with an infinite number
of points along this wave. Now take digital formats. Remember Real Player?

I think the original format was 16 bit; this meant that only 16 points
were taken over one cycle of the wave and then on reproduction
straight lines from point to point were used to fill in the gaps. No
longer a smooth wave form. And not a true copy of the original.

Now just like DVDs and Bluray you could not tell the difference. But
unlike movies in the digital domain music is getting more compressed.
Thankfully film is going the other way.

I think the new for MP3 will be the last format. In order to keep this
file still sounding good, artists have to keep in mind the limitations
of playback when composing new music. So of the older tunes say pre
1990s sounds just aweful in a MP3 file.

So it goes to prove the the originally recoreded records need to be
stored as they are a better copy of nature.

Film on the other hand can still be thought of as a digital format all of
its own. It is call film grain. This grain in effect does the same kind of
bit mapping that is used in music. However, lets just say on a much
larger scale. But new digitals formats compress this down still. So once
again one must keep the most high def version around. And still the
best copies would be made from the original version back to the original
high def material ie Film. Such as a fine grain print.

Digital will be a great tool but still will not be a replacement.

Pookybear

fwtep


Ah yes I was most interested in your floating numbers on the cost
of digital pereservation of film. Just as thought it seems that you
like pulling numbers from the air. Here is the real cost of digital.

"While a theater can purchase a film projector for US$50,000 and expect an average life of 30–40 years, a digital cinema playback system including server/media block/and projector can cost 3–4 times as much, and is at higher risk for component failures and technological obsolescence. Experience with computer-based media systems show that average economic lifetimes are only on the order of 5 years with some units lasting until about 10 years before they are replaced.

Archiving digital material is also turning out to be both tricky and costly. In a 2007 study, the Academy of Motion Picture Arts and Sciences found the cost of storing 4K digital masters to be "enormously higher - 1100% higher - than the cost of storing film masters." Furthermore, digital archiving faces challenges due to the insufficient temporal qualities of today's digital storage: no current media, be it optical discs, magnetic hard drives or digital tape, can reliably store a film for a hundred years, something that properly stored and handled film can do."

Further more the same study put real numbers down for you as well

"a report recently released by the Academy of Motion Picture Arts and Sciences, digital film storage costs $12,510 per year, compared with $1,059 for celluloid. More dramatically, source materials -- those outtakes and audio recordings that often make up bonus content for special edition products -- cost 429 times as much to store, a whopping $208,500 per year for digital materials vs. $486 for film."

I have not found yet the cost of taking film stock to a digital format
but I am sure either of your figures would not cover the costs.

But do not worry you can just write a larger check to the Library of
Congress. Just make sure to add lots more zeros to the original 30,000,000. Nothing like more zeros we all love them.

Pookybear

Digital media is such a poor long-term storage media, that digital files are probably going to be stored in an analog medium in the future. A few months ago I read that a group of computer scientists are developing an analog method to store digital files. Their idea is to print out something like barcodes, and then lock 'em in a closet and forget about them. The most amusing part of the article was when they said they were considering microfiche, since it has a proven track record of lasting!

For a start, no-one knows how long digital data can be preserved.

I think this is a misunderstanding of the digital medium. When you're talking about film, you need to preserve the physical object because the picture and the substrate (film) it's printed on are inseparable. But digital can be "printed" (stored) on any media device that's large enough to hold the file. So, while DVDs, hard disks, and Flash drives certainly won't last forever, it's not a big issue to move their contents to some new media device before the old device fails or random cosmic rays somehow corrupt the data.

I think you're misunderstanding the digital medium. When film is about to decompose, there are all sorts of warning signs that tell the archive that they have a clock running down on their opportunity to rescue the information. With digital media, the first time you know there's a problem is when it's failed. And when it's failed, you are going to be extremely fortunate to rescue anything. Rumour has it - from impeccable sources - that the DVD versions currently available of some very high profile, award winning, computer animation features are not the originals, but partial remakes; when they went back to recover the data for the domestic releases, great chunks were inaccessible and had to be remade at huge cost. It was only because it wasn't live action that this could be done at all....

I think you're misunderstanding the digital medium. When film is about to decompose, there are all sorts of warning signs that tell the archive that they have a clock running down on their opportunity to rescue the information. With digital media, the first time you know there's a problem is when it's failed. And when it's failed, you are going to be extremely fortunate to rescue anything. Rumour has it - from impeccable sources - that the DVD versions currently available of some very high profile, award winning, computer animation features are not the originals, but partial remakes; when they went back to recover the data for the domestic releases, great chunks were inaccessible and had to be remade at huge cost. It was only because it wasn't live action that this could be done at all....

Now this is an interesting statment.

It does make me wonder at least what one of those effected title was
called?

Pookybear