Media Formats Update Panel

    Mona Jimenez and Steina Vasulka
                  Mona Jimenes and Steina Vaulkka          Steina and Mona

JIMINEZ:  When I first was starting preparing for this, I thought of doing a definitive format discussion: what the sizes are, their composite, component, and all the technical details. I turned myself around, because that is now something that drives me a little crazy about media formats and I can only take it for so long. If you go to AMIA, the Association of Moving Image Archivists conference or events like that, there are a lot of projections about what will happen and the details about various formats. What I wanted to do, actually, is come up with five areas I think might be useful to look at when considering media formats, and then look at the various media formats and apply this to see what happens. I will mainly talk about analog to analog, and analog to digital. The digital to digital is another day; I will mention it, but I feel that is a big can of worms, and what we are really talking about is older work that is not being moved forward. That is my main emphasis. I won’t have much to say about audio, but hopefully somebody in the audience will.

Tape is storage medium for the signal; that is what it is, whether it’s a digital signal coming out of a tape, or whether it’s the analog signal. As far as I’m concerned, that is the bottom line. I came up with these five things to consider, and they are the things that I have come upon. Though I do not claim to be an archivist or a conservator in the traditional sense, I have listened to a lot of them so this is informed by folks who have brought these things to my attention.

In terms of a format, traditionally one would think of a format - an archival format - as lasting for over a hundred years, but obviously that is not where we are at. It is a process of reformatting; we don’t know how long, maybe every ten, fifteen, or twenty years. The first thing I will mention is tape durability: in terms of the material the tape is made out of, the size of the tape and the amount of information that can fit on the tape. That is the way I would think of tape durability, size of the material. For instance, format like Beta SP or a larger format is usually made out of a better, more durable material, a thicker substrate. You have your base, you have some kind of binder, and you have the coating--the oxide-- that the information is stored on. We have not spoken yet about degradation, what happens to tapes. The problem is the binder breaks down and the information falls off, or gets gummed; the binder basically gets “gummed up”. This is much of the problem. When you have a smaller tape, it is more likely to stretch. It is not made out of the same material. If you remember, the metal evaporate was a real big problem with high-8 tapes. That has to do with the actual corrosion of the binder. So this stuff is one place to look at.

I was interested to read recently that they say there’s no binder with mini-DV, and that the information is imbedded in the base film. Or rather, the medium that takes the information is imbedded in the base film. That is actually a shift. That is the thing about some of these materials; they keep changing, and people who want to get into the study of that material [have trouble] because things are changing and it happens so fast.

The second thought I had was of the deck durability or the construction of the deck: the heads, the electronics of the deck, and how all of this stuff holds up. Some people feel that there’s more that can go wrong with the smaller tape paths, that they are basically less stable and there are fewer people who will work on them; that’s a problem with smaller ones. There is obviously a lot of expertise as to which of the decks really hold up well and which require a lot of maintenance. Luke helped me a lot in the last few days, humoring me by talking through all of this stuff. One of the things he was saying was that the digital Beta decks, at least in the beginning of when BAVC had them, required something like a five thousand dollar head change each year. In addition to the cost of the decks, which are anywhere from, used, I guess, in the thirties to fifty thousand dollars, you have an additional huge repair bill. That may have changed as the decks have been out for a while, but it is an issue.


COMMENT:  It’s gone up, Mona.


JIMINEZ:  Oh! It’s gone up… So the issue is which ones are reliable? How can they handle the stuff on the tapes, the orange peel or whatever it is Kim was talking about? How do they handle the dirt they come across? Do tapes get stuck in them? Questions like these.

Then the characteristics of the signal, and the ability to carry the signal forward; the archivists and conservators say you want to carry forward the maximum amount of information. Obviously, with some of the digital formats, there are issues around compression and the compression rate: how much information might be “lost” in the transfer, the sampling rate of that particular format.

I want to mention the difference between tape and data, because I think there is a lot of confusion. You can’t really just come off a tape and go straight to a hard drive without a conversion process. What you have on a tape is not data that is on a hard drive; there’s a difference. If you come out of a tape deck and it’s converted to data, the data goes out and the other deck reconverts it and then puts it back on the tape. Sometimes there is confusion when we talk about analog to digital stuff; “Can’t we just put everything on hard drives?” We can certainly discuss this, but it doesn’t happen that simply. There are issues concerning standards, when the information comes off the tape and becomes data, and then goes from data back on to the tape, or goes from one digital tape to another digital tape. We can address these details later. That is another thing that people are looking at in terms of getting and moving that signal around. You also have the issues around error rates and the decoding. Yet another thing to look at is whether the signal is component or composite--component being the signal that is separated, the luminance from the chrominance. An SVHS deck or a Beta would be a component. There are some composite formats, three-quarter inch and others like some from the D family; I always have to look at my chart for that, but it’s like D2, D3, D3, and D2. That is another thing to think about: what we want to do, and what we need to do to make it happen?

Then there is the deck obsolescence. We don’t have the 8650 anymore, so this relates to is the deck still being made? Is it still being serviced? Are there parts? Are there people that are knowledgeable that can fix the decks? There is all this stuff going on with Beta SP. Some cameras are going out— they have stopped producing the cameras and certain decks. What they have said (I don’t know if you can trust it) is that Sony will make service available for seven years after they drop a product. I don’t know about Panasonic or JVC or what is happening with that. Traditionally, we’ve been in this industrial or prosumer line, as opposed to the broadcast line. For instance, if you go outside San Francisco and other places like New York, there aren’t too many media arts centers, libraries, or museums with digital BetaCam. In the media arts community we tend to be working in that industrial and prosumer line, as opposed to the highest quality broadcast. Obviously, these products are different, between what’s available at the industrial and prosumer as opposed to what’s available at the other levels.

Broadcasters and large clients really drive the market, but this is nothing new and we’ve know this twenty-five or thirty years. The three major manufacturers are Sony, Panasonic, and JVC.  If you get into the newer prosumer gear, then you’re dealing with companies like Canon, but mainly it’s the big ones. We are in this changeover for broadcast, at least in the public television realm, from Beta to digital Beta. I don’t really know what is happening with all these broadcasters; maybe other people do.  But there does seem to be a changeover and a commitment to digital Beta, at least for a while.

In terms of going from uncompressed to a server, I think only CNN and a couple places have started to think about going uncompressed to a server. It’s really just for them to store information that they’re going to dump back into an Avid and re-edit. It really is not for long term storage but for media use. It’s like a network with the data; not the tape, the data. The data can get pulled back to go into the Avid and be re-edited. So in terms of “Can’t we just put it on a server somewhere?” Again, I imagine broadcasters are moving towards being able to deliver and store digitally, but that’s a totally different mindset from what we are trying to do.

One of the positive aspects is backward compatibility and playback. I am aware of the Sony J series, that playbacks Beta, Beta SP, digital Beta, and the newer, compressed format that is around. We have been dealing with obsolescence for a while, and the cost relationship of resources to production. Can we afford to buy a deck? Can we afford to buy the tape? Is it more than two hundred dollars a tape if we go with this certain format? What will it really cost us? Again, in different regions there are different resources available for transfer. I remember being at Visual Studies Workshop, some years ago, and somebody offered me a Beta SP to shoot on, and I was like: “I live in Rochester; I can’t edit on Beta SP; just give me a good three-quarter Portapac or whatever!” It is going to really differ, in terms of the resources that are available.

We have talked about this relationship, at least with the media arts centers and individual producers, between production and preservation. This needs more exploration. I’ve always imagined there being more preservation done in media arts centers and I think that it is at media arts centers in addition to our great BAVC. But it has been haphazard so far and I wonder about centers that are focused on production working on preservation or focused on distribution. Steina brought up the fact that using DvCam makes a lot of sense for a person who’s producing and using these tools. This obviously has to be weighed out. What [available tools] do we have? I feel like we need to collaborate a lot more. There are some partners here that are not traditionally [linked] with media arts crowd, which is great; libraries, university media centers, museums, public television, municipal archives, and a whole host of people in the region.  I was just at the Grand Rapids Community Media Center, which is going to become part of this municipal archive-- city, county, and the museum. In terms of storage I don’t know, but in terms of duplication, I think it makes a lot of sense. When I was at Visual Studies Workshop in 1990, I first started to pull out those decks again. Just for me. Now it’s 2002, twelve years later, and how many more opportunities are there for me to transfer that tape? If we go this slow, we’re going to be in trouble! I think we have to be realistic about our resources.


Now let’s take a look at some of the formats and just sort of think. There’s no one right way to do things, which is great, because heaven forbid! There will be a range of solutions in different places. I included SVHS as an analog format, because some people ask about it. I’ll be answering questions about these, but first let’s just consider what they mean. These (Beta SP and SVHS) are two are component, or analog formats. Beta SP may be a stronger, more durable tape stock. There are plenty of decks, I would say, of both out there, probably more SVHS decks than anything because they’re making a million a day or some outrageous number.

I think we may have more expertise, in terms of repair, with the analog decks than we have with digital. I think we have a broader knowledge base with that. They are certainly more affordable than some of these others. Except for the digital— or I should say prosumer digital formats. We probably have a sense of which ones are good, and which don’t work as well.

One of these formats, I think it’s Beta SX, people are saying was produced at the request of CNN.  It’s a half-inch Beta-type and a very compressed format, so it would probably make sense regarding acquisition. In terms of going toward digital delivery, I don’t know what they’re thinking, but it is more compressed. I’m not sure about digital S, but definitely digital or Beta SX.

Then there is the D family. These decks are, I think, in the thirties used, fifty thousand or so new.   As I said before, with the D family there is component and composite; some are compressed, some are uncompressed. The one people are drawn to the most is D-1. This is an uncompressed component format. I actually don’t know the exact size of this tape (maybe three quarter) but they are huge. They are expensive; with these decks, you’re looking at an eighty thousand dollar price tag. Some people have gone to that when they’ve done remastering to Beta SP and then to D-1, think that that, you know, was an uncompressed component format, and as an archival master that’s tucked away. I don’t think you’ll find the same market penetration with the D family, and so you won’t find as many decks. Again, this affects the future of being able to retrieve the signal.

Then there is the DV family, or DVC Pro, which I think is a six millimeter format. Some of the things I mentioned about tape durability and the tape path size would apply to that — although the price is really great and it’s compressed.

For our comparison, I didn’t want to talk too much about optical, but I just want to say a couple things. Everybody asks me, “Why can’t I just put it all on DVD?” I do not think it is a good idea for a number of reasons. First, we are still in the midst of the DVD wars. Secondly, it’s highly compressed. Third, you can’t fit much on it. And fourth, I don’t think it’s going to be around long. That is my reading, but as you know, things are changing and who knows what will happen. The other thing about optical media in general is there has not been a lot of testing of the material, and there are frequent changes in terms of what the CDs and DVDs are made of. They think they are quite affected by pollution and other chemical changes — for instance, writing on them with a Sharpee. It is a whole different thing; it is a substrate, and there is dye and glue involved. To be technical, there is glue and there are other chemicals on there. There has not been a  lot of information about their durability. They say they last for a hundred years but again, they’re fragile in other ways, so I think it’s something to keep an eye on.


Just in the last few days, thinking about coming here, I’ve just been wondering more and more why don’t we want to use the Beta SP format, and for a number of different reasons. One is that it’s an analog format. I know there are problems with that, such as the generational loss. But that’s what we’ve been dealing with. We like it better when it doesn’t happen, but I feel like I’m being sucked along in the whole digital thing. I go to these conferences, and there are tons of predictions and information about what’s going to happen. Often it’s talking about media-less archives or storing things on servers. I don’t think they are very applicable. But the issues around the tape are really hard to track. Not that I necessarily think that things are going to settle down, but I wonder about going to a digital format if it’s not needed. Partly because of the cost, partly because of feeling sucked along and that we can’t really depend on any of these being around very long; wondering what’s going to happen, and hearing a lot of predictions.

I’m also feeling like we have a lot of people who can actually help— that we can really help each other understand the mechanics of these decks, the maintenance and repairs of these decks. We can get parts; we have some time to gather things. I’m not talking about forever. But why can’t we pull that analog signal off later, and not worry about pulling it off now? I just wonder about this. Regionally there are very few places that have digital decks available with the bigger tape size, the bigger tape path, and all those issues. But I don’t know about what the predictions are about how quickly that format’s going to change with DvCam, mini-DV and DVC Pro. So now I’m going to turn it over to Steina.


STEINA VASULKA:  I feel differently about this analog and digital thing. Once it is in digital form, the formats become very irrelevant. The compression is still very relevant. I take tapes I’ve made, I play them on a lot of other formats, and they play back for me. But if I take an hour of recording on this deck and go to a DvCam, it records only forty-five minutes because it is a different speed. When I go to the DvCam and put the tape in here, I can hear my noises from the recorder. Then it goes to the speed of that format it was recorded on. We also have a multiple playback; besides this SP and LP--there are at least three speeds built into this camcorder and you only know when you go from one to another. I was in Iceland, my native country, and they had this damn camcorder that always shut off-- there was no way to keep it on. I just stuck a tape into it, and it taped always for an hour; then we rewound the tape, because we needed a live feed and that was the only way to do it. I take this tape home, I stick it in here, and although that’s a Panasonic PAL, it plays back on my Sony and TLC. We have come to a point where this paranoia of VHS and SVHS, component and composite, all those things are rapidly becoming irrelevant. Obviously there are different size formats now. There is the long one, and Hollywood is going to come up with gazillions of them I’m sure. But as long as they all playback, they are inter-compatible and it’s not that important. This is just my opinion. It is that it is important to get the old, and I don’t know how important it really is to get the old stuff off right now, or if you can wait another twenty or thirty years. I don’t see the deterioration going like that. (Yesterday people were talking about mold, and that scared the hell out of me!) But otherwise, can’t they just sit on the shelf a little longer, until this format war calms down a bit?

I used to be very apprehensive about compression and especially these ideas that you compress more or less, according to the information on a particular frame. I have worked with the Sony codec for a while now, and I’m very impressed with it. At least if you are coming from anything that is black and white, it’s plenty. You never see any deterioration or any difference, really. I’m sure there are fanatics, and I might become one of them myself, because remember the Lionel people — who are still around? I mean, if a musical recording isn’t a Lionel, it’s no good. [They consider] CD stuff not as good, but thank God for the CD; it has change our whole culture. When we were battling with this scratchy records, dust on them, fingerprints and everything--and they were short. You could only play what, twenty, thirty minutes? See what has happened to us as people since this CD audio? Not that anything is available, and maybe there are plenty of rewards for a lower quality. But I don’t know if it’s true, and it’s the same with this. Once it is digital, is it so retrievable. We don’t really have to worry about the material it sits on, because every time we change it to another material, we are transferring over all of the information--every zero and one that is on it. It is not a battle about degeneration anymore. “I’m sending you a third generation tape--please keep it in mind;” you don’t have to say that anymore.

I want to talk about things that I do not know, and that is how this is encoded and decoded. It is a mystery even to my genius friends; there is an ancient Japanese secret and they will not let us know how they do it. We know it is encoded, not like a film, which is like a frame— or even video, which is scanned. After it is scanned, there is a black, and then the next scan. We understand this, but we also don’t quite understand this. I’m especially intrigued by the decoding. If you have done your edit and put a black on the end, and if you are fooling around with this last frame of video and black, you get what you see on the black, some blocks of the last frame. Suddenly you understand what you are getting is partially the last frame of video and partially this black. That is because when it decodes, it just take information and this is your field. Then comes the next one, and this is your field. This is why we don’t see analog-type dropouts on digital. If you take the information directly into digital and then play it back, it is restored (to a large extent) because the dropouts that you looked at going in don’t come back to you because there’s so much intelligence in the retrieval process. Those little things, they decide to skip and not show you the drops, but they take some information from the previous or next frame and put that in there instead.

We were talking about methods to restore and retrieve. My method is restore, retrieve, and restore. It is to save my material to digital, which I’m some sort of half way through. I must say my biggest stumbling blocks are the playbacks, these old tape recorders break down all the time. I have, like, ten of them, so I throw this one out and take the next one; it is very problematic to get the information onto a digital with no signal restoration. But it can then linger in this digital forever, or until you need it, as far as I’m concerned. That is when you go into this second part of the restoration, which is to set your black and white levels which Heather discussed. If it is not already a fact, there will be software that does this. I know there is already some software that resides in those editing programs, but it’s not good enough. BAVC discussed how they have somebody physically look at the tape and take down the numbers where somebody flipped the light switch and the whole situation changed. This is what computers are so good at; it holds it at certain levels, until everything changes, and then instead of a human having to change it, the computer can change it. I haven’t heard of software that is specifically written to restore this, but it will be there soon, without question. I was apprehensive about signal processing in digital, because it was clumsy and expensive. It was hard to throw in a whole hour into the computer, but it isn’t anymore; it’s a piece of cake. You throw in several hours of digital recordings into the computer, and then you just let the computer work it out. I think once it gets to a digital form, we have plenty of time to think about the next step, the second part of the restoration. As I say, it might linger there forever or we might need it today to look at some visual materials. I want to have it restored right now. So the priority goes according to need. The most important aspect is always the retrieval. We have gotten along with linear, instant retrieval because, let’s face it, tape is a drag. It is ten minutes in, play for three minutes, ten minutes in. But instead, we just go on the keyboard and “Here it is.” Or with whatever other means to access the information. It is all there and instantly accessible.

That is one part of retrieval. The other is the cultural retrieval. I think we all know that we are at an unbelievable cultural revolution and for a lot of us, it’s already old hat. But for others, we are just getting into it--and that’s the web, and the very idea that you can access all human knowledge on the web. It is so preposterous that it’s hard to adjust. I have taken a long time to adjust to this idea that it is all accessible. We now have website that maybe will also become obsolete; we don’t know how this is going to go. I got a little depressed when we were talking about retrieving a high school drama, but then I thought: God, if I could see a high school drama that they did in the times of Leonardo or Aristotle, it would be so fabulous, you know? It costs so little to restore all of this. In bits and bytes, it is a shrinking field and it is getting down to nothing. 


SELSLEY:  A question is if we see signal deterioration after we clean the tape?

Well, that is difficult to answer, because when we get a twenty or thirty year old tape, we don’t know because we can’t really do a before and after. Some tapes have a little more deterioration than others, but I can’t objectively say.


VASULKA:  I do the cleaning physically, because I don’t have a large stock, like you do. I have maybe three or four hundred tapes I have to do. So I just put a cotton glove and I run it through one of those obsolete machines, that don’t play back anymore, and clean it up. I get this incredible goo; it is black and ugly, sometimes oily and sometimes dry. You can see all the stuff that has fallen into the recorder when you clean the path; there is a lot of stuff that comes off. But in spite of all the stuff that comes off, once the tape plays, it plays perfectly. It plays like the day I first saw it. But I’m not sure.


LUKE HONES: From our experience early on, we decided not to transfer tapes without cleaning because we just had so many bad experiences with heads clogging and tapes sticking in the tape path. We did do a test where we did one pass through twenty passes, all in one day, and then compared them. In our online suite, we did a split screen between the first pass and the twentieth pass; looking at the scopes, there was no real difference between the two.


VASULKA:  The other thing is about speed of deterioration. Do you think it is important to save them now, or can they sit in those boxes for a few more years?



HEATHER WEAVER:  I think it depends a lot on how they’re stored. As you’ve said, you’re in an arid climate and perhaps you aren’t having as many problems with mold. From what we’re seeing humidity seems to have a big effect. Jon: since you first started doing preservation, have you noticed if things are getting dirtier and taking longer to clean?


JON SELSLEY:  Yes, Heather. When I first started seven years ago, it seemed that on average, tapes took two to three passes to clean. Now the average seems to have gone up to four or five passes before a tape is clean. That’s just a personal observation.


HONES:  I think part of this discussion, and some of the exciting information Bill Etra brought in, will help us start to think about other routes besides the cleaning route— or besides what we started out with. Basically, the process we’re using now is the same process we started out with.


ETRA:  There are several things I’d like to make clear. My first videotape has disintegrated, probably because it wasn’t stored right. The only thing I have left from that era is something that the Vasulka’s still have, which I shot with an Orecon onto film. That is the only piece that still exists from two years of early works during the late sixties. Regarding people’s worries about the quality or the compression in the digital realm (something I know a lot about); there are two things you must remember. One is the bandwidth is extremely low on CV and AV format, half-inch reel-to-reel tapes. There is nothing terrible that is done on the analog side when you’re digitizing it; you don’t get to the point at which you’re losing information on any of the digitization of the half-inch tapes. I happen to know this, because at the time I was looking at digital information, most of the three-quarter inch cassette decks had a cutoff at 2.8 megahertz. Again, for color or black and white, there is nothing--if the digitization process is done properly-- that would lose information from those tapes. Even the Sony one-inch reel-to-reel, in the days they were first broadcast with those big, standup machines that were thirty thousand dollars, had a cutoff at about 3.4 or 3.5Megahertz. It was all under 4 megahertz. All the compression methods will now deal with that; I cannot believe anyone could tell if there is any signal loss.

In addition, when you retrieve the information from the digital domain in the compression methods, there are different methods of expanding the area you have to work with: like going to a higher bit depth when you reconstitute them, which allows you when you’re correcting it to play with--instead of a 24-bitX signal--a signal with more room so you can correct the gamma and the color path. There are all sorts of techniques. Lucas did some of it on the computer effects for this last Star Wars, when he bumped them up before they actually went out filming. Without getting too technical, there are techniques in existence that allow for all those early tapes to be put into digital with virtually no loss from compression.


COMMENT:  There is a certain paradox when talking about early tapes. When you take the brown coated CV tapes, you don’t need to clean them. In our little paradise, it plays every day without cleaning. There were these upgrades, like the black coated tape by Sony, which don’t play. Steina does it by rolling them; she adjusts the deck and rolls them horizontally, until they dry out--maybe ten or twenty times. Then she goes through and collects the glue or whatever. They eventually play. For the professionals, this is no way of doing it, because they have to make money. If you had allow for it and there is no other choice, you can do it by hand and eventually you will get every goddamn sucker to play! Then there are intermittent programs, like in U-matic, where we have no problem. There are certain types of metals that are on the head that repel this goo or dust. Others freeze immediately. So actually, there are a number of irregularities. We are battling it in a pragmatic way, pushing the most bedeviled tapes into the corner. Like when Etra was looking at Sony stuff, you just go directly with the tape and extract the information digitally as a map and then that goes to a computer where it remaps it into whatever format there is, without going around the head. Because you can’t play a tape until it goes around the head.   That means they’re on the drum, and that’s what we are talking about, post and drums. So we are stuck in the same level as anybody. But once we get over that, of course, there’s no discussion left.


PIP LAURENSON:  This has brought up a lot of different issues. I think we need to have a different discussion about the involved risks. It’s one thing to be able to create miracles. There are people who have spent a lot of time with this material in the field, and who are able to do miraculous things, almost bring things back from the dead! But there is a point that most of the people dealing with this material should be minimizing the amount of change that is introduced in terms of the signal, and should have an accurate transfer of material. It’s very difficult for most people to make judgments about what change is acceptable as you go through all these different possibilities; maybe they aren’t the author of that material. It’s very hard to get good criteria for success when you’re doing all of this. Therefore, I would argue for a much less risky strategy, and to be much more cautious around issues dealing with compression.

The other thing is when we talk about “forever,” I think we really need to know the risks, in terms of how long that is. Your tapes are still on a very, physically, vulnerable material. I think we have to be careful with these sorts of systems, and that is by assessing your risks. You have to be realistic about what you’re committing yourself to, in terms of regularly, systematically transferring that material when the stock deteriorates. The one-shot wonder of being able to transfer onto a format and then not worry about it is really dangerous. You have to make an ongoing commitment about the program, in order to take into account the risks of the formats you’re transferring onto.


JIMINEZ:  I think it is really different for an archive that handles materials for individual producers. I think that is what is great about having both of us here. I have my own stuff, mainly a lot of community documentation, in the hallway at home and God knows if it plays! But I also work with a lot of archives, and I wouldn’t really recommend they go to DvCam. As I was saying about the Grand Rapids Community Media Center, they are mastering on DvCam now. It’s a different story for the digitally born stuff, but I think that’s another discussion. I said to them, “Look, just do two masters when you do it, and put one away someplace safe. At least do that as a preventive measure.” We are dealing with the same old thing and we are going to have the same old problems, because lots of people are mastering to DvCam now. It’s just one of those things


Looking Back/Looking Forward: A Symposium on Electronic Media Preservation

May 31 – June 1, 2002

Media Formats Update    1