Issues in Physical Preservation Panel

    Luke Hones (Artists Television Access), Jon Selsey (Bay Area Video Coalition), Heather Weaver (BAVC) and Kacey Koeberer (BAVC)

            Kacey Koeberer  

          Jon Selsey   

Luke Hones and Jon Selsey



SHERRY MILLER HOCKING:  I just want to welcome you, first of all, and thank you all for coming. We really appreciate the patience you’ve shown as we’ve tried to reschedule this meeting, originally scheduled to be held in Buffalo last fall. We rescheduled, of course, because of September eleventh. So thank you for your understanding.

I want to thank our funders, the New York State Council on the Arts, specifically Debby Silverfine, who is the Deputy Director of Visual Arts at the Council; also, Karen Helmerson, with us today, who is the director of the Electronic Media and Film Program there. I want to thank them not only because of their support for this particular symposium, but also because of their support over the years for preservation activities. I think they have shown remarkable foresight, figuring out early on that this was something that needed to be addressed. I also want to thank our supporters and our partners, IMAP, Independent Media Arts Preservation, under the direction now of Dara Meyers-Kingsley, and their steering committee and board. IMAP has agreed to help us to put up the results of this symposium on the TV Center’s History Website. I thank BAVC for not only their support for this symposium, but their assistance with the development by Luke Hones of “Reel to Real”, and which is also up on the Video History Website, under the preservation area.    Thanks to Mercer Media and Bill Seery, who will host a small reception after today’s proceedings. I want to give a big thank-you to David Jones, from Dave Jones Design, who has helped me enormously with the website, and will help me again to get the transcripts up on the website.  Thanks to DCTV for hosting this event, and to Mona and to Luke for all their work.  So thanks again for coming. Catherine Martinez, from DCTV, will now say a few words.


CATHERINE MARTINEZ:  Yeah, I just wanted to welcome everyone to DCTV [and] specifically to our cyberstudio here. We’ll be taping much of the event, I understand. We also have a cyber-studio show every Monday night, so if anyone is in the neighborhood and wants to come down and see how the whole studio works — there’s an interactive component, performance, and a documentary. If you’re interested in the programs at DCTV, we have our new issue of “Scanlines” out and I wanted you all to take a copy before you leave. Thanks.


MONA JIMENEZ: I was just talking to Kate Horsfield, and we were reminiscing about how we were in New York a couple years ago and we said, “We’ve got to figure out how to get more tapes remastered. How BAVC did it, and how Luke set up that remastering center in the beginning?” And we said, “Well, you’d need to get together around a couple decks and talk about it.” So that is how this started, a conversation. Then as soon as I spoke with Sherry, we had a full-fledged program! So we’ve expanded beyond that first “let’s have a discussion around an AV deck.”

As a result, the spirit of this symposium is really a working group as opposed to those of us who happen to be at the front of the room giving information to those of you who happen to be sitting over here. Basically, what we’re trying to do is not just hear and learn about, for instance, the BAVC model, but to come out of here with some idea about what needs to be done next: what research needs to be done? What questions need to be answered? What projects could be started to help move this whole issue along--to get more work remastered, highest quality, lowest cost? Although we’re focusing mostly on analog video tomorrow we’ll be talking about Capturing Related Histories. We wish we could be talking more about audio today, too, because that’s a big component. So I would encourage you to think broadly, even though we’re mostly focusing on analog video today.  At the end of every part of the program, we will try to ask, “What out of this can we capture, in terms of what we need to do research about, or what questions we need to answer?” and then come up with some pretty specific next steps. In addition, there are many different levels of technical expertise in this room, and you may need to rely on each other to clarify things that are said--as opposed to us continually going over the minutiae about basic technical questions.


LUKE HONES:  Good afternoon, my name is Luke Hones. I’m from out in San Francisco. I work for Artists’ Television Access now but at the time that I was working with Bay Area Video Coalition. As Jon and Heather and I were talking about this session back in San Francisco, one of the images we came up with is when we discuss what it is we did, it’s sort of like describing to someone what it’s like going to the moon. It’s a very detailed process, and it takes a lot of perseverance to set up the kind of center we did. But it’s also like you have the background of the Wright brothers. In some ways, we were always searching for more information--for some way of learning more about what it was we were doing. When I got into video, three-quarter inch U-matic was the format that everyone was using. I never saw a half-inch open reel deck until 1994, when I started doing transfers. The first transfers we did were test transfers; so we learned the systems before we started doing transfers for clients. The summer of 1994 was learning about half-inch open reel for us. We tried to find some good literature from the period when half-inch open reels were the current technology. In my paper, I mention one that I think was particularly useful to me, and that’s by Charles Bensinger. It’s called The Video Guide.  We also got some information from NASA on different cleaning machines. An issue that was being dealt with in another discipline was all of the data tapes that people were using to keep data had to be cleaned, so we looked into what people used to exercise and maintain their data tapes. We talked to everyone we could who used to work with half-inch open reel, and a lot of our first contact was with the folks at Video Ant Farm, with Chip Lord and Star Sutherland --basically, all the collectives around San Francisco during the seventies. After gathering that information, we felt it was something we could do, and we had to take a chance. So we started looking into it, asking around to different funders to see if they’d be interested.

What I’m going to talk about is not only our cleaning process, but how we came about deciding what cleaning process we would use. It was the first one we chose to use because it seemed like it would offer the most success for the most number of tapes. But at the same time, part of us would love to experiment with other processes. If you’ve visited the website, this will sound familiar. The first cleaning process for any tape, of course, is no cleaning. Any time you have an opportunity not to run a tape through a tape path and just transfer it that would be fantastic. But that’s not really a choice we had with half-inch open reel. Any time we would try and run a tape that hadn’t been cleaned yet, either it would stick in the machine — and the machines are very hard to come by, and very hard to repair — or it would start playing and then basically, material off of the tape would start collecting on the heads; and eventually, all you would see is snow, you wouldn’t see the image anymore. So for us, not cleaning the half-inch open reels didn’t seem like a viable alternative.

I got a call one day from a fellow from Emerson University. Somehow, he’d gotten my name, and he was describing how he set up a reel in his bathtub and was running it through some solution, and that’s how he was cleaning his tapes. He said he’d had some success, but he couldn’t really play the tapes back after one time. I heard about other people who were doing sort of a washing process. We didn’t try that, just because it seemed a little risky. It sounds like there may have been some folks in New York who were using different washing processes, and it would be great to hear about those in more detail.

The other thing that we heard about from engineers at 3M and from engineers at Ampex, as well as from folks who were doing audio preservation, is the idea of baking tapes. When I talked to the folks at 3M and Ampex, their idea was they put it in an oven at, like, a hundred and twenty degrees, or some low temperature. I think Ampex would put it in for three days, and 3M would only put it in for eight or ten hours. Then they’d try transferring the tapes. They said that was pretty successful for them — except they also said that they couldn’t play the tapes back again. They could only play them once. So we veered away from that idea.

The one we decided to go with was one documented in BAVC’s magazine of the seventies and eighties called “Video Networks.” We had a professor at San Francisco Art Instituted, Sharon Grace, contact a company called Recortec. Recortec created the data cleaning machines, and she had worked with them to modify a machine so it would work with half-inch open reel videotapes. She’d clean the tapes, and she had used the copies of those tapes for a presentation that she did at the Art Institute. With that idea, and because Recortec was so close, we contacted them and started doing tests with their organization, using this cleaning machine.


HONES:  The machine has two reels, the source reel and the takeup reel. The first module that it goes through are these two little spools, like thread spools, expect what they have on them is like cleaning tape.  Actually, I think there’re four spools that the tape passes through. While the tape is passing through them, the two sides of the tape rub up against the cleaning tape. The dirt might collect there, but these spools actually continue to spin during the whole process. And so essentially, the tape is always touching against a clean piece of cleaning tape.

Next there’s a burnishing sapphire — a lot of times they call it a sapphire blade — that the tape passes over and burnishes. What I think is the coolest part of this system is there’s this big old chamber that the tape goes into, and it kind of hangs down as it’s coming off of these reels. What’s different about this cleaning machine versus a videotape machine is on a videotape machine, the takeup reel is dragging the tape over the tape path. On this cleaning machine the reels are spinning independently — they’re going the same speed. What this vacuum chamber does is it makes sure that, though there are some variations in the speed, the tape that bounces up and down in here and stays even. For the most part, there’s very little tension in this tape path. It’s funny; I saw this machine working for probably a year, year and a half, until I got a chance to actually clean on it. That’s when I realized how cool this chamber was.

On the other side of the chamber are these slotted grids.  I guess the best way to think about them is it’s like an electric razor, and there’s a vacuum behind it. And I know it’s probably a little kinder to tapes than an electric razor would be, but similar to that. Then any debris that may come off in this area is basically sucked out by a vacuum that’s behind those modules. I like that there’s vacuuming going on through here; but I love that there’s no tension. I like that these cleaning tapes are constantly moving, so the grit doesn’t build up. We decided to go with the Recortec machine, because they were so handy to us; they were right down the street. I’d also read, in the NASA book, about BOW Industries doing cleaning machines, and at least one other person was documented. But Recortec was one of their preferred providers as well, so it seemed like a good bet.

When we first started working with them, Recortec was going through a business transition, and we were fortunate enough to have a couple of engineers from Recortec who did our cleaning for probably the first year or two. So we had some real experts working on those machines. When it came time for us to buy the machine, they also put our staff through a lot of training so they would know how to use the equipment correctly.


JONATHAN SELSLEY:  I’d like to say a couple things about the cleaning machine. I’m Jonathan Selsley; I’m the technician at BAVC that actually operates this machine. I sort of took it over from Luke. I want to point out that there really is, first off, no tension on the tape. But also, it’s a non-invasive process. On the cleaning tape, it’s just a slight offset; it’s not really pushing on the tape at all. Right next to the right of the chamber, the speed is controlled by a fixed post.  When you have on a feed reel a whole lot of tape, and on a takeup reel no tape, if the speed is relative to the takeup reel or the feed reel, you’re going to have a different speed on each end. Like, when you’re rewinding your VHS. The speed is actually sensed by a fixed post that has a fixed diameter. See, if the diameter on either side is smaller, or larger, you’re going to have two different speeds. So we want to keep the tape at a constant speed as it passes through the cleaning tapes and as it passes over these vacuums here. So the speed is on a fixed post. There is one item that we removed, which is a box that kind of has a straight blade that goes down also, that acts as a scraper. We removed that, because we were finding a lot of edge damage, a lot of creases, wrinkles, tears, slices and these sorts of things that I really hate to remove during the cleaning process.


HONES:  That is worth saying. When we first set up the system with the folks from Recortec, they had to take out a number of modules that would normally go on these systems. The data tapes were sitting in clean rooms and only trusted engineers touched them, but the half-inch open reel tapes are coming out of public access channels and out of a variety of places where they’ve been played over and over again, so sometimes there was edge damage. Some of the modules wouldn’t allow for the amount of-essentially-usage that the videotapes had gone through. This is the process we started with, and this is the process we still use.

When we sit around and get misty-eyed, we talk about how it would be great to have test tapes and have the time to be able to try out baking some tapes to see what exactly that process is like—to see how we could refine that, or if that would play a role in some of the more difficult tapes we have to transfer. We haven’t really had that opportunity. For that matter, this area of the video preservation we do, or the video restoration, we haven’t been able to explore as much as we’d like. Probably about a year into working with this process, we started to talk with conservators. The person I talked to first was a fellow named Bob Tudenet, out in San Francisco. We said, “Well, what do we have to do to give ourselves as much comfort as possible with this process?” He went down this list of basically, Herculean tasks that we had to go through. We had to clean the tapes as many times as possible--until the tape failed--until we’re comfortable with this process. We tried to do that and we tried to take the input that we could get from conservators because they’re really the ones that will allow us to have comfort with the system that we’re using, and comfort with any system that we may try using in the future. So that is our cleaning system.


QUESTION:  How many times before the tapes failed, did you—?


HONES:  We cleaned the tapes twenty times. They hadn’t failed at that point, and—how long does it take it to run a tape through there?


SELSLEY:  Well, generally, we run tapes a minimum to two passes of cleaning. And one pass is sort of a forward and reverse process of cleaning; that’s one pass.


QUESTION:  How long does that take?


SELSLEY:  It takes about ten to fifteen minutes per pass.


HONES:  So we spent a lot of time doing those twenty passes, and at that point, we just kind of exhausted the process.


SELSLEY:  Well, we do two passes, because sometimes the first pass will actually uncover mold. The white mold tends to basically dissipate when it goes through this process. What we’re removing is really debris, or if the carbon backing’s starting to break down--any sort of airborne volatiles. What we’ve seen is the first pass will remove this mold, and then the second pass will actually start removing debris. So if you’re looking at the cleaning tape and saying, “Hey, this tape looks clean,” because the cleaning tape, the tape path, everything looks clean,  “Let’s put it on a deck,” but the [tape] will say, “Hey, no, no, no; we’ve got to do another pass.” So really, a dirty tape--a really, really dirty tape—is about eight to ten passes of cleaning, if it takes that much time and that much effort to remove all of the debris. Something like twenty passes, if a tape goes that far, is really beyond what we’re really comfortable transferring on our machine. However, our tests have shown on a tape that wasn’t dirty to begin with, going through twenty passes of cleaning, we could still transfer.


HONES:  At the time when we did that, we videotaped it, and we recorded the tape at the different passes — because we were going to a conference in Boston. And so I showed that at an AMIA conference, where we basically had a split screen of pass one, pass twenty; pass one, pass fifteen. And it seemed like a pretty noninvasive process.


QUESTION:  What’s the cost of the Recortec machine, and do they have a website?


HONES:  We bought it secondhand, and I think it was fifteen hundred dollars when we bought it.


QUESTION:  What year was that?


HONES:  I’d say probably ’95, ’96. Is Recortec still in the business of making cleaning machines?


SELSLEY:  No. It’s interesting. They are still in business, and they’re still in the same location, but Recortec has sort of changed. They’ve sold the company, sold the name. They’ve changed what they do, and they actually are more into custom sort of rack mounts and slide shelves and things for decks and keyboards. However, they do still have a lot of spare parts. They’ve said that they can sort of build a few more machines to spec.


COMMENT:  As part of the website at ETC under Preservation Resources, we included a URL for BOW Industries, because they’re still in the business.


COMMENT:  There’s a lot of information on the Experimental TV Center website under the Preservation area, specifically about Recortec and some other companies. Addresses and URLs are available.


SARAH STAUDERMAN:  I actually did some testing of cleaning using some analytical tests that conservators use. I made a presentation about this at AIC, and also later on, at AMIA. One of the things that is certainly true about this cleaning technique is that it does not seem to alter the chemistry of the tape — which is really important. But one of the things that I’ve been debating in my own mind about cleaning techniques is the idea of using, instead of a baking system —to add to a list of possible cleaning or pre-preparation activities — is the idea of desiccation. By putting a reel-to-reel tape, into a very low humidity environment, you can create a microclimate, which essentially reduces the humidity in the space. It does the equivalent of baking, without adding heat. Or maybe even lowering the temperature and removing the moisture. The issue we’re trying to get at here is the hydrolysis of the binder. If people were washing these materials, they were literally adding more moisture to a binder which is beginning to deteriorate — unless they were washing it with a solvent. But I can’t imagine why anybody would wash a tape, if they were actually using water — or a surfactant, for that matter. But this idea of desiccation has not really been discussed anywhere, and hasn’t been explored. And it would be probably the most passive way — especially with very vulnerable tapes, where you were fearful of putting it on even this very delicate system — perhaps of reducing the hydrolysis, maybe for that one last chance of pass-through. So it’s another area that needs to be looked at. The whole cleaning thing desperately needs to be addressed by research institutions somewhere in this country that can look at the tape surfaces before and after cleaning, and make a determination about not only the chemistry of the tape, but what does it look like when it gets played back? What difference does it make to the system? If it actually improves the output, if it actually permits that tape to live for a longer period of time because it’s gone through this process — those are all really good reasons to proceed with cleaning.


HONES:  I’d like to second that point. One of the things that we would love to have is someone who could have scientists work at it. As I say, in a sense, it felt like we went to the moon, but we’re still the Wright brothers in our bicycle shop. We didn’t have the advanced knowledge one would expect. We’re a production house. And it would be great to have that input, not only to help BAVC with its processes, but to help, as you say, move along the whole issue. I’m curious, as far as the desiccation process, how long a process is that, would you say?


STAUDERMAN:  I’d say three days.


QUESTION:  I was going to say a lot of this type of information might be gathered more from the audio world because— partly because with the binders in audiotape there’s much more variety and much inferior quality. A lot of this has been addressed, I think, earlier on by record companies and professionals in that field. When you get a bad tape, an audiotape can be really gummy. I’m just curious how gummy videotapes can get? Before you answer that, the desiccating oven concept is something we’ve used with audiotapes, using a convection oven with the door open, at a very low temperature — usually a multi-day process. That can take a lot of the gumminess out of a tape. But we’ve gotten some in that were almost like a mass of solid material.  My impression is that videotape is just, in general, a higher quality tape, and less variation, than what you can find with audiotape.


SELSLEY:  Quarter-inch was made by some real sketchy people, so… Something I’d like to say at the Bay Area Video Coalition, we’re concentrating on videotape. We haven’t necessarily gotten into preserving audiotape. But there are a couple of differences I’d like to point out.  As you mentioned, there are a lot of proprietary blends of binders, and in videotape, there are still a lot of blends and things. We’re actually finding different tapes that are holding up better or worse over a thirty or thirty-five year period. But I’d also like to step back and say that with our preservation process, we’re cleaning the tape and remastering. We are not necessarily making it so that you bring in a thirty-five year old tape, we clean it, and you can store it for another thirty-five years; that’s just not where we’re at. You know, we would like to see more research or something to that effect. But also, with audiotape, there are a lot of differences. I mean, you have sort of a fixed head, versus rotating head. There are different things that the tapes are going through. So we really have to sort of focus our time and energies with videotape at BAVC.


BILL ETRA:  Have you looked at microscopic inspection of the tape, so that you have a standard for when you go to try to record from it? Have you used tape developer and the microscope and has anyone made a record of this, so it could be, conceivably in the future, image processed? Have you thought of putting a digital camera in the path of the tape to detect when it’s roughly done? Or have you done any work along those lines? And the last thing I want to say is that there is a lot of desiccation information in, believe it or not, the cracker business, for packaging crackers. There are very well designed machines that Nabisco and other people have designed specifically for the problem of keeping crackers from looking wet.


HONES:  I hear that more often than not, that the answer lies in food preparation and storage. As far as film goes, I recall they were talking about not freezers, but the coolers they have--


ETRA:  That was the way CBS used to do this in the days of two-inch tape. They still edited some videotape with a razor blade, two-inch tape with a razor blade. This is no longer practical—and I would suggest nobody try it, it’s highly illegal — there was a freon spigot next to every videotape machine at CBS, which both desiccates and chemically cleans and gets rid of the gunk. However, you’re also poisoning the environment. But that’s how they used to do it in the old days.


HONES:  We stopped using freon in, like, 1990.


WEAVER:  We don’t do microscopic inspection at BAVC, no.


QUESTION: The problem, I would say, with microscopic studies is that the level of deterioration and the level of indications we’re looking for actually requires extremely sophisticated equipment —like a scanning electron microscope or an Atomic Force microprobe — to get to the level of information we’re trying to get. Because I think at the sort of microscope or just sub-microscope level, you’re still not going to find what you’re looking for, because we’ve looked at surfaces of tape using very sophisticated equipment and that wasn’t strong enough; there was not enough amplification of the surface. It’s not enough to demonstrate what we’re trying to look for.


SELSLEY:  Yeah, we’ve looked at tape under a microscope, but to be honest, we’re really not prepared; we don’t really know what we’re looking for and looking at. Before we transfer a tape, it goes through this cleaning machine and the determination is made by the technician; the only chemical that’s used in the process is denatured alcohol, to clean the tape path. So it’s just really a visual instinct: Ok, everything has been removed through these various processes; the vacuum’s clean, the wipe’s clean, the posts are clean; let’s go to the next step and try it on a transfer deck.


HONES:  I think that’s why a great outcome of this meeting would be how do we bring in the scientists? How do we bring in the people who are on a level above us to deal with those issues?


SELSLEY:  I mean, it’s actually kind of awkward being in a room, to be honest, with a lot of the sort of sixty and seventy-year old Ampex and Memorex folks. And I just— I ask tons and tons of questions: How do you do what it was that you did back then? And how can I do what I do now better? How they made tape, how they tested tape, how they determined whether a tape was good tape or bad tape. So that’s really, I think, my personal challenge: getting as much information from these people as I can before they’re all retired on some island somewhere.


COMMENT:  There was a whole generation of Ampex tape that basically nobody in broadcast used, in two-inch or anything else. And this is as late as in the seventies— late sixties, early seventies. They made an entire generation of tape that Ampex said was basically bad, you couldn’t broadcast on it, the two-inch tape wasn’t made correctly.


SELSLEY:  Well, there is over fifty tape formats that have been made, manufactured and sold, marketed since 1956. There’s only a handful that really became any sort of standard. I can think of about a half a dozen. First one that comes to mind is the M2, a very beautiful, beautiful tape, wonderful. But the problems that they had with the M1, that Panasonic had with the M1, it just didn’t take, and they said, “Hey, no, no, no, it’s better. Look at this, this is much better.” And it was. But it just didn’t take on. The BetaCam SP is what took over from there. We can go ad nauseam about just all the different tape formats. To back up a bit, this cleaning machine is for the half-inch open reel. We have a RTF cleaning machine for the three-quarter, the U-matic, the three-quarter SP, and we also have a VHS cleaning machine. So really, we cover just a handful of formats that are out there.


DARA MEYERS-KINGSLEY:  I’m from IMAP, and I have two questions. One is you mentioned that there are these different brands and formats of tape that seem to be stronger, or more hardy, than others. Are you in some way recording and documenting those, so we could have some kind of report, if you will, to the field on the history of the strength of tape? I actually wanted to ask the three of you to speak about your own personal professional training that brought you to do the video preservation work that you’re doing. Part of what the field is asking for and we’re all here about is to develop scientists and engineers and technicians to further preservation efforts, both in the academic world, as well as in the conservation/academic world. I wonder, Jon, are you an engineer?


SELSLEY:  No. No, I’m not. My background is in communications and Eastern European history, so I feel I’m uniquely qualified. My relationship with the Bay Area Video Coalition started about eight years ago, as just sort of being the handyman, pretty much. Since then, I’ve been taking engineering courses, and I also do systems integration. I consult--digital newsrooms to big servers--and just the whole transition from linear to nonlinear. I’ve worked at a whole laundry list of post houses (SGI, KGO, and Oracle). So my background is really connecting and documenting everything. I mean, as far as the tape formats that sort of have come to the forefront or fallen by the wayside. But we deal, or try to deal with the tapes that people bring to us. We can’t transfer everything and we can’t clean everything; we’re really sort of handcuffed. You can’t bring us a tape and we don’t have a deck; you have to bring a deck, too. You can bring in a machine from 1956— or a tape from 1956, you’ve got to bring us the machine, as well. Really, what we deal with is trying to figure out what was the dominant format at the time. There are so many formats out there now. But they were used for mainly news gathering, or this was a consumer format. It just really depends on how the companies were marketing these.


HEATHER WEAVER:  Well, we’re primarily a nonprofit arts organization, so we work with independent documentary and filmmakers. What we have the most experience with is what they’re shooting on — which ranges from VHS, hi-8, mini-DV, BetaCam SP, all the way up to Digital BetaCam. And occasionally, we do have clients who have to bring in the odd M2 deck that we can use. I’m primarily a linear online editor. And my experience with format stability and how tapes hold up over time is just shuttling through hundreds of them as I’ve been putting together documentaries. I guess it’s bad that the record we have about what we find is mostly in our heads, and not on paper. I have been collecting a nice gallery of dropouts. What a dropout looks [like] on hi-8 versus what it looks like on DV. And I would say that BetaCam SP does experience probably more dropout than DV, from what I’ve seen — although with DV, it hasn’t been around as long, so it’s not really a fair comparison. But in an online suite, where we’re also doing the finishing, I find that dropouts on analog formats, they’re easier to fix. When DV drops out, it usually alters three or four frames, and the dropouts are very big and chunky. On a BetaCam SP, or even a hi-8 — although hi-8 tends to have excessive dropout — it’s just a little line, and it’s pretty easy to cut and paste, if you will, to fix the image.


HONES:  I just wanted to say something about Heather’s background, as well, because she didn’t mention it. Heather trained at the national PBS, with PBS engineers, to make sure that what was coming out of BAVC was meeting the specs before it got to PBS. Essentially, she edits the PBS programs that come out of BAVC — which, there are a lot of them. She’s edited a lot of the programs that go onto film. And I certainly don’t know all the details, but I’m certain you’ve done a number that have gone on to be at least Academy Award nominated, or maybe won an Academy Award.


WEAVER:  Yeah, we do well in Sundance, when we get pieces in there.


HONES:  Essentially what that means is she’s in there working with the tools of the video trade. That’s the waveform, the vectorscope. Those are probably the big ones.


WEAVER:  Also the color corrector— I do a lot of that. A lot of image repair, as well, which is how I know what’s harder to fix and what’s easier to fix, and what lasts longer, I guess.


HONES:  I think Jon’s training has really been key to us because of his incredible ability of understanding how to put together good signal flow, and understanding, you know, what it looks like when equipment is working right together. Heather’s experience is that she knows, with a great deal of experience working as a production editor, how to put a piece together and what a good image looks like, you know, once we have a system set up correctly. Essentially my video background was at BAVC, and learning at BAVC how to put together the sort of equipment that we have, and how to assemble that sort of facility. Before that, I was a computer programmer, and I built computers systems, as well, back in the mid-eighties. I’d say more my experience was making sure we put together a good team, and a good team that could do what we needed to accomplish. That being said, I certainly did some of the early transfers. Mostly, I worked on the test tapes to get a sense of what we could expect out of the half-inch open reel decks that we’re working with, and what we could expect out of the cleaners that we’re working with. I had a hundred tapes that were donated from Redwood High. So I saw a lot of bad theater!


KIM TOMCZAK:  Hi, my name is Kim Tomczak, from V Tape. I have a number of questions. I understand that all tape manufacturers produce tape with an orange peel effect on it, so the tape bumps; it actually skips along the edge and doesn’t touch so there’s a certain kind of texture to them. In your experience, does the blade or the burnishing process remove any of that orange peel effect? That’s the first question.


WEAVER:  I would say not as far as we can tell. But again, we’re not looking under a microscope.


HONES:  I would say one other thing about the half-inch open reels. When I was first setting them up, I worked with a Sony engineer. What he was saying, the connection between the Sony half-inch open reel heads and the tape is a little more rough and ready than that. There’s actually a lot of contact.


KIM TOMCZAK:   The problem with baking and the desiccation concept to me is that it doesn’t remove the dirt.  There’s also hair, skin, smoke, dirt on the tape. So I’ve often thought that that was a non-issue, because really, you’re trying to get the stuff off the surface. That’s a statement. I am trying to think when they remaster ancient films, you know, they actually go back and they aesthetically decide on the way they’re going to look, based on what they think the filmmaker thought they should look like in 1920. Are you at a state now where you’re going to take an Ant Farm tape and, you know, put it into a noncompressed digital editing system and recreate it the way you would think of Chip Lord…


WEAVER:  Actually, Chip Lord’s probably coming in to do that with us. But it’ll be in the linear suite, actually. We don’t digitize and compress.


HONES:  It’s a very good point. When you’re talking to us, we’re the technicians and, you know, we would have someone like the folks at Video Data Bank come in to put together a program from those tapes. The other issue that’s got to be addressed with baking or with desiccating, if we’re talking about three days or eight hours or whatever, versus half an hour. And part of the whole issue of dealing with the clients that we deal with is how do we keep the price down as low as possible? So I think when we think about baking or desiccating, if there were tapes that couldn’t be transferred in the way that is most successful is there an alternative to go?


QUESTION:  How much problem do you have with the actual tape degenerating? And how much of a problem is control track, a loss of control track? How much problem do you have with the tape not playing back, with timebase error that’s not easily correctable? And how many  tapes that you absolutely can’t play have a problem with timebase or control track, and how many have a problem with disintegration of the actual record medium? Like what are the percentages?


HONES:  As far as the timebase error, when we first started out, one of the TBCs I was looking at was from Prime Image. I thought it was a pretty good bet, because the guy who owns Prime Image has a long history working with TBCs. But we could not feed half-inch open reel through that TBC without the image freezing. Essentially, what we settled on for a full frame TBC was the DPS TBC IV, and it passes video through really well. The way that we have our system set up, we have an alternative. At the same time, we’re passing it through a BVT-810, which is not a full frame TBC. For those things that are still causing problems in the full frame system, we can go through the BVT-810.


QUESTION: And that takes care of most of it? Or do you have any tapes that absolutely won’t play back for control track reasons? What is that, relative to the percentage that won’t play back from actual loss of record medium?


SELSLEY:  We have three TBCs and we also have three play decks, so if we’re having a problem on a tape, we’ll go through all TBCs. We’ll go through a process of verifying that our test tapes work, and try all the different play decks.  It has been such a small loss because of those kinds of errors. The biggest thing why we can’t transfer a tape is if a tape is at what we consider end of life — something that has gone through these twenty passes of cleaning, and it’s just not coming clean….But some of the problems we’ve also had are actually if the tape was recorded on a deck that really wasn’t up to par; it was in poor shape to begin with. So that’s a process where the tape is clean, the image that’s on the tape is really fine, it’s just it was recorded where maybe the heads were worn, where the heads were out of sync, out of alignment, or it had any number of a hundred and one things wrong in the deck. So actually, in certain circumstances, since we are lucky enough to have three play decks, three AV play decks and the one CV play deck, and also a nice gentleman that can tinker on the decks. He has actually modified our deck to reproduce the sort of horror recording that was done initially in the early seventies.


WEAVER:  It’s also moving the heads, because if the heads were out of alignment on the tape path when it was originally recorded, to get the head to misalign our aligned heads to get it to play a tape back.


COMMENT: That’s also do-able. I mean, it would occur to me (but only if that’s a major problem) that today you could build a deck fairly easily and inexpensively, where all of that is variable under computer control and re-memorable…


HONES:  Yeah. I mean, I would love to do that. My only concern is would it be actually affordable? We were talking with someone about building a cleaning machine. And I think they were going to charge us thirty thousand? I don’t remember the numbers, and I’m glad I don’t because it was a lot of money, just to build the cleaning machine. Now, if we had them in there actually making sure— I mean, if we’re going to build a new machine, are we going to have a really nice tape path, and take into consideration some of the technologies that have come since these machines were first built? Are we going to make it a much nicer machine?


COMMENT:  Well, if you do that, you have some problems. But if you just want controllability of all the elements that could have gone wrong in the original deck, it’s not prohibitive. I guarantee it’s not a prohibitive enterprise.


QUESTION:  I was just wondering if I could ask a questions about the RTI machines, if anybody knows of any research or anything that’s been done on cleaning tapes through those RTI machines? RTI, it’s a company that makes machines that are really tape evaluators; but they also have some of the same components as the Recortec. They are used in a lot in libraries, for instance, to run the tapes through. They check for certain errors.


HONES:  And the formats are U-matic, VHS, and probably Beta.


COMMENT:  Yeah, they have a Beta. And one-inch.


COMMENT:  There was a time when broadcast insisted that a tape be evaluated before they used it. So if you were at CBS in the seventies or sixties every tape that you put up would be an evaluated tape. You paid less for an unevaluated tape. There were machines that actually did that. Ampex, I think, built one; several companies built one. They built one for whatever format they brought out, because broadcasters would not use tapes that couldn’t be evaluated, and therefore, you needed a standards machine to evaluate it.


HONES:  And those broadcasters were reusing their tapes so much. I mean, definitely, the local news stations.


QUESTION:  In the same vein of talking about RTI machines, has anyone thought about taking one of the half-inch RTI machines, with a cassette mechanism—an open-able cassette mechanism, to load half-inch open reel in, and use the RTI for that?


WEAVER:  Yeah, but they don’t make one specifically for half-inch open reel.


QUESTION:  No, no, I’m saying to take a cassette—like a pre-loaded cassette that screws open?


WEAVER:  That would be something that we would be definitely interested in partnering with them or trying. But right now, I don’t believe that we have the resources at BAVC to do something like that.


QUESTION:  With the Recortec, the hubs and the transport, the takeup reels, how did you have to modify those — or did you have to modify those — for videotape?


SELSLEY:  Well, it was initially for a one-inch computer tape. Actually, the entire tape path, the posts, the vacuum, everything was modified per our requests to Recortec. So we just told them pretty much what we wanted.


HONES:  Yeah. If you look at the machine, you can see essentially, the modules and everything are just screwed in, and they just built it to spec.


QUESTION:  I’m going to go out on a limb here with regard to the RTI tape deck, especially the error evaluation. I have observed them a great deal, and I don’t think that they are worth anything. I think it is bells and whistles. I think the people who rely on them to determine the condition of their tape are being misled. I don’t think that it does anything. We looked at tapes that had been cleaned by the RTI tape check that had, you know, various amounts of bells and whistles coming off of them for, you know, very dirty, edge damage, blah-blah-blah. The tapes were fine when you look at them under these more sophisticated machines, such as FTIR. They don’t have the damage. I don’t know what those bells and whistles are doing. I think that the tape cleaning part is fine. And no one has ever been able to actually tell me what supposedly they’re detecting, because they’re not expensive machines--in comparison to what we’re truly trying to determine about tape damage of the surface.  Sometimes they have a blade that you want to take off because some of the RTI tape decks come with a blade you take off. But in essence, there is polyester, non-woven webbing that the tape runs up against, precisely as yours does in your record one-half inch cleaning material. And it does the thing.


SELSLEY:  Just a couple quick things about the RTIs. We just purchased their VHS cleaner, and we didn’t go with the evaluator. First, for money, and secondly, we were unsure. What are we really doing here? So we sort of did without it. And also, we’re very happy to have that machine; however, it was three or four thousand dollars, and I’m not going to take our VHS machine apart and try to adapt it for half-inch open reel. However, I’d certainly be happy to take somebody else’s machine apart!


HONES:  Then you have to put it back on the reel because of the tape path difference for the two machines. I did kind of think about that. I was looking to see how close the tape path was to half-inch open reel. It’s very different. As I understand it, there’s a process in tape duplication, where you run your master tape next to another tape, and magnetically, the signal is transferred. I was thinking: Boy, if you could do that with half-inch open reel, that would be the most interesting, if you were transferring it to brand new stock.


SELSLEY:  Now, I’ll basically go over how we have things set up at BAVC. This is an engineering program. It’s a CAD program. But it also has a database of about thirteen thousand pieces of equipment that are commonly used — distribution amplifiers, timebase correctors, decks, monitors. And it’s about seventy-five hundred dollars. And we really don’t want to just suddenly, you know, make this information, like, you know, start giving it out. It’s really, like, a twenty-four inch by thirty-inch piece of paper that three or four people can look around and say, “I’m having some problems with the signal; where is it?” And it really allows you to track down your signal and find exactly which cable is giving you the hard time. This is an AV-8650, and that is one play deck.


HONES:  Let me ask you, Jon, in red are basically boxes, so over here is a VTR, here is a switcher?


SELSLEY:  Yeah. This is a deck and this is a deck. This represents two AV-8650 half-inch open reel decks, like we have up front here, that are mounted in a rack, so we don’t have to keep switching the cables around, we don’t have to move the deck around; it’s in the rack, in our control room, with all of our equipment. Here is the audio for one deck, here’s the audio for another deck, and it goes into a switcher. What the switcher allows us to do is take one tape and switch back and forth between the decks without any recabling or repatching; all we do is just hit a button on the switcher. Then we have the output of the switcher you can really get down to the exact wire and say, “Ok, this is the one.” If we’re having a problem, we can troubleshoot. This is incredibly valuable for troubleshooting. There is a line conditioner, because the audio coming out of the play decks is unbalanced audio. Professional audio tends to be balanced. It’s coming in unbalanced, and goes out balanced. Also, with the audio, we have this mixer we can bring in, and that’s where we can really tweak the levels. You can mix it, bring up the levels, and bring down the levels. But we really make sure that our audio levels are just within basic broadcast specs and guidelines. We’re not going to sit there and ride the levels through the tape; there are just too many questions and concerns about artist’s intent and it’s not really for us to decide. We basically clean the tape and get the best possible transfer that we can off that tape. In post, you can always go back and make changes, or you can say, “Hey, can you retransfer this tape, riding the levels,” or… But really, our concern is getting the original master cleaned and transferred, and then essentially, that’s the end of our preservation.

And then there is an audio distribution amplifier, so we have coming out of the mixer, patch, into a distribution amplifier. It comes “normalled” out of the mixer, which basically means you don’t have to patch anything, but you can use any of these other patches to go straight into patch to the distribution amplifier. The important thing is that you can have one source and six destinations, all at the same time, so really what we’re doing is we have the one master, and we can remaster to up to six different formats with the same audio. So when we say we can go to three formats at a time or six formats at a time, or if you drop something off and you want two formats, it’s in the same pass. And then here’s an audio monitor so we can monitor the levels, Techtronix, and that’s normalled out of the sixth output of the distribution amplifier.  Then coming out of the mixer we’ve got some speakers.


HONES:  This represents essentially our attempt to try and set up a system that would strategically help us be as efficient as possible in working with half-inch open reel. And that means that we have three decks here. If a tape arrives, we don’t know if it’s going to work on this deck; these decks are essentially identical, they seem to play a little differently. We don’t know if that tape’s going to be skip field or if it’s going to be EIAJ half-inch open reel. So that was the key, is we have this switcher and essentially, we have everything down the line. So we put it on this deck, play it, and everything else is patched, and so it’s going to the record deck. If it doesn’t work on that deck, we move it over, play it, and all we have to do is flip a button and it goes to all the places it was going from on this deck. It’s a little different than most of the other installations that we’ve done, in that our key thing is we need to move from deck to deck to deck, with basically only pressing one or two buttons.


QUESTION:  I’m curious to know why you’re using 8650s. I mean, this is the question. Why wouldn’t you be using 3650s?


HONES:  8650s give us color. There are plenty of tapes that have part color, part black and white on it so we had no way of knowing, with tapes coming in, whether they were going to be color or black and white. Essentially, we had to go with color. It’s not like we can just throw the black and white switch in the middle of a program.


SELSLEY:  Well, we keep the color switch on until we’ve determined that it is a black and white tape. Because if it’s a black and white tape and you leave the color switch on, sometimes you get false color from time to time.


WEAVER:  Well, prior to the transfer, we do spot check the tape for several minutes at lots of different places, to get our video levels at that time, you’d know if there’s color.


HONES:  If it was half and half, would you keep the color up nowadays?


SELSLEY:  No, I’d keep it black and white on the black and white part, and then color. And then, also, we’ll get into the documentation, as well, later. But that also goes into documentation. That would be noted, so that that can be carried to future generations. So fifty years from now, you don’t come back and say, “Why is this black and white? Make it color. Somebody please do something.” No, it was black and white when it was recorded, and when it was transferred; and then the second part was color. We get that when there are edited pieces, which we actually see a lot in the education industry. People take something that was color and tape over it, tape their own program over it. So somebody had a program on it, it was color; they take it for their own use, and say the first twenty minutes of the tape is their own black and white footage; the last, say, twelve minutes on a thirty minute tape is color. I’m going to transfer everything, because that’s everything that’s on the tape. I don’t worry about, you know: Ok, well, this is some TV show from twenty-five years ago. You’re going to get everything that’s on there, and I’m going to note that the first twenty minutes is black and white then it switches over to color.


HONES:  Yeah, we’ve run into that, with part of it being EIAJ and part of it being skip field, as well.


SELSLEY:  And also, on these 8650s here, which is actually about 99% of our half-inch open reel tapes are the 80 series. We do have three, but we keep two up and running. And then when one goes out to service, we sort of rotate it in. They never go out to service all at the same time. So there could very well be one deck that’s literally fresh back from service — fresh alignment, fresh heads, it’s got a new scanner and capacitors all over the place — and we’ll start with that. Then we rotate through, and that’s just really to match up the right tape with the right deck, because the tape might not have been recorded on a perfect deck to begin with. It might need something where the heads were a little worn, where it had a different signal-to-noise ratio, or it had the tape contact.


HONES:  The other thing I wanted to mention about the signal flow is coming off of these decks —I think it’s probably minus ten dB, as far as unbalanced audio goes. The matchbox corrects it to plus four, which is usually the input that you essentially have on Beta SP.


SELSLEY:  That’s automatically in the system; we also have our BVU-800 and VO-9850--three-quarter decks. They have audio, and time code is an audio signal… Those don’t go through the switcher, because this audio’s actually coming out balanced, and coming out two channels; whereas on the other decks, it’s coming out one channel that we mix, and it’s coming out unbalanced so what that does is it allows us to take the audio out, patch, and then actually go into this distribution amplifier, straight in out of the decks. Then once again, coming out of the distribution amplifier, we listen to it with our ears, and we monitor it. This Tectronix 760 audio monitor is a very sensitive piece of equipment that tells us: Is this within legal limits? In this industry, everybody’s always told to think outside the box, and that’s wonderful--that’s how innovation comes up. Except at this point, it’s where we really have to think inside the box, because you’re going to be looking at it through a television, and we don’t want audio levels bleeding into the video; you don’t want to have things way out of whack. 

Now let me bring up the video. Once again, we have the AV-8650 play decks, CV-2100, BVU-800, and the VO-9850; these are the three-quarter decks. This is video only. For the video of the half-inch decks, once again, we go to the switcher; we can carry the video of the three-quarter decks also to the switcher, because it doesn’t have the same limitations or differences that the audio has--where the audio coming out of the decks is a different signal type. It’s also one channel instead of two channels. This is the same kind of video that’s coming off the half-inch open reel decks. We have the composite video out, and then there’s also the monitor out, which carries the time code on the video signal.

Then we have our first TBC, the DPS ES-2200. It’s normalled out from the switcher. So really nine out of ten times, our transfers go into this particular TBC. Luke goes into it in depth in his paper. Regarding the nature of how TBC handle an image: there are different proprietary cone filters, and various proc amps that are used. This is basically matching up the right tape with the right deck, then the right TBC, to get the best possible image.

HONES:  A TBC is a timebase corrector. Very quickly, when a tape is going through a tape path there is a little bit of flutter, and what is created is what’s called timebase error. A TBC was created so the video signal goes into the TBC, it’s corrected and, the timebase errors that come from dealing with the real world, they go through and they are cleaned up and then sent out. It’s essential to getting a signal--not only for preservation but for broadcast as well--from a deck to another deck. Most of the equipment nowadays--if you’re buying broadcast equipment--has TBC built in, so a lot of times you don’t see them. But with older equipment, we have external TBCs.


ROBIN SCHANZENBACH:    Aren’t you going through some sort of waveform and vectorscope, before you get to the timebase corrector?


WEAVER:  It’s after the TBC.


SCHANZENBACH:  And my other question related to that, you  had said you work within broadcast specs. Are you always assuming that there are bars and tone at the head of the tapes to do setup with?


WEAVER:  I’m going over that in my presentation. But if there are bars and tone, we will use them. A lot of times on the half-inch open reel, there are not. We set to provide a good transfer without losing detail, highlight or shadow areas of an image.


HONES:  That is why it’s very important to have someone like Heather, who’s had the level of training she has, to work with these tapes that don’t come to us with standards to help us figure out how things work. She’s able to work with the tapes and come as close as, I’d say, is humanly possible.


SELSLEY:  In this process is a patch point. What we would do, our standard procedure, is to take bars and patch it in to this patch point. Then we’d set the TBC to bars, and then we’d also set the audio record levels to tone. Before we do a transfer, we’ll spot check the first few minutes of a tape, and make sure that everything is within legal broadcast specs, using the Techtronix audio monitor. We have two TBCs, and our HR600. We have what’s called a transcoder, because the signal that’s coming out of the switcher is a composite video signal, and if we’re going to a machine that has component video, this is what we go to. Then there is a Techtronix 1765 waveform and vectorscope. The whole preservation process, it is very exciting--what we’re doing, this whole transfer and recovering this media.  First off, it’s kind of a chore to match up the right deck and the right TBC, and it really is a monitor process throughout.  My job is to monitor, using this waveform and vectorscope.  What we’re monitoring is what’s coming off the tape, what’s coming out of the TBC, and what’s going in and out of all of the record decks. Then we’re taking all that information through a waveform and vectorscope. The whole time, I’m bouncing back and forth with all the patches, monitoring all the various signals, to really ensure that what’s coming off the tape and what’s going to two or three or four record sources is within legal limits for your television, and it’s also theoretically, the same image going to and from all the various play and record devices.


HONES:  Essentially, the waveform vectorscope is serving the same purpose as the color monitor. It’s looking at the video.   The vectorscope allows you to look at the color signal, to see where the color signal is falling. The waveform, for the most part, I use for looking at the black and white part of the signal. Think of this in the same way that you think of the color monitor; it’s helping you to look at what you’re doing.


WEAVER:  It’s just another much more precise way. Because you can actually measure the voltage levels of the electronic signal, so you know exactly what you’re getting.


SELSLEY:  We have a transcoder that will take a composite into a component signal. Then we have an ADA, which can take an analog signal and transfer it and turn it into a digital signal, as well. We have this device if we want to go digital into DigiBeta or a DvCam, it’s what does. Then we also have a Techtronix 601, which is a digital waveform. A digital signal is different from an analog signal, and it therefore has different measuring equipment--that’s the quick and simple answer to months and months of talk! Also, there is a distribution amplifier we have, coming out of the switcher, is normalled to this timebase corrector. However, we can patch into any of the other two timebase correctors, and it’s normalled into this distribution amplifier. We can patch anything into this, and it does the same thing as the audio distribution amplifier, which basically takes one signal in to six signals out and is normalled into a monitor.

Another important device is what’s called a “feather”. Every device that we have at Bay Area Video Coalition, the thirty decks that we have, the edit suites, every monitor, and every timebase corrector — everything we have shares the same reference. Heather will get into exactly how important it is, but please understand that this is incredibly important.  What we have is our Grass Valley; it’s a sync generator. From one source, one sync generator, we have the black  reference signal coming out to distribution amplifiers. These distribution amplifiers take one in and six out, and distribute the same black signal throughout the entire facility. I can’t stress enough how important that is. And then also, another thing about the Grass Valley is it has color bars here and the color bars hit here, which is patch. It also has tone, a one kilohertz sine wave, which is an audio test signal. It hits the distribution amplifiers that it hits. Everything is sharing the same reference signal. Our record decks, a BVW-75 Sony, that’s a Beta SP; another BVW-74 Sony Beta SP; and another BVW-75, which is actually an Ampex CVR-70. Then we have a DigiBeta, a DvCam, and a DVD burner.


HONES:  And what happens if it doesn’t [share the same reference]?


SELSLEY:  You get sync shifts. You’ll have a big black line across your image. On the top of your screen, you’ll have the bottom of the image; on the bottom of the screen, you’ll have the top of the image.  Essentially, reference is the heartbeat; it’s the heartbeat that all of your equipment is listening to. What is going down the cable is just electrical impulses. It’s not like film.


WEAVER:  There was actually a good article on the Experimental Television Website that Sherry Hocking wrote about sync.


SELSLEY:  This is an awful lot of information. But the important thing here is this is what ties everything together--how everything is connected. This gets down to the actual cables, so that if you’re having a problem with anything, you can say, “Hey, this cable here, let’s check it.” At any point, you can check your sync. You can say, “Well, how do you know that you’re sharing the same sync?” You can check it on a waveform and say, “Oh, well, that sync is off.” So rather than rewire your whole facility, or freak out because you’re having a sync shift or some sort of problem, you can test every single cable in this whole system.


HONES:  As complex as it looks, it’s what allows a facility to run. It keeps the downtime down, when you have this level of planning that has gone into place before cable is even laid. Drawings actually come to us before the system is set up, and we approve them.


QUESTION:  Was your system designed primarily for postproduction, and modified for preservation? What is the relationship between these two things?


WEAVER:  Preservation was added, so it was just more equipment in the racks. We did already have the infrastructure; we already had the patch bays.


HONES:  The design for the preservation center is more than just the decks. A lot of thought went into. It was very new to me, and unusual to me, to think of putting a switcher before a DA. But we did that, because— unlike postproduction — we potentially needed to try a tape out on three machines. We needed to do it as efficiently as possible so we could worry about the stuff that the client would want us to worry about.


QUESTION:  What percentage of time on this system is actually used for preservation?


SELSLEY:  Really, that’s about a fourth of the time, is actually transferring the tape. Somebody drops off a one hour reel, a large open reel--that’s about three to four hours of handling by the time that it’s cleaned and we’ve found the right deck and the right timebase corrector. Then there’s also all the prepping of the record tapes (Heather will discuss test signals on the record tape) and getting everything ready, and any patching you need to do. Of course there’s documentation for the whole process. The transfer is such a small part of the process. There is a technician there the whole time to make sure that nothing happens--no patches get pulled and nothing gets stuck. That’s almost like the hairiest process, because you’re monitoring what’s coming off the tape, audio and video; what’s coming out of the timebase correctors; what’s going into the record decks; what’s coming out of the record decks. Even though everything’s in the same room in eight racks wide, you’re running back and forth checking everything.


QUESTION:  Would there be any difference in the system if you only did preservation--If that was your main task? Would the components and the things that you put together be different?


HONES:  We’d have a smaller patch panel.


WEAVER:  Yeah, we’d just have a smaller patch bay and a smaller physical room. Everything on there, I feel, is essential.

HONES:  I was just listening to Jon’s description, and it’s true, a lot of that stuff gets used for postproduction.  Pretty much most of what we have in that control room is a lot of video DAs and audio DAs to pass signals, and they’re being used for video preservation.


QUESTION:  I’m interested in knowing what people can do on a smaller scale that will still be effective.


WEAVER:  Well, I guess it depends on what format you want to transfer to, because if you’re not going to transfer to Digital BetaCam, you don’t need the box that changes an analog signal into something digital. It really depends on what formats you’re going to go to. I would still highly recommend a patch bay, timebase correctors, and the scopes are essential.


HONES:  For preservation, we need three TBCs, where normally you’d need one.


SELSLEY:   A lot of play decks have an internal TBC. If somebody brings in a BetaCam, Beta SP tape to you or to your facility, your deck will most likely have a timebase corrector in it.


WEAVER:  But that only works on the output of the deck, not the input.


HONES:  The other real value of this sort of setup and this big control room is that we run one signal; it could go to six decks, if you have six decks to record to. And so it’s one pass, and you have your VHS copy, your Beta SP, or whatever you’re transferring to.


WEAVER:  So you might not need as big a DA, distribution amplifier.


HONES:  You won’t have to run it as many times, so in that way, there becomes a lot of efficiency in that size a system.


QUESTION:  Are you planning to add non-compressed digital video as a recording stage?


WEAVER:  Well, Avid’s not really non-compressed.


SELSLEY:  Even putting it through a timebase corrector does digitally alter the image, so you are digitally altering the original analog image to a digital type of image.


WEAVER:  We can go to an Avid right now. We can go to a Smoke, or an Octane SGI right now.  That’s not how someone would walk away with it yet, at this time, but for postproduction we can do that. If you transfer to Beta SP or Digital BetaCam, we can edit in the linear suite, as well.


QUESTION:  I’d be surprised if you’ve got a D-1 machine, which would be the digital uncompressed, really the best quality.


WEAVER:  We’re not really looking at that right now.


SELSLEY:   It’s time and it’s space.  Our relationship with people that come in with their collections-- it really is a long relationship. People will come in and it will be a year or two, three, five years. “What do I have? What do I do with it? Does it need transfer? What do I transfer it to?”  Uncompressed digital tape format is very, very expensive for us, and to say, “Ok, now by the way, we’ve got to get a hundred dollars for just for the tape stock on this one tape.” Like our Digital BetaCam decks, we have them because we use them for postproduction. Our cost is fifty to seventy dollars just for the tape, and when somebody has a hundred tapes. I mean, we are nonprofit, but we’re not free.


HONES:  It’s pretty expensive, even Beta SP stock.


WEAVER:  Yeah, it’s thirty-eight dollars, about, for a one hour tape. It’s still a cost.


STEINA VASULKA:  I am very sympathetic with everything you’re doing, and I’ve tried the same thing. I have also several timebase correctors and CV tapes and AV tapes, and I’ve found a shortcut for me. But then I’m lucky because I lived in New York State for ten years, then I moved to Santa Fe. I’m realizing from the discussion here now, I moved into the baking oven, so my tapes are in a good shape. First I clean them. This is, again, something we haven’t talked about here much--stock. The CV tapes almost don’t need any cleaning, because there is no black coated Sony. It didn’t exist then, they hadn’t started manufacturing it, and so all those tapes are in a better shape. I just use a tape recorder to go back and forth. Then I take them just directly and play them into a DvCam — no timebase correctors, nothing. You look at your scope and you can’t believe it--it just looks so bad. It’s just wobbling this way, it’s wobbling that way— your picture, as well as your signal— as you see it. What comes out of the DvCam has added burst, color burst to everything. You don’t have to worry about black and white additional information. It is, for me, an incredible shortcut. And it works. I don’t think it would work for the worst case scenario that they get, but I find that we are going into such an over-engineering in this discussion of the worst scenario tapes. Some tapes are in a very good shape. They have been sitting in baking ovens, in boxes. You take them out, you do a little cleaning, and you just transfer them straight to digital. And I’m very happy with the result. I find there is no deterioration, because first of all, reel to reel wasn’t such a high definition to begin with, so it is never really going down. So relax a little bit, some of you. It isn’t all that horrible.


SELSLEY:  I’d like to add that we at BAVC do sort of anticipate worst case scenarios, so I think that’s already built into our process. There are many things that we anticipate, like different formats to transfer to and what formats come in the door. We’re always trying to work with people for that. But with individual collectors and collections, you may have a pile of tapes that were recorded on the same deck, maybe over a short period of time, and stored in the same way. You’re going to have a much different scenario than what we deal with. I guess you can simplify the process. We would discuss with you folks what you are trying to do, and where you want to be in ten or thirty years. What do you want to do with your material, if you’re setting up a facility to remaster your tapes?


WEAVER:  One collection versus setting up another preservation remastering center, where you can have tapes from all over being sent to you.


HONES:  I know, having dealt with a lot of insurance salesmen, how for health insurance and they start telling you how terrible it is be without life insurance. They go down the list and they paint these horrible pictures. I think what Jon is saying is right. Really, where we’re coming from is we’ve always tried to anticipate whatever may be coming down. But in no way do we want to suggest that it’s scary or horrifying. Jon also said, and we all agree, that we have a wonderful time not only working with the tapes and seeing what comes out, but also working with the clients. It can be a really wonderful experience.


WEAVER:  After a tape is cleaned, that tape is mounted onto a playback machine, and we transfer the video image to the format of the client’s choice. I’m going explain the process that we use at BAVC to transfer the tapes. When we’re mastering a videotape or making dubs, it is important to route the video signal through a device called the time base corrector. The time base corrector stabilizes the video image going through it by replacing the synced signal of the tape with a cleaner house reference signal. Most modern machines have built in time base correctors, but the half-inch open reel machine and three-quarter inch machine do not. This makes having an external TBC a necessity. At BAVC, we have three external stand-alone TBCs. Because TBCs can sometimes add unwanted artifacts to an image, we use the TBC that provides the cleanest transfer. A TBC will also usually include a processing amplifier, or proc amp, which allows the user to adjust the video signals going through it. We can adjust the video level or the brightness; the setup or the black level; chroma, how much color’s in an image; and hue, what color makes up the image.

Because the video signal can be altered in so many ways, we want to make sure that the video system is just stabilizing the signal and not actually altering it. We also want to make sure that the signal we are creating is within standard specifications for a video signal, so videotape machines will be able to properly play the signal back and display it. Some equipment is less tolerant to signals that fall above a hundred IRE and below 7.5 IRE on a vectorscope. Video travels as an electronic signal, and the components of this signal can be measured with special oscilloscopes called the waveform monitor and the vectorscope.

It’s important to note that the scopes do not alter the image in any way, or alter the signal in any way. They just monitor the signal. If you move the knobs on the scopes the image is not affected. Only the TBC alters the image on demand. The video goes from 7.5 to one 100 IRE, which stands for Institute of Radio Engineers, and it corresponds to millivolts. What we see on the scope is every single line of video that makes up the image.  Darker areas are out on your 7.5 IRE, and the brighter areas are near a 100 IRE. One field of the video frame is on the left, the other is on the right. In the center is the horizontal blanking information; that’s the area of the video signal that holds the synchronizing pulses, which tells an electron gun when to stop drawing one line of video and to go back and begin drawing the next. On the waveform, you can see color burst, depending on the way the waveform monitor is set up.

The vectorscope only describes the color information. The phase, or angle of the signal, tells us what color something is.  The amplitude, or how far from the center the signal reaches, tells us how much color there is.

What we see on the normal test pattern are the pluge bars. When you raise it, there are three bars on the color bar pattern. You use the setup on the TBC to adjust it, so that the middle bar falls at 7.5 IRE. Then on your TBC, when you’re adjusting your video level or your brightness, you use the white bar at the bottom of the screen, and you use the video level knob to set that. To set the hue and saturation, we use the vectorscope, that’s set up to easily read the colorbar test pattern. Each bar of color corresponds to a box on the scope. The scopes are labeled: yellow, cyan, green, magenta. We adjust our chrominance, giving us how much color, so that the little dots line up in the correct box. Then, as far as hue, we use our phase knob to adjust it so the colors align properly.

So now we know that our TBC is set properly, we go ahead and look at the video signal. We play the tape in various places for a few minutes and monitor it, both in monitors and scopes, to ensure that the video signal is within the limits that machines can tolerate without losing detail in the shadow or highlights area. When you have detail under 7.5 IRE, you lose that detail. So when we do a transfer, when we watch the tape, we make sure that we don’t have dark areas below 7.5 IRE — otherwise, they’ll be gone and won’t be in your remastered tape. The same thing with the brightness levels--when we have brightness over 100, we’re losing detail in the highlight areas of the image. So we have to bring the levels down to make sure that it stays below 100 IRE. Occasionally with moving video, you get peaks above 100 and below 7.5 IRE. That’s ok. The occasional peaks, while they’re not ok for broadcast, most modern equipment will tolerate it. If you have black that’s too far, the black can interfere with that synchronizing signal, which tells the electron beam when to start drawing on your monitor. What will happen is you’ll get an image disturbance or an error, because it can’t read the sync. The other thing is if you have white levels over 100 IRE, in a broadcast situation, it can interfere with the audio levels. And some machines are unable to accept such a high level, and you’ll lose the detail that you once had. So at BAVC, we don’t sit and adjust the TBC levels as the transfer goes. We simply look at the tape and pick one best light, basically best TBC setting, to do the whole transfer. That way, the integrity of the original scene to scene relationship remains intact. If you have something that’s going bright, dark, bright, dark, you can’t really sit there and manually adjust, because once you see it, it’s too late. So rather than risking altering the original image, we go ahead and just use one setting.


SELSLEY:  We make a note of it as part of the documentation.


WEAVER:  While the tape is being transferred, a preservation technician sits with the deck the entire time, within arms reach of the deck, and monitors the entire transfer process. It’s a labor-intensive, time consuming process, because it’s not like the technician can just put the tape on there and walk away; just in case it’s still sticking, or in case something happens, like we have a power outage. You don’t want anything bad to happen to your tape. So while the technician is sitting there, they’re using this form that we call the preservation dub watch form. (Note: This form is posted on the Experimental Television Center’s Video History Site, in the Preservation area. See Reel to Real.) The form has a little column where you can jot down a time code number. And then you would jot down if you notice any drastic changes in the video signal. It’s unrealistic for a technician to be able to note every dropout or every flaw, but the technician will write down if there’s a drastic audio or video change, or if there’s excessive dropout, or if there’s tape damage. So the new transfers are then spot-checked in at least three spots to make sure that we did a proper transfer. After that, we label the tape and send it to billing and that’s about it.


QUESTION:  Do you do any kind of documentation during the cleaning aspect of it also? We didn’t seem to talk about that.


WEAVER:  We do, and on the form there is also a space to say how many passes we had to run it through.


QUESTION:  Do you get feedback from your clients on the dub watch form? Do you send the tape back to your client with the dub watch form?


WEAVER:  I know we send it when it’s requested.


SELSLEY:   We take the notes that Jon makes on it and put it into a template and send it off if it’s requested. But it’s always kept on record, so that if a client…


QUESTION:  Would you have an example of a time when the documentation form was useful for questions for a client?


WEAVER:  Yeah, it’s very useful. Because if a client doesn’t understand why their transfer went the way it did, we can go back and refer to it and refresh our memories. Not all of our clients understand dropout and skew and sync problems, so it’s helpful for us to be able to communicate better with clients.


SELSLEY:  The dub watch form does several things. If audio or video levels go way up or way down during a transfer, it really allows the client in post to correct it. The last thing that we want to hear is that our equipment is screwy. Basically, we want to assure ourselves, and also assure the client, that we’ve done everything that we can to get the best possible image off the tape as it exists--picking the right deck to play the tape, and picking the right time base corrector. Still, we will have tapes, like edited pieces, that just have problems. Or if the tape has a lot of dropout. It’s really a note for the client. Then also, if the tape has problems, we can always go back at the end. When the tape’s recorded, maybe halfway through the tape it looks like the head’s clogged, but our heads are fine-- that’s a very, very important note. At the end of the transfer I’ll immediately put a test tape on. If the test tape plays fine, then that’s a very important notation. Then we can say that yes, without a doubt, that problem is on the tape. We can also go to the same spot on the master and put it on a different deck. That’s really what the documentation process is all about, did we get the best possible image? We want to give you information that’s helpful, not only if you want to edit, but also, in twenty, thirty, forty, a hundred years down the road— my name goes on the tape, and I don’t want somebody cursing it! A hundred years from now, “This guy Selsley, what was he thinking here?” We want to assure, in every way that we can, that we’ve done everything that we could, and that’s really what the dub watch is all about.


HONES:  A good early anecdote that led to the dub watch form, and also really was a great client interaction, was I was working with the Minnesota Historical Society. We sent back a tape, a new copy to them, and they said, “Well, what’s this going on?” When you looked at the tape, there was this moiré effect, essentially. I’m looking at it like this, saying, “Well, what is that?” As we’re standing there, this one guy who worked for them said, “This is a tape that was shot off a TV.” You know, that was an example of an artifact that we had to sort through, and it actually helped sorting through it with the client, as well.


WEAVER:  As far as the labeling goes, the other thing we do when tapes come in — because we don’t have much room here — we look at the tape and we’ll assign it a number, like “one of twenty, two of twenty, three of twenty,” just to make sure we don’t lose any tapes. We make an Excel spread sheet, and we’ll note what the label is, the label information on the reel and the label information on the outside of the case, and then any other notes, like, “Case was ripped,” or “The reel was damaged.” Those notes are mainly for us, just to make sure that we don’t misplace cases or do anything horrible like that.


SELSLEY:  Regarding the preservation process, I have to stress that if a collection comes in — if it’s one tape, if it’s twenty tapes, if it’s a hundred tapes — we handle one tape at a time.  That’s through the entire process of cleaning and transferring. This is because as tapes come in, there are just so many times the case doesn’t match what’s on the reel. Or what it says is on the case and on the reel, is not what is on the reel. So we document as much information as we can, just to make sure that we have the same tape and the same case.  Part of the documentation is also, if it’s totally off— like, it says this is show A and this is piece B, but it’s something entirely different, I’ll say, “Hey, this is absolutely not what it says.” Because yet again, I don’t want somebody coming back a hundred years from now to whomever replaces us, and says, “Hey, you know, these clowns at BAVC lost our tapes, or got them all mixed up.” That’s part of the process, but it’s important to stress it is one tape at a time through the entire cleaning and transferring.


WEAVER:  What I wanted to do was stress the importance of a processing amplifier, a TBC, and a scope when you do the transfer, because if you do have detail under 7.5 or above 100, some systems just clip that information right off. If you don’t have that information when you do your original remaster, it’s gone forever--unless you go back to the half-inch open reel.


QUESTION:  On your preservation dubs, the remastered tapes, do you routinely, or on a standard basis, do you put colorbars and tone? And what is it that you do?


WEAVER:  Yes. We set it up how we set up every other postproduction tape. We have time code starting at 58 — 58:00:00. It’s generally nondropframe. From 58 to 58:30, you have black. 58:30 to 59:30 you have SMPTE bars. And you have to be careful with bars, too, because there are 75% bars, 100% bars, and SMPTE. SMPTE’s the best, because you have the pluge, so it helps a lot. Those go from 58:30 to 59:30. And then from 59:30 to 01:00:00:00, hour one, you have black again. Sometimes a client requests information slates. And if that’s the case, we leave six to eight seconds, depending on if we have to put one slate or two, or how much information it is, at the head. Then we assemble edit the transfer. All the tapes are prepped prior to the transfer.


HONES:  Just a real quick thing about colorbars with pluge, the SMPTE bars. What that is most useful for is adjusting color monitors. Speaking for myself, I would tend to trust waveforms and vectorscopes before I’d ever trust color monitors. You can set up color monitors so they can be relatively useful, and the way to do that is with this pluge — what Heather’s calling SMPTE bars.  Essentially, it has a black that’s darker than official video black, which is 7.5 IRE so you can adjust the monitor and see SMPTE black. You can see the black that’s just a little hotter than SMPTE.


WEAVER:  You can see the 11.5 IRE pluge.  The 7.5 is the one right at black, but then there’s one at 3.5 IRE. The 3.5 and 7.5 will just look like its own little box, and then you’ll just see this little rectangle of the 11.5 IRE.  When I’m in the online suite working with clients, color’s very important… it’s so the way an artist intended something to look can be carried through when it’s projected somewhere else.


HONES:  And because we didn’t set up the projector using SMPTE bars—it was a little darker than we than we would prefer it.


WEAVER:  No, we did not. It was much darker.


QUESTION:  And your audio tones?


WEAVER:  At the one kilohertz sin wave.


QUESTION:  Is that during bars?... Thirty seconds?


WEAVER:  No, from 58:30 to 59:30, so one minute.

SELSLEY:  With the bars.


WEAVER:  With bars. It’s bars and tone; it usually goes together.


QUESTION:  Do you do channel one and channel two separately, or do you do them together?


WEAVER:  We generally do them together on the half-inch open reel and three-quarter transfers, because that’s helpful when you want to mix the audio after the fact and it’s stereo. But usually, that’s not an issue with those.


QUESTION:  I understand what you’re talking about, and it’s certainly appropriate for 90% of what you’re talking about. But if it’s an experimental work, where the limits have been pushed all over the place, how do you handle that and keep the intent of that intact?


WEAVER:  Well, one thing we do is we look at the tape coming directly off the half-inch open reel machine. It goes up to a patch bay, so we can patch straight out of the half-inch open reel into our monitor. We can also look at the output of our TBC into a monitor. So not only do we have scopes, but we can match our eyes, we can use our eyes to help that. But if you have something that’s really pushing those limits, it’s not going to play back on a monitor. It’s going to upset the sync. Or you just lose everything that’s there. Modern equipment is not set up to be able to handle stuff like that. So if it’s going above 100 IRE, us bringing it down really doesn’t affect the overall intent.


SELSLEY:  Some of you might be aware of the amount of video art that was going on, when it became affordable, with the Portapac in the late 1960s--when suddenly, people that didn’t have access to video suddenly had access. The first thing you would do is just take the camera and shoot it into the monitor and start doing weird things; it’s just the early sort of graphics that people were doing. We get tapes like that from time to time. So I think your question’s more like how would we handle that? We see what’s coming off the deck, we see what’s coming through the TBC, we refer back to our test tapes to verify that our equipment’s in good working order, and then we just make sure that the audio and video are within the acceptable levels.   William Wegman, before he  struck it with the gray dog, he used to do experimental things, like with noise. Those are sort of particularly difficult to handle. For instance, he would be in a room such as this, and take a bookcase, knock it down, pick it up, knock it down, and have this reverberating room noise. So how do you handle that? And how do you handle when he’s taking a broom and sort of sweeping the microphone around the room? Then all the various feedback that people were experimenting with in video. It’s an interesting question, but  I guess my point is we trust our equipment, and make no judgment on the content of the tape. Once again, making sure that our equipment — our time base correctors, and our play decks, and our monitors, and waveform/vectorscope — are all calibrated properly. Then we also refer to the test tapes, and trust the waveform to tell us where the levels are.


COMMENT:  I guess what I was getting at is that you also have to do some detective work sometimes. Often it’s the producer or whoever, the curator who’s bringing that in, or even the person who created the tape has that kind of chain of understanding.


QUESTION:  How often do you have artists bringing in tapes?


SELSLEY:  Occasionally. An interesting one was a few months ago, where we actually transferred a few tapes. An artist sent a modest amount of tapes just to see: Ok, I’m going to send you five tapes of this large collection. He received them back and said, “You know what? I remember them being a little sharper.” It’s sort of is interesting, “Ok, sir… We’ve been doing it for thirty years…” I mean, you know? “What do you remember?”


WEAVER:  The other thing that’s really difficult with that is with every monitor. If I gave everybody a VHS tape to take home, you’d all have a very different experience of what something looks like, unless everyone sets to bars and tone every day--which I doubt.


COMMENT:  We want to just kind of recap. Again, our mission here is to come out of this with some very concrete things that we could do, or need to be done, in terms of this work. I was taking notes, and others probably have notes about some of the things that were mentioned, in terms of possible work. I have: Testing of baking, as a method of preparing tapes; Testing of the failure of a tape — when does a tape fail on the cleaning system? Creation of a microclimate with a Desiccant, and possibly documenting points in the tape path through some kind of photographic process; Getting information out of Heather and Jon’s heads; Replacing motors on VTRs; Loading half-inch open reels into cassettes; modifying the RTI; and clarifying or exploring the relationship between postproduction facilities and preservation facilities.  Are there other things generated out of this discussion that people think we need to look forward — specifically around the cleaning and remastering section?


CAROLE LAZIO: I wanted to repeat Sarah’s observation that we should talk about how to engage scientists in doing research in this area.


LINDA TADIC:  I don’t know if anybody’s already mentioned this, but the Image Permanence Institute, which did all the research on vinegar syndrome for acetate degradation, and also developed the acid detection strips, they’re beginning work now on something for video, for magnetic media, from what I understand. Or they’re creating a device like an acid detection strip that will be able to detect when hydrolysis is starting to affect binder in video that would be great if they can develop it. It’s a multi-year project that they’re just beginning to work on right now… So you can test so if you have, you know, a huge room of videotapes, you don’t have to pull it out to test if it’s starting to deteriorate.


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

May 31 – June 1, 2002

Issues in Physical Preservation     1