sd@sdchema.UUCP (Steve Dempsey) (11/09/85)
What's the latest and greatest in film recorders? We have a Silicon Graphics IRIS 2400 from which we want to make 35mm slides and 16mm films. Steve Dempsey Chemistry Dept., B-014 University of Calif. at San Diego La Jolla, CA 92093 (619)-452-4016 ...!ucbvax!sdcsvax!sd@sdchem
papke@dicomed.UUCP (Kurt Papke) (11/12/85)
In article <471@sdchema.sdchema.UUCP> sd@sdchem.UUCP (Steve Dempsey) writes: > >What's the latest and greatest in film recorders? >We have a Silicon Graphics IRIS 2400 from which we want to make >35mm slides and 16mm films. > > Steve Dempsey > Chemistry Dept., B-014 > University of Calif. at San Diego > La Jolla, CA 92093 > (619)-452-4016 > ...!ucbvax!sdcsvax!sd@sdchem Well, I swallowed the bait..... This is an incredibly broad question. What are your requirements for: o Resolution (8k, 4k, 2k, screen res) o Geometric accuracy o Speed o Consistency, RAS (reliability, availability, serviceability) If all you have is an Iris, and you need to take advantage of your Geometry Engine you are going to end up taking pictures of the screen with a hood, not exactly "latest and greatest" stuff !! A brief overview of the Film Recorder marketplace: Manufacturer/model Price Comments Polaroid Pallette $1500 1k res, limited colors Matrix PCR $10k 2k res, newly introduced 4k model Matrix QCR $15-20k 2k or 4k Celco $150k 4k, newly introduced 8k Dicomed D48 $250k 4k/8k There are a few others rattling around in the marketplace, but the list above includes the majors. There is obviously a wide discrepency between the price at the bottom of the market, and that at the top. This arises from the cost of pushing the performance envelope in the 8k region -- film recorders are complex electro/mechanical/optical devices. Kurt Papke ihnp4!dicomed!papke "The views expressed herein are not those of Dicomed Corporation. The only views of Dicomed are those of downtown Minneapolis from the rear windows."
dave@onfcanim.UUCP (Dave Martindale) (11/13/85)
In article <648@dicomed.UUCP> papke@dicomed.UUCP (Kurt Papke) writes: >A brief overview of the Film Recorder marketplace: > > Manufacturer/model Price Comments > Celco $150k 4k, newly introduced 8k > Dicomed D48 $250k 4k/8k A comment on "8000-line" film recorders: The newly-introduced "8k" resolution Celco is not new at all. The CFR-4000 already had a 13-bit vertical DAC and so could plot 8000 lines; calling the new model the CFR-8000 is just marketing. The real difference is that the 8000 mounts the electronics in a full-height rack, rather than the stubby rack designed to fit under a table. The older Celco was called the "4000" because its useful resolution was on the order of 4000 distinct lines - this is limited by CRT spot size, not DAC resolution. I believe the primary reason for plotting 8000 lines was that if you plotted on very large sheet film (8x10) you could start seeing spaces between the lines if you plotted only 4000. But you still couldn't resolve 8000 distinct lines, since the CRT image isn't sharp enough. The Dicomed, I believe, has 15-bit vertical and horizontal DACs. These allow precise formatting of the on-screen shape of "pixels" and allow the drawing of very smooth vectors. But the number of pixels it can resolve on screen is also in the same range as the Celco's - again limited by CRT spot size. >There is obviously a wide discrepency between the price at the bottom of >the market, and that at the top. This arises from the cost of pushing the >performance envelope in the 8k region -- film recorders are complex >electro/mechanical/optical devices. The amazing thing is how well they are built. Our Celco seems to use mostly military-spec parts. The weakest link in both the Celco we have and a Dicomed I've seen is the camera.
gwyn@brl-tgr.ARPA (Doug Gwyn <gwyn>) (11/15/85)
Does anyone know of a computer-driven camera (digital film recorder) that makes images of high enough resolution for making holograms? Note that resolution on the order of the wavelength of visible light is required. Thanks..
papke@dicomed.UUCP (Kurt Papke) (11/18/85)
In article <14752@onfcanim.UUCP> dave@onfcanim.UUCP (Dave Martindale) writes: >In article <648@dicomed.UUCP> papke@dicomed.UUCP (Kurt Papke) writes: >>A brief overview of the Film Recorder marketplace: >> >> Manufacturer/model Price Comments >> Celco $150k 4k, newly introduced 8k >> Dicomed D48 $250k 4k/8k > >A comment on "8000-line" film recorders: The newly-introduced "8k" resolution >Celco is not new at all. The CFR-4000 already had a 13-bit vertical DAC >and so could plot 8000 lines; calling the new model the CFR-8000 is just >marketing. The real difference is that the 8000 mounts the electronics >in a full-height rack, rather than the stubby rack designed to fit under >a table. The older Celco was called the "4000" because its useful >resolution was on the order of 4000 distinct lines - this is limited by >CRT spot size, not DAC resolution. ============== True, but not complete. CRT spot size is the one of the most important factors contributing to the actual "res": resolvable line pairs/mm on the film plane. Other important factors which can compromise the spot size are the quality of the optics in the lens, the color filters (Dicomed filters are actually lenses to correct for chromatic aberrations), and mechanical stability. At 8000 lines the spot can be blurred just by someone walking across the floor unless the tube, optics and camera are mounted well and properly damped mechanically. >I believe the primary reason for >plotting 8000 lines was that if you plotted on very large sheet film >(8x10) you could start seeing spaces between the lines if you plotted >only 4000. But you still couldn't resolve 8000 distinct lines, since >the CRT image isn't sharp enough. This is perhaps true on the Celco, but Dicomed does not have problems with inter-line gaps at 4000 lines. >The Dicomed, I believe, has 15-bit vertical and horizontal DACs. These Correction: 16 bit !! These suckers are expensive !! >allow precise formatting of the on-screen shape of "pixels" and allow the >drawing of very smooth vectors. But the number of pixels it can resolve >on screen is also in the same range as the Celco's - again limited by >CRT spot size. This can be hotly debated. The primary difference between Celco and Dicomed in this area is that Celco uses an intensity modulated beam, whereas Dicomed uses time modulation with constant intensity thus maintaining uniform spot size. The Dicomed spot is under 0.8 mils, but Celco's is tough to measure. It is a distribution based on spot intensity. >>There is obviously a wide discrepency between the price at the bottom of >>the market, and that at the top. This arises from the cost of pushing the >>performance envelope in the 8k region -- film recorders are complex >>electro/mechanical/optical devices. > >The amazing thing is how well they are built. Our Celco seems to use >mostly military-spec parts. The weakest link in both the Celco we >have and a Dicomed I've seen is the camera. What weaknesses do you find in the camera. Marron Carrel makes some pretty good stuff !! "The views expressed herein are not those of Dicomed Inc. They are my own personal, biased, slanted opinions".
dave@onfcanim.UUCP (Dave Martindale) (11/19/85)
I hadn't intended my comments about Dicomed and Celco film recorders to turn into a debate about their relative merits. And I'm neither in a position to compare them fairly (since we don't have a Dicomed and the only one I've used is rather old but we have a new Celco) nor to champion Celco (since I don't work for Celco, haven't read all their sales literature, and don't have circuit diagrams). So I'll just provide some more information. First, I should point out that the principal difference between the Dicomed and the Celco, in terms of driving them, is that the Dicomed is fundamentally a point-plotting display (with spacing of points that is much finer than the CRT spot size) while the Celco is a pure raster device - it draws scanline after scanline on the CRT face. Thus the Dicomed can draw random vectors that look very smooth on screen, while to get vectors on a Celco you have to do antialiased scan conversion in software. So if you want to plot line drawings, the Dicomed has a clear advantage. But if you're going to plot images that have anything other than lines or areas with strictly horizontal and vertical edges, you have to do the scan conversion yourself anyway. For our application (animation), vectors are almost useless. The other major difference is that the Dicomed changes the brightness of a pixel on film by varying the time that the beam is unblanked, while the Celco modulates the beam intensity to directly change the phosphor's light output. The Dicomed's system has the advantage of constant spot size, but the brighter the image the longer it takes to plot. The Celco's plotting time is constant and, for reasonable images, apparently significantly faster than the Dicomed. For a best-case example of plotting speed, the Celco will plot a 1536x2048 pixel resolution full-intensity white frame in about 20 seconds real time, including the time spent rotating the filter wheel. This is at the fastest pixel clock rate; there is enough light to use this rate on 5247 colour negative film (rated at ISO 80 daylight) with the lens stopped down one stop, when recording onto a standard 4-perf 35mm movie frame. I'm curious how long the Dicomed would spend on the same frame. >CRT spot size is the one of the most important >factors contributing to the actual "res": resolvable line pairs/mm on the >film plane. Other important factors which can compromise the spot size are >the quality of the optics in the lens, the color filters (Dicomed filters >are actually lenses to correct for chromatic aberrations), and mechanical >stability. I know - I was just pointing out that you shouldn't expect to get 8000 resolvable lines with either a Celco or Dicomed, but that 4000 was achievable. >>But the number of pixels it can resolve >>on screen is also in the same range as the Celco's - again limited by >>CRT spot size. > >This can be hotly debated. The primary difference between Celco and Dicomed >in this area is that Celco uses an intensity modulated beam, whereas Dicomed >uses time modulation with constant intensity thus maintaining uniform spot >size. The Dicomed spot is under 0.8 mils, but Celco's is tough to measure. >It is a distribution based on spot intensity. Celco claims to have 7500 resolvable pixels across the useful diameter of the 7 inch CRT (this would be across the image diagonal, not one edge). Doesn't the Dicomed use a 5 inch tube, requiring a smaller spot size for the same resolution? In any case, if you divide a .0008 spot size into a 5 inch tube, you come up with a number that agrees with my original comment - both film recorders can plot 4000 resolvable lines, neither can do 8000. >>I believe the primary reason for >>plotting 8000 lines was that if you plotted on very large sheet film >>(8x10) you could start seeing spaces between the lines if you plotted >>only 4000. But you still couldn't resolve 8000 distinct lines, since >>the CRT image isn't sharp enough. > >This is perhaps true on the Celco, but Dicomed does not have problems with >inter-line gaps at 4000 lines. Is this because of the Dicomed's pixel formatting? As I understand it, each pixel is plotted as several overlapping points. This should give you the ability to "size" the pixel so that adjacent lines overlap. Or is it simply that the resolution of the system (CRT spot size plus lens sharpness) is not high enough to show interline gaps? :-) Definitely positive features of the Celco: Both film recorders have circuitry of some sort to help set the "brightness" control to compensate for CRT characteristics drift. The Dicomed one involves a manual adjustment, performed by a human reading a meter. The Celco's is an automatic adjustment triggered on command from the host, and thus can be done every few frames during filming, improving exposure repeatability. The Celco mechanical assembly allows easy access to the CRT face for focusing adjustments, and the heights of the camera and lens stages are easily adjusted. This makes changing film formats reasonably painless. I've never seen a Dicomed camera stand of comparable design. The Dicomed recorder I've seen has a tendency to drop bits of the filter-wheel drive belt onto the CRT face; the Celco filter wheel is enclosed within the lens assembly and gear driven. Also, service from Celco has been (so far) impressive. One of the company vice-presidents, who is an engineer, came here to do the installation. Both he, and the regular service person, know the system well enough to do troubleshooting at the chip level. The experiences of the Dicomed owner I know of have been considerably less favourable. Now, some of my impressions may be based on a long-obsolete version of the Dicomed. But every year I wander past the Dicomed exhibit at SIGGRAPH, certainly interested in taking a look inside the film recorder, and find that the film recorder itself is closed up, off in a corner somewhere, while all of the booth's attention is being focused on Dicomed's slide-making turnkey system, in which I have no interest. >>The amazing thing is how well they are built. Our Celco seems to use >>mostly military-spec parts. The weakest link in both the Celco we >>have and a Dicomed I've seen is the camera. > >What weaknesses do you find in the camera. Marron Carrel makes some pretty >good stuff !! In the following, keep in mind that my application is animation, where very accurate registration of film and long-term durability are important. For slides, almost any camera would do. The Dicomed I've seen is an old one, and the camera (made by Electro-Optical Systems, or a name similar to that) was an absolute joke for animation use. It had no sprockets at all, moving the film with a rubber capstan. There were no registration pins, and the only side-to-side guidance the film had was a roller with edge flanges. The film would weave from side to side about 4 thousandths of an inch. Good stability is somewhere around 4 ten-thousandths of an inch. There was no aperture plate, just a hole that the film travelled over, and the CRT blooming caused by a bright area in one frame would show up in the adjacent frame. It may have been OK for some slides, but nothing more. The Marron-Carrel cameras on present Dicomed and Celco film recorders are a considerable improvement. They are pin-registered to provide much better frame-to-frame registration accuracy. But when you compare it with the Oxberry animation-stand camera we have sitting around here, the differences are apparent. The registration pins seem not to be as well-machined as they are on professional equipment. We've had some problems with oversize pins that generate small pieces of emulsion chaff when the film is pulled off the pins - this chaff accumulates inside the camera. Also, the position of the pins relative to the film frame seems to be whatever was convenient during design. In professional cameras, the position of the pins relative to the frame is standardized so that the *same* film perforations will be used to register the frame later during optical printing. (This may not matter if you only intend to project the film without further optical effects). In the Oxberry, the film is removed from the registration pins by a special carrier in the transport that touches the film only by its edges, in the sprocket hole areas. The Marron-Carrel uses a thin strip of steel that goes all the way across the width of the film, with the potential (at least) of touching and scratching the emulsion in the image-forming area. The metal parts that contact the film in the Oxberry are all highly polished and most are chromed, and all appear to be made of steel. This provides smooth surfaces to avoid scratching the film, and long wear. The MC makes no use of chrome, makes the aperture plate and pressure plate of aluminum, and generally does not have as fine a polish on surfaces. The Oxberry camera has a number of "buckle switches" in it. They are long pins that are placed just out of the film path in various places in the camera. If the film breaks or jams, it will soon touch one of these switches, and shut down the camera. This is a Good Feature for long unattended shooting sessions. The MC will just keep on merrily running, potentially making a real mess of the camera innards. This is made more important by the way the magazines are loaded. When a film magazine is put on the Oxberry, it automatically couples to the film spool drive motors. On the MC, you have to remember to run a belt from the magazine motor to the magazine's spool. If you forget the belt, you guarantee a camera jam. The Oxberry uses standard Mitchell film magazines. The MC uses magazines of their own design which are intended for 46mm film. For 35mm use, they supply spacers, but there is a metal plate supporting the film pack on only one side. The film is free to weave as it winds on the core, and if you take a magazine off the camera and lay it on its back, the film will "spill" off the core. The Mitchell magazines have none of these problems. (For people used to shooting slides: The normal film load in an animation camera is 400 feet on a core; there is no spool to provide side support). To be fair, the MC does have some advantages too. It seems to use sealed bearings (probably ball bearings) everywhere, so it should never need disassembly and lubrication. The Oxberry is much more complicated mechanically, and uses many sleeve bearings and sliding surfaces, so it does need to be serviced every once in a while. The MC's viewing system uses a reflex mirror that slides into place, allowing a reticle to be projected onto the tube face. This is *much* more convenient for setting up the film recorder than the rackover viewing system in the Oxberry. This has gotten long enough already. I do *not* want this to turn into a long argument, just provide information useful to other people considering spending large sums of money on a good film recorder. Some of the information presented here was learned the hard way, and I hope it will be valuable to others. Dave Martindale Anti-disclaimer: Everything I say is my own opinion unless I claim to represent someone else. Why should anyone expect anything posted to usenet to reflect the "official policy" of any organization on any subject?
renard@lll-crg.ARpA (Paul Renard) (11/20/85)
It's worth the big bucks for a decent film recorder if you intend to make film output of lasting and durable quality. For quick turn-around or throw-away slides, stick to the cheaper models. I like Celco, except for the camera change mechanism. If you do more than one type of film format (35mm vs. 16mm, say) you have to mechanically move the film plane. I've seen focusing problems as a result. I'm biased towards Dicomed (we have serial #1) but some problems to look out for are: 1. Best results for different film types requires separate optics assemblies for each type. At $30,000 a throw, that can get expensive. (We have 3 assemblies for doing 16mm, 35mm, 35mm-slide, and 4x5 sheet film.) 2. For any recorder, make sure you have adequate controls for proper color balancing. The Dicomed has several controls that are set once (never correctly by the service people - so be prepared for some agonizing months of adequately balancing your recorder). We've noticed little problem of color-drift once the recorder is set up properly. 3. The software for controlling the recorder can be expensive too. Write your own! (You can surely do a better job and allow for more flexible data formats.) 4. We produce 8x10's by enlarging our 4x5 film. I've seen 8x10 enlargements from 35mm done offline on a 3M machine that are quite acceptable. Dicomed's 8x10 optics are not quite up to snuff. (To get enough light onto the film requires 2-passes over the image, and thus increases the time to plot so much that it becomes impractical to produce 8x10. I'm not particularly impressed with the quality of Dicomed's 8x10 - Celco's is better - which is surprising, since Dicomed's smaller formats are generally better.) If you get a recorder of Dicomed's caliber, it's worth providing your own film processing, or make sure you have a good photo lab available. I've seen lots of potentially good film produced on the recorder, but ruined by the photo lab. We just switched from our faithful Automax and Acme cameras to Merron-Carrel. It's too soon to judge the new camera but a few things I've noticed: The film magazine covers 'clip' shut, rather than 'screw' closed, so they open too easily at inopportune times. The same camera body is used for all roll film types - you just change the aperture plate - and I'm expecting the aperture plate to eventually wear out just from changing it. They 'forgot' to put handles on the camera body, so there is no convenient way to pick the bloody thing up. And we've had the camera installed for just over two months, and it broke today! (At the aperture mechanism!) My advice: Shop prudently. Establish what kind of quality you're expecting from film output. And don't forget to consider the cost of controlling software. (And maintenance! - Dicomed probably makes more money from maintenance contracts than from sales of recorders!) That's all.... P. Renard
papke@dicomed.UUCP (Kurt Papke) (11/21/85)
In article <1004@lll-crg.ARpA> renard@lll-crg.UUCP (Paul renard) writes: >I'm biased towards Dicomed (we have serial #1) but some problems to >look out for are: > 2. For any recorder, make sure you have adequate controls for > proper color balancing. The Dicomed has several > controls that are set once (never correctly by the service > people - so be prepared for some agonizing months of > adequately balancing your recorder). We've noticed little > problem of color-drift once the recorder is set up properly. Out latest work on our slide-producing software will greatly reduce this problem. We have spent a fortune in the last couple of years on R&D concerning color problems: setup, matching CRT colors to film colors, color consistency between films types, etc. > 3. The software for controlling the recorder can be expensive > too. Write your own! (You can surely do a better job > and allow for more flexible data formats.) There is some merit to this statement if you are not using the slide production software. Dicomed has done much incremental improvement to its Precision COM software in the past few years, but has not made any quantum leaps. > 4. We produce 8x10's by enlarging our 4x5 film. I've seen > 8x10 enlargements from 35mm done offline on a 3M machine > that are quite acceptable. Dicomed's 8x10 optics are not > quite up to snuff. (To get enough light onto the film > requires 2-passes over the image, and thus increases the > time to plot so much that it becomes impractical to > produce 8x10. I'm not particularly impressed with the > quality of Dicomed's 8x10 - Celco's is better - which > is surprising, since Dicomed's smaller formats are generally > better.) The reason for this is that the 8x10 optics were designed for overhead transparencies !! Our staff artists use 4x5's for going to print. Also of very high quality is our new 70mm assembly. This is due to high-quality lenses, and the fact that the smaller film can be rigidly supported to stay flat on the focus plane, and the result of less diffusion from a thinner film backing. > If you get a recorder of Dicomed's caliber, it's worth >providing your own film processing, or make sure you have a good >photo lab available. I've seen lots of potentially good film >produced on the recorder, but ruined by the photo lab. This is all-to-true. Our FE's spend a lot of time setting up recorders for color calibration, only to have the lab change their process negating all the careful tuning !! >wear out just from changing it. They 'forgot' to put handles on >the camera body, so there is no convenient way to pick the bloody >thing up. That is why we mount a winch on the top of the D48 cabinet !! > (And maintenance! - Dicomed >probably makes more money from maintenance contracts than from >sales of recorders!) Not true. We charge the same basic percentage for support as any computer manufacturer does. One expect to pay for maintenance on a precision piece of equipment just as you do for the disk drives on your computer.
julian@osu-eddie.UUCP (Julian Gomez) (11/21/85)
*** rubber baby buggy bumpers rubble buggy baber bumpys ruggy oh line-eaters..
A couple of comments on Celco from where I used to work (Cranston/Csuri
Productions, Inc.) They've had a Celco for about two years.
Film format changes are nontrivial. It is considered a semi-major
operation, and if anybody wants something exposed in a format other
than what's currently mounted, it has to be requested a few days in
advance. There's a lot of jiggling and tinkering that has to be done,
including a test exposure to make sure the mounting was done correctly.
The other comment is that Celco support has been less than great.
--
"If Chaos himself sat umpire, what better could he do?"
Julian "a tribble took it" Gomez
Computer Graphics Research Group, The Ohio State University
{ucbvax,decvax}!cbosg!osu-eddie!juliandave@onfcanim.UUCP (Dave Martindale) (11/23/85)
In article <860@osu-eddie.UUCP> julian@osu-eddie.UUCP (Julian Gomez) writes: > >Film format changes are nontrivial. It is considered a semi-major >operation, and if anybody wants something exposed in a format other >than what's currently mounted, it has to be requested a few days in >advance. There's a lot of jiggling and tinkering that has to be done, >including a test exposure to make sure the mounting was done correctly. There are actually several ways to do a format change. If you want the maximum resolution, you want to be using as much of the tube area as possible for the image, and so you have to change camera and lens positions, refocus, reset image sizes and positions. Yes, it's a lot of work. And, of course, since each format is a different magnification, you have to redo your exposure tests and thus have different film calibration data for each format. On the other hand, if you just want fast format changes, you set up the mechanical part of the system to handle the largest format you do (say 36mm slide) and then shoot the smaller formats using the same setup, using smaller areas of the tube face. You can use all the same exposure data for all the formats. A format change is just entering a different bunch of numbers to reposition the image and changing the camera aperture plate. (If you have a fixed-format camera stand, this is the only way you can go about changing formats). The Celco makes this easy by having the camera aperture plate in use cause the recorder to select between two different sets of size and position adjustments, and rotate the image on the tube, so you can have the system permanently set up for 35mm slide and cine, or 35mm slide and 16mm cine for example. So you have a choice. If you're fussy, you do it the hard way; if you want to switch formats easily you can avoid the hard work. Of course, if you're making a drastic switch in format (35mm to 70mm or 4x5), you probably have to switch lenses too and you can't avoid all the readjustments. The above comments apply specifically to a Celco; I don't know what Dicomed does, but if a variable-format stand is available, it probably works in a similar manner.
johnf@apollo.uucp (John Francis) (11/25/85)
Here at Apollo we have a Matrix QCR D4/2 - Oxberry animation camera combination. For those of you unfamiliar with this hardware, the Matrix is a dual resolution device - either 2K or 4K resolution (software programmable). The amount of user control on the Matrix is good - you can load compensation tables of your choice as well as the "standard" sets supplied for Ektachrome, as well as setting the beam intensity for each exposure pass. (The Matrix is a constant beam intensity device, with the exposure being varied by lingering the beam on each pixel). You can also rotate or reflect images to deal with horizontal or vertical formats. It also has internal calibration logic, and you can tell it to perform a calibration scan whenever you want (between frames). It is a raster-scan only device, but it does have a form of run-length encoding to reduce the amount of data to be sent. You can get several different camera modules as well as the Oxberry camera - we also have the standard 35mm slide camera. Changing modules is simple - all you do is pull one module out and plug in the other one. The system can even tell the software what camera module is loaded (in fact you can read back just about everything you could ever want to know, and then some!) We have just had some of our images of parts of the Mandelbrot set enlarged to 20x30 (inches). These were exposed at 2K resolution, but only computed at 1K resolution, so we had to do software pixel (and scanline) replication. We were using standard Kodak 100 ASA negative film, and the results were impressive! There are really square pixel blobs visible on the prints, so after a few days of CPU time we are now ready to try a 4K resolution exposure, with every belief that the camera is really able to achieve this degree of resolution. For what it is this device is also fairly cheap - the Matrix is about $20K, and the Oxberry about the same. I do not know enough about animation cameras to compare the Oxberry to other possibilities, but it is the standard Oxberry pin-register animation camera adapted for use with the Matrix (presumably the adaptation is simply to allow computer control of the shutter). Enough of the sales pitch - now for the bad news: The Matrix camera really only has 64 different pixel intensities. This leads to very obvious contouring artifacts when the image is a computer-generated picture (such as a ray-traced image) that has smooth variation of intensities. This can be fixed by adding random low-order noise to the image, but of course this means that run-length encoding goes out of the window! As the interface to the Matrix is over an IEEE-488 (GPIB, or HP instrumentation) bus, and the maximum data rate the camera can accept is about 50K bytes/second, a 4K by 2.7K RGB image takes about 12 minutes. Even in 2K resolution mode the time taken is 3 minutes/frame. This is fine for the occasional slide, but trying to shoot animation sequences (at 2K by 1.5K resolution) requires 24 HOURS of exposure time for ten SECONDS of running time. There are rumours that Matrix are working on a faster interface, and/or a lower resolution mode (512 by 384 is often adequate for animation), but at the moment I have heard nothing more than rumours. When changing camera modules, dust has a nasty way of creeping in and settling on the CRT. This is exacerbated by the fact that the Matrix has to be mounted vertically for use with the Oxberry (that thing is HEAVY), so dust settles on the screen by gravitational as well as by electrostatic attraction. This means that after a week of continuous shooting you may well find everything ruined by a black blob of dust right in the middle of your image! Mounting the Matrix vertically also requires a modification to the filter-wheel assembly, since normally the drive capstan for the filter wheel is held in place by gravity. When we first got our unit from Oxberry this modification had NOT been made. (Still at least the Matrix DID tell us that it could not rotate the filters, rather than blindly charging ahead). The version of microcode in the Matrix we got with our Oxberry camera does not assert the "not ready for data" signal during the RESET sequence, and so we have to put in a software delay to allow the RESET to complete. Although this only occurs at the start of an exposure sequence (rather than for every frame) the delay needs to be about 40 seconds to allow time for initial calibration! An older Matrix that we borrowed as an evaluation unit did not show this problem, but that had an old version of the microcode that did not know about the extra commands added for the Oxberry. And finally : Dunn Instruments have just announced a new system - the DFR 8000. This allows for resolutions from 512x512 to 8Kx8K (including run-length encoding), several different interfaces (including offline use from 2400' 800/1600 bpi tape), and can expose an RGB 8Kx8K image in about 20 seconds! It does, however, cost $115K.