The official DANCAD3D (tm) Beta Test Web site.

Click on these thumbnails for related information about my current software, features, and IM support. Hold [Shift] and click [Reload] to refresh IM thumbnail.

If are helping "Beta Test" you should report bugs and feel free to ask questions about the program's commands, it is best to email me at tempnulbox (at) yahoo (dot) com and put "DANCAD3D (tm) 24x7 SUPPORT SUBMISSION" in the email subject line so your mail is not deleted as junk. See Section: 8 for more information about support related issues. I want you to ask questions so long as they are on the subject and relate to the current program's commands. If you do not get some kind of reply assume I did not get your message and resend it and also try voice mail, postal form, etc. All submissions and correspondence become the sole property of Daniel H. Hudgins to do with as he sees fit, so stay on subject.

DANCINES.TXT (tm) Copyright (C) 2008 All Rights Reserved. This document may not be reproduced or shared without first obtaining a written distribution agreement signed in ink by Daniel H. Hudgins. This file is Preliminary Documentation for DANCINES.EXE (tm) v1.00 BETA. This Document and the program DANCINES.EXE (sm) are governed by the End User License Agreement a.k.a. "EULA" located at the Author's web site www.DANCAD3D.com (sm). You must read and agree to the terms and conditions of the current "EULA" before using or testing this document or the programs. DANCINES.EXE (tm) is made available free of charge only for "Beta Testing" with the current revision of DANCAD3D.EXE (tm) and may not be distributed for other purposes. DANCINES.EXE (tm) uses the Borland (tm) files DPMI16BI.OVL and RTM.EXE to load the 16 bit code under DOS, and will not run without those files being in the same directory as DANCINES.EXE (tm). If you have problems with the program loading see file DPMIFIX.ZIP (tm) in DANCAD16.ZIP (tm), it may help you configure the memory on some older computers. Please do not re-distribute DANCINES.EXE (tm) or DANCINES.ZIP (tm), if you know someone who would like to help "Beta Test" them, please have them visit www.DANCAD3D.com (sm) and download the current version after they agree to the "EULA". You should also check that you are "Beta Testing" the current version. Trade marks mentioned belong to their respective trade mark holders. Information about products does not imply an endorsement or recommendation for purchase, you should make your own purchase decisions based on information taken from other sources including but not limited to the manufactures literature. DANCINES.EXE (tm) has been undergoing development and testing and is not finished yet, so features described in this document may not have been tested adequately or may not have been tested in the current revision and therefore may not work as described here. If you find any discrepancies between the program operation and the descriptions in this document please report them to the Author, instructions for reporting things are located in Section: 8 at www.DANCAD3D.com (sm). --- OVERVIEW: WHAT DOES DANCINES.EXE (tm) DO DANCINES.EXE (tm) is a computer utility used to automate the exposure of Digital Images of motion picture film negative or positive by triggering a high resolution Digital Camera to make frame shots of about 2K or 4k resolution. The numbering of the frame shot files is taken care of by the Digital Camera or the software that downloads or acquires the Digital Camera images. The Digital Camera used can be a DSLR such as the Canon XTi (tm), a high megapixel industrial CMOS monochrome camera that uses some kind of computer interface and image capture utility, or a USB camera like a Web CAM that has 1600x1200 pixel resolution and manual exposure. Because of the limited tonal range of CCD or CMOS Digital Cameras, DANCINES.EXE (tm) has a provision to shoot more than one exposure of each frame of Movie film, these "duplicate" exposures can have slightly different exposure to extend the tonal range of the Digital Camera or can have the same exposure and still result in improvement because of random noise in the images of the frames shot. The Fuse option of the Insert command in the Edit list command in current versions of DANCAD3D.EXE v3.7 (tm) and DANCAD87.EXE v3.7 (tm) would then be used to "Fuse" the group of exposures of each frame to make a single higher quality frame image, normally the Fused frame image would be saved as a 48bpp file to ensure that the high quality is available for color correction and filtering. DANCINES.EXE (tm) is a program to automate the operation of a home made motion picture film scanner. In order to be able to edit motion picture film in current DANCAD3D.EXE v3.7's (tm) Edit list as a Digital Intermediate you need high quality high resolution digital image files of each frame of your shot and processed motion picture film. Since you would make mistakes shooting the 300,000 or more frame images manually, DANCINES.EXE (tm) was designed to automate the task of advancing the stop motion film projector, shooting the images of the film frame, then advancing the film again. DANCINES.EXE (tm) also adds some more advanced features such as the ability to skip over frames in a film roll for bad takes or parts of a take that do not need to get scanned, and to be able to switch the color and brightness of the stop motion film projector light house to give the best exposure for movie film frames that may have been over or under exposed when shot in the movie camera. To get the best color it is better to use Red, Green, and Blue filters of lamps and make separate exposures with the Digital Camera for the Red, Green, and Blue pixels in the digital scan of the motion picture film. The Fuse command mentioned above would then be used to Fuse the separate Red, Green, and Blue exposures into a single 48bpp RGB digital image file, such as the *.NIP file formats supported by DANCAD3D.EXE v3.7. Because of the limited tonal range setting DANCINES.EXE (tm) to shoot three Red, three Green, and three Blue exposures, nine digital exposures for each movie film frame, should be used to improve the results. In fact if you have the time and computer power and space, better results might be obtained by exposing 16 exposures in the Red, Green, and Blue, but such over-sampling would slow down the film scanning process and take a great deal of computer power and space to process. After the RAW scans of the movie film have been Fused into a single 48bpp RGB file of good quality you could delete the RAW scan files to recover disk space. DANCINES.EXE (tm) also has a provision to shoot an IR or Infrared exposure, although this feature and how it will be used has not been fully worked out as of the initial release. The IR or Infrared exposure would be used when scanning color motion picture film to make images that isolate the dust and some other blemishes on the motion picture film as an aid to spot or remove dust from the RGB exposures. The color dyes in color motion picture film are mostly transparent to IR or Infrared light, so if you put an IR or Infrared filter over the light house, or use IR LEDs, the Digital Camera which has some IR sensitivity will make an image of everything IR opaque on the film other than the image. Since silver is opaque to IR light, if the color film has been "bleach bypass" or "skip bleach" the image cannot be "removed" in the IR exposure. If the bleach was exhausted in the film lab there may be a "ghost" image in the IR exposure since some of the silver did not get removed. And if you have Black and White film the IR exposure is not of much use since the IR exposure would look like the Red exposure just maybe a little softer and out of focus. The lens used to shoot the film needs to be stopped down since most lenses are not designed to focus IR light and will make an out of focus image if used wide open. If you re-focus the IR exposure the image will not be the same size as the RGB images, and would need to be re-sized to fit. DANCINES.EXE (tm) automates the movie film scanner and shoots the Digital frame images of each movie film frame, DANCAD3D.EXE (tm) is used to color correct, edit, mix sound, preview at 24fps, and output the edited frames order list. DANCINEL.EXE (tm) operates the film recorder to put the edited frame images scanned with DANCINES.EXE (tm), or obtained in some other way, onto motion picture print stock by controlling the 35mm animation camera to shoot off a monitor displaying the digital frame image usually by making separate Red, Green, and Blue exposures to get better results in the finished 35mm motion picture print. The source code for DANCINES.EXE (tm) was derived from an earlier program I was developing for in-house use to control an ACME step contact printer with a Bell and Howell dichroic additive light house. The idea was to zero cut the negative into A, B, C, and so on rolls as needed, have the software turn the light house light on and off as needed to avoid having to frame line cut the negative. The A roll would be printed the light would come on for the frames that where to print and off for the frames to skip. Then the B roll would be printed so that the light would be on for different frames on the print stock. Other rolls would be printed when there were conflicts between the A and B rolls to avoid cutting the negative close to an edit cut (shot change). This would allow the film print to be printed from the camera negative without destroying the negative by cutting on the frame line at the edit points thereby allowing edit changes later. The contact printer light and the ACME animation motor mounted in place of the continuous ACME sync motor where (are) controlled by pins of the computers parallel port going to solid state relays and such. Two problems arose with this setup, the first was that the shutter shaft and transmission in the ACME motor have too much play so the exposure frame to frame is not uniform enough and a constant flicker results in the print. The other issue is the print stock cannot be contact printed onto print stock because "Newton Rings" occur due to the surface of the films being too smooth and shiny. A third issue is that the light valves in the Bell and Howell light house are very old and do not always set the right light colors since their parts stick and fail to move on each command (I wonder how much film was wasted in film labs because of those cantankerous light valves...). To overcome the serious problems with contact printing it was decided to alter the code that became DANCINES.EXE (tm) to operate an Optical motion picture film printer made with parts from an Oxberry (tm) optical printer. This required two motor pulses to be sent from the parallel port, one to advance the projector's animation motor and another to advance the camera's animation motor. To get rid of the flicker a third motor hooked to the camera would be required to work as a capping shutter triggered by a cam on the cameras shutter shaft so that the capping shutter would open when the camera's shutter was also open and the capping shutter would close before the camera's shutter would start to close. The Optical Printer plan was put on indefinite hold because of problems with the Bell and Howell additive light house not working properly on every light change. Since DANCINES.EXE (tm) can operate the projector and camera advance, it was decided to go 100% Digital Intermediate, this solves most of the image quality issues, 1) Optical shooting of the film frames does not produce "Newton Rings". 2) The DSLR shutter does not flicker like the 35mm printer camera. 3) Color correction can be done in DANCAD3D.EXE (tm) rather than with the Bell and Howell additive light house for each shot change. So the light house can be set for "one light" printing and color changes made using DANCAD3D.EXE (tm) after the frames are shot. DANCINES.EXE (tm) can also make light house color changes from light numbers in the *.SCF cue file for the film roll being scanned, if needed. Also the dust and scratches can be re-touched manually if need be on the scans of each frame, something that is not practical for film to film printing. Also the digital color correction can compensate for over or under-exposed shots better, and can also correct for bad processing and foggy out of date film much better than film to film printing. The main drawback from going to a Digital Intermediate is the problem of storing the many TB of frame files, and the long times to process, backup, and move the many very large files. To keep as much quality as possible in the film scans so as not to lose too much quality over film to film printing the work-flow from DANCINES.EXE (tm) to DANCAD3D.EXE (tm) to DANCINEL.EXE (tm) is fully UN-COMPRESSED, this avoids double compression losses and other compression artifacts and provides a TRUE 24fps un-compressed work-flow system. Because DANCINES.EXE (tm) supports A, B, C, and so on, rolls it requires a *.SCF type Scan Cue File to tell it which frame numbers to expose or shoot and which frame numbers to skip over. You can look at the sample file DANCINES.SCF to see how the dot codes are used, the help screen in DANCINES.EXE (tm) lists the dot codes. The SCF file can be named to go with the roll you are scanning, and you can burn a CD of SCF files to keep with film rolls in case you need to re-scan the same roll at some point in the future. If you are editing the *.SCF files under a Windows (tm) OS you may want to rename them *.TXT since NOTEPAD.EXE (tm) and the folder display may not show the *.SCF file extension. Although DANCINES.EXE (tm) is primarily intended for use in automating a motion picture film scanner now, it might still be possible to control a step contact printer or step optical printer with it, if you have specific questions about the options of the commands in DANCINES.EXE (tm) you can contact me, see SECTION: 8 at http://www.DANCAD3D.com for support details, email may work best for such questions (make sure your subject line is clear since I get a great deal of junk mail and if you do not get a reply try again since your mail may have been deleted as junk by accident.) --- SYSTEM REQUIREMENTS Since DANCINES.EXE (tm) is a 16 bit program, its requirements about the same as what DANCAD3D.EXE (tm) v3.7 requires. OPERATING SYSTEM: DOS 6.22 or equivalent, with memory manager configured. Because this program requires close timing, even though it might run under W95OSR2 or W98SE it would be better to boot those in "Safe Mode" command prompt or with a system formatted floppy to reduce the noise in the CPU. To make a DOS95 or DOS98 floppy disk to boot your computer with use the DOS command FORMAT A: /S under those OS. FREEDOS (tm) may work as well with DANCINES.EXE (tm) if configured for the computer's memory. VIDEO BOARD: Should support CGA, EGA, or VGA 80x25 color text mode. SYSTEM MEMORY: At least 16MB of system memory may be required. 128MB or better may be required under Windows (tm) if any other programs are loaded. HARDDISK: A Harddisk is not required, a single 1.44MB floppy may be all that is needed if you can put DOS 6.22 and the program and other needed files all on one floppy disk. CPU: 80286 or better. There may be issues with CPU faster than 2.8GHz, but I do not have specific information as I write this. Please report such issues. MOUSE: Mouse support is not needed since many of the commands are by key-press. Under DOS a mouse driver would be needed to get the mouse to work in the CAD programs. PARALLEL PORT: LPT1, LPT2, or LPT3 can be used for the control signals. For PCI cards you may need the cards driver, for ISA cards no driver is usually required. If you are running under WIN98SE (tm) you would have installed the PCI Parallel Port driver already. Watch out for Windows (tm) (or other) resident printer drivers that capture the parallel port data, you may need to install a second parallel port and tell all the printer drivers not to use it. Do not use parallel port pass through devices, connect directly to the port pins. MONITOR: Monitor capable of CGA, EGA, VGA or better. DANCINES.EXE (tm) does not display graphics images of the frames shoot off the film in v1.00 since the RAW frame images need to be processed and color corrected before they can be viewed. If you wish to view the edited frame images at full resolution up to 2048x1536x32bpp or so use DANCINEL.EXE (tm) not DANCINES.EXE (tm), see the download file DANCINEL.ZIP (tm) and read DANCINEL.TXT for more information about displaying the edited frame images at full resolution. Depending on what digital camera you use to shoot with you may be able to use a third party program like DCRAW (tm) to convert the camera's RAW mode frame files into *.TIFF or *.TIF images you can look at with some third party image viewer. To view frame images at 2048x1536 with a third party image viewer you may need to set the Windows Desktop to 2048x1536 24bpp or "32bpp" resolution beforehand. The computer running DANCINES.EXE (tm) only needs a monitor showing 640x200 to 640x480 color 80x25 text mode. --- INSTALLING AND CONFIGURING DANCINES.EXE (tm) DANCINES.EXE (tm) is supplied in a zip file named DANCINES.ZIP (tm). In this zip file are these files: DANCINES.EXE Digital Cinema motion picture film scanner automation program. DANCINES.CFG Default configuration file, may be changed using menu. DANCINES.SCF Sample Scanner Cue File for testing your port interface. FPLUSF.SCF Sample SCF file using 35mm Feet plus Frames values. SMPTE.SCF Sample SCF file using SMPTE Time plus Frames values. SPERF.SCF Sample SCF file using audio Samples per Frame values. DANCINES.TXT Documentation for DANCINES.EXE (tm). DMPI16BI.OVL Needed by DANCINES.EXE (tm) for 16 bit memory use. RTM.EXE Needed program used by DANCINES.EXE (tm) to load. DCRAWBAT.COM Small program to make batch file to process raw scan files. Normally you would unzip these files directly into your DANCAD3D.EXE (tm) sub- directory and run them from there. If you need them in another sub-directory remember to change to that sub-directory before running DANCINES.EXE (tm). You need to change the program directory with the DOS CD command before you run the program. EXAMPLE: C:\>CD C:\DC37 C:\DC37>DANCINES DANCINES.EXE (tm) has its own configuration file named DANCINES.CFG. If you are updating or want to back up your current configuration use the COPY command at the DOS prompt to make a copy with another name, such as DANCINES.C01 and such. Future versions may use different configuration files, so if you copy an old configuration to use with a newer version some of the configure options may need to be manually configured. You should not copy a configuration file from a newer version for use with an older version since some of the command options may have been expanded and the older version may not work with those changes. --- IMAGE FORMAT AND DIGITAL CAMERA ISSUES The size and shape of the sensor should be kept in mind when selecting the Digital Camera for your motion picture film scanner. A camera like a Canon XTi (tm) has a sensor that is about 3:2 ratio, so if you scan 4:3 movie film you will need to crop of the sides of the digital frames shot, and if you scan your movie film for 16:9 you would need to crop off the top and bottom of the scanned frame digital images. If you have shot 'Scope anamorphic 35mm film you need to think about if it is better to optically unsqueeze the film image onto the digital sensor in your camera to use its full width, or to shoot the movie film images squeezed and use the sensor full height. If you keep the images squeezed 2:1 they will take less disk space, and not need to be re-squeezed for output in DANCINEL.EXE (tm) if you are going to make a 'Scope print. You can maybe turn down the monitor height if you are using an analog monitor to view the 'Scope images about in their wide screen format even though they are still in squeezed image files. DANCAD3D.EXE (tm) can crop, re-size, and squeeze or unsqueeze the scanned images as needed, it can also crop and letter box, so you do not need to have the film frame fully fill the digital camera sensor all the time. To get high quality from a monochrome camera you can scan at exact pixel dimensions by adjusting the scanner image lens to get the film image exact size on the sensor. When using a digital camera with a Bayer filter the OLPF filter degrades the resolution and the camera has Bayer filter artifacts so you will probably want the image larger than needed on the camera sensor and then down res the digital frame images after scanning to get the required resolution frame files for use in color correction, editing, and for output in the film recorder. --- FILM ISSUES The best color results might be obtained by making "monochrome" color separation exposures. Color separation exposures can be made in two forms, in the first you run the film scanner/printer to just shoot/expose the red images from all the frames, then just the green images for all the frames on another roll of film, then just the blue images for all the frames and end up with three sets of frame files, three exposed black and white film rolls to be used for printing in a step contact printer as A, B, and C rolls, or three exposures per frame on the same roll of color film. This makes a sequence like: RRR...R, GGG...G, BBB...B. The second separation method is to scan/expose the red then the green then the blue image frames in sequence as single roll/frame sequence three times as long, that roll/frame sequence would then be printed in an optical printer using a color filter wheel and print three exposures for each frame in the optical printer's camera advancing the camera once for each three advances of the projector, or have the three color frame files fused together into a single 48bpp RGB frame file with the Fuse command in DANCAD3D.EXE v3.7N (tm). This makes a sequence like: RGBRGBRGB...RGB. Using a stabilized power source for the scanner/printer Light House is recommended since variations in line (mains) voltage can produce flicker and density variations in the scanned/printed frames. Sequential separations can be scanned/printed using a dichroic or wratten filter wheel, this can be a very slow process since the motor needs to rotate the filter wheel. If Light House mode 3 is used, colored LEDs can be switched to the right color very quickly, so the Light House might switch on Red snap an exposure in the camera, then switch on Green snap an exposure in the camera, then switch on Blue and snap an exposure in the camera. To get higher quality in the Digital images of each frame you can snap several exposures of each color then use the Fuse command in DANCAD3D.EXE (tm) to fuse the color images together to get a better tonal range, the more exposures the better the color quality in the resulting scan generally, though at the expense of greater time and wear on the equipment. Three pass roll separations can be scanned/exposed in turn on (triple scanned/exposed by rewinding the film to scan/expose twice) a fixed pin movement in the scanner projector or printer using the appropriate separation filter (wratten 99, 98, or 70) in the printer Light House filter slot. The use of a narrow band absorbing filter to block yellow and cyan light from the white light Light House may also improve primary color saturation. The use of an ultra-violet and infrared blocking filter may help in printing from a tungsten light. Digital cameras are usually IR sensitive even if they have some kind of IR filter built in, so a extra IR filter may help, especially if other filters are used in the light path since filters that block visible light may also transmit IR. The color dyes used in color motion picture film pass Infrared or IR light which Digital CMOS or CCD cameras can be sensitive to which will de-saturate colors or cause colors to be off balance, so heat absorbing glass and interference filters such as hot or cold mirrors should be used if the light source emits Infrared or IR light for the Red, Green, and Blue exposures. For the Infrared or IR exposure some Infrared or IR light would be needed to make an Infrared or IR exposure. Making the IR exposure as a forth pass of the roll to scan is not a good idea since the dust would be in different places, it is better to scan all four colors while the film frame is in the projector gate. A wratten 2E filter can block UV if you are using DANCINES.EXE (tm) to operate a film to film printer such as a step contact printer or optical printer for zero cut printing from A-B rolls and such. DANCINES.EXE (tm) can be used to shoot images that are Mirror (flipped) images in the scanner, these can later be Mirror (flipped) back in the Insert command of DANCAD3D.EXE (tm)'s Edit list to come out the right way around. Generally you would want to scan the camera film going through the scanner film projector from bottom to top, but you would also want the film emulsion on the camera lens side and the film base (cell) on the Light House side, so if you are scanning rolls that are prints and such the image may be reversed side to side even when right side up in the projector. Since marks, scratches, and dirt on the film base can degrade the image of the emulsion in the scanner you would want to scan the emulsion side, always scanning the emulsion side also helps reduce the number of times you need to re-focus the scanner lens since the thickness of the film base would make the images out of focus. Be careful when you focus your scanner that the constant heat from the light house has not made the film pucker and cause you to set the focus off of where it should be when cool frames are loading into the projector gate while the scanner is scanning. Always advance a few test frames and re-check the focus to make sure that the scanner is focused all right for un-buckled frames. --- ADJUSTING EXPOSURE IN THE SCANNER OR PRINTER Keep in mind that ND filters may pass some IR light, and so using heavy ND or other gel filters when scanning to a Digital Camera may require additional IR blocking filters. Because Digital Cameras that use a Bayer filter on their sensor may be balanced for daylight 5500K the light used in the scanner may need to be extra blue green to compensate not only for the Daylight balance of the sensor's Bayer filters but also for the orange mask tint of the color negative or master positive film stock. Although the Digital Camera may have a auto white balance or 3200K setting, using those may result in increased noise in the blue pixels because the blue gain would be increased when in 3200K setting over the 5500K setting. To get the lowest noise the EI setting of the Digital Camera should be set to one of the lowest EI values, like EI 100, and the color temperature set to the natural balance of the Bayer filters blue to red balance. Then use colored filters or adjust the color of the light house to get a neutral balance rather than use the camera's adjustments for color balance. If you are scanning a color negative shot of a standard 18% Gray card shot under the right color light, your digital image of that film should look neutral gray when viewed and read equal values with the digital densitometer in DANCAD3D.EXE (tm)'s Grading command (in the Edit list). If the Red, Green, and Blue histograms are unequal or the gray value is unbalanced, adjust the color of the light used for scanning with CC filters or by adjusting the voltage to the LEDs. If you have the vane values from the SCF file automatically adjusting the Red, Green, and Blue LED brightness in the light house (type 3) you can adjust the values in the SCF file to bring the scans to neutral color balance. You should expect motion picture lab consistency to produce some variation from day to day so tests of exactly the same studio setup run through the cine film scanner may not match. On the head of each roll of motion picture film you shoot and you send to the lab you should shoot several frames of a gray card and gray step card that you know how they look, so that you can figure out if the timing errors are do to some mistake you made, or the lab processing. There are some variations in film stocks from one emulsion batch to another. Film manufactures mark their film cans with the emulsion batch number so you can use rolls from the same batch to reduce variation when you change from film shot on one roll to another roll. Processing variation is usually greater than film variation. Lab films such as print stock do not have a fixed color balance, so rolls you purchase one year will probably be slightly different from rolls purchased another time. You need to keep notes about what CC filters or other timing changes you are making for a given emulsion number. Film color balance changes as it ages, so you should do tests if the film has been sitting around for a while in a warm place. Film should be kept in a dry cool place, but freezing it may produce spots from ice crystals, try 40F and 50%RH or follow the manufactures recommendations. Film can develop shiny spots if kept in a damp place for prolonged periods, these make an uneven density to the image and such film is not much good for anything, you can see the shiny spots if you look at the film by reflected light, they are generally larger than several mm and can cover most of the film when very bad. Shiny spotted film may also make a ticking or clicking sound when unwound from one roll to another on rewinds in the dark. You should keep your processed film in a humidity controlled room to avoid getting shiny spots, which can happen both before and after processing. If the light house is hooked-up to change the amount of light it puts out according to the light number vane values you entered in the *.SCF file for the roll being scanned or printed you can do some scene to scene color correction with the light house automatically, otherwise you can insert console messages into the *.SCF file with the .ms dot command that tell you what CC filters to insert into the light path. See the information below about the light house hook-up options. If the film you are going to scan is consistent you may be able to adjust the light house once, then scan the frame images and do all the color balance and correction using DANCAD3D.EXE (tm). --- SPEAKER ISSUES DANCINES.EXE (tm) can produce sound effects for each operation that give you an audio cue as to what the scanner is doing. You can turn off the sound effects with an option in the config you get in the DANCINES.EXE (tm) main menu. Since the sounds come out of the computer's small speaker the Windows (tm) volume control will probably not affect the volume of the sounds made. Also if your computer's case did not have the small speaker installed, you will not get sounds even if you have configured them on since there is no hardware speaker, DANCINES.EXE (tm) does not use the sound board output, just the hardware speaker inside the computer's case. --- PARALLEL PORT CONNECTIONS The pin numbers referred to here are the pins on the 25 pin parallel port connector on the parallel port card, not the pins on the 32 pin end of the printer cable. Because DANCINES.EXE (tm) can be used with more than one type of light house, and can be used as a step film scanner, step optical printer, or step contact printer the use and meaning of some of the Parallel Port pins can vary with some pins having different use when different modes are set in the configuration file, so if signals do not seem correct check the configuration settings. Do not run DANCINES.EXE (tm) on a computer that has a Windows (tm) printer driver installed, since some printer drivers loaded resident into Windows (tm) can capture signals going to the Parallel Port or coming back from the Parallel Port and prevent DANCINES.EXE (tm) or your circuits and devices from operating properly. It is best to run DANCINES.EXE (tm) on a clean DOS 6.22 computer system. The parallel port connections are similar to those used by DANCAM.EXE (tm) and DANPLOT.EXE (tm), see SECTION: 3.2.40 at www.DANCAD3D.com (sm), but are used in different ways. The chief difference is that there were not enough output pins to control the Bell and Howell additive light house, so the six lower bit pins of the output data byte were used for two functions, those six bits being stored in two latch memory chips to get twelve outputs from the eight data bits, that is the lower 6 bits are stored in two six bit latches, and the upper two data bits are used to select the latch the lower six bits are to be stored in. Type D flop flops can be used to latch the twelve latched data bits. The other output bits and input bits are not latched in external chips. The input pins should be pulled up to a +5 volt DC supply through a 2.2K ohm 1/4 watt resistor. Because of EMI/RFI you should low pass the data set latch pins of the D type flip flops and the six data lines going into the latch chips with 0.1mf through 100 Ohms. If you do not do this the two latchs may not operate reliably, and reset at random times with random values causing harm or havoc. Because the meaning of the pins on the parallel port change depending on how you have DANCINES.EXE (tm) configured, mostly regarding the light house type selected, more than one pin out is shown below, be sure you understand which pin out your scanner or printer should be using, and have DANCINES.EXE (tm) configured and running before you turn on any power to your scanner/printer. In other words run DANCINES.EXE (tm) from the DOS prompt and wait until its main menu is displayed on your computer monitor before you turn on power to any part of your movie film scanner/printer. If you turn the power on before you have the program running the relays may be in the wrong state and could cause the scanner projector to start running or the camera to shoot. The computer boots with the printer port bits in some random or undesirable state, so the DANCINES.EXE (tm) must be run in order to clear the parallel port bits to set a stable state to the pins so that when power is turned on the relays will, hopefully, be in their ready or standby state. You can configure DANCINES.EXE (tm) to use any of the three LPT parallel printer ports for its basic I/O, but LPT1 would be the usual choice since many computers only have one parallel port. On the parallel port there are three bytes that the computer can access, two for output and one for input. These will be called DO for Data Out, CO for Control out, and CI for Control In. The pins on the 25 pin parallel connector associated with the port byte will be indicated as a number after the port two letter name, i.e. DO02 or IO10 and such. Other pins on the parallel port are not used or serve as a common ground connection. The data bits used to read and write to and from the parallel port are: Pin No. 13 * CI13 Control Input Bit 4 of 3 of 7 TIE TO COMMON - 25 12 - CI12 Control Input Bit 5 of 3 of 7 TIE TO COMMON - 24 11 - CI11 Control Input Bit 7 of 3 of 7 TIE TO COMMON - 23 10 - CI10 Control Input Bit 6 of 3 to 7 TIE TO COMMON - 22 09 - DO09 Data Out Bit 7 of 0 to 7 TIE TO COMMON - 21 08 - DO08 Data Out Bit 6 of 0 to 7 TIE TO COMMON - 20 07 - DO07 Data Out Bit 5 of 0 to 7 TIE TO COMMON - 19 06 - DO06 Data Out Bit 4 of 0 to 7 TIE TO COMMON - 18 05 - DO05 Data Out Bit 3 of 0 to 7 17 04 - DO04 Data Out Bit 2 of 0 to 7 CO16 Control Out Bit 2 of 0 to 4 - 16 03 - DO03 Data Out Bit 1 of 0 to 7 CI15 Control Input Bit 3 of 3to7 - 15 02 - DO02 Data Out Bit 0 of 0 to 7 CO14 Control Out Bit 1 of 0 to 4 - 14 01 - CO01 Control Out Bit 0 of 0 to 4 * Pin 13 is used for both Output and Input, Output Control Bit 3 of 0 to 4, and Input Bit 4 of 3 to 7, so it is both CO13 and CI13. The 8 data bits are marked 0 to 7, their data value for I/O bit masking is: Bit 0 of 7 = 0 or 1 Bit 1 of 7 = 0 or 2 Bit 2 of 7 = 0 or 4 Bit 3 of 7 = 0 or 8 Bit 4 of 7 = 0 or 16 Bit 5 of 7 = 0 or 32 Bit 6 of 7 = 0 or 64 Bit 7 of 7 = 0 or 128 The data values for the three bytes used to read and write to the parallel port pins are: Pin No. 13 * CI13 Control Input = 16 (Out = 8) TIE TO COMMON - 25 12 - CI12 Control Input = 32 TIE TO COMMON - 24 11 - CI11 Control Input = 128 TIE TO COMMON - 23 10 - CI10 Control Input = 64 TIE TO COMMON - 22 09 - DO09 Data Out = 128 TIE TO COMMON - 21 08 - DO08 Data Out = 64 TIE TO COMMON - 20 07 - DO07 Data Out = 32 TIE TO COMMON - 19 06 - DO06 Data Out = 16 TIE TO COMMON - 18 05 - DO05 Data Out = 8 17 04 - DO04 Data Out = 4 CO16 Control Out = 4 - 16 03 - DO03 Data Out = 2 CI15 Control Input = 8 - 15 02 - DO02 Data Out = 1 CO14 Control Out = 2 - 14 01 - CO01 Control Out = 1 * Pin 13 is used for both Output and Input, Output Control Bit 3 of 0 to 4 (=8), and Input Bit 4 of 3 to 7 (=16), so it is both CO13 and CI13. The pins of the parallel port output about +3v to +3.5v without a pull-up resistor to a +5 volt supply. This is enough to control some solid state relays that have a rating of +3v input. Although the input pins tend to float high, they should be pulled-up through a 2.2K ohm 1/4 watt resistor to a +5 volt supply. It should be all right to pull-up all the active data pins on the parallel port through 2.2K ohm 1/4 watt resistors to ensure both logic high and logic low voltages. The meaning of the "Data Out" pins changes depending on the type of light house used in the scanner or printer. Also the "Data Out" pins need to be wired to D-flip-flop type latches to hold the data value of 6 of the data bits. Two 6 bit latches are required for the data out bits to latch, the other two data out bits (of the 8 bits total on the parallel port) are the latch strobe for the two six bit latches. In other words, the 6 data bits go to the input pins of the two 6 bit latches (three 4 bit chips) and the 7th data bit goes to the set pin on one of the two latches, and the 8th data bit goes to the other of the two latches. The set pins on the two latches are never active at the same time, so they control which latch will hold the current value on the 6 lower data bits when the 7th or 8th bit is strobed by DANCINES.EXE (tm). This gives you a total of 12 data output bits that can be set 6 at a time, that is one half can change at a time. The 12 data output bits from the latches go to relays or solid state relays to control the lighthouse and or scanner/printer along with the other output pins on the parallel port that are not latched with external chips. Each latch chip has four type "D" flip flops, so you need three chips to latch 12 bits (3x4=12). In DANCINES.EXE v1.00 four types of light house are supported: Light House 0, this is for a single lamp light house probably using CC filters. Light House 1, Bell and Howell additive using automatic light valves. Light House 2, Bell and Howell additive manual vane type. Light House 3, Colored LEDs using relays and or D2A converters. Light House 3 mode can use other types of colored lamps or solenoid shutters over colored lamps to get more light than you could get from LEDs, but since LEDs switch on and off fast they are the intended type of lamp for that light house mode. The information below goes into more detail about how the pins on the Parallel Port are used for each light house type. Future revisions may alter this hook- up information or allow for additional connections, please report any issues you find with this hook-up information. The pins used to read and write to the parallel port pins are for Light House type 0 (Single White lamp) are: Pin No. 13 * CI13 - Not Used TIE TO COMMON - 25 12 - CI12 - Not Used TIE TO COMMON - 24 11 - CI11 - Not Used TIE TO COMMON - 23 10 - CI10 - Hold While Not Ready TIE TO COMMON - 22 09 - DO09 - Not Used TIE TO COMMON - 21 08 - DO08 - Not Used TIE TO COMMON - 20 07 - DO07 - Not Used TIE TO COMMON - 19 06 - DO06 - Not Used TIE TO COMMON - 18 05 - DO05 - Not Used 17 04 - DO04 - Not Used CO16 - Light House Lamp On/Off - 16 03 - DO03 - Not Used CI15 - Not Used - 15 02 - DO02 - Not Used CO14 - Camera Shoot Expose Pulse - 14 01 - CO01 - Projector Advance Pulse So you can see from this that you only need to use CO01 and CO14 signals plus ground to run your scanner if you are sure that the camera will always be ready to shoot or expose. To save on the Light House lamp life you can wire a solid state relay to CO16 to have the Light House lamp turned On and Off as needed. If you are going to use DANCINES.EXE (tm) to do zero cut printing in a contact printer or optical printer where both the projector and camera advance one frame regardless of if that frame is to be printed, such as in A-B rolling, you need to wire up the Light House lamp relay so that the light can be blocked for the frames that are not to be printed. Also when printing A-B roll the room the printer is operated in should be dark to avoid weak ghost exposures, that is why their is a delay after the lamp goes dark, because the high wattage lamps take a few seconds to go dark. The CI10 hold not ready input is used to hold the program until the red light on the back of a DSLR goes out to make sure that the next camera shot expose pulse will be clear to make another frame image. Since the red light on the back of the DSLR does not stay on while the DSLR is busy there is a delay and digital filter used to try to keep the program in hold until the DSLR is finished saving the last exposure frame and is really ready. Because the DSLR red busy light is not always on while the DSLR is busy some extra delay needs to be added in the program configuration so that you do not lose frames or exposures, i.e. that the camera is always ready for the shoot expose pulse. Solid state relays can be used on the CO01 Projector Advance and CO16 Light House Lamp On/Off signal pins to amplify the signals to switch 120/240VAC directly, or DC as needed. For inductive loads such as the Projector Animation Motor you should use a Solid State relay with a rating of 240 Volts at 30 Amps even though the motor does not need that to reduce the impact of the inductive spikes caused by switching the projector motor on and off hundreds of thousands of times. Because you do not know the voltage used in the DSLR camera's remote trigger switch you should use a solid state relay to drive a mechanical relay to open and close the mechanical relay contacts for the DSLR. Because DSLR sometimes have two switch contacts that close in the order: both open then ring closed tip open then tip and ring closed then ring closed tip open then both open, you may need to wire up two mechanical relays with some delays and logic to make sure that the ring and tip on the DSLR remote trigger input operates as was intended. If your camera on the scanner is shooting by pressing a key on the computer's keyboard or mouse, you may need to wire a mechanical relay across the contacts on the keyboard or mouse so that the additional computer running DANCINES.EXE can trigger the shoot key to make the cameras software shoot a frame exposure and save it to the computer's harddrive through the USB cable or whatever is being used. This method could be used to wire to the shoot contacts on a Web CAM to shoot frames as well, maybe for doing a fast workprint scan. The pins used to read and write to the parallel port pins are for Light House types 1 and 2 (Bell and Howell additive light house) are: Pin No. 13 * CI13 - Not Used TIE TO COMMON - 25 12 - CI12 - Not Used TIE TO COMMON - 24 11 - CI11 - Not Used TIE TO COMMON - 23 10 - CI10 - Hold While Not Ready TIE TO COMMON - 22 09 - DO09 - Latch Brightness Pulse TIE TO COMMON - 21 08 - DO08 - Latch Relay Pulse TIE TO COMMON - 20 07 - DO07 - Blue or 20 Vane TIE TO COMMON - 19 06 - DO06 - Green or 16 Vane TIE TO COMMON - 18 05 - DO05 - Red or 8 Vane 17 04 - DO04 - Unlatch or 4 Vane CO16 - Light House Lamp On/Off - 16 03 - DO03 - Zero Close or 2 Vane CI15 - Not Used - 15 02 - DO02 - Vane Select or 1 Vane CO14 - Camera Shoot Expose Pulse - 14 01 - CO01 - Projector Advance Pulse Note that this is mostly the same as for Light House type 0 except that the Data Output pins DO02 through DO09 are used to set the light valves in the Bell and Howell light house. Note the unusual brightness code used, values from 0 to 31 are Binary, but since the next bit codes for 20 rather than 32 light vane values above 31 follow a special sequence of bits that use 20 plus some of the lesser bits in the 6 bit brightness value for the vane opening in the three light valves. For Manual Bell and Howell additive light houses, the program should stop for light changes and display the vane settings that you should set the knobs on the light house to manually. In the config you select light house type 2 to have the program stop and tell you the vane values read from the *.SCF file. The lighthouse trim values would normally be set before the *.SCF cue file is run in DANCINES.EXE to compensate for the normal lab or stock variation, the trim values need to be written on a piece of paper and kept with the roll of film so that you can reset the light house trims to the same value if you need to re-scan or re-expose later. For automatic operation 12 solid state DC relays are used to control the solenoids in the light valves and to select the light valve to operate. Nine relays are used for the light valve's nine solenoids, D0 to D5 plus Unlatch, Zero Close, and Vane Select. The three other solid state relays are used in series with the ground on each light valve so that the Red, Green, or Blue light valve can be selected. Since the nine solenoid wires from all three light valves are wired in parallel to the nine solid state relays that control them, DC blocking rectifiers and reverse diodes need to be used to keep the spikes and stray signals from going into the light valve that is not grounded by the Red, Green, or Blue select solid state relays. High voltage 0.1mf ceramic capacitors should also be arranged across the solid state relays along with reverse diodes to bypass the high voltage spikes coming out of the light valve solenoids when the power is switched on and off. The pins used to read and write to the parallel port pins are for Light House type 3 (Colored LEDs or maybe filter wheel) are: Pin No. 13 * CI13 - Not Used TIE TO COMMON - 25 12 - CI12 - Not Used TIE TO COMMON - 24 11 - CI11 - Not Used TIE TO COMMON - 23 10 - CI10 - Hold While Not Ready TIE TO COMMON - 22 09 - DO09 - Latch Brightness Pulse TIE TO COMMON - 21 08 - DO08 - Latch Relay Pulse TIE TO COMMON - 20 07 - DO07 - Blue or 32 Vane TIE TO COMMON - 19 06 - DO06 - Green or 16 Vane TIE TO COMMON - 18 05 - DO05 - Red or 8 Vane 17 04 - DO04 - IR or 4 Vane CO16 - Light House Lamp On/Off - 16 03 - DO03 - Not Used or 2 Vane CI15 - Not Used - 15 02 - DO02 - Not Used or 1 Vane CO14 - Camera Shoot Expose Pulse - 14 01 - CO01 - Projector Advance Pulse The pin connections for Light House type 3 are like those of type 1 except that DO04 is used for IR lamp relay rather than Unlatch, DO02 and DO03 are not used for relays only brightness, and the brightness range for the Vane settings is 0 to 63 rather than 0 to 51. If the D6 value was kept 20 rather than 32 a look up table would be required before the brightness data was sent into the D2A converter to control the voltage of the LEDs used for each color. Because the brightness range is linear Binary 0 to 63, that does not correspond to the Log type brightness range of the Bell and Howell additive light house. If you want the brightness values of the LED output to correspond to the Bell and Howell light valves you would need to burn a 16 bit EPROM for a look up table and have that drive a 16 bit D2A converter. For simple LED brightness control you only need a 6 bit D2A or to use the upper 6 bits of an 8 bit D2A with the lower two bits wired low. The output of the D2A can be scaled with resistors (a pot) and feed through a power darlington wired as an emitter follower to give automatic voltage control to the voltage feed to each LED color. To save on parts you might be able to use one D2A and use the color relay lines to switch the ground leads like was done to save on relays for the Bell and Howell light house, i.e. one D2A, one power emitter follower, and three (R,G,B) or four (IR,R,G,B) ground relays. It may be possible to wire up stop switches on a filter wheel and use Light House type Mode 3 by connecting relays to the IR, Red, Green, and Blue color select relays such that the filter wheel rotates until one of four micro switches trips on a notch in the outside edge of the filter wheel. By switching to the next of four switches on the color change the filter wheel will rotate one color. Since the location of each switch relates to a specific color filter in the wheel the filter wheel will rotate until the right color is selected if enough time is allowed for a full rotation of the filter wheel before the next exposure is made. In other words, one of the four color select relays connects one of the four micro switches spaced 90 degrees around the color filter wheel, and that micro switch passes power to the motor that rotates the filter wheel until the switch opens when the single notch on the outer edge of the filter wheel rotates to be at the selected one of four micro switches. If only Red, Green, and Blue filters are used then the filter wheel could be surrounded by three micro switches spaced 120 degrees. If only IR and White filters are used then the micro switches could be spaced 180 degrees. The hold pin on the parallel port should be held in hold state while the filter wheel motor is powered on rotating the filter wheel so that the camera on the scanner would wait if need be. Another configuration possible is to change the brightness of a white LED or other voltage or current controllable light source by using the Green vane value in the *.SCF file (rather than the average of the Red, Green, and Blue values). The Parallel Port Data Byte mask values for the Latched bits are: For White light house (type 0): The bits masks 1 through 128 are undefined for type 0 light house, and may change so should be ignored by the scanner/printer circuits in this case. For Bell and Howell additive light house (type 1 and 2): 1 = Vane select (moves light vanes to current vane setting) 2 = Zero close (moves light vane setting to smallest opening) 4 = Unlatch (lets light vane data keys move to new setting) 8 = Red select relay (not part of light valve, just ground for Red valve) 16 = Green select relay (not part of light valve, just ground for Green valve) 32 = Blue select relay (not part of light valve, just ground for Blue valve) 64 = Store control bits listed above in 6 bits of 12 bit latch circuit. 128 = Store brightness data for vane in other 6 bits of 12 bit latch circuit. The bit masks for the relay latch in the Bell and Howell light valve are: 1 = Vane select, Pin 2 latched by Pin 8. 2 = Zero close, Pin 3 latched by Pin 8. 3 = Unlatch, Pin 4 latched by Pin 8. 8 = Red select relay, Pin 5 latched by Pin 8. 16 = Green select relay, Pin 6 latched by Pin 8. 32 = Blue select relay, Pin 7 latched by Pin 8. The bits masks for the brightness latch in Bell and Howell light valve are: 1 = +1 value, Pin 2 latched by Pin 9. 2 = +2 value, Pin 3 latched by Pin 9. 4 = +4 value, Pin 4 latched by Pin 9. 8 = +8 value, Pin 5 latched by Pin 9. 16 = +16 value, Pin 6 latched by Pin 9. 32 = +20 value, Pin 7 latched by Pin 9. Note odd counting system 32 bit equals +20 value, not like binary. For LED or Filter Wheel type light House (type 3): 1 = Undefined for DANCANES.EXE v1.00 2 = Undefined for DANCINES.EXE v1.00 4 = IR select relay signal, Pin 4 latched by Pin 8. 8 = Red select relay signal, Pin 5 latched by Pin 8. 16 = Green select relay signal, Pin 6 latched by Pin 8. 32 = Blue select relay signal, Pin 7 latched by Pin 8. 64 = Strobe pulse to latch the color select signals above. 128 = Strobe pulse to latch the 6 data bits above in the brightness latch. The bit masks for the relay latch in the type 3 Light House are: 1 = Undefined in v1.00, Pin 2 latched by Pin 8. 2 = Undefined in v1.00, Pin 3 latched by Pin 8. 3 = IR select, Pin 4 latched by Pin 8. 8 = Red select relay, Pin 5 latched by Pin 8. 16 = Green select relay, Pin 6 latched by Pin 8. 32 = Blue select relay, Pin 7 latched by Pin 8. The bits masks for the brightness data for D2A in the type 3 Light House are: 1 = +1 value, Pin 2 latched by Pin 9. 2 = +2 value, Pin 3 latched by Pin 9. 4 = +4 value, Pin 4 latched by Pin 9. 8 = +8 value, Pin 5 latched by Pin 9. 16 = +16 value, Pin 6 latched by Pin 9. 32 = +32 value, Pin 7 latched by Pin 9. Note the 32 bit is not the same for Bell and Howell light house, for the D2A it is simple binary counting from 0 to 63 for the vane value. The vane value can be put through a hardware look up table to get an "exact" 16 bit brightness value for Log and other brightness curves to compensate for the type of lamp used in the light house. The way the latches are used is that the lower 6 bits of the Data Output a.k.a. DO byte on Pins 2 through 7 are set all together with the values to be latched, then a strobe pulse is sent to Pin 8 or Pin 9, to Pin 8 if the particular lighthouse relays are to be changed, or Pin 9 if the light house brightness or vane value is to be changed. The Bell and Howell light valve key values (that set the vane plus trim values to the light vanes) may be inverse so the values in the 6 vane value data bits may also be inverse, ones equal to LOW and zeros equal to HIGH. The common output signals used for the scanner/printer that do not require a latch circuit are on the Control Output Port byte: 1 = Projector motor advance pulse, port Pin 1. 2 = Camera Shoot pulse, port Pin 14. 4 = Light House Lamp ON or Lamp OFF relay state, Pin 16. 8 = Undefined in v1.00 16 = Undefined in v1.00 32 = NC 64 = NC 128 = NC Bit masks 1, 2, and 8 are set to default to logic 1 to output TTL LOW on the output pins when the program boots since those pins have a hardware invert. Bit masks 4 and 16 default to logic 0 since those pins on the TTL output are not inverted on program boot. Because the state of all of the output pins of the parallel port is not safe or correct, you should not turn on the scanner circuits or motor power until after DANCINES.EXE (tm) has booted and the main menu has finished loading. The common input signals used for the scanner/printer that do not require a latch circuit on the Control Input Port byte are: 1 = NC 2 = NC 4 = NC 8 = Undefined in v1.00 16 = Undefined in v1.00 32 = Undefined in v1.00 64 = Camera not ready when HI, ready when LOW, Pin 10. 128 = Undefined in v1.00 If you find any errors in this hook up information for DANCINES.EXE (tm) be sure to let me know about them, this information is preliminary and subject to change without notice of any kind. The input pin 10 may need to be pulled up to a 5 volt supply through a 2.2K ohm 0.25 watt resistor. A micro switch or relay can be used to short pin 10 to ground in order to signal the computer. You could also use a 7414 or other 74xx series TTL IC to buffer the signal going to pin 10. If you do not need to use pin 10 to hold the computer for camera ready, you can disable the pin 10 checking in the General configuration options of DANCINES.EXE (tm). To ensure good signals both the input and output pins used should be pulled up to +5 volts through 2.2K ohm 0.25 watt resistors. Pins 18 through 25 are used as common ground. The +5 volt supply should be isolated, three "D" cells in series may work for a while since +4.5 volts will read as logic HIGH on the parallel port, but do not use more than +5 volts since that may damage the parallel port. Making connections to your parallel port can damage your computer's mother board, so you may wish to install a second parallel port card board and use that card's 25 pin connector for the port to connect to, that way if you destroy that port board you can replace it without replacing your mother board (provided the damage did not extend beyond the port board and damage the mother board.) You can add opto-isolators or TTL buffer chips between your parallel port and the film scanner/printer circuits if you want more protection to your computer. Be careful that all power supplies are isolated since you do not want AC line current being accidentally sent into your computer through the parallel port. It may be a good (or bad) idea to plug your computer into an AC isolation transformer rather than the wall to have the computer ground isolated. See configuration commands in the DANCINES.EXE (tm) Main Menu for information about the values that affect the Parallel Port pins. The delay values are approximate, so when the camera exposure is controlled by the computer, such as when printing film-to-film, it may be better to control the exposure with a capping shutter operated from a crystal timer. Variations in exposure can cause flicker in the scanned or printed film at sound speed. Driving cine cameras one frame at a time can produce flicker because of backlash in the shutter shaft and animation motor, these backlash problems do not show up much if the camera runs continuously, only when it is made to start and stop. A capping shutter can be used after the camera shutter has opened in this way, close the capping shutter, advance the cine camera until its shutter is open, stop the cine camera, open the capping shutter, time the exposure with a crystal timer, close the capping shutter, advance the cine camera until its shutter is closed, stop the cine camera, set pin 10 for clear to change the image frame to scan/print, and so on. When using a CMOS or CCD camera and the exposure is controlled inside the camera, DANCINES.EXE (tm) only needs to be configured so that it holds the exposure pulse and post exposure pulse delay long enough for the CMOS or CCD camera to finish each exposure and saving the exposures, that is delay errors from DANCINES.EXE (tm) do not matter as long as they are too long when a CMOS or CCD camera is used with internal exposure control. There is an option to use pin 10 as a hold for the program to keep the program from getting ahead of the cine projector or scanner/printer camera i.e. closed loop, if this option is not used be sure you give everything plenty of time to settle before the next frame image is advanced in the projector. If you set the delays too long that generally just makes the scanning/printing total time longer, but if you set the delays too short you may get errors, bad frame images, frames not saved, or skipped frames. --- IMAGE FRAME FILE CONSIDERATIONS For editing images that are scanned or taken from nature it is best to use a file format that stores the brightness values as 48 bpp even though the noise from the sensor will make the color resolution less than that. Because the image contrast in the mid-tones of the image will normally have its contrast increased storing images in a single 24bpp frame image will result in gaps in the tonal range, for instance if mid-tone contrast is doubled then only every other image tone would be present doubling the stair-stepping of the image tones, and since 256 image tones per primary color is just barely adequate decreasing that to 128 image tones may result in a visible degradation of the frame images. To reduce the increase in stair-stepping of the image tones when the contrast, gamma, curves, and color balance are adjusted with DANCAD3D.EXE (tm)'s Grading commands more than one 24bpp scanned image of each frame can be Fused together to increase the color bits. For instance two 8 bit images fused give about 9 bits of color, then doubling the contrast in the mid tones would result in that part of the curve having about 8 bits rather than 7 bits resolution. So it is best to make three or more exposures of each frame with the Digital Camera, then use the Fuse option in the Edit list Insert command to store the frame images in a 48bpp file format such as the NIP file format. Other frame file type options may be introduced later. The file numbering format is that used by DANCAD3D.EXE (tm). The Short format ranges from SOMEFILE.-99 to SOMEFILE.999. Since Windows (tm) programs like the file extension BMP on BMP files they will not read files numbered using the Short format. The Long frame format ranges from -9999999.BMP to 99999999.BMP, but the OS will not let you store that many files in the same sub-directory, so in practice each 35mm 2000 foot reel's frames would need to be in a separate sub-directory, so you would number Long format frames 0.BMP through 32000.BMP. If you want the whole 88 minute or longer film to have its frames consecutively numbered you can use the Divided numbered name format, it breaks the frames into groups of about a thousand frames per sub-directory and makes a series of numbered sub-directories to hold the frame sets in. When you enter a Divided name in you use the Long format to the master directory that the Divided sub- directories are in, not the path to the actual files, e.g. if the Divided files start with C:\FRAMES\BMP0\1.BMP you would enter C:\FRAMES\1.BMP at the prompt for the BMP filename. The Padded file naming format is like the Long format but the filename numbers are Padded to the left with zeros, that is useful in two ways, Windows (tm) directories sort the Long format out of order, Padding with leading zeros helps get the folder directory to display sorted in order, and also when doing drag-n'-drop into some Windows (tm) programs of frame sets the Padded format may be more likely to have the frames loaded in the right order since if you are making an AVI file or something the Long format frames may get sequenced out of order whereas the Padded format frame images should load in the right order, perhaps. There are utilities in the Files Utilities sub-menu in DANCAD3D.EXE (tm) to convert numbered filenames used by other programs into the Short, Long, Padded, and Divided numbered filename formats used by my programs. Likewise those utilities can do some backwards name conversions as well, in case you need to edit frames in my numbered name format in a program that uses a different numbering scheme. Most Digital Cameras or image capture programs used with Digital Cameras can save shot image frames in some kind of a numbered format like, IMG_4567.RAW or IMG_8765.CR2 and such. To convert such mixed type numbered filenames into one of my numbered file name types you can use the commands in DANCAD3D.EXE (tm)'s Files Utilities sub-menu. For use with the Edit list Insert command most frame sets should be renumbered to start with 0, so if you scan frames IMG_4567.RAW to IMG_5567.RAW those would be renamed and renumbered to 00000000.RAW to 00000999.RAW since there were 1000 files scanned. The utility program DCRAWBAT.COM (tm) included in the DANCINES.ZIP (tm) archive can do the renumbering when the X.BAT batch file it makes is executed to convert Canon *.CR2 files into *.TIF files using the third party program DCRAW (tm). You may need to use some third party programs to convert RAW scans from your Digital Camera into BMP or whatever my programs can read. You can use www.Google.com to find download links for Freeware image conversion programs. Three programs you can download from the Internet you should look into are: DCRAW (tm) to convert *.RAW or *.CR2 image files into *.TIF or *.TIFF type, IRFANVIEW (tm) to batch convert images files like *.TIF to *.BMP, and VIRTUALDUB (tm) to convert *.BMP frames and *.WAV sound into an un-compressed AVI file that you can convert with other Freeware into *.MPG MPEG2 type Media files for viewing of your frames in Media Player. --- PROCESSING RAW FRAME IMAGE FILES DANCINES.EXE (tm) automates the shooting/exposing of your cine film scanner or printer, it does not in v1.xx process the image files shot. A third party program you can Google (tm) on the internet called DCRAW (tm) is designed to convert RAW Digital Image files shot with the DSLR camera you are using in your cine film scanner into image files of a type that can be loaded into many kinds of image file conversion software or image editing programs. I would advise against using the JPG image save mode in your DSLR to shoot film frames because the compression of color negative images can result in artifacts showing up after color correction, and RAW mode images are generally sharper and do not have the tone blocking that JPG images can display. A small utility program called DCRAWBAT.COM (tm) is included in DANCINES.ZIP (tm) to generate a batch file called X.BAT that automates the conversion of up to thousands of motion picture frame images by making instructions in the batch file X.BAT that invoke the third party program DCRAW (tm) like a command line utility as many times as required for the number of frames to convert from raw to a readable type, in this case from IMG_1234.CR2 to IMG_1234.TIFF the X.BAT file then renames the IMG_1234.TIFF files to 00000000.TIF so that some programs that do not recognize the TIFF file type can read them as a TIF file. For use with my programs Edit list Insert command it is best to have image frame sets start with the count of zero in the filename of the first frame. Because DANCAD3D.EXE (tm) v3.7 may not read TIF files another image file conversion program may need to be used to convert the TIF frame image files into 24bpp BMP type image files for import through the image Insert command in the Edit list. The third party program IRFANVIEW (tm) can do batch conversion from *.TIF to *.BMP, you can look into downloading it from the internet. If more than one exposure was made of the motion picture frame to get extended or better tonal range, the Fuse option of the Edit list image Insert command would need to be used to Fuse or combine the separate exposures into a single image file that can be used for color correction Grading and re-sizing. The Fuse command can take three separate exposures made through color separation filters, i.e. Red exposure file, Green exposure file, and Blue exposure file and make them into one color file. The Fuse command can also take more than one color file and fuse them to get a longer or better tonal range, e.g. you can set the camera to "auto bracket" so that it shoots sets of three exposures, the first +1 stop, the second on exposure, and the third -1 Stop. The "auto bracket" feature is found in your DSLR's built in menus. If you combine the "auto bracket" feature with making Red, Green, and Blue filter separation exposures you can then get 9 exposures per frame, three Red, three Green, three Blue and thereby get better tonal range and color purity, perhaps. The over- sampling can be done further by making three sets of bracketed exposures in each primary color to get 27 exposures per motion picture frame. Going beyond 27 exposures per motion picture frame would probably going beyond the point of diminishing returns as far as image quality is concerned unless your Digital Camera has a high internal noise level. To reduce the internal random noise of your Digital Camera set the gain or EI index to the lowest or a low level. Fusion of several exposures can help reduce the random noise in Digital Camera exposures, but may not help with "fixed pattern noise" such as burned out pixels, or with film grain in the motion picture film being scanned, although getting a longer tonal range can render the grain smoother since the Digital Camera noise is single pixel and the film grain spans several pixels. To get the third party program DCRAW (tm) to be able to convert frame files shot with a Canon XTi (tm) DSLR in its CR2 filetype using my DCRAWBAT.COM (tm) program you need to make a sub directory (folder) on your C:\ drive in the root directory called C:\DCRAW and put DCRAW.EXE (tm) in it along with any other files it requires such as cygjpeg-61.dll, cyglcms-1.dll, cygwin1.dll, these or other dll files may also need to be in the Windows (tm) directory where dll files belong. If you do not have the needed dll files you may be able to find them on the internet. File conversion of large file sets of frames can take several days or longer, and require a great deal of harddisk space. Storing all the frames from a feature in their RAW format, converted formats, color corrected and re-sized forms, and the final edit and Blu-ray (tm) or DVD size can take 25TB (25,000GB) or more. To be able to process and store all the frames you will probably need several computers loaded with large harddisks and maybe wired to a network to be able to swap files around. To store the frame image files made while scanning, you first should use the Files Utilities Kinema Structure command to create a film project directory structure. You can then use your memory card reader and the Windows (tm) file manager to copy the CR2 files from the memory card into the I01 resolution level for the shot the frames belong to. At some point you need to rename the frame files so that the frames for each shot start with frame number zero. Normally frames in the shots of the project structure are saved in the Divided file name format, but since that format is not supported in Windows (tm) programs the Padded numbered file name format should be used, unless the Long numbered file name format is required. Remember what file name numbering format is used for frame images at various levels of processing and resolution since you may get "File not found" error messages if you select the wrong file name format. To convert the TIF frame images you get from using the third party program DCRAW (tm) to into 24bpp BMP frame files third party programs like IRFANVIEW (tm) might be used. IRFANVIEW (tm) has an internal batch process mode and can automatically renumber the frame images. IRFANVIEW (tm) can also batch re- size, sharpen, gamma adjust, negative to positive, and such but may be limited in its ability to do full color correction which is probably best done with DANCAD3D.EXE v3.7 (tm) for use with my system. There are many other programs that can convert frame image files from one file type to another, you can search for "Freeware" using Google (tm) and such. When converting frame images you should use UN-COMPRESSED image types such as 24bpp BMP rather than compressed image formats like JPG, or JPEG since compressed formats degrade images and alter the brightness and color of individual pixels in the frame image files. If you are short on disk space you could shoot JPG images in your DSLR then convert those to 24bpp BMP for loading into DANCAD3D.EXE v3.7 (tm) but if you try to lighten the shadows or sharpen compressed images the compression artifacts may become more noticeable. Scanning to RAW sensor images can let you use more extreme image tone manipulations before sensor or histogram artifacts become visible. If you are making a Black and White film saving the edited image frames that will go to the film recorder in 8bpp Gray scale BMP can reduce the disk space required for that frame set to 1/3, but at the expense sometimes of a little tonal quality since 24bpp BMP files can have noise in the tones between the Red, Green, and Blue exposures, so the eye sees the mixing of color values that are not the same for all three primary images, like in an 8bpp file R=128 G=128 B=128 but in a 24bpp file you might get R=128 G=127 B=128 so the brightness range on average would be a little smoother using 24bpp BMP files to store the edited graded frame images in. When you re-size frame image files it is best to use the highest quality method that the software supports since the "nearest pixel" or "quick" method can make reduction images look noisy, grainy, and jagged. Some graphics programs may not do an acceptable job of re-sizing images. If you need to reduce the size of images by a large amount you can sometimes get better results by using the image soften or blur command with a radius of one to maybe five pixels, then re-size in reduction, then sharpen the smaller image. Another trick is to reduce the images in steps of only making the image 2/3 of original size on each pass until you get to the size you need. Enlarging small images such as 720x480 up to 2048x1365 should turn out well if you use the "Quad pixel" re-size option in DANCAD3D.EXE v3.7 (tm)'s image Insert command in the Edit list since the output will interpolate between pixels rather than show blocks of pixels like some other programs might. --- ISSUES AND PROBLEMS If the program is run under Windows (tm) some programs can force a pop-out to the desktop, that might ruin your scanning or printing, be sure not to have any programs running or installed on your system that can force a pop-out from a DOS program running in a DOS window. The same goes for the "Windows (tm)" key on the keyboard, if this gets bumped while the film scanner or printer is running your job could be ruined. You should use an older AT style keyboard that does not have the "Windows (tm)" key on it. The [Pause] or [Ctrl]+[NumLock] keys on the keyboard could also cause problems if pressed while film is being exposed. Programs or things running in background in Windows (tm) may alter the timing and cause flicker or cause frames to be exposed incorrectly. Running the program under DOS may reduce some of the Windows (tm) issues, but may slow down some operations. If you accidentally have the program select a video mode your monitor cannot display and you see scrambled or distorted images on your monitor turn the monitor off immediately. If the monitor is Plug-and-Play it may go into "safety protection" and then give a black screen, under Windows (tm) you can press [Ctrl]+[Alt]+[Del] and then tell Windows (tm) to force a shut down of the program. Only use the configuration to select a video mode or resolution values to ones that your video card and monitor can safely support. Under DOS you will need to reboot your computer if you select an invalid video mode. Do not select the Parallel Port Pin 10 pause mode if you do not have Pin 10 connected properly since the program will wait forever for the correct signals which will never come. Power going to your cine film scanner/printer may fluctuate causing flicker and image color changes in the scanned or printed images You should use a regulated inverter operating off of a bank of storage batteries being continuously charged to operate the film scanner/printer in order to have a UPS that gives very constant power. Harddisks that automatically park or power down can freeze the system, which might affect the exposure of a frame. To avoid problems related to system timing issues it is best to have the film advance and exposure triggered by the computer but timed by an external crystal timer. Care needs to be taken to make sure that the camera or projector do not get ahead of the computer, or that the computer does not get ahead of the camera or projector. The parallel port signal mode has an option for a hold in pause until OK signal on Pin 10 of the parallel port, so you can wire up your film scanner/printer to take advantage of that option. Be sure to disable any programs running in background, especially anything that can write to the screen. Be sure to disable any task scheduling program. Be sure to disable any power saving features. Be sure to disable any screen saver program that might blank the screen after a period of time. Be sure that there are no pending print jobs that could activate the pins on the parallel port. Do not turn on your film scanner/printer control electronics until after you have booted DANCINES.EXE (tm) and configured the program, since tones may be generated by various programs, or the state of the parallel port pins may be such that the film scanner/printer could start advancing film before the projector has any image loaded, and maybe ruin a whole magazine of film or set of digital image frames. The same problem can happen if the computer reboots, the parallel port pins can get set to various patterns, which might start the film scanner/printer running wild. So always turn off the film scanner/printer when the last frame is shot, and do not turn the film scanner/printer on until DANCINES.EXE (tm) has cleared the parallel port signal pins. You can put LED's with series resistors on the parallel port signal pins to check and see if they are clear (all outputs set low, and the input is set as you need). DANCINES.EXE (tm) may be able to be stopped by pressing [Ctrl] and [C] at the same time, i.e. ^C aka Control C. On some keyboards this may be done with the [Break] key. The ability to "break" DANCINES.EXE (tm) may vary from one revision to another, and when it is working you may need to press ^C more than once and or hold it down for it to work since the program only checks the keyboard after certain tasks. --- SCF FILE SCAN/SHOOT CUE FILE DOT COMMANDS To control the frames that are scanned or shot from the roll of film in the projector DANCINES.EXE (tm) reads a cue file with the extension *.SCF that stands for Shoot/Scan Cue File. Under Windows (tm) you may need to rename that *.TXT for editing with NOTEPAD.EXE (tm). The commands in the *.SCF file start with a dot or period and are called "Dot Commands". They all have a dot followed by two letters, the case of the letters does not matter. You should not put anything on lines in the *.SCF file other than the dot commands. The dot commands available in DANCINES.EXE v1.00 (tm) are: .ms [message text to console, i.e. displayed on screen] .lv [initial IR] [initial red] [initial green] [initial blue] .pa [startframe] [endframe] [IR] [R] [G] [B] [F1] [F2] [F3] [F4] [F5] [F6] [F7] .pb through .pz like .pa, rolls A to Z, start and end are in frame numbers. .cc [comment text notes just in file not displayed on screen] .zz ; End of file. Dot command .ms must be followed by a space then a text message to display on the computer's screen when the *.SCF file is executed in DANCINES.EXE (tm). .ms can be used to confirm that you are executing the right *.SCF file for the roll of film to be scanned or printed. The dot command .ms can also be used to give instructions to the scanner/printer operator to change settings such as the lamp voltage, etc. Dot command .cc is like .ms in that it must be followed by a space then some text. If you just want to add some space to the *.SCF file to make it easier to read when you are editing it you can use a period or dash as the text. Dot command .lv sets the initial light valve settings, that lets you follow it with a .ms dot command to tell you to check the light with a light meter or to let you make a test exposure or something with the light vales set to a selected value. Dot command does not advance the projector or make an exposure. Dot commands .pa through .pz are the "print" dot commands, the p stands for "print", the letter after the p is the film roll letter for A/B rolling, such that .pa prints or scans frames from the "A" roll, .pb prints or scans frames from the "B" roll and so on. Since in the real world you would never have a "Z" roll the dot command .pz can be used to tell DANCINES.EXE (tm) to advance a number of frames without printing or scanning, i.e. fast forward over so many frames as indicated. Since there is one *.SCF file for printing all rolls you cannot double expose, in fact the program simulates the *.SCF file looking for errors such as gaps or overlaps before it will execute the scanning or printing. If you need to double expose you would need to make a copy of the *.SCF file and edit the in and out frame numbers for the rolls you wanted an overlap at. If you are printing and want black frames onto print stock you could print a "X" roll without a negative in the printer, i.e. the lighthouse would go directly onto the print stock (with a filter added to the filter pack so that the light would be the right color even if you print black to avoid over exposure and flair in the sound track area.) When the dot commands .ap through .pz are executed the program checks to see if any of the filter names have changed, if so it will stop the scanning/printing and put up a message onto the computer's screen asking the scanner/printer operator to change the filter. The operator then changes the filter and presses the key indicated to have the program continue executing the *.SCF file. Normally for making up *.SCF files that will just be used for scanning or workprinting film rolls you would only use .pa for the parts you want to scan or print and .pz for the parts you want to skip over. You would only use .pb, .pc, and so on if you are zero cut printing film to film since when you scan you will be using the Pick command in the Kinema Edit list to set the first and last frames in the editing, although there is nothing to prevent you from A/B rolling in the scanner if you have a reason to do so such as saving time by not scanning the frames at high resolution that will not be used. Dot command .zz marks the End-of-File and should be on the last text line in the *.SCF file. Do not put anything after .zz or on the lines below it. Avoid using a text editor that inserts ^Z (Control Z) at the end of the file, since if you open the file several times it might add ^Zs several times and if you get a ^Z in the middle of your *.SCF file the execution might stop before all the dot commands have been read. You can look for ^Z and other "invisible" characters with the Binary Editor in DANCAD3D.EXE (tm), and remove them if you need to. Some example *.SCF files are included in DANCINES.ZIP (tm) for you to use in testing your scanner/printer. Because DANCINES.EXE (tm) can accept the frame numbers in different formats, Frame numbers, SMPTE tome and frames, Feet plus Frames, and audio Samples, more than one example is given. Frame numbers are always right, but the other frame number formats require you to have the values in DANCINES.EXE (tm)'s configuration set correctly. Before the *.SCF file is executed with command 2 from the DANCINES.EXE (tm) main menu the scanner/printer operator would set the counters on the camera and projector to zero. The first frame counted is the frame in the projector gate at the time the *.SCF file is started, the projector is then advanced the first time to frame 1 which is the second frame. The *.SCF file always starts with frame 0. When making the *.SCF file you would need a sync block with a frame counter and need to note the frame count at the ins and outs as you wind through the film roll or rolls. If your sync block has a footage counter plus frames marked around the sprocket you might find entering the values into your *.SCF file easier as Feet plus Frames. If you have transferred a workprint scan of your rolls to video tape with SMPTE time plus frames window dub burn in you may find it easier to use the SMPTE option of the *.SCF file for the ins and outs of the rolls. Making a fast "workprint" scan of your rolls can be easier on them than running them through a sync block and maybe get less dust and scratches on them. If you use DANCAD3D.EXE (tm) with a SVGA to composite converter to make the video tape of your workprint you can select Frame numbers, Feet plus Frames, or SMPTE Time plus Frames for the count displayed over the frame during the 24fps playback (the video rate is not important since the SVGA to composite converter can make NTSC or PAL/SECAM from the 24fps since the video card will be running at 60Hz or faster probably, giving you one or more fields that are right to look at on the single frame advance of the video tape. You should check the ins and outs in DANCAD3D.EXE (tm)'s Edit list by single frame advancing over the edits to be sure the correct frame numbers are being selected. You do not need to make a video tape, it just might be easier to give that to the director and editor to play with, they can then give you back the edit points on lined paper that you can enter into the Edit list to make them a confirmation video of the edited footage, or they can have a duplicate set of proxy frames and their own computer setup to run DANCAD3D.EXE (tm), but using a VHS playback is cheaper and simpler is several people are going to be making edit reviews. You can also make a MPG2 or DVD but I am not sure how many dropped frames you would get in a compressed format, VHS is not compressed. To make understanding what the *.SCF file looks like so you can may your own their text is reproduced here, Example file: DANCINES.SCF 964 Bytes 03-12-08 11:34p .ms SCF (Scan Cue File) test file #1 for DANCINES.EXE (tm) .lv 25 25 30 35 ; set initial light vanes .cc . .cc First part, .pz 0 11 20 25 25 25 . UV 2EPACK . ND0.20 . . ; run in .pa 12 26 21 22 23 24 . UV 2EPACK . ND0.20 . . ; shot 1 comment space .pb 27 40 25 0 26 51 . UV 2EPACK . ND0.40 . . ; shot 2 comment space .cc . .cc Start of next part, .pb 41 45 25 25 25 28 . UV 2EPACK . ND0.20 . . ; shot 3 comment space .pc 46 50 25 29 25 25 . UV 2EPACK . ND0.60 . . ; shot 4 comment space .cc . .cc Start of next part, .pa 51 55 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 5 comment space .pc 56 70 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 6 comment space .cc . .cc Last shot, .pa 71 73 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 7 comment space .cc . .cc Run out, .pz 74 80 25 25 25 25 . UV 2EPACK . ND0.20 . . ; run out .cc . .cc End SCF (Scan Cue File) test file #1. .zz Example file: FPLUSF.SCF 1,141 Bytes 03-12-08 11:35p .ms For 24 frames/sec, SMPTE SCF test file #1 for DANCINES.EXE (tm) .lv 25 25 30 35 ; set initial light vanes .cc . .cc First part, .pz 24T00:00:00:00 00:00:00:11 20 25 25 25 . UV 2EPACK . ND0.20 . . ; run in .pa 00:00:00:12 00:00:01:02 21 22 23 24 . UV 2EPACK . ND0.20 . . ; shot 1 comment space .pb 00:00:01:03 00:00:01:16 25 0 26 51 . UV 2EPACK . ND0.40 . . ; shot 2 comment space .cc . .cc Start of next part, .pb 00:00:01:17 00:00:01:21 25 25 25 28 . UV 2EPACK . ND0.20 . . ; shot 3 comment space .pc 00:00:01:22 00:00:02:02 25 29 25 25 . UV 2EPACK . ND0.60 . . ; shot 4 comment space .cc . .cc Start of next part, .pa 00:00:02:03 00:00:02:07 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 5 comment space .pc 00:00:02:08 00:00:02:16 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 6 comment space .cc . .cc Last shot, .pa 00:00:02:17 00:00:03:01 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 7 comment space .cc . .cc Run out, .pz 00:00:03:02 00:00:03:08 25 25 25 25 . UV 2EPACK . ND0.20 . . ; run out .cc . .cc End SCF (Scan Cue File) test file #1. .zz Example file: SMPTE.SCF 1,141 Bytes 03-12-08 11:36p .ms For 24 frames/sec, SMPTE SCF test file #1 for DANCINES.EXE (tm) .lv 25 25 30 35 ; set initial light vanes .cc . .cc First part, .pz 24T00:00:00:00 00:00:00:11 20 25 25 25 . UV 2EPACK . ND0.20 . . ; run in .pa 00:00:00:12 00:00:01:02 21 22 23 24 . UV 2EPACK . ND0.20 . . ; shot 1 comment space .pb 00:00:01:03 00:00:01:16 25 0 26 51 . UV 2EPACK . ND0.40 . . ; shot 2 comment space .cc . .cc Start of next part, .pb 00:00:01:17 00:00:01:21 25 25 25 28 . UV 2EPACK . ND0.20 . . ; shot 3 comment space .pc 00:00:01:22 00:00:02:02 25 29 25 25 . UV 2EPACK . ND0.60 . . ; shot 4 comment space .cc . .cc Start of next part, .pa 00:00:02:03 00:00:02:07 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 5 comment space .pc 00:00:02:08 00:00:02:16 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 6 comment space .cc . .cc Last shot, .pa 00:00:02:17 00:00:03:01 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 7 comment space .cc . .cc Run out, .pz 00:00:03:02 00:00:03:08 25 25 25 25 . UV 2EPACK . ND0.20 . . ; run out .cc . .cc End SCF (Scan Cue File) test file #1. .zz Example file: SPERF.SCF 1,039 Bytes 03-12-08 11:36p .ms For 2000 samples/frame, SAMPLE SCF test file #1 for DANCINES.EXE (tm) .lv 25 25 30 35 ; set initial light vanes .cc . .cc First part, .pz 2000S0 S22000 20 25 25 25 . UV 2EPACK . ND0.20 . . ; run in .pa S24000 S54000 21 22 23 24 . UV 2EPACK . ND0.20 . . ; shot 1 comment space .pb S56000 S80000 25 0 26 51 . UV 2EPACK . ND0.40 . . ; shot 2 comment space .cc . .cc Start of next part, .pb S82000 S90000 25 25 25 28 . UV 2EPACK . ND0.20 . . ; shot 3 comment space .pc S92000 S100000 25 29 25 25 . UV 2EPACK . ND0.60 . . ; shot 4 comment space .cc . .cc Start of next part, .pa S102000 S110000 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 5 comment space .pc S112000 S128000 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 6 comment space .cc . .cc Last shot, .pa S130000 S146000 25 25 25 25 . UV 2EPACK . ND0.20 . . ; shot 7 comment space .cc . .cc Run out, .pz S148000 S160000 25 25 25 25 . UV 2EPACK . ND0.20 . . ; run out .cc . .cc End SCF (Scan Cue File) test file #1. .zz Note that the filter names need a place holder in the empty filter slots, you can put a period or dash there, but keep a space on both sides of all values and strings. The IR vane value needs to be included even if it is not being used so that the *.SCF file will be valid for later use. Do not use light vane values over 51 if you are going to use the *.SCF with a Bell and Howell light house, otherwise you can use light vane values 0 to 63. Mid-light vane value is normally 25 but some labs use other numbers since they are running their printers faster than they have light for, they might use 30 as mid light and put a 1200 Watt lamp in the light house to run the printer faster. Notice that you can put comments to the right of the .pa through .pz dot commands as notes to yourself as to what shot those printing lights are for. There may be some changes to the contents of the *.SCF files in future versions, so some editing might be required for use with future versions of DANCINES.EXE (tm). When you edit your *.SCF files you should avoid getting Control Z ASCII values in the files or other non-standard ASCII 32-127 values since only LF and CR, #10 and #13, are defined, others such as form feed are not defined. If you get problems try retyping the *.SCF file using the Write command in DANCAD3D.EXE (tm), since the text editor you are using may insert formatting codes into the text file, also save as DOS *.TXT type text file not a *.DOC or other type. --- SCANNER DIGITAL CAMERA ISSUES Ideally you would want a Digital Camera, CMOS or CCD, for your movie film scanner that has a resolution of better than 4096x3072, 4:3 ratio, 12 bits of true tone, and is monochrome without a Bayer filter. There are industrial cameras like that that can download images at maybe 2fps through USB or some other cable. The sensor should also be about 22mm wide so you can use a 1:1 printing lens on your scanner. If the industrial camera does not have enough dynamic range to capture the full density range of your camera negatives you will get a loss of tonal values in your scans. If the industrial camera has a high noise level you can try making several exposures, maybe with a filter wheel to shift the exposure a half stop between exposures, and use the Fuse image command in the Insert command in the Kinema Edit list command in DANCAD3D.EXE (tm) to extend the tonal range of the Fused image beyond what a single exposure can obtain and to reduce the random noise from the sensor through pixel averaging. If the industrial camera has fixed pattern noise such as burned out pixels or uneven columns or rows that are the same in all the exposures using the Fuse command will not help much. You would need to subtract the fixed pattern "blemishes" in the image, and interpolate over the bad pixels. The software that came with the industrial camera may have fixed pattern noise removal built in, you should ask about the fixed pattern noise issue before you purchase the industrial camera. The reason you would want a monochrome camera is that you will be making three or more exposures through color separation filters or with colored LED lamps so having the Bayer filter on the sensor just reduces the resolution of the camera to about half. Not having the Bayer filter on the sensor can also reduce some of the fixed pattern noise since the filters are not perfect and may introduce some speckle to the image. Another advantage of using a monochrome industrial CMOS or CCD camera is that it does not have a mechanical shutter like some DSLR cameras. If you use a DSLR the mechanical shutter and reflex mirror will probably wear out before you have shot even half the frames from a feature film. To scan a feature film you will need to make 300,000 to 600,000 or more single exposures, and if you are making 9 exposures per frame you would need 2,700,000 to 5,400,000 exposures. If your DSLR is rated for 100,000 exposures and you need 5,400,000 exposures you would need to purchase 54 DSLR cameras to get all your footage scanned for just a single feature. If the DSLR costs $600 then 54 of them would cost $32,400! If the monochrome industrial camera costs $5000 it may end up being cheaper in the long run. You could sell the DSLRs on ebay.com when they get to maybe 75,000 exposures, but you would still have a loss from the new to used price drop. That said, the dynamic range of a DSLR like the Canon XTi (tm) is quite good and more than enough to get adequate movie film scans if you Fuse three or more exposures. Also the Canon XTi (tm) has built in exposure bracketing so you can turn that on and when you make three exposures of each film frame the camera will take one on exposure, one at plus a stop or so (as you set the bracket range in the camera's menu), and another minus a stop or so. The electronic cable release is easy to wire to two relays that can be controlled by the parallel port pins, since the current draw is low you can use some TTL buffer chips with open collector and two 5 volt DIP relay. You need to add some delays and logic so that the ring on the sub-mini phone plug closes first then the tip closes while the ring holds then the ring opens. If you are using a DSLR to shoot in your scanner you can save the frames to the camera's Compact Flash card (or SD etc.) or use a USB cable (or FireWire etc.) to download the images to your computer's harddisk. DANCINES.EXE (tm) has a built in exposure counter so that it will pause after a configured number of frames are shot and wait for you to do something like change the Compact Flash card in the DSLR. On the Canon XTi (tm) a 2GB Compact Flash card will hold about 170 RAW images, so you set DANCINES.EXE (tm) to pause after each set of 170 exposures so that you can swap the memory cards. If you use the JPG file type you could hold more frames but you would get compression artifacts in your scanned frames. The time to download the frame exposures from the memory cards to the harddrive in the image processing computer may be longer than the time to shoot in the camera so you may need several computers downloading and organize the file sets later on a network connection or by putting two or more harddrives in a single computer so you can move the files into a common directory. If you use the USB feature in the Canon XTi (tm) when DANCINES.EXE (tm) sends the shoot signal to the parallel port on the computer that it runs on, the relay for the shutter release fires, then the image is saved not in the camera but on the harddisk of another computer. Because of the numbering system used in the DSLR you may be limited to 10,000 exposures before you need to reset the camera to count from 0, i.e. it numbers the frames IMG_0000.CR2 through IMG_9999.CR2. You can configure DANCINES.EXE (tm) to pause after shooting 10,000 exposures. You should use two computers one running DOS for DANCINES.EXE (tm) and another running Windows (tm) for the image download utility that comes with your DSLR. Since you are going to be shooting TB of RAW frame exposures you should arrange the computer you will be saving frame exposures to to have hot swap harddrive bays so you can keep feeding in more and more 500GB harddrives without having to turn off the computer and fiddle with the wires inside the computer. The harddrives with the RAW frame exposures need to be processed into RGB files for Color Correction in DANCAD3D.EXE (tm). To keep the workflow going as fast as possible you should have separate computers for the scanner frame exposure downloading, the RAW to RGB conversion, and the Color Correction, with the harddrives with the frame exposures moving from one processing step to another. After Color Correction of the Key frames, you need to make Proxy resolution frame images that can be used with the Pick and View commands in DANCAD3D.EXE (tm)'s Kinema Edit list. The Proxy frames are required to display the real time play back at 24/25/30fps while editing, and being smaller allow all the frames scanned to be accessed at random while editing so you can see all the frames and shots without having to swap disks or move files around (once you have all the proxy frames in a Edit list structure for access while editing). The proxy frames are not used for output, except maybe for conversion to BMP to make a quick check un-compressed AVI with VIRTUALDUB (tm) to check the sync of a sound mix or to burn a DVD so you can show someone the edit/mix away from your editing computer. The Canon XTi (tm) seems to be able to download its RAW images through about a 30 foot USB cable, how much longer of a cable might work I have not tested yet, but you should keep the cables as short as you can normally. Since it takes about 4 to 5 seconds to download the RAW images to the computer's harddisk using the utility that comes with the camera you may need to configure extra time after the post exposure delay to make sure the projector does not finish advancing until after the last image has finished downloading. --- SCANNER LENS ISSUES The best lens to use on your scanner is generally a lens off of an old 35mm Optical Printer such as an Acme (tm) or Oxberry (tm). The Kodak Ektar 103mm (tm) was popular before the Nikon Nikkor 105mm and Nikkor 150mm were introduced. The Printing Nikkors are about the best lens you will find to put in your scanner, but they are not cheap expect to pay more than $1000 each. The Nikon Nikkor 95mm was a blow-up lens for 16mm to 35mm, the other mentioned are for 1:1 printing, so the sensor in the Digital Camera should be the same size as the film frame to get best results. If you need to enlarge or reduce you should stop the lens down and not shoot wide open. The f-stop rings of printing lenses may be marked like a normal lens, except that they are never used for infinity focus, they are used at equal distance 1:1 magnification so the f-stop markings are two stops larger than what is the real f-stop, in other words if it says f/8 on the lens the real f-stop is f/16. Because of diffraction you do not want a real f-stop smaller than f/11 most of the time, unless the film is very grainy and you want to use f/16 as a optical low pass filter to blur the image a little. To get f/11 you would set the lens to f/5.6, so a printing lens marked f/2.8 wide open is really f/5.6. Because it is a problem to get film to lay flat in the projector gate in your scanner/printer you should not use the printing lens wide open, stopping the lens down as much as you can without getting resolution reduction from diffraction will help keep all parts of the frame in equal focus, another reason for setting the lens to f/5.6 or f/8 to get f/11 or f/16. The highest resolution of the scans may be on setting the printing lens between f/4.5 and f/6.3 but you may just be focusing on the film grain better and making your film scans more noisy, particularly if you need to sharpen the scans a little because high speed lenses were used and were a little soft in focus on the negative. If you stop the printing lens down too much you can reduce the grain in your scans so that if you sharpen later you will sharpen the image and not mostly the grain which is smaller than the circle of confusion of the camera lens when high speed camera lenses are used, such as at f/1.2 camera lens. If you cannot afford to purchase a true printing lens, you can use a Macro lens made in a mount that will fit on your Digital Camera. A 60mm Macro lens will be 120mm when extended to 1:1 magnification, and when set to f/8 will really be f/16. Some Macro lenses fit into bellows that have rack-and-pinon focusing, while others have a double helix lens mount. The double helix lens mount may be a little floppy when run out to 1:1 magnification, which could be a problem since any movement of the lens will show up as misplacement of the image on the sensor which could be bad if the image moves between frames from vibration of the movie film projector advancing, or the lens getting bumped by the scanner operator while doing test exposures or changing the Memory cards etc. If you are doing composites you want your scanner rock solid so that the foreground with scan in register with the plate, if anything moves like the floppy lens your scans may not register later or the edge of the gate will drift into frame and such. You may be able to get better results than using a Macro lens by putting two f/1.4 or f/1.2 58mm SLR lenses face to face. This will give you a twelve to fourteen element 1:1 lens, a bit like the Printing Nikkor. You would leave one of the lenses wide open, and stop the other one down to f/11. When you have two lenses face to face like that one acts like a close up lens for the other one, so there is no "bellow extension" to reduce the f-stop, i.e. f/11 will be f/11. Both 58mm SLR lenses would be focused at infinity. If 58mm lenses do not give you enough "working distance" you could use 85mm or 135mm SLR lenses just as well if they are very sharp lenses to start with. You should make some kind of ring to hold the two fronts of the two lenses together so that the optical axis of the two lenses is straight. If you are going to blow up or reduce you can use two lenses face to face that have different focal lengths. Very good enlarger lenses can also be put face to face. To blow up 16mm to a Digital Camera sensor about 35mm movie size you might put a 58mm lens on the 16mm side and a 135mm lens on the Digital Camera side with "film" end of the two lenses away from each other. To scan Vista- Vision (tm) or 65mm camera negatives you might put a 150mm 2 1/4 camera lens on the film projector side and a 85mm or 58mm on the Digital Camera side. If you are using a Digital Camera with a sensor larger or smaller than the film being scanned, you can adjust the focal length of the two lenses to make up for the size difference, in other words if the sensor is half as big as the film you use a lens half the focal length as on the film size, if the sensor is twice as big as the film size you use a lens on the Digital Camera side that is twice as long. You just need to use lenses that are open enough that you do not get vignetting or dark corners. The lenses should also have enough "coverage" to allow some shifting of the lens to center the image on the sensor and still have sharp corners. Putting the two lenses so that their "front" elements are almost in "contact" can help reduce light loss in the corners of the frame. If you use a single Macro lens to "blow up" and the sensor is larger than the film, such as going from Super16 to DSLR, you would do better to "reverse mount" the Macro lens so that the end that normally fits to the camera now points to the film projector. Most Macro lenses focus up to 1:1 but not past that, to go past 1:1 you need to turn the Macro lens around. Some Macro lenses made for use on Bellows are already mounted "backwards" in thei