Producing commercial quality animated films on the SC4020 presented a variety of problems. The Laboratory worked with Peter Hadingham of Swift Film Productions and other experts to provide some guidelines to users. The set of papers below give some idea of the problems encountered.
The aim of this meeting was to investigate some of the problems that have been around in the past and some we will no doubt be encountering in the future.
Considering our existing equipment, Peter Hadingham thought that it was almost impossible to record directly onto a film's magnetic stripe using the Bell and Howell projector without getting projector noise. He felt that the noise might be kept to a reasonable level by putting the projector outside a soundproof booth but even then some noise would still be present. He suggested that it would be simpler to record the soundtrack on the tape recorder without the film, running at the same time. This would of course need some means of cueing the person producing the soundtrack. He suggested that this could be done by some system using a synchronised motor and footage counter (or reed switch with frame counter). The user would then work out the footage positions where the various speech paragraphs started using the projector and synchronised motor. The soundtrack could then be made later and transcribed across. Unless lip synchronisation was required, this should be adequate for speech tracks of this kind. There should be no problem in transcribing from the Revox to the Bell and Howell although a matching transformer might be required.
The soundtrack should be recorded in a soundproof room. He thought that anywhere in the new building was out of the question due to large amount of background low frequency noise which would be impossible to eradicate. In the office blocks the most suitable room was the archives due to the lack of windows and also being away from the road. By putting suitable soundproofing on the walls this could be made into a reasonable room.
To make soundtracks with reasonable synchronisation would require the purchase of additional equipment. Basically the requirements are:-
The first item is most likely a tape recorder using sprocketed stock. Soundtracks would be recorded in bits on a conventional tape recorder and transcribed to the sprocketed one.
For editing and playback purposes, the simplest equipment would be a manual editor with a small viewer and possibly one or two magnetic sound tracks. The soundtracks would be edited by hand and good synchronisation could be achieved. It might be useful to provide a motorised version of this as it is difficult for an untrained operator to run at uniform speed using a manual control. Alternatively a separate playback device could be purchased.
Mixing the soundtracks and transcribing onto the film track requires expensive equipment and, at least initially, this could be done using a photographic house. Peter Hadingham suggested that his own facilities could be used by us if we came to some arrangement.
Obtaining commercial quality prints will require a certain amount of care in the handling of the initial film output. Given two copies of the original film undeveloped, it should be possible to generate a good master assuming only a small number of random SD4020 errors.
Basically the process suggested is as follows :-
In generating the master, Peter Hadingham suggested that he might be willing to provide such a service for us. He would get the processing of the original copies done and, assuming he had a sample print from us, could make an initial judgement as to whether errors existed. If they did then a work print would be obtained from the originals and sent to us. We could indicate the errors and he would then do the necessary editing and optical work.
He was quite prepared to take on the job of supplying copies from the master on a commercial basis. He would charge his own costs plus a handling charge together with any profit we wanted to add. An arrangement whereby British Educational Institutions paid a different price from the general public did not appear to create any real problems. He already provided such a service for some of his own films that had been commissioned by industry.
To see how feasible the above setup is, we have arranged a pilot exercise where a single film is sent to him for processing.
Drafted after consultations with various members of the Atlas Laboratory staff and. examination of a selection of animated films.
Primary investigation - Animated film production
Camera original - sensitive material to be used, negative or reversal processing of this.
Intermediate stage -for insurance of camera master where many copies are to be produced.
Final printing - form of final output and method of production. Examples:-
Visual image quality - adequate contrast, size. of readable characters.
Image positional stability.
Retention of fine line detail OR
Satisfactory broadening of lines generally in the photographic process (Effect of this on programming) OR
Generation of broader lines on Charactron display.
Cleanliness of background in final result.
Methods of editing out errors of SC4020 operation, programming or film processing.
Registration of colour images when produced by multiple printing.
Suitable materials and processing at all stages.
Relative cost of white/black, monochrome or colour systems and final prints.
This recommendation is to give white on blue (can be modified in final stage to produce white on other colour backgrounds) release prints of commercial quality, and covers all stages from camera exposure on SD4020 to final printing.
The computer program (recommendations for programs for animated films are given separately) and quality of magnetic tapes should be proved using the less critical recommendations for microfilm production (general) before this method of final film production is embarked upon.
It is assumed that the SD4020 is operating efficiently. Steps should be taken to ensure that the tape deck is operating reliably particularly on frame repeat sequences - eg cleaning of pinch rollers as recommended by the service engineer! Particular care should be exercised in the loading and handling of film for animation film purposes.
l6mm Track Chamber film as recommended for microfilm production (general) ie f8 at standard tube intensity
Two consecutive complete runs should be exposed if there is reason to expect very much in the way of SD4020 error. (This allows a second copy for patching purposes). (I do not really think this should be necessary and would like to see it eliminated from the routine. Future experience will show whether this can be done reasonably safely.)
The exposed film should be carefully unloaded from magazine, the end secured with adhesive tape, wrapped in opaque paper and put into metal film can. This film is to be sent to Filmatic Ltd for processing - spec to follow.
Compatible with AERE positive 50ft/min.
From the processed negative Filmatic should be asked to make, immediately, 2 white on black intermediates, on Agfa/Gevaert 553 film. (The background density of these should be of the order of 2.2.)
One of these intermediates should be preserved, without unnecessary handling, as Master intermediate for final printing stage, the second is to be returned to Atlas for examination and checking by the originator. This is known as the work copy or cutting copy.
Any editing which is necessary will be done on this copy (originator and photographic staff together), but notes will have to be made indicating exactly any alterations from original which have been made (this is necessary because of the lack of edge numbers on this film).
After editing of the work copy to complete satisfaction this is passed to photographic staff (or P Hadingham if he is going to serve in this capacity) and the Master intermediate is cut to match the working copy. This must only be carried out by properly equipped photographic staff or consultant/agent.
Straight cut splices (negative width) will normally be used although A + B roll checkerboard cutting may be used in exceptional circumstances. Any non-computer titling etc (if required) will be added to this Master Intermediate at this stage.
Release copies (white on blue) will be printed by Filmatic Ltd from the Master Intermediate using a separate exposure run (with clear or no negative film) to give background colour. These release prints are to be made on Eastman Reversal Colour print film type 7387 (NOT 7389).
This system allows lines of any colour or combinations of colours to be formed on a black background.
The SD4020 program must be written so that lines of each different colour are isolated and can be imaged alone on one length of film negative. If 3 different colours are required (eg blues yellow and magenta) then three separate records will be exposed in the camera one after the other each containing only the lines to be printed as one single colour.
It would aid subsequent printing if, at a fixed distance of 5 ft (200 frames) from the start and end of each colour record a register frame could be exposed - possibly with the intended colour written on it.
All colour records should be exposed successively on the SD4020 to minimise the risk of drifting of image position on the screen between records.
These should be as outlined in Section 1.
It is hoped that this could be reduced to absolute minimum (preferably none) for multicolour work.
Examination of the work-copy intermediates would show if any was necessary but examination of all three together is difficult. If the program has been properly proved beforehand simple comparison of record lengths and examination for blank or other rogue frames should suffice.
It would help checking of the program if this could be arranged in such a way that the various colour records could all be exposed together on a single length of film.
Intensity setting of the 4020 is standardised to give a reading of 8.2 from the standard long vector patch in centre of frame 10 on daily test tape, on Gossen Lunarsix Meter located centrally.
l6mm and 35mm Track Chamber Film (at present).
F8 for image composed of single stroke vector. For images with very fine detail or of hardware characters f11 may be preferable (but not for animation sequences).
For the present in Hadland processor for non-critical work. It is hoped to recommend a replacement for this processor. Alternatively,' critical work can, at present, be processed at AERE photographic group. (Process spec;- Positive bath 50ft/min). AERE processing will give lines of higher density than that from Hadland processor.
When loading the camera the operator should ensure that feed and take up are operating satisfactorily and that camera gate is properly closed. At no time, either before or after exposure, should the. roll of film be tightened by pulling on the outer end - if the roll is loose it should be handled extra carefully but left loose. Tightening is likely to cause short abrasion marks parallel to the length of the film throughout its length. Loose winding by the camera should be investigated.
In the production of distribution copies of animated film the photographic stages (carried out on 16mm film) inevitably introduce a widening of the lines drawn on the 4020.
The extent of this widening depends on the original camera material and exposure, the number and type of intermediate stages in the process and the method of final, release printing. The amount of widening can be limited to some extent by the acceptance of lower image contrast but since the lines are by nature very fine, low image contrast makes them visually unacceptable.
The recommended system made for the production of animated film represents, in my view, the best compromise between image contrast, resolution, handling and editing requirements which is possible at the present time. This system involves an increase in width of single stroke vectors by a factor of about 3 times (from 7 µm to 20 µm approx. on l6mm film image and allowance must be made for this increase when programming for animated films.
The following recommendations take account of this situation and apply to single stroke vectors unless otherwise indicated:-
