Dirk DeJonghe

A worse problem might be that the negative may be notched. A small shape may be cut out on the side of the film to trigger the scene-to-scene correction on the printer. This practice was later (mid 1970'ies) replaced by metallic foil patches and still later (1980'ies) with FCC ( computerized frame count cueing). When this notch passes the telecine gate it will give a noticeable sideways jump. Easiest but not cheapest would be to strike a low contrast 35mm print and transfer this to video. You could even do a separate print from the A and B-roll, transfer to video and make the final color correction and dissolves on a digital color grading system such as Baselight.

Most of it was done on a flatbed table like a Steenbeck, with the sound transfered to 16mm SEPMAG. In a way this kind of editing was much superior to the then current linear video editing. You could extract a scene from anywhere in the film and move it, or shorten or lengthen it, it was already some form of non-linear editing but without the instant access that we are used to now. The negative was conformed on a synchronizer with the edgenumbers as a guide (machine readable keykode numbers were not yet invented). From the cut negative a blow-up was made (direct or via IP/DN). Some film schools still teach this editing method and I think it is valuable to have had physical contact with pieces of film at least once in a carreer before jumping into cyberediting.

My scanner is like an optical printer, with the projector movement (XY adjustments) on one side, a lens carrier (adjustable in XY axis) and a camera on an optical bench type mount. Very easy to zoom in to whatever framing desired on whatever pixel size desired, mechanical autofocus, camera can be tilted. If I wanted I could extract a frame smaller than 16mm from anywhere in a 35mm frame and scan it to 2K. To set up a 1:1 scan/recording I scan a 35mm ruler film (marked in mm/inches) and then record it back to film. I record maybe a hundred frames of this film at slightly different magnifications and then select the one that comes closest to the real mm/inches. This particular frame is then put in the scanner and while looking at a split-screen I zoom in or out on the original ruler film until it matches within one pixel usually. Can't do half-pixels yet.

I just received digital images scanned on a Spirit from S35/3P for output to Cscope on my Lasergraphics recorder. Compared to my own scanner, the Spirit has noticeably more grain in blue skies etc. I feel that for S35/CScope 2K is not enough vertical resolution, HD is even less. In this case 817 pixels high are resized to 1742 high (2048 wide). Just last week I finished another job also on S35 (4P) for optical blow up to Cscope. FX were scanned in-house to 3K and resized to 2K for filmout (1850 pixels high to 1742 high). Very big difference in quality and practically undistiguishable from the optical blow up parts (80% of the film). I always use the full 2048 pixel width even for Academy framing.

David, there is a contradiction here, 3Perf 35 is Super 35 by definition, no sound track involved, so the full picture width would be 2048 pixels wide, not 1820. In my lab 2K S35 is scanned at 2048*1157 pixels if 3Perf (1.77) format.

Since the original is 16mm and the final print also, some of the solutions proposed use stocks only available in 35mm. My suggestion is to do a 2K scan from 16mm and a digital recording to 16mm. Would come close to the 16mm to 35 blow up Pan IP and reduction to 16mm 7242 DupNeg. For about one minute we usually try to match the price of our optical printer work since it is much easier to do and doesn't tie up the optical printers for several days doing blow-up and reduction. I have a S16 gate for my film recorder and it comes in real handy for these jobs.

I think professional negative cutters are some of the least appreciated jobs in the industry with a very high responsibility. I know that producers pay more per hour for labor to get their kitchen painted than to cut their precious negative that may be worth ten times their entire house. It is a job with no glory if all goes well but with plenty of room for errors if it doesn't.

John, you are quite right about the Kodak stocks giving very neutral and natural color reproduction especially flesh to neutral balance, difficult to achieve when starting with a Fuji negative. However, it is good to have choices of looks. The world wouldn't be the same if Van Gogh would have painted his sunflowers in natural colors. On the older Agfa stocks, prior to the CP20, I was not so happy with the reds, they tended to turn to orange, no deep reds possible even if you wanted. The current CP30 is so close to the Kodak look it could have been made in Rochester by someone who mixed half an emulsion of Vision with half an emulsion of Premier (just kidding of course but you get the idea).

We use three stocks in our lab (excepting Kodak low-con and Fuji for special orders). They are the Kodak Vision , Kodak Premier and Agfa CP30. When a customer hesitates or asks our advice about which stock to use, we print a section of his film on all three stocks and show him the result as A, B and C. The customer doesn't know which stock he is watching (this comes with experience) and makes a selection based upon look. In 90% of the cases the CP30 is the preferred stock. It sits nicely between the Vision and Premier in contrast and saturation. I think it is more stable in processing and printing. The Vision is nice for high contrast originals where you have too much contrast to start with, Premier is for special effects, and the CP30 is just right for a look with just a little extra punch without overdoing it.

Probabaly the Canada balsam used to glue glass element together yellowing. Quite common in older lenses. I don't know if the glue can be unglued and redone. You need to talk to a qualified lens technician.

The decision is also influenced by the shooting ratio, 35mm runs 2.5 more footage per minute than S16 and is about twice as expensive to buy. Some items are charged by the footage (processing, printing, etc) while others are by the minute (telecine transfers). The whole production must be taken into account: actors, crew, equipment, etc. I know of several productions that wanted to shoot 35mm at all cost with the result that there was not enough money to finish the film. Might as well throw away the money.

if the meter looks trough the filter it will "see" it and take into account the 2/3 stop loss in light because of the filter. In your case you overexposed by 2/3 of a stop. Nothing to worry about. Don't ask for a pull process, this will bring down the contrast and will be harder to match with other negatives. 2/3 of a stop is well within the latitude of the negative film and may actually be beneficial for grain and saturation, it depends what your "normal' exposure is.

I am glad you liked the film. 'Calvaire' was shot entirely on Vision 2 7218, some of it pushed one stop. No other film stock was used. it was a very small budget production (are there any other here?). Several minutes of the film were treated digitally because of severe scratch problems in the camera. After making a S16 answerprint (bleach bypass) a Super 35mm blow-up Interpositive was made, then an 35mm squeezed duplicate negative, both on 5242 stock. Finally a bleach bypass print was made on Agfa CP30. The DoP wanted a very gritty look to fit the story. A similar film 'Ultranova' was shown in competition in Berlin this year, bleach-bypass in the Interpositive stage to make the look semi-monochrome. Also S16 to CScope. Last year "25 degrees en Hiver" was shown as closing film at the Berlin festival, also S16 to CScope, this film had a mixture of Standard 16 shot on 800 ASA stock, pushed and with anamorphic lens attachment on the camera for the opening sequence, rest of the film S16 with masked ground glass mainly on 7274 (Vision 1 200T). Very nice cinematography by Walter Vandenende. Two more S16 to CScope feature films are in the pipeline.

I am amazed that the US negative cutters haven't discovered the german Hammann film splicer. We cut all our S16 single strand mostly, only AB rolls for dissolves when needed and then only the dissolving shots. Much easier to grade and to print. No problem at all making fine invisible splices on CScope negatives. It will also allow scenes to be cut together without losing frames or even redoing a splice on top of a splice. When I started my career, in the early 70'ies an old man from NYC gave me his 'personal' B&H hot splicer since he was retiring. He had used it most of his professional life. I think the design is from the mid 20ies. Some people are still using it now apparently.

I would say that Kodachrome is far to contrasty for normal telecine transfers. Strictly for direct projection. Very unlikely you can keep the highlights and shadow details in the same shot at the same time.

Shooting on sound recording film has been done, but it may have some unexpected side effects. One of them is that the sound film doesn't have a black antihalation layer. This means that light rays can penetrate the film and be reflected off the shiny pressure rails on the film pressure plate. Don't ask me how I know. The solution was to remove the film pressure plate entirely. Not good for other issues.

The answer will depend on the hardware used. In my case a telecine from a low contrast print matches the projected film look very closely. IP can be nice but is less film-like but this is very subjective.

I don't think the picture would become any sharper. A full frame 1.37 scan to 2K would be 2048*1536 pixels (3145728 total). A 1.85 frame is 2048*1108 and a pillarboxed 1.37 part of that would have an active picture area of 1518*1108 or 1681944 pixels total or about half. Once you get below 1800 pixels wide, aliasing may become a problem. On the other hand an optical reduction from 1.37 full frame to pillarboxed 1.37 inside a 1.85 would give a very sharp image (but smaller) since you gain resolution in an optical reduction. In digital you only use part of the same 'raster' to make the reduction effectively throwing away pixels and resolution. We are just now doing an optical reduction from Super 35/3 Perf to 1.85 4Perf and it comes out very sharp, also due to the edge effects in the intermediate stock that are enhanced by the optical printing.

May I recommend you read 'The Negative' by Ansel Adams, a classic textbook on B&W exposure, processing and printing (if you include 'The Print'). The 5222/5231 stocks have hardly changed in 50 years so there is no reason you can't get the same results. First, get the processing and exposure right. I am now working with Kodak Chalon to establish the reason for several customers having underexposed B&W negatives even if our development is at the proper gamma. Most likely the given exposure ratings are too optimistic, will know more in a couple weeks. If you do the speed point test as described by Adams everything will fall in place nicely. Second, not many labs still do proper B&W with sensitometric tests for each production, controlling gamma to the final print. Third, learn to "see" in B&W for lighting and composition. I would say that of the B&W short films we do, they seem to win more festival awards than the 'others'.

Inside the can, your film should be in a folded balck plastic bag. If this is the case then there will be no fogging. I hope they didn't use transparent plastic bags.. Seriously if it was a lab doing the splitting of the cans they should have plenty of black bags and know how to fold them.

Why don't more people shoot 2 perf: very simple: there is hardly any infrastructure for that non-standard format. Just providing 3perf facilities is a major undertaking, it took about 18 months before the first productions in 3Perf started to come in due to lack of cameras. Currently two 35/3P S35 productions underway at the same time. Serious lack of information too, some DOP's were convinced that 3Perf could only be digitally postproduced, this is not true we do both optical and digital. 3 perf is probably 10 times more widespread than 2 perf and it is already having a hard time.

To have full effect of pin-registration you need both the camera and the projector movement with pin-registration. The equipment shown belongs in the DIY class of equipment which may be fine for some people but not for a serious lab.

Of the whole Kodak negative range, I think 7245 is the one with the least tolerance for underexposure. In bright sunny conditions, no better stock can be found. I can show you helicopter shots in the high Alps, amazing detail in the sunlit snow, yet shadows look neutral and not blocked. Interior with tungsten lighting? Forget it! Waste of money on good stock.

I have Baselight since version 1. One big advantage is that it removes the pressure from the telecine suite. It is much more comfortable to do disk-to-disk grading than a direct feed from a telecine. The images already on disk are just zeros and ones and won't change unless ordered to. The work on a Baselight can be resumed anytime, in six months .. the images will be identical, try that on a telecine. Also I can jump instantly to the first or the last frame of the show, put 9 different parts simultaneously on the screen side by side, etc.. The future of telecine is as a pure technical scanner where D-min is measured and set to Cineon values of 95 with no further intervention of the operator.

Just another datapoint: to be able to do 3 perf in my lab was an investment of about 35000?, not including the testing we did, just to allow our customers to save money on stock. We charge a small fee for any work in 3perf that requires special adapted equipment. On the other hand, 3Perf is really taking off for feature films here, of 4 productions currently shooting we have two on 3 perf, one in CScope and one in 1.85 4 Perf as well as one S16. ALL are shot on Aaton.