Robert Hart

Dom. Thank you indeed for the correction and my apologies to other readers for misinforming you. Age does not necessarily weary us but the brains sometimes addle. The light loss "seemed" to be the case with my specimen back in 1980s and I think now I have confused my recollections with also using the tungsten balanced reversal film I was supplied with for the news work and having to use a colour correction filter for which exposure had to be adjusted. My camera may have some of the light-stealing characters induced by surface contamination of the splitter optics described in an earlier post. I certainly did not then nor have since opened it up to check. I should dig it out of storage and have a look. Advice which was given to me at the time was that the lenses for RX had been adjusted so that the focus and aperture marks behaved as if there was no light loss. This was word-of-mouth and not a document so it is a fair bet I was misinformed way back then. Groundglass textures are a curious thing. They do steal light but apparently not in a linear fashion related to the lens aperture setting.

The splitter optic, from memory, takes away about 1.5 stops but I could be wrong. It has been a long time. The groundglass itself will take another 1.5 stops of brightness from your eye view. So darkness in the viewfinder image is its normal state. The lenses specifically for use with the RX camera have apparently been re-marked so that the aperture settings are "as if" there is no splitter brightness penalty for a given light meter reading. The lenses are also collimated to this specific camera type and will be off if used on another non-reflex type. They will focus but the witness marks on the barrel of the lens will be wrong. If using lenses not specially calibrated for the RX camera, you then need to add another 1 to 1.5 stops of exposure to your light meter's recommended aperture setting and focus with the groundglass viewfinder, not the witness marks which will be wrong. A common practice in this circumstance was to set the light meter "as if" the ISO of the film was 1 to 1.5 stops slower and then use it normally without having to remember to count back stops on the aperture setting. It was too easy to muck up and go 1.5 stops the wrong direction in haste and end up three stops wrong. A necessary normal practice when using these cameras is to open your lens aperture wide for both viewfinder brightness and the shallowest depth-of field which pretty much guarantees a sharp focus at a tighter aperture. You set focus, then reset your aperture to correct exposure, then frame as best you can with the groundglass finder or revert to the much easier parallax side-finder which should be in your kit and attached to the left-side door. Forgetting to reset the aperture after a focus trim is the common trap which will bite you. There is an adjustment knob on the parallax finder to adjust the correct offset for distance from subject.

I too can go the mirror and lens route with an old Sony 90degree telecine adaptor for old 2/3" EIAJ video camera-recorders. It is okay but not outstanding and a frame sync issue exists. The SI2K is essentially a 2/3" sensor with a few extra pixel rows around the outside. Because it is a CMOS camera and rolling shutter is an issue, the prism shutter has to remain in the Steenbeck. - Once the camera is slaved to a Steenbeck generated trigger pulse, then the shutter speed can be made faster so that the blurry frame transitions are no longer visible. There remain optical internal reflection artifacts from the prism. These do not permit scanning of neg film because of the heavy-duty contrast which has to be added after inversion. This amplifies them to the detriment of the recovered image. Direct imaging to the camera is achieved by removing the screen box, adjustable surface-coated mirror tower and erecting prism from the Steenbeck's optical path. The small 50mm projection lens is replaced by a Fujian 50mm camera lens set well forward of the camera as a camera macro lens. I stacked CS/C-Mount adaptors for a ballpark forward macro offset and shimmed between a few of them for a final trim. The Steenbeck lens has no controllable iris. With this arrangement and the prism stopped and centred to static tests, the sharpness is acceptable and with careful focus adjustments matches much other HD scanner/telecine imaging I have seen published on Youtube. With the prism removed, the best sharpness is achieved by using the Fujian lens as a projection lens with the tail of the lens facing the film plane. I cannot post an image here but here is a link to a post on dvinfo.net regarding the arrangement. If you look at the still image, you will observe a couple of 4x4 filters improvised into the optical path to bring the Steenbeck's lamp down to acceptable brightness and a IR-750 filter to take out the infrared which is very severe from that lamp. http://www.dvinfo.net/forum/silicon-imaging-si-2k/533332-si2k-telecine.html

FOOTNOTE TO ABOVE. I have also been using a SI2K in conjunction with an old Steenbeck ST16 flatbed editor as a sort of improvised telecine. I have yet to invent and build a trigger generator to sync the frame rate to the Steenbeck. It is not outstanding but for the time being is adequate.

Bolex H16. - paid for itself with stringer news work. Bolex H16RX5 - paid for itself with stringer news work. Auricon Cine Voice. Sold back to the original seller as a museum piece. Good image but nightmare to use. CP16A - Didn't pay for itself with stringer new work as video which was horrendously unaffordable took over. CP16RA - Likewise as a news cam. CP16RA - Bought as a spares donor but bore little resemblance mechanically to the other "RA". Sony DVR-PD150P. Sony HVR-Z1P. - Still used for some events work. Sony PMW-EX1. - Still used for some events work with large SxS cards or BM SDI recorder. Silicon Imaging/P+S Technik SI2K. I bought it as a barebones ex-demo to avoid it being lost to local production. The Cineform-based ecosystem did not take off here then along came RED and the system probably became the most rapidly depreciated camera system of all time. - I still use it for some long form events work with ENG lenses, now that SSDs have become fast, reliable and of higher capacity. Second U/S SI2K as a parts donor but I actually got it working so it became a second events cam. This was a freebie for the local community college. Students operated the second tripod cam and the roving cam ( EX1 ) The SI2Ks are not ideal in low-light situations. For those interested in sound, the desk audio feed did not eventuate so the camera ambience audio was used. This was fed to the camera from a DECCA tree array of three mikes mixed down to a stereo pair through a SD-302 mixer. Mikes were three Rode NT2a, the wing mikes each to individual L and R channels as figure 8 and the centre mike as omni shared to both L and R channels. I likely will not buy any furthur cameras as they become redundent feather-dusters far too soon these days.

You can run full magnetic film through a CP16 and record to it, even the thin mylar but there is a real crisis of confidence in using the thin stuff. I have done so but it is a really poor substitute for a proper recorder. On using film. I love film but unless you have a really good film-trained camera operator / DP you could be on a hiding to nowhere. A critical shoot with film is no place for a learner driver. There is also so much which can go wrong as the skillsets and intellectual property associated with 16m film diminish, shipping, processing, shipping back, telecine/scanning to digital. Your cast will need to be solidly rehearsed. You will need to be on the page with your actors, directing, solidly briefing with and then delegating to and trusting your DP, not taking refuge in the tech, not freaking out over the tech and standing over your camera operator "just in case". In some circumstances, probably many, over-rehearsing can can hurt the spontaneity of a performance but with film, you cannot afford to roll endlessly and wait for the golden moments. If you have inexperienced actors who are perhaps a little too much "into the zone" or "the art", they may not project their voices adequately or may be arrogant enough to mumble low and blame the sound recordist when it turns out to be "via the armpit through a sock" dialogue quality. Take after take trying to get them to get it right and stressing them because you are stressed as you watch the dollars worth of film stock sprout wings and fly away may send you all slightly insane. Sound is also a craft you cannot afford to neglect. With film, the vision media is not reusable. My vote would be with much regret, going the digital route. That said, two winning little short films shot by some locals here in WA, "Luger Story" and "Good Pretender" were deliberately shot on 16mm film - with a solid and experienced veteran DP recruited for the job. "Good Pretender" was also the launch pad for a wonderful little actor, now wonderful bigger actor Olivia DeJonge. The mechanism of the CP16 cameras with their magnetic recording head was styled after an earlier analogue optical sound recording system in the Auricon series of cameras. Optical sound on purposed film-based sound recorders was an actual double-system recording method as well. Digital optical would not be impossible to implement but the economies of scale just are not there.

I have gone about it mean and cheap in two ways. One was to add some extra pieces to an old West Coast style heavy truck mirror, lay in on its side in the window space with a couple of expansion struts as well as the original telescopic stays. There are more modern mirrors these days but you might find one in a scrap yard. The other was to make two broad plywood hooks contoured to the shape of a light sedan door, inside down to the armrest and outside to the bottom of the door where there is a solid lower structure inside the door to prevent the plywood from denting the panelwork at the lower edge. The plywood should reach the door bottom edge but be clear of the sill panel below the door. It is only good for the car you choose so be sure what sort of car it is to be and stick with it. My particular specimen was for a Toyota Echo two-door hatch. It rides quite well as it is but with four crewies weighing it down it will ride softer, The inner surfaces of the hooks need to be padded with thin firm stick-on foam strip to protect the paintwork. This needs to be kept clean of sand and road grit which may work underneath. Both hooks are spaced as wide apart as the lower window edge will permit. There are three 30mm by 50mm pine ties, one on the inside below the window bottom edge with a camera attach point, another on the outside about the same height below the window edge with a camera attach point. The lowest tie has a hole drilled through it allow a padded gauge-plate hook and tiebolt with a wingnut to tension onto the lower door edge. The pine ties are attached with five long coach hex-headed screws at each end. This is necessary because drill holes into end-grain strip a lot more easily than across grain or on the radial. You also need to use glue on the joints so that there is no flex of the structure. The glue has to be added into the joints with the screws backed off and the gaps opened, the mount assembled to the car so that the joints firm up in exact compliance to the shape of the door when the coach screws are pulled down and the glue sets. Do a dry run with the screws tightened first to be sure that the hooks will slide off the door when you want to dismount the rig. This arrangement has to bonus of allowing the door to open and close with camera mounted. For the mount I chose a Manfrotto geared head. These can be had on eBay for about US$400 last I checked. However a Miller DS10 head with a longer stud and a wingnut would function equally as well. This arrangement permits cross cab interiors, car to car and interior to exterior views. It is not as versatile as a fancy rig but you can drill as many holes in the wood as you want for camera positions.

CONT'D FROM ABOVE. Here is the address of the quadbike rig. http://exilemovie.com/gallery/main.php?g2_itemId=165 And here is the website gallery which remains extant. http://www.exilemovie.com/gallery.html

I would have some concerns at following immediately behind them with a pickup vehicle any time forwards or reverse. If they bail sideways in front of you it is "thum-thump" a little lurch and game over. A pickup simply is not agile enough for dodge-em. My personal preference might be use a farmer's quadbike. With their balloon tyres they are a softer ride. The downside is possibly more noise. Park your cammo on the front carrier and another person on the rear carrier as ballast for the follows and your cammo on the rear and the person on the front for ballast. The driver is in a much better position to be aware and react. The overloaded quad might roll though in a desperate dodge. This method was used quite successfully and effectively for a leading shot of a running teenager in a pre-prod teaser of an unrealised feature titled "Exile", about Cuban rafters. There was quite a good behind-the-scenes clip published for that project but it is a few years since and it might have gone by now. I'll have a look. In meantime I found the trailer.

If you are not using the lenses' own reflex finder and have removed it, you may need to cover the exposed viewfinder port to keep light out or you may get flare.

I guess you are already doing it but yes, any newly exposed bare metal needs to be coated with a matte finish. I used black marker ink ( parcel pen or "Sharpie" ) and found that was too shiny. The flare only showed when a really hot pinpoint highlight moved in the image until it hit frame edge, then there was only a faint flare into the opposite edge of the neighbouring frame. It can happen with unmodified cameras. You sometimes see it in some older movies.

No doubt you are already onto the following subjects, however another mention can't hurt. When you dress the inside of the extended gate, take care to do the final sizing gently with a piece of smooth machined bare metal and not so vigorously as to induce friction galling of the alloy gate which will ruin the smoothness of the visible frame edge. Like the salt in the stew, you cannot take an error back out. Also recoat the worked surfaces or you may get internal reflections off the top and bottom edges onto adjacent frames.

When applying movement to the dolly, use a drawstring attached as near to table level as you can to avoid a tipping moment also being applied. Adding weight should help. The easiest method might be to beg some wheel weights from your local tyrefitter and gaffertape those on to the structure at as low a position as you can. Make sure the wheels are scrupulously clean so no bits of grit cause the wheel to hop. Spin the wheels to check for runout. Silicon tyred rollerblade wheels may develop flat spots if the dolly has been stored on the wheels and not propped upright. If moisture has entered the bearings, they may also be running out or momentarily catching, causing jolts. To control the movement, maybe tie another drawstring on the rear and attach to it, a weight with a flat surface wrapped in cloth to provide a dragging load which is easier to controllably pull against. Don't cut the drawstring too short or it will pull upwards and will jump. If the tabletop is too small then hang the load on a long string over the edge.

To my eye, the artifact seems to be entirely heat shimmer which you will get even on a cool clear day in a city when there is a mix of hot steel roofs, metal sidings, concrete roofs, tarred roads and airconditioning towers or boiler/heaters putting out concentrated volumes of warm air.

Dom. I wonder if letting the lens roll to a near-death experience is a method to create dramatic tension in what might otherwise be a dull technical presentation.

Ken Hale at Whitehouse AV used to make and install stainless steel PL-Mount replacements for the CP-Mounts on some Angenieux lenses. I don't know how well he is travelling these days. He was sick quite a while back. There was a guy with an english accent who used to work at Whitehouse AV when I visited there in 1996 and had modified a CP16R for PL and Super16mm. I think he may have moved to Visual Products. An enquiry to either might find a solution for your lens.

Peter. I think the ARRI mount you refer to is the ARRI "B" or bayonet development of the original ARRI standard mount. From recall and checking the Wiki on the mount, the flange to focal plane is the same as PL at 52mm with a shoulder diameter of 41mm. The distance along the should from flange face to front and rear of key and from flange face to circlip channel I would have to measure. The original ARRI standard mount design was dreadfully imagined with the lens free to turn inside the mount. I can only assume that the original reasoning for the free-turning mount and the "ears" on the lens barrel may have been that by turning the actual lens body and holding the front focus ring to refocus, that any optical defects in the image would not turn with focus adjustments. The image shift during focus pulls must have been horrendous. A couple of old fungus-rescue Cooke Series 2's with aluminium mounts have considerable wear on the mount shoulders. Different lens manufacturers seem to have evolved their own methods of creating the bayonet key in their ARRI B mount tails. One example I have seen was setting an entire key shape into a woodruff key style slot with the outer piece cut to the ARRI profile and secured with small screws. Because of often small workspaces within a design, that option mat not always be available. Another style seems to have used a swageing technique to deform the key outwards from the base material and machine that to final shape. Another seemed to have turned and milled the entire key from the base material. An old fungussed Angenieux 25-250mm I repaired had been modified by somebody back to the original ARRI standard shape so I no longer have a reference to measure for you for the ARRI B. I added a chinese PL-Mount adaptor to that one and hit lucky gold in that it required no re-collimation to zoom true. Keeping the fungus under control is another matter.

There are some really low-cost C-Mount lenses out of China which wear the brand name "Fujian" in focal lengths 25mm,35mm and 50mm. The image circles on these are suitable for Super16mm. My examples which I bought in for a home-build scanner yield comparable sharpness numbers on the SI2K camera to the Optar Ilumina and CP Ultra T* lenses. The machining work is good but the actual focus movement is rather "basic" relaying upon a grubscrew registering in a single helicoid channel to slide the entire group of optics forwards and rearwards for focus adjustment. However, the fit is snug and with the dampening grease used, there is no observable "hop" or image shift. That might change as they age and the lube thins or dries out. The focus ring is retained from moving off the front of the lens barrel by a screw-in front which may be loose or barely hand-tight when received. There is no secondary grubscrew retention of this threaded ring. When loosening or tightening this front threaded ring, or removing the lens from a baulky C-mount, EXTREME care needs to be taken that no rotating force is applied to the aperture ring or it may ride over the internal actuator or damage the iris blades. There is very little lens body to grab hold of. Your lens focus ring may be found to be immovable when you receive the lens. This will likely be due to the lube having dried near the gaps and hardened. Unscrewing the front a few turns and pulling the focus ring forward will draw the internal lens block forward, crack the lube and it will move freely and smoothly thereafter. Take care not to use too much force or the grub screw may be broken off. If the lens block is really frozen in dried grease, it is better to screw the focus ring off entirely and then attempt to push the lens block forward from the rear taking care not to damage or soil the rear optical element by protecting it with cloth or tissue. http://www.ebay.com/itm/FUJIAN-35mm-F-1-7-C-mount-CCTV-lens-for-SONY-E-Mount-NEX-5-NEX-3-5N-C3-hood-/281880872114?hash=item41a168c4b2:g:l~UAAOSw-4BXYOzQ If you buy on eBay from the chinese vendors be careful to additionally state in your payment message that a specific focal length is required. When ordering a 50mm lens I received a 35mm lens in error. The costs involved do not justify sending one back.

I second Satsuki Murashige's suggestion over anything I have suggested.

This suggests there may be insufficient pressure from the pressure plate springs but you may be also getting a short loop. My sense is that there are two things happening. A short loop which is being tolerated by the camera may pull the film slightly off the focal plane. More commonly, you will get pulled or vertically smeared frames and noisy operation from this condition The film's normal state as you have it running is slightly out of focus. Temporary ripples which normally form in the film when it is parked in the transport will lift a "soft" pressure plate as they pass through and change the focus momentarily. In your example, those ripples may be pulling the film into correct focus momentarily. This hints that your lens optical focal plane at infinity may be rearwards of the camera's mechanical focal plane. Momentary focal shift on startup is a near-to-normal condition with the CP16 cameras but I have not seen it on a Bolex H16 type. Let's first assume you may just have a loading problem and your camera is otherwise normal. Do you use the little chrome scissor-guillotine blade inside the bottom of the camera body to cut the required shape on the head of the film before you push it into the first guide when "autoloading"? ( H16RX5 ). On earlier H16 models, the "autoload" feature may not exist. Do you "autoload" the film by running the camera continuously at normal shutter speed with the guides in the load position. The better method is to turn the governor back to its slowest frame rate and progress the film through the loading guides in very short bursts. Once you have the film threaded through, there is a pin near the centre which is contacted by the door when it is closed. This normally retracts the guides as the door is replaced to the body but will not set the pressure plate in its correct position. It is highly likely that Tarzan has been at play with your camera at some point and forced the lid on and latched it. If the inner panel has been warped or the little levers have been bent, it is possible that the door will fasten with no hint that something is not in its correct place. On the inside of the door are donkey traps, little holes into which knurled knobs on the loading guides and pressure plate fit when they are in their correct positions for running film. If the door "rocks" when you are trying to close it, then it is being held high buy one of the knobs which is not in its correct position. There are other reasons for the film being out of focus. The position of the viewfinder groundglass focal plane may have moved out of correct position and may be causing you to incorrectly focus relative to the film focal plane. You need a camera tech to fix this. If you are using a tape measure and witness marks on the lens barrel for focusing - and using a lens not correctly collimated to a Bolex with a reflex prism viewfinder, the focal will be off. Lenses specific to the reflex Bolex are slightly off-focus when used on a non-reflex Bolex and vice-versa. To help others here with better skill than I drill down to some problems and solutions can you reply to the following questions. Does it still have a viewfinder on the top of the body? What type and brand of lenses do you have on your camera? Does your camera have a three-lens turret on the front that you have to rotate to choose your lenses? Do you have the little chrome thumbwheel and retainer on the left lower side of the three lens turret tightened to prevent the turret from falling forward and moving the taking lens out of focus. With the lens directly in front of the viewfinder covered, can you still see through the top viewfinder? That will do for starters.

If you have tried another monitor and it works but the troubled monitor does not, swapping out cables does no good and swapping cameras does no good, you might try applying alternating twist pressures to the monitor body itself very slightly in degree of twist and very gently with a flat hand palm on each side. If the screen image restores or jumps about more, this suggests there may be a dry joint on ribbon cable conductors or other connectors or around the display panel within the monitor itself. Please be exceptionally gentle doing this or you may cause greater damage. One of my SI2K monitors plays up like your example. The LCD panel itself was the culprit for me. Short of replacing it, there was no cure. The temporary fix was to establish positive pressure to the back of the LCD by adding a cube of black foam between the rear of the panel and the rear casework. This worked for a while until the foam eventually slumped one warm day, then it was back to square zero.

I have recently joined this forum to seek a bit if tech advice. It is only fitting that I give something back. Here are some links to some home-baked tutorials on the CP16R film camera. I expect there will be hardly anyone shooting with one these days but you never know. The methods of loading film are my own. There may be better methods so I will appreciate any corrections any more competent folk than I can add to the mix.

I am restoring a complete clean lens of this type from two damaged examples. Before I go to the torment of designing and building a PL or direct IMS mount for this lens, does anyone have a clue on where to find a machine design diagram for a PL mount for this lens? The two specimens I have wear the Cameflex and Aaton mounts. I can calculate and draw my own design but machining a new mount will be easier from a proper design. My intention was to take the easy way out and machine up an ARRI standard tail, attach a PL-Mount adaptor and use shims to get the collimation right. Making the PL-Mount is somewhat of a challenge. I have made them for other lenses and the similar IMS-Mount for a Speedbooster adaptation to the SI2K camera IMS-Mount. Small hobby lathes struggle a bit with the wider diameter of the PL-Mount. Any advice will be greatly appreciated.

Hello Peter. Welcome to the throng. I have not long signed up myself and am nowhere near your league of achievement. Hopefully we may plumb your knowledge for good advice. There are two SI2Ks with identical lens sets over here in Western Australia and they have been put to an un-natural relationship for some local 3D tests. Here are two links to a short clip I extracted from one of the shots. The other more elaborate shots with greenscreen I have no clearance to manipulate and publish. These clips, one an anaglyph render, the other a side-by-side view (crosseyed) render are not optimised for 3D viewing on youtube. Side-by-side, the cameras can be mounted with optical centres about 72mm separated when the IMS mount rings are taken off and C-Mount rings screwed into the fronts. With adjacent side-walls milled down, I have worked out that two heads, wearing small C-mount lenses could be just made to work at the human interocular distance of 66mm or thereabouts. I don't think anyone is going to mutilate their SI2K mini heads anytime soon just to do this. A more practical separation of optical centres which permits acceptable convergence adjustment is about 82mm. Steve Rice who owns one of the cameras and a slumdog rig is off today to Germany to have a look at the Weisscam, the P+S 3D rig and to talk to a 3D shooter/steadycam op in the UK. There is a new product under refinement called the cinedeck which will be about the size of a small laptop and will work with two SI2K Mini camera heads via data cables. This unit is already sold for other consumer/prosumer camera-recorder types. Ari Presler at Silicon imaging can tell you all about it and how far they have progressed. A version has already been improvised and used in the UK. One of those commonly mounted with two camera heads on a Slumdog rig with two eyepieces and follow-focus commonly geared to both lenses would be the ducks guts and we are hopefull that is where the evolution is headed. A geared convergence system would be a bonus. Probably all very pricey too. Currently we are going the cumbersome route of using two Mini heads tethered to their heavy docking recorder units by data cables and linked to a sync box. We bolted the recorder units down to a plate on a post which socketted into one of the seat sockets for his Elemack.

When testing rather unthoroughly I admit, some Nikons against CP Ultra T* lenses on a SI2K, the ones which came closest were the 28mm f1.4 and 58mm Noct Nikkor f1.2 and surprisingly a 55mm? Micro-Nikkor f3.5. This not to say that other Nikon lenses are less sharp but that is how it panned out for me with the specimens I had access to. An older Cooke Speed Panchro Series 2 50mm, effectively a ruined lens due to tropical fungus, after being pulled apart by myself and cleaned, came up to the sharpness of the Noct-Nikkor. It seems that by the time you get into the zone with 35mm stills lenses it seems that you are approaching the cost ballpark of cine lenses and you still don't get lens gears or long control throw on the focus rings and of course they spin the wrong way as well.