Dom Jaeger

I'm glad you had a good time with the D-21 Adrian, I've been championing that baby for ages! Also like your "foot-held" technique, people often underestimate the sensitivity of their lower prehensibilities. Should be more foot-held going on. On a technical note, the Alexa Studio is really only a stop faster than the D-21, since its mechanical shutter halves the native 800 ISO down to 400.. not that Arri like to admit it.

Hi Aaron, I've pulled a lot of wind-up cameras apart, but not that one (yet), so this is just generic advice.. :) If you can still inch the camera via the sprocket wheel when you press the release trigger it suggests the clockwork isn't jammed by debris in the gears or a bent shutter etc. It also suggests the problem isn't the trigger, since it's allowing the mechanism to turn over. What sometimes happens with spring motors is that the connection at one of the two ends of the spring breaks and the spring can no longer drive the mechanism, but in that case the winding handle will just keep turning without tensioning the spring, or it winds up for a bit and then the tension slips. If that's happening then it's not really worth fixing. If the spring seems fully wound and the winder won't turn any more, inching the camera by hand for a while should relieve the spring tension and let you wind it up a bit more. In this case, the problem could be fixable. You might just need to clean and relubricate the mechanism. Or possibly the gears that stop the spring before it slows down have slipped and are stopping it early. Get out your screwdrivers and have a look.. could be a fun weekend project! Feel free to PM me if you have any questions about the procedure, I might be able to help.. or not! But I have a lot of time for people who still value their old wind-up film cameras. :)

I'm sure Deakins knows what he's doing too, but the fact remains that IR pollution can be controlled with IRNDs and testing, and any green colour shift can be corrected with a white balance on a grey card and/or in post. Maybe you should blame the colourist not the camera? I haven't seen Skyfall so I can't comment on it in particular but can you name any other movies shot on Alexas that show the problem? One point of interest, Skyfall was I believe one of the first movies to use the Alexa Studio (including prototypes), which has a built-in ND filter that can be moved in front of the sensor when more than 4 stops of ND are needed. The filter has some IR filtration, but gives a slightly different colour shift than a standard Alexa using IRNDs. Alexa Plus and M were also used on the shoot. May or may not be a factor. There were also apparently issues with the Imax conversion affecting the colours. Deakins' own website has some interesting discussions about Skyfall, including his disappointment that RED footage had to be used (for 2nd unit aerial shots), because it shows clipping. Not our experience either, the rental house I work for has 6 Alexas including the first one in the country and no major problems encountered. No camera system is perfect, and coming from a background of servicing Arri film cameras I've had a few issues with the Alexa build quality (mainly the EV), but in this age of lowest denominator disposeable digital products it seems to be the best and most reliable product out there. Certainly preferred by virtually every DP and operator I talk to.

Hardly an issue that only affects Alexa! Usually solved with the right IRND (or Tiffen's T1, T1/2) filters.

The rental house should be able to check your camera's flange focal depth and ground glass focus (and their lens) to make sure everything is within spec. Takes about 15 minutes on a bench collimator. At the rental house I work for I usually do that for free for students renting our lenses who bring in a film school camera body.

There's not much between them really, both are top notch S16 zooms with barely any breathing and minimum focus of less than 2 feet. On projection the 7-63 looks a smidgeon sharper across the field when wide open. Both have an image circle of at least 16mm diameter through their range, the 7-63 degrades less past the S16 boundary. The 8-64 is about an inch shorter and slender all the way til the very front, whereas the 7-63 carries the same 95mm front diameter halfway down the body. Interesting how much extra girth they needed for just one more mm of focal length at the wide end. When you need to go wide though, there is a surprising difference between 8mm and 7mm, more than you might think. The multi-coatings are similar, 8-64 looks a little more purple. The 8-64 rear element protrudes about 24mm behind the PL mount, compared to nearly 27mm for the 7-63. The 7-63 ramping is barely noticeable, but yes, if you zoom in all the way at T2.6 it will darken a quarter stop or so beyond 50mm.

Hi Deborah, one of the forum rules here is that you need to use your full name. Ernemann were one of a number of photographic firms active in Dresden in the early 20th century, many of which merged in 1926 to form the legendary Zeiss Ikon. They made a few early movie cameras and projectors for amateur use, in 17.5mm and 35mm gauges, and many plate and folding cameras. In the early 20's they released what would become their greatest legacy, the Ernamox, a smallish camera taking 4.5 x 6 cm plates. What was revolutionary was not the camera but the lens, a 100mm f/2 Ernostar (very soon improved to f/1.8), which was astonishingly fast and sharp for the time. A range of focal lengths followed. The Ernostar was a variation on the Cooke triplet, designed by Ernemann's precocious young lens designer Ludwig Bertele when he was only 23. He went on to work for Zeiss Ikon and produced many other great lens designs, including the Zeiss Sonnar, which was a refinement of the Ernostar. The high speed allowed candid photography that was impossible previously, and probably marks the beginnings of what we'd now call photo-journalism. The manufacture of Ernostars continued under Zeiss Ikon until the early 30s, branded Zeiss Jena. The only movie cameras I know of that used Ernostars were Ernemann's Kinnette, and Ica or Zeiss Ikon Kinamos. I've seen pictures of earlier Ernemann 17.5mm 'Kino' cameras retro-fitted with them, and also certain Debrie Parvos which could take virtually any cine lens re-mounted in the Debrie mount.

Sure it's parfocal, it's really only still photography zooms that you need to worry about. One of the main design parameters of zooms made for cine cameras is that they hold focus through a zoom. How well a particular cine zoom holds focus depends on a few factors - the quality of the lens, the zoom range, but most importantly whether the back-focus is exactly correct, and whether there is any wear to the zoom barrel cam tracks or followers. If the back-focus is even slightly out the zoom will progressively lose focus towards the wide end. When assessing a zoom, a technician can make a graph of the zoom curve, mapping how far the focus drifts in both directions of the zoom travel. There are always certain tolerances allowed, often the focus is good at both ends but may form an S-shape drift through the range. On higher end zooms an adjustment can be made to the spacing between variator and compensator to correct the curve. From my experience the Kern zooms are usually pretty good at holding focus, but wear or dried lubricant in the zoom mechanism can cause a focus jump when the zoom direction is changed.

If the camera has a PL mount with '1' and '2' printed on it switching between Super and Standard is easy, the 6 mount screws just need to be undone, the mount rotated 180 degrees and then re-secured. When the '2' is at top right the mount is centered for Super. (The PL locating pin also needs to be shifted back to its original position, the baseplate adjusted, and maybe the video tap re-centered, all easy to do.) It's a good idea to check the camera flange focal depth afterwards, even though most of the time it stays the same. Converting between 4-perf and 3-perf is a much bigger job, definitely not something that could be done on set. The whole movement needs to be replaced, plus modifications done to the sprocket drive timing and footage counter. I'd hazard a guess that most BL4Ss would stay permanently converted, unlike Arricams which can be converted in-house in about half a day. I'd also guess that the standard option for BL4Ss was originally 4-perf.

The lens itself doesn't have filter threads and the front diameter is only 75mm. Angenieux made a clip-on hood (engraved R7) that fits over the front of the lens which has a 93mm front thread. It's not pictured in the Visual Products ad.

Yes, the high speed mags have a single-piece pressure plate, compared to the two-piece plate on standard mags, and the high speed camera flange depth is set 0.03 mm shorter than in a standard camera. These differences allow more 'breathing' space for the film as it rushes through the gate at high speed. It's not recommended, but if for some reason a high speed mag was used on a standard body it wouldn't break anything.

It's a very simple conversion, just the lens mount gets swapped over. On long focal lengths with a large focus travel it's possible to adjust the back-focus instead, bringing the collimated focal plane 2.2 mm forward. I haven't done that with those Yvars personally, so I'm not sure if you lose some range at minimum focus. Swapping over the mounts means you don't have to adjust the back-focus. Any lens tech should be able to do it easily, they just need the part. Here's what the 2 mounts look like: The H8 RX mount is longer - it effectively shims the lens 2.2 mm out from the mounting surface, bringing the focal plane forward so that the lens images 15.3 mm behind the mount instead of 17.52 mm as all standard C mount lenses do. The 20.76 mm figure for H16 RX cameras is the physical flange depth of the camera, because the prism in the path diffracts the light rays and extends the focal plane back, but in air H16 RX lenses are still set to image 17.52 mm behind the lens mount, just like non-RX C mounts. The only difference is the RX lenses are corrected in their optical design for the aberrations introduced by the prism.

This is a partial list of S16 lens image circles that I've measured, using a projected reticle like Matt. I haven't found any focal lengths under 16mm that would cover the Black Magic sensor (~18mm image circle). Optex 5.5 ~15 mm Kinoptic 5.7 ~14.5 mm Angenieux 5.9 ~14 mm Zeiss Mk3 9.5 ~14.5 mm Zeiss Mk3 12 ~16.5 mm Zeiss Mk3 16 ~19 mm Kowa Ultra T 9 ~14 mm Kowa Ultra T 12.5 ~15.5 mm Kowa Ultra T 16 ~16 mm Switar 10 ~14.5 mm Switar 25 ~15 mm Cooke SK4 6 ~16 mm Cooke SK4 9.5 ~16 mm Cooke SK4 12 ~16mm Out of that list, Zeiss Super Speeds are the only option, with the 16 (and 25) covering. From Matt's description it sounds like the Zeiss Ultra 16s all have an image circle of ~16mm, similar to the Cooke SK4s. With zooms, you can get partial coverage at the long end, and go a bit wider if you don't stop down past T8: Zeiss 11-110 covers from 50-110 Canon 11-165 covers from 50-165 Canon 8-64 covers from 30-64 Angenieux 7-81 covers from 25-81 Canon 7-63 covers from 20-63 It's possible to increase the image circle of a lens with a focal length extender, which is basically how old Standard 16 lenses were converted to S16. I tested a Zeiss 10-100 with a doubler for example and found that it would cover the Black Magic sensor through the whole range (20-200).

The paint is a type of wrinkle finish enamel used on a lot of older photographic and optical equipment. I believe it was a two-pack paint, hardened by baking. Later Arris used a 'spatter' paint finish. For cleaning you can use a damp rag with a mild cleaning agent, window cleaner works well. Some people like to use shellite (napthal or Zippo fluid in the US). For tough stains try a citrus cleaning agent, but old tape residue in particular can be a bitch to clean off. A good quality corrosion inhibitor like CRC 2-26 or 5-56 lightly applied with a cloth, then wiped down with a clean rag is a good idea if metal is starting to tarnish, I also use it with a cotton bud on screw heads. Avoid acetone, thinners or other strong cleaning agents that can discolour the paintwork. You can use automotive touch-up paint to fill in scratches, or wrinkle finish spray enamel for larger areas but it can be tricky to match the colour exactly, and the surface needs to be thoroughly clean and de-greased. For a proper restoration you need to strip the camera right down, a very big job.

OK, a 2x Mutar clears the mirror but since you can't blimp the 10-100 it was made for I guess Arri decided it was suitable for SRs only. Without the blimp support, a heavy zoom could also put pressure on the rubber shock mounts that hold the floating mechanism and lens mount in the body. Open to other theories though...

The flariest (is that a word?) anamorphics are ones with the anamorphic elements right up front, like Lomo square-fronts and other early anamorphic systems. They also tend to mump because as the anamorphic group is moved to change focus the squeeze factor alters, especially at close focus. Round-front Lomos have a fixed anamorphic group in the middle, in front of a spherical lens set near infinity, with a spherical focusing element at the very front, so not quite as easy to flare, no mumps and less breathing. Of the anamorphics I've seen, square-front Lomos, Arri Ultrascopes, Cooke Xtal Express tend to flare white, Panavision, round-front Lomos, Bausch and Lomb Cinemascopes more to blue. The colour of the lighting can also have an effect, from memory under tungsten the Lomo round-front flares are more blue, under HMIs they can go to orange. You'll get the least flare from modern lenses with highly efficient multicoatings like the Hawks, also far sharper edge to edge than Lomos. If you can afford them they fit your brief. I don't have any experience with the Optica Elites, but I've heard they're reasonably flare resistant. Judging from their other lens lines I'd expect the build quality to be better than Lomos, but well below Hawks. Virtually every anamorphic zoom works by having an anamorphic adapter at the very back, so no horizontal flaring there either. Looks more like cropped spherical. On a budget, old Angenieux HRs and Cooke 18-100s work well, but you lose a few stops and sharpness (esp the HR) could be an issue. Good luck with the job Evan, it's nice to hear of people still shooting 4-perf anamorphic. You shot a Grimes vid on 35mm if I recall? I've had quite a few people mention that one to me as a standout. Well done!

Hmm, tricky one. All the literature I've got for 2x Mutars specifically mentions it being for the SR, the rear baffle has "only for Arriflex 16SR" printed on it, but the rear element doesn't protrude very far, and when I carefully fitted one to a 16BL the mirror cleared. I'll do some more digging.

Hey Adrian, I'd love to hear how that shoot goes. We have a D-21 in our rental fleet which just never goes out, the perception seems to be that it's not up to Alexa standards, too slow etc, but aside from slightly less DR and a slower native sensor speed the thing makes beautiful images, quite filmic, as long as you light for it. Anyway, I'll be curious to hear how it cuts with S16, sounds like an interesting project. Good luck with the move, too!

Unfortunately there's not a lot of choice in lenses for the H8 RX. Apart from the 5.5, 12.5 and 36mm Kern Primes, the 8-36 Vario-Switar zooms and the Berthiot Pan Cinor 40 RX, there were H8 versions of the 100mm and 150mm Kern Yvars. They can also be converted from H16 to H8. A new mount can be fitted, which is about 2.2mm longer and is engraved 'H8 RX'. More tricky to identify are those that have simply had the back-focus adjusted, but may not be marked as H8. With shorter focal lengths the problem is not just the different back-focus from H16 RX, but the difference in prism thickness. The H16 RX lenses are designed to correct for a thicker prism, and need more correction for the introduced spherical aberration than the H8 RX ones. I'm guessing that's why Kern didn't offer any other 16mm RX lenses for conversion. But it's possible to get them to infinity focus on a reflex H8 by shimming out the mount by 2.2mm (though the lens is then only screwed in by one or two threads), and by stopping down a little the aberration will diminish (if you even notice it). Some older (non-RX) C-mount lenses have longer mount threads, so you could shim them out and still have enough thread for a more secure mounting. The effect will be a reverse spherical aberration to using H16 RX lenses (ie positive vs negative), but again a little stopping down will reduce it. It's a shame C-mounts are becoming so pricey though, I think it's a bit mad spending hundreds of bucks on a single lens for Standard 8mm. In particular I don't know why the H8 RX primes are so expensive. There can't be that many H8 RX users out there, so it must be digital photographers, but why do they want to pay hundreds of dollars for a tiny image circle with spherical aberration? :blink:

I just 'down-loaded' this great exposure calculating app from e-bay ;) : Despite being around 90 years old, it fits snugly in the palm just like an iPhone. Samsung should have brought one along to their court case with Apple, who obviously stole the iPhone form-factor from Pathe.. The six different sliders along the edges all combine via a complex series of interlocking levers to output an aperture f-stop. Anything below f/3.5 gives a reading of "insufficient light", which gives you an idea of the lens speeds available in the day. There's also no provision for film stock speed, since only one speed was available. It still works though, which after 90 years is more than you could hope for from your iPhone. B)

Interesting thread.. i think it's a problem that faces any artist who knows their craft. I'm sure novelists have trouble enjoying books without critically analysing them, or painters find themselves studying brushstrokes and technique instead of just taking a picture in. As others have said, it both enhances and detracts from the experience. I sometimes have to force myself not to analyse art and just enjoy it, but I prefer that (slight) burden to the alternative of being ignorant of the skill behind its creation.

Also possibly some flicker from the arc lighting? Certainly not shutter speed variations, even a large variation of 1 or 2 fps would only be a tiny fraction of a stop in exposure. Not sure what a 'faulty shutter' means in this context, if it was out of sync you'd get constant smearing not exposure variation, the shutter angle wouldn't change frame to frame. Uncoated lenses might cause some flare or reflections or loss of contrast but again it would be consistent, not frame to frame.

I've collected some image circle data on 16mm format lenses, but it's not always cut and dry. The circle often shrinks when a lens is stopped down, so for example a 25mm Kern Switar will cover the roughly 18mm image circle needed for the Black-Magic sensor when fully open, but stopped down past about f/2 you'll get more and more vignetting. The 50mm Switar covers easily. I haven't checked a 12mm Zeiss Super Speed, but the 16 covers (just), the 8 and 9.5 do not. Most of the wider focal lengths I've measured won't cover an 18mm image circle. That includes Kowa 16, 12.5 and 9, Angenieux 5.9, Kinoptic 5.7, Optex 5.5, Switar 10 and Cooke SK4 6, 9.5 and 12. Anything 50mm and over shouldn't be a problem, but there's larger format options for that end anyway. Zooms tend to be worse than primes, I can't find any that wouldn't vignette on the BMCC under 40mm. With zooms the image circle changes more with focus distance, so at 40mm they might cover at infinity but vignette when focussed under 10 feet. One possibility is using zooms with a doubler. A Zeiss 10-100 with a 2x Mutar for example would cover an 18mm image circle all the way from 20 to 200, providing you didn't stop down too much past T8. Of course even if a 16 or S16 lens covers there's no guarantees that the image won't be degraded in the corners, since they were only optimised to cover that size image circle.

Congratulations Brian, I see you were centrally involved in the restoration. Great job! I also noticed that Edward Turner now has a freshly referenced entry in Wikipedia as the earliest known producer of colour motion pictures.

4 perf Arricams and Moviecam Compacts can be easily converted to 3 perf if the rental house has the conversion kit. Check with your local rental houses.