Sarah Thompson

Arm is now sold, pending being checked out by Robert Luna. Sled is still available: $3900 for a quick sale.

Update: the vest is now SOLD. The arm and sled are still available, no open offers yet.

Thomas + others, I'm basically looking for offers around the following: Vest $3000 Sled $4000 Arm $3500 Buyer pays shipping. I can do paypal, cashier's check or bank transfer. I'd like to move this quickly, so if someone was to make a sensible offer for some or all of the gear and do Paypal, I can potentially ship same day via FedEx. I'd like to see the gear used rather than sitting here, and I'd love to pass it on to a newbie who will hopefully fare better than I did -- I'm willing to do a very good deal if someone wants the whole rig, for example.

PS: Correction -- I typo'd when I wrote the description of the sled. It's set up for 2 AB batteries not 3. Also, photos for the rig are here: http://findatlantis.com/rig4sale/rig/

PS: I am located in Cupertino, CA, just south of San Francisco. I'm willing to ship anywhere however.

Hi folks, It seems I'm one of the ones that got away. A back injury basically means I can't fly anything for more than 2 or 3 minutes without being in extreme pain, so the gear has to go. :-( I'm selling the following: Series 3 arm with Pelican case. Cosmetically it shows its age, but it works fine (35lbs up to way more than you'd sensibly want to fly). It comes with a set of 4 posts of various lengths with the right diameter for a 3a sled (or equiv). 3a sled. Good condition, has a few modifications including a TFT monitor and 12/24V power. Comes with a variety of plates, cables, etc. Modified to take up to 3 Anton Bauer batteries. Comes with a really solid metal-clad flightcase. I can throw in a Mitchell stand if you want one. Ultra vest. Very good condition. Sorry, I don't have a flight case for the vest. It almost but not quite fits the pelican that the arm lives in (another vest might fit, or it would fit without the arm). I need to sell relatively quickly, so make me an offer. I'm happy to sell the whole lot together or split it out as necessary. I'd also consider taking a Pilot in partial trade (my back is probably fine for that weight, but not for a big rig sadly). Let me know if you want more detailed information/photos/etc.

http://www.youtube.com/embed/WoZ2BgPVtA0 A story about a set of very Steadicam-like arms made for a 2-year-old suffering from arthrogryposis. Some of you may have seen this already, but I was sufficiently boggled by it I had to post it here. :-)

I just bought a used Sony clamshell monitor that happened to use the same batteries. It came with a power supply/charger with a fake battery. I popped it onto the Ninja, and it powered it just fine, so I therefore have a mains power solution. I'll probably modify the cable and build an 8V regulator which should make it possible to power it from about 10V to 25V. I don't know if the Sony parts are still available, but I can dig out the part numbers if anyone wants them.

Did you remember to black balance? ;-)

I'd underscore the awesomeness of McMaster-Carr. I've built Mars robots and specialized cameras for work, and use them all the time because they are incredibly good at having stuff show up next day. They also carry most of the bizarre things I need. They aren't so good on module 0.8 gears, but other than that...

Hi Eric, Who is making the adapter? Atomos or AB? Thanks, Sarah

Hi folks, I've been experimenting recently with an Atomos Ninja HD recorder. Seems to work really nicely, I'm happy with the results, and going straight to Prores is really nifty. However, its out-of-the-box power solution is kind of nasty. It has two Sony NP-F style battery mounts on the back and no other power connectors. The standard (small) batteries supplied with it are just giving about an hour of record time each running ProRes HQ, which (duh) isn't remotely enough to be useful. It can accept larger NP-F style batteries, but since I'm trying to move my rig to AB gold mount for everything, I'd rather power it from that. Thing is, I can find gadgetry that will let me hook an AB battery to it, as if I was powering a camcorder, but I'd rather be able to power it from the same battery as the camera (which might be on a steadicam rig or mounted to the rail system with a cheese plate if I'm using sticks). My Google-fu is failing me -- does anyone know of a suitable adapter (i.e., 12V in, fake 7.4V NP-F battery out)? If it was not for the weirdness of needing a fake battery rather than a plug, I'd probably have just built my own little power regulator, but... Thanks, Sarah

Very nice work -- I went through the whole site this morning. I'm a newbie starting out, piecing together my first rig currently, so I am probably in your exact target audience. I'll PM you with my reactions rather than post them here, but good work!

I've experimented with Mamet's approach. It works, but I'd be uncomfortable calling it better or worse than the alternative (i.e., what he'd call an inflected shot -- a long master, long Steadicam tracking shot, etc.). Realistically it's just another tool in the toolbox for constructing a visual story. It leans less on the actors (so might help if your actor isn't too great, I suppose...) but much more heavily on the editor -- if the rhythm is off, the sequence can stop making sense. It can also help to sell an emotional idea that isn't directly playable. I've recently been doing rewrites on a feature script that was mostly pretty good, but the writer strayed into some lazy action ('Jack notices that <something emotionally complicated happened off screen that you don't see directly>') that translated fairly straightforwardly into juxtaposition. I haven't shot it yet, so I'm crossing my fingers that it'll work in the edit, but it does seem to make some kind of sense. I don't think I could have done the same thing adequately with a long master approach because it would just be no way for the actor to get the information across. OK, here goes, I'll stick my neck out and link an example. First thing I'll say is that Spin was never really intended to be a short film -- it was shot very quickly (couple of hours) mostly as a source of test green screen footage for checking out a post toolchain (compositing/CG/sound). I can list a lot of things I'd reshoot or do differently if I was really aiming for it to stand up as a 'real' project. However, another thing I was trying was Mamet's approach, partly because I'd just read a couple of his books -- the short has no dialogue, and in fact its story never really appears directly, it's all inferred through juxtaposition (of sound as well as images). Its nearly all composed of locked-off shots because every shot has a green screen comp going on with some combination of matte painting or CG background, and I wasn't feeling like spending a week or several doing matchmoves.

Two pieces of advice SEARCH WORKSHOP then come back and ask your questions <3-year-old-tantrum-mode> But I wannit an I wannit NOW! </3-year-old-tantrum-mode> Good advice, thanks. Really. The workshop part won't happen until next year most likely -- it would be awesome if there was one I could get to sooner (or, listed at all currently, actually), but I'll have to wait. I am doing my homework as best I can, however -- the archives here are very useful, as is the handbook, of course. All I'm currently trying to do is piece together a rig without dropping a large amount of money on something that won't work well and/or that will end up swapped out sooner than I'd like. Thanks again!

Yes it does, hugely in fact! Are there any arm/sled gotchas to know about? Weight capacity is kind of obvious and easy enough to check, but things like 'don't ever use a G50 arm with a blahdeblah sled because it makes your legs turn backwards and your nose go green' would be useful to know. :) As I said, I don't care much about needing a custom made post because I can make one myself in an hour. Similarly, electrical/electronic issues don't worry me either, so picking up a half-dead sled and rewiring it from scratch would be something that wouldn't concern me over much. Thanks again, Sarah

Thanks -- that helps. So as it seems, the interface between various arms and sleds is effectively a pin/dowel, so machining each end to different appropriate diameters is all that is necessary. Am I right in assuming this? I have a small lathe, so I could make something like that trivially. Next question would be the connector between the vest and the arm -- how standardized is this? I could if necessary fabricate a mount, but it would be nicer not to have to! I'm something of an odd shape, so it wouldn't surprise me if I needed to modify a vest to fit me properly, so I'm expecting to have to make at least a few modifications along the way.

Something that has been puzzling me a little, thinking as I am about what I'm going to do for a first rig, is interoperability. Looking at a lot of the posts on here, it seems that a lot of the 'full size' (for want of a better description) rigs are interoperable in terms of arms, vests and sleds. What are the limits of this? That is, what works with what? I was just wondering whether it might be feasible to up my budget a bit and look seriously at piecing together a full size rig from older parts, and since it seems that on the marketplace people generally will sell just an arm, just a vest, etc., this might be an option, but knowing what works with what would help, obviously. It also occurs to me that buying into a system that could be upgraded might make some sense longer term. This seems to me to be not crazy, but I know enough to know that I don't know anything, so I'm asking for help! (I do plan on doing a workshop, but it'll most likely be a few months before one happens somewhere I can get to. I'd love to do the 6 day workshop in October, but the dates clash with something else, unfortunately)

The standard mechanical engineer's trick is to heat the side that's the nut -- this will free it up very quickly. Boiling water generally works (be careful!). In a really bad case you can use a blowtorch, but that can discolour the metal.

I picked up a mint AF100 at just over $3k recently. This is an overlooked camera, I really don't understand why. Maybe because it was first, maybe because Panny didn't do as good a job at PR as Canon or Sony. Image-wise, it's extremely similar to the FS100, but its ergonomics are far superior -- built-in NDs so shooting in bright daylight is no problem, usable audio, variable frame rate (with 1080p up to 60fps), intervalometer for time lapse, long exposure up to 0.5s, HDSDI, really solid top handle with 1/4 and 3/8 threads, autofocus if you're using m43 or 43 lenses, etc. My thinking was that it's good enough (easily) for the stuff I do, and if I need something significantly better for a specific job, I can just rent it. I've always preferred to own my own cameras because whilst I can pick up just about anything and manage to make sense out of it, there is nothing quite like really getting to know a camera when you want the most out of it. With the AF100, the scene file system is extremely tweakable and can get a lot more out of the image, but that's something I'd never attempt just picking a camera up for the first time. The way I see it, $3k doesn't buy you much rental time, so this was a nobrainer for me.

Interesting comments, much appreciated. In response, am I interested on a hobbyist basis in using a small rig on a few student shorts or friends' weddings? No, that's not for me. Apart from the fact that I've always hated working on weddings either as a sound engineer or stills photographer, and gained a policy never to agree to it with or without being paid for it, no, I'm more serious about this than that. I'm also not interested in specializing in camera operating, either flown on Steadicam or otherwise. I'm interested in the art, the final image, and can't really see myself only being involved in part of that chain. I'd likely never earn an SOC because I could never imagine only operating the camera. If I'm involved in a project that has the budget to use an Alexa, F65, etc., I would most likely be hiring someone with an SOC after their name, and would be able to explain what I want because I understand what they are doing. This is also why I've learned to do my own matte paintings, match moving, CGI, grading, etc. -- not because I necessarily think I'd be the best matte painter/match mover/modeler/rigger/animator/compositor etc., but because by actually understanding the complete chain I get the chance to genuinely spot when someone is better than me, so I can hire them. And, if they aren't better than me, or if I don't have the budget, I can do it myself. Being realistic, I doubt anyone would hire me to work on their movie. I'm too weird. I'm not Hollywood. I don't fit the profile of any particular job. I'm too old (44) to work my way up, regardless of how good I might be, or not. That's not going to stop me, however. As for the follow focus, or any of my other equipment, I can see where you're coming from, but like I said, I'm not likely to be hiring myself out as a Steadicam operator, so if I am operating, I'll be doing so on my own productions. This is fine with me. I have more ideas with respect to the follow focus than I've talked about here -- I'm actually waiting for a provisional patent to be filed before I want to talk about it online. What I'm building is actually not comparable with a Bartech because it's not actually a follow focus in the usual sense. Yes, the first prototype will look pretty bizarre, but prototypes generally do because the aim is to make them work. Making them look cool or making them manufacturable comes later. The first prototype won't be water resistant, because that adds complexity that is irrelevant to getting the results.

Yes, all good points, I think. The Provid I'm most likely buying is a heavier weight class, more like the current Zephyr, which was a conscious choice. I think the AF100 is a bit heavier than the FS100 by default (probably the same if you take the handles off), but I didn't want to be jamming up against the top end of the weight limit on the rig. I'd like to be able to experiment with antlers, etc., even gyros. As for focus, all reasonable points. If push comes to shove, I could probably get away with using my old HMC150 on some shots. I'm actually not a huge fan of shallow DOF anyway -- my stills photographic influences were not at all in that direction (I'm a huge f/64 group fan). If I can get enough light, I'd far rather shoot at f/8 or f/11 than at f/1.4, but this really is quite a different aesthetic to what seems to be popular these days. It makes composition way more critical, for sure -- you can't rely on just throwing the background out of focus, so it has to formally be part of the composition. Overdone, I think it can pull you out of the experience (some of Barry Lyndon does that to me), but got right it can be awesome. To use a couple more Kubrick examples, The Shining (of course), and Clockwork Orange really nail it, for me anyway. I'll most likely heavily modify the Provid, partly because I just can (and tend to have fun doing that kind of thing), and partly to get it doing what I want.

One other thing as regards 1 or 2 axis, it makes no difference because the IMU basically needs full 3 axis data in order to work properly, so the tilt axis data will be there anyway whether you display it or not.

Few things. 1) why? Tilt is free 4) no can do with a outboard hd-sdi level it's a minimum of three frames delay in order to put a overlay on the screen 8) bad idea the reason you do the two second hold is so that you can't randomly set a new level 1/ because it's not free. you have to use a second set of gyro&sensor pair which, i think, will increase the price more than it increases the practical usage. if not i'm all for more data. let's also put a altimeter so we could know about our boom height! 4/ this was my desire as a potential custommer, i'm not qualified to say it is possible or not. 8/ it all depends on where you put the button and the shape of it. on Ultra2 the placement is very handy but at the same time you never push it by accident. And even with a one touch calibration, i'm stressed thinking about changed level when i calibrate it. a 4 sec. calibration will upset the balance. How about this: Hit the button for less than 4 seconds, and you calibrate to the average over the time you have the button pressed. Hit it for longer than 4 seconds and you go into a calibrate routine where it waits for the sled to completely settle, then takes a very accurate reading. Sounds like best of both worlds - super accurate neutral balance, whilst retaining the ability to quickly lock to a nonstandard angle.

Few things. 1) why? Tilt is free 4) no can do with a outboard hd-sdi level it's a minimum of three frames delay in order to put a overlay on the screen 8) bad idea the reason you do the two second hold is so that you can't randomly set a new level Actually it should be possible to do HDSDI overlay with a delay in the microseconds, but it requires a bit of careful design. The 3 frame delay comes from triple-buffering, as traditionally done by old-style timebase correctors. It's a powerful technique because the output frame rate phase need not have anything to do with the input phase, making it possible to do overlay without genlock. (I once helped to design a genlock circuit, must have been in 1986 -- this brings back memories). Anyway, there is a second approach that could work here. Basically, the overlay box sits as a 'man in the middle' between the camera and the monitor, essentially passing through the HDSDI signal unaltered, with a few pixels delay (inevitable due to the way that the signal has to be received and sent). An overlay circuit could piggy back on this signal, replacing pixel information on the fly as necessary. If the overlay is in the form of another unsynchronized video signal then it would need a 2 to 3 frame delay, but (critically) if the signal was also generated on the fly, in sync with the outgoing pixel data, then there need not be any extra delay at all. For practical purposes, this would give you overlay with zero perceptible delay. As regards an inertia compensated level, there would be an inevitable need for a small amount of delay there due to the need to filter the outputs of the IMU, though the data rate is probably going to be high enough that this will not be noticeable. So, doing overlay is possible, but it will need an FPGA and quite a bit more development time. Not quite a weekend project. What would someone expect to pay for a bit of hardware like that?