Interesting new rig

[David Hughes]

I have seen the AR on numerous occasions and flown it myself a few times. It is a truly remarkable thing of great engineering beauty, a lot of thought went into Howards' design and we can all thank him for that. Where does the C of G sit within the rings of an AR? From what I remember it is difficult to adjust this because of the rings. It should be easier to adjust this axis to almost perfect balance and make less work for the rotation motor.

From what I remember Lyns' rig also had a motor at the rear for rotation around the lens axis and was held in place by some carbon fiber rods similar to this.

I just think unfortunately for some the writing is on the wall. Open source will give us all the option to fly our sleds this way in very short time, sure It won't be applicable to all cameras and the mechanisms wont be as elegant as and AR.

 

just because it's open sourced doesn't necessarily mean it has to be half arsed, just look at the Multi copter progress in the past few years.

Dave

[Thomas English]

Are we going to get the grips to run around with this thing whilst we operate? Or are we going to run around whilst the DoP operates?

[Mariano Costa]

Didn't Garrett Brown put a Stabilized head on his Steadicam Rig for some chase sequences in Star Wars?

 

http://www.theasc.com/magazine/starwars/gallery/jedi/stcm/img3.jpg

Edited April 6, 2013 by Mariano Costa

[Eric Fletcher S.O.C.]

Didn't Garrett Brown put a Stabilized head on his Steadicam Rig for some chase sequences in Star Wars?

 

No he had a clockwork to counterbalance the film shift in the vistavision mags

[chris gonzalez]

just because it's open sourced doesn't necessarily mean it has to be half arsed, just look at the Multi copter progress in the past few years.

Dave

 

agreed. Actually it was because of open source that all of this innovation came about. This new gimbal technology is a direct descendant (literally the same hardware) of the multi copter, which started out as open source as well. I'll tell the story if you want to hear it, but if it were not for the RC community, we may still be in the stabilization stone age.

 

i myself have been chasing RC aerials for years, starting out with servos. mechanical gyros, electronic gyros, helicopters with 6ft rotor spans, then muliticopters. i personally never thought we'd achieve this level of stabilization (ie, Steadicam stable). even now, i catch myself gawking at the stuff that is coming out of my little nex camera.

 

it will not replace the Steadicam (Garret Brown got it right, it's like trying to obsolete the hammer). It will not stabilize an 80lb 3D camera rig, but it definitely has its place and applications.

 

I don't know what to build next, a mini Revolution with the nex is theoretically possible, or a bigger 5D or Sony EX3 class rig. just wish i had access to a proper machine shop, but for now the old drill press and calipers will have to do.

Edited April 7, 2013 by chris gonzalez

[Charles Papert]

 

Didn't Garrett Brown put a Stabilized head on his Steadicam Rig for some chase sequences in Star Wars?

 

No he had a clockwork to counterbalance the film shift in the vistavision mags

My recollection from the AC article was that it was a servo cage that allowed for remote operation of the roll axis to bank the camera when it took a turn.

[Eric Fletcher S.O.C.]

 

 

Didn't Garrett Brown put a Stabilized head on his Steadicam Rig for some chase sequences in Star Wars?

 

No he had a clockwork to counterbalance the film shift in the vistavision mags

My recollection from the AC article was that it was a servo cage that allowed for remote operation of the roll axis to bank the camera when it took a turn.

Both Charles. He had to have the mag shift clockwork and the bank cage

[Michael Tsimperopoulos SOC]

A few grabs from the BTS DVD.

 

[AndreasKielb]

I'm working on a jib solution for my steadicam with a brushless remote head. The tubes will be 2 meter in length so it'll be possible to go from ground level to about 3 meters in height. Pretty much a cross between the Portocam and Tango.

 

Don't have much to show for now unless a video off the motor for a bigger head I'm working on and a few mirror tests of my first DSLR Brushless Gimbal. A lot of these mirror videos can be watched on youtube if you search for "brushless gimbal" so I don't want to bother you with just another one. Unfortunately we didn't have time to dry some mikrokopter flying with the smaller brushless gimbal before my partner left for NAB (I'm only the camera operator when we go out to fly).

 

But this is the video of my first rewound motor, the gyro sensor is just taped to the outer ring keeping it level:

 

There will be a 3 axis version of the controller I use soon and then a handheld rig like the MöVI will be possible as well. The jib will be only 2 axis stabilized for the moment with a normal gearmotor at the third axis but I'm sure inertia on the long system will help against undesired pan moves. Can't tell much about the possible weight but I guess it might allow for an Epic with larger lens and remote focus as well.

 

Best regards,

Andreas Kielb

Edited April 7, 2013 by AndreasKielb

[AndreasKielb]

There is a video on youtube which describes the balancing process of brushless gimbals and it also shows some of the physics behind it an why it actually works. Most of the stabilization comes through gravity and inertia with some help of IMU and motors.

 

 

You might skip the fist part and start watching towards the end. The device is silent with PWM frequencies over 20 kHz. The open source version has 32 kHz.

Edited April 7, 2013 by AndreasKielb

[Charles Papert]

I think this technology (or a variation on it) can lend itself to solving a lot more than "how can we get Steadicam-like shots without the learning curve" and "how can we pass a camera through a hoop" (the latter literally and figuratively). Having an inexpensive three axis stabilized head that lives on the truck is a pretty powerful concept. I'm thinking of not having to lay dance floor for a dolly move, or vehicle mounts (anyone who has shot a long lens single from camera car to picture car on process trailer knows how dodgy that can be, vibration wise) to avoiding judder on a Technocrane--basically, a Scorpio/Libra/Stab-C etc. without the budgetary hit.

 

I was talking to one company a few years ago that had shown a concept of an encoded motorized head that was roughly the shape and footprint of a fluid head. I talked to them at length about incorporating a set of wheels so that you could operate it locally to recreate the geared head experience, except with 90 degree tilt in both directions like a 2575, and also remove the gears and operate remotely within seconds. That particular company turned out to be most skilled at hype since that and several other of their products remain vaporware, but moving forward from that concept, now add the notion of three axis stabilization. I'm not suggesting that the Movi is the be-all and end-all but think ahead a generation or two. Even the time-hit of having to rebalance when you change a lens or add an accessory could be solved by a self-leveling system.

 

And how about this: since the handheld look seems to be a long running trend, and who hasn't had to recreate that look on a dolly by "worrying" the frame (and/or putting the camera on a bag or a squishy head or whatever else); having a variety of handheld simulations built into the software, so that you can dial one in to your liking and let the head itself make it look all shaky/indie/immediate/real/organic blah blah blah while you sit comfortably operating the shot on the dolly, or remotely from a nearby plush chair. The irony would be that it is doing exactly the opposite of what it was designed to do, but who cares? It's the future, it's our robot and we are its master (at least until Skynet comes on line, and the head will turn around and destroy everyone on set).

 

Just spitballin'.

[AndreasKielb]

This technology is indeed capable of beeing a small and affortable Gyro head solution with similar results as those Scorpio/Libra/Stab-C. Actually it's pretty much like a cineflex in performance. The IMU mostly used has a resolution of 250 deg/sec. At 12 bit processing this is 250 / 4096 = 0,06 deg/sec and pretty close to the cineflex. The motors greate a spinning magnetic field which is also very similar to the torque motors of the cineflex. The only problem is that the motors have very limited holding torque so it will be very difficult to create bigger heads with them holding a Alexa with Optimo Zoom and everything. A striped down Alexa or Red One might be possible, though. And of course Epic, Alexa M, Blackmagic etc.

Car to car shots and shotmakers are of course possible with it as well (this was shot with a 2 axis system so jerks in the pan axis are not corrected in this video):

 

I think this technology (or a variation on it) can lend itself to solving a lot more than "how can we get Steadicam-like shots without the learning curve" and "how can we pass a camera through a hoop" (the latter literally and figuratively). Having an inexpensive three axis stabilized head that lives on the truck is a pretty powerful concept. I'm thinking of not having to lay dance floor for a dolly move, or vehicle mounts (anyone who has shot a long lens single from camera car to picture car on process trailer knows how dodgy that can be, vibration wise) to avoiding judder on a Technocrane--basically, a Scorpio/Libra/Stab-C etc. without the budgetary hit.

Edited April 7, 2013 by AndreasKielb

[AndreasKielb]

I can't edit my last post but I was wrong about the resolution thing. At the lowest rate of 250 deg/sec the sensitivity of the gyro output gives a resolution of 0.005 degrees or something at 12 bit data calculation. The Cineflex works with 0.003 degrees resolution if I remember correctly. Denny Rowland somewhere made the full derivation but I can't find the link. Of course I don't know anything about which IMU the MöVI uses and how they do their calculations. I can only speak for the open source solutions I use.

Edited April 8, 2013 by AndreasKielb

[Alan Rencher]

I just tried it in majestic mode. It has a delay in the pan, and it didn't allow me to tilt. My general impression was that there is no way to be precise in your framing... It's like using autofocus.

[Michael Hauer]

I just tried it in majestic mode. It has a delay in the pan, and it didn't allow me to tilt. My general impression was that there is no way to be precise in your framing... It's like using autofocus.

What is 'majestic mode'? Sounds like a 1970s washing machine setting. Either way if thats true its the best news i've heard since friday