My scripts are online now. They can be downloaded on www.luxmemories.lu under Software.
The documentation can be found in the script itself.
This script is event based, not file based. You select the events you want to have deshaked on the timeline. The script will not deshake the entire sourcefile but only the selected events. In order to do so, the script renders every selected event to a file, runs Virtualub with Deshaker over it, and adds the resulting file to the event as default take.
Now I understand: your script runs VirtualDub and Deshaker from within Vegas. That is too cool! I've downloaded it and will try it when I get a chance. Not long ago someone was talking about deshaking totally as a script from Vegas. I was confusing your effort with that.
Only been in copter once myself, Vegas down into the Grand Canyon, best way in the world to discover the joys of travelling in a copter.
I'll query the guy again about what he said and try to remember more of it this time, also something about tail rotors might have crept into the conversation, or was that about how to land a copter that's lost a tail rotor....
Also his copter is pretty big, big enough to lift a lot of water for water bombing.
Either way I'm not going to knock back the opportunity to shoot my first aerials just cause the free flight's only in a 2 blader :)
Now if only CASA would let us fly a glider over Sydney, they've got to be the best platform for shooting from.
He doesn't, doesn't own a Squirrel but he claims that's how they're designed and built. His problem is that because the Squirrel is the new kid on the block everyone wants to hire one even though the Bells are better helicopters. The new Squirrel with the enclosed tail rotor sure looks sexy but I watched two of them landing and it sure looks like he's right, they're hard to hold in the hover, they seem to wobble around a lot.
Well I'm thoroughly into the whole process of stabilization now, but let me take a moment of suggestions to Farss, re Tuesday's heli flight.
It's been quite a few years since I did heli work, and those years were essentially with a Tyler rig and Kenyon stabilizers - I forgot that -until I looked at my new footage (whoops) - so first off, you are right - get the stabilizers if possible or resign yourself to less than perfect footage.
Next - the adreneline is pumping when you're new to heli flight. So try to remember SLOW DOWN. I'm embarassed to say that a bunch of my shots look like amateur time at the Little League game. A good deal of that was due to my pilot who was "zipping along" at 60 to 100mph. My bad. I should have insisted on "Slow Down" - but this first flight was a test of many things - HDV camera, total handheld (with image stabilzation in mind), small chopper, and just a look at the general area to see what I 'really want' for the final movie. The pilot had no experience in video work, and was trying to get us to as many sites as possible within the allotted time period.
A good, though expensive, learning situation. I did get quite a bit of "fixable" and usable footage, and I did get to scout a bunch of sites, however. But not the quality I want for the main aerial part of the movie, for sure. More along the lines of cutaways.
My next flight will definitely include the Kenyon stabilizer, and I'm considering giving up the HDV footage and going with a standard stabilized news chopper. Definitely the "best way" except I'm still longing for HD finish footage. The only local chopper set up for HD acquisition is $2,500 an hour - too much for this film.
But cruising around in a helicopter is an experience not to be missed, if you get the chance. Just take a deep breath as you're shooting and remember to slow down the chopper and slow down your pans and zooms - it's a lot quicker once you're back in the editing suite.
I've read the js text. So it looks like an event is rendered out to Cineform (the included Sony version or a latest update?); the two-pass stabilization is completed (is this with a saved VDub "processing settings" that allows the Pass 1 and 2 settings to be preconfigured?); and the resulting file is saved back as a new Cineform file (again, does the VDub "processing settings" specific the compression settings?)
I love the new event showing up as the active take.
You have to define a template in Vegas named DESHAKE, the event is rendered out in whatever codec you want (but Virtualdub needs to be able to read it). This codec is only used for a temporary file and thus for this codec, I prefer something lossless like Lagarith, Huffyuv or uncompressed. The vds file is configured for cineform export (here I prefer cineform as this file is needed by your project, so I prefer not to waste space), but you can simply select another codec in Virtualdub, do a save processing settings and replace the second VirtualDub.video.SetCompression line from deshake.vds by the VirtualDub.video.SetCompression line from your processing settings. Deshake.vds is preconfigurged for the deshaker settings I prefer. If you want other settings, simply open a HDV avi in Virtualdub, do a save processing settings for the first and the second pass, and replace the
first VirtualDub.video.filters.instance[0].config line in deshake.vds by the corressponding line from your first pass settings and the second VirtualDub.video.filters.instance[0].config line in deshake.vds by the corresponding line from your second pass settings.
How hard would it be to swap the SD camera in the SD Chopper for a Z1?
Surely only a few minutes work!
The Z1 with a battery I'd bet will be less weight than the existing SD camera. Take the composite video feed from the Z1 out into the exisiting monitoring setup, put the cam into record just before takeoff, so long as the flight is less tha 60 minutes you should be set to go. With a bit more effort you might be able to get a firewire cable from the mount into the cabin and then into a M10U battery powered VCR, then you can change tapes inflight.
If this chopper has the 'bubble' on the nose rig it can get you shots that you cannot easily get any other way, looking forward, down low over the terrain. The opening sequence for one local production used exactly that, if it wasn't for the types of trees trees you could have mistaken it for the Everglades. I'm sure you know the kind of shot I'm talking about. Large water birds taking off in front of the camera etc. Awesome stuff.
Actually I'm wrong it could have been Florida!
Some kind person donated a few of our local trees to Florida and they've taken over the Everglades!
BTW, have you looked at www.coptervision.com?
Looking at their demo reel, their RC copters can get you shots that nothing else can. Although I'm certain some of the stuff on that reel must have given their insurers heart attacks.
Just my worthless spin on this but the copter shots are the hero shots, if they look low rent on the big screen it'd kind of blow the look of the rest of it. The local movie I was talking about at the start, the print I saw had serious color timing issues in the opening sequence, for me at least that set a bad vibe for the rest of the movie.
The first thing I asked was could we swap out the cameras.
No... at least not in this one (and most of the others I think). It's the lens and sensor that are at issue here. The bubble on front houses the lens and sensor and 5 gyros, etc. It sends the image back into the copter. The problem is, how do you get a HD image from a SD sensor. He's recording to a BetaSP deck so it's not chopped liver, but now that I'm totally in love with HD, I'm trying to find a way.
Have you got any experience handholding a Z1 type camera with a Kenyon 6 hanging below it? I sort of remember 2 stabilizers in the old Tyler rigs I was using. All I want is to end the bounce.
I also agree - better no heli, than bad heli. But I'm not even close to giving up yet.
Just a reminder that the FAA (in the US) gets into the picture the moment you *attach* anything to "the airframe" - which is to say, mount a camera to the outside of the bird. Once the paperwork has been filed and approved for a specific camera and mounting location and style, it takes another round of paperwork to modify anything, else the pilot and the bird owner face losing license to fly.
Interestingly enough, you can legally dangle by your spurs, upside down, with untold amounts of weight, shooting to your heart's content (thus comes the "lean out the door and hang on" technique). But put a 3oz bullet cam on a skid without permission and paperwork, then suddenly your pilot is in for severe penalties.
I just had lunch with one of Florida's foremost heli cameramen, and he told me of a wonderful gig he just completed. The client wanted "just over the water" shots without rotorwash to disturb said water.
They HUNG the camera on a control platform off of a 50' umbilical. They attached a vertical "rudder" to the back of the rig to keep the lens pointed forward. They then pointed the camera in the correct direction by steering the copter. He said it worked wonderfully. He did state that a gyro rig was necessary in most if not all camerawork from a copter - a lesson I "relearned" last week.
The pilots I've worked with have learned to be "DPs" - the camera is *hand-held* stationary (the operator is the "tripod" and "fluid head" and "shock mount" but not the "pan-tilt servo control.") Then the pilot aims the bird. The pilots here in S.E. MA are extraordinary with dangling platforms (their big payday job is hauling 500lb cranberry harvest crates from the bogs to the loading areas, and *stacking* them real time). Yeah, I can imagine how great the shot would be that you described.
But that dangling rig is supposed to have its own paperwork, serial number, etc., and carry only cranberries (no cameras)! Bummer.
Keep in mind that hovering uses close to twice the fuel/hour as does moving. Pilots around here hire two "hook men" and pay them good money just to minimize the time hovering (about 3 seconds seems to be the rule) while the crates are hooked to the cable and then unhooked at the drop area (even though they have in-cabin release mechanisms, the pilots *need* a hook man to be sure that sucker is unhooked). Any way, asking to keep the bird "moving around a bit" while you shoot will make the pilot happier.
The cardinal rule is: What the pilot says is law. Do whatever the pilot says. It really helps the editor if you come back alive.
I promised an update on this heli work, so here it is.
What I wanted was lingering beauty shots from the skies above.
At first I went out with my Z1, held my breath and just shot. I was strapped into a Raven R44 helicopter without doors on it. ($500 per hr) The Raven is like a golf cart with a rotor - not the greatest, but within the production budget I can afford for this film.
Using DeShaker I was able to get about 3 seconds or so of usable shots before the camera would jerk beyond repair. Workable but not glorious.
Heeding Farss' advice I rented a Kenlab KS8 stabilizer to hang on the bottom of my Z1. It's a super high spinning gyro that totally minimizes shake contrary to the direction it points. By pointing it forward, it virtually eliminated up and down "jars".
By itself the footage looks quite nice - but once I again "DeShaked" it (thanks Lawrence for your help) - the footage achieved the glory I was after.
Now there are definitely even better ways to do this. A Tyler rig will actually use two or more Kenyons plus gimballed suspension rigs to totally isolate the camera from the aircraft - something I try to do by hand. (another $500/day plus installation and removal by expert technician) You can't really zoom my handheld shots the way you would with a Tyler rig. And now, of course, the news camera rigs have a 5 gimballed and stabilized lens setups which feed an internal deck, controlled by a video game-like controllers. But that's serious money ($850 to $2,500 per hr). And these hourly costs don't include startup and travel to the location fees (same money).
Having rediscovered the Kenyon stablizers, I would say if you're doing any kind of hand held action from heli, boat, or car - well, it's a worthwhile investment to rent for the day.
Probably the gyro got rid of enough vibration so the encoder wasn't destroying the image. Then the image stabilisation software had a chance to do its thing.