[kalthans]

you can set the filter controls of your animation UI to expand/collapse animated channels automatically

k

[kalthans]

tracking markers would be a good idea no matter what

i can’t speak directly to the Milo/Flair system but i did do a bit of work on the Kuper system years ago and you basically have two options:

#1 take the data in directly and recreate the physical geometry of the moco rig in maya and plug in the channel data to each axis. this presupposes that all the axes are calibrated accuratley in degrees/inches.

#2 have the moco software convert the multi-axis camera data into straight xyz cartesian data (with camera pan/tilt/roll as well). on the kuper system this was pretty easy and i’m sure the guys at mark roberts moco have a similar gizmo available to them in their Flair software.

the milo/cyclops system is a pretty wild rig…it has lot of variable geometry axes and can really get into some weird confiurations! on the one occasion i had a chance to visit with them their lead programmer seemed very on top of it.

you might want them (the on-set moco driver) to give you this data both ways and provide you schematics of their rig dimensions and travel ranges. i’m sure they have done this a hundred times before so lean on them and let them help you.

when you place your tracking markers you might want to make sure to measure the distances between some of them. having these “known measurements” will help if you go to 3d track the scene later. i find that if you’re pretty sure you won’t be using the shot plates to 3d track later on you might want to use green “painter’s tape” to make your tracking marks on the greenscreen. that way the marks are there but they will be more likely to be keyable and you wont’ have to paint them out by hand when they cross actors, etc. (on the other hand if you think there’s a good chance you will 3d track it in boujou i’d go ahead and use their 2d3-approved “triangle in a circle” tracking marker)

depending on the nature of the move you might also want to shoot a tracking pass (no actors, etc) with a known wireframe object or objects (say a 1’x1’x1′ cube or such) in the scene that you can use to check your mayamilo registration

make sure you make careful notes of all the camera optical data: lens, film back, centering, etc.

you might want to call the company who is providing the moco system (rental co?) or the parent company (MRMC) who built the rig before going on the shoot. i’m sure they would be very happy to give you any info you need.

good luck

k

[kalthans]

i stand corrected. thanks, deke

it has been only a year since they have rebranded most of the discreet products with the autodesk label

[kalthans]

first of all, in response to the posts about how Flame/Inferno R&D has suffered at the hands of Autodesk…

i don’t think there has been enough time under the Autodesk banner to make that claim. it has not even been a year since Autodesk acquired Discreet. there just simply isn’t enough evidence to support an analysis. all of the technology demos we’ve seen so far have been fairly impressive, pointing to increasingly higher levels of optical flow integration. and judging by their marketing presence (and $$$ spent) at NAB they are definitely in it for the long haul.

second, as for all the people freaking out about the possiblity of Autodesk killing off the Linux platform just because their CAD/CAM stuff and 3DS are Wintel…..

i really don’t see that as being an issue. the only reason 3DS is still wintel-only is because that is how it came to them in the acquisition of Discreet. discreet’s other product lines (smoke/flint/burn) all have Linux-based versions that are up and running strong and will be around for a long time if you believe the technology gurus at discreet who are all but shunning the big-iron boxes of sgi. even if inferno gets an update it will most likely be on a Prism box running some variant of linux. so i really don’t think Maya guys need to worry about losing the stablity and robustness of the linux platform.

as for the coexistence of two 3D products in the same company: that is the million dollar question. if any product suffers it will most likely be Max because it represents the smaller capital investment made by Autodesk. the most likely scenario would be a marketing-driven one: return Max to it’s roots as game-developer-centric product while continuing to position Maya at the high end. i think this would be unfortunate because of how fast Max has grown in the past few years as a ready-for-prime-time player, but when big $$$ are at stake hard decisions are often made.
k

[kalthans]

you mean sorta like DeepPaint?

that would be cool!

but i think the batch paint will probably be a very stand-alone node at first. i can imagine the technical limitations of trying to use a context view as you are working in paint. not a trivial undertaking.

[kalthans]

i think you’re talking about danny yoon’s stuff:

http://oyoon.home.comcast.net/

fxguide has a bunch of his tips & tricks in their archive:

http://www.fxguide.com/fxtips-135.html
http://www.fxguide.com/fxtips-127.html
etc

k

[kalthans]

using a single frame of noise as a displacement source works well, too. scaling & softening the noise along with some smoothing of the displacement will yield various results.

k

[kalthans]

you can definitely do it in after effects

think of how you’d do a sky replacement in photoshop as a still then apply that thinking to AE where you’ll add motion

you can even animate the clouds in AE with just a still (a large one) of some clouds

k

[kalthans]

are you looking for general tips or tips pertaining to a specific application/system (flame, shake, etc)?

i assume the image is a still

you will need to do some extensive rotoscoping and do a complete sky replacement. the only thing you can salvage from the original image is the frames of the windows (the borders)…everything else must go. replace the sky with a new one (timelapse maybe?) and use that same clip as the reflection element. some distortion/displacement will be in order for the reflection to make it look like it is passing over the uneven surfaces of the glass windows.

shifting and strobing of specular detail as the (timelapse?) clouds cycle on the glass will also be an important bit of detail that will sell the shot.

does that help?

[kalthans]

jersy….

my biggest apologies….i didn’t realize you were in Combustion….i thought it was a flame/inferno question. i don’t think C4 has displacement yet.

k

[kalthans]

{post deleted}

[kalthans]

see this previous post:

http://www.fxguide.com/postt1472.html

[kalthans]

another point to consider when you do your 16mm -> tape transfer:

as with most things in life you get what you pay for. given that you are on a bugdet your two options when contacting transfer facilities will be:

#1 they give you a good deal on a decent colorist and decent equipment because you are a student and they want to help.

#2 they give you your money’s worth and give you a crappy colorist using crappy equipment that they reserve for cheap dailies runs.

obviously you’ll be better off with the #1 scenario. every xfer facility i’ve dealt with has been very cool and approachable with helping students. they won’t give you their $1500/hr suite+colorist, but they will help you as much as they can. go there with your hat in your hand…have very good estimates on the amount of footage you’ll shoot and what your target budget for telecine is and they will probably work with you as much as they can. it’ll probably end up being a lot more than what you want to spend, but you get what you pay for.

now, most colorists should automatically know that when timing bluescreen&greenscreen material they need to lay off the heavy color-correction and just balance the primaries in the scene to match the other footage. this is critical because any heavy color correction can introduce unwanted secondary and tertiary colors on the transition region between the key-source background and the foreground that can be very difficult to remove and keep good edges. on 16mm this problem will be additionally magnified by the small size of the neg compared to the final output.

good luck.

k

[kalthans]

to the original poster:

first, do you realize that TriX is a black-and-white stock? i assume you misstated your stock choice (maybe you were talking about Tgrain stocks or maybe Vision or Vision2?)

i can completely understand your budget limitations and why 16mm is a requirement. if you are locked into using 16mm then i would suggest using one of the Vision or Vision2 stocks with a speed no higher than 100ASA (7212 or 7245). This is especially crucial if you are doing any keying. the size of the 16 negative is drastically smaller than the 35 and you will be fighting grain & detail on your key the whole way.

before you make a final committment to 16mm might i suggest trying HD? the price of HD has come down dramatically and if you are xferring your final product to NTSC (betacam, digibeta) then you might end up with an image comparable to 35 but with a acquistion cost similar to 16.

as for xfer format for your effects work: again, budget needs to be considered. i would strongly advise against DVCAM or any highly compressed format. Digibeta should be your baseline choice. if you do not have access to Dbeta then straight BetacamSP is your next best bet. (Actually, Paul….betacam is a component format, just analog component at a low sample rate). you can get fairly decent color-signal separation from BetaSP sources (in fact, back when the only choices were D1, D2 and BetaSP i often xferred by greenscreen/bluescreen material to BetaSP over D2 because it was MUCH easier to key from since D2 was composite digital and D1 was expensive and not every facility had the decks).

good luck on your project

let me know if you have any other questions

k

[kalthans]

when you first start CCing your matte you’ll need to crush the blacks and the whites a bit

again, do it in layers…don’t try to make one image/key work for the entirety of the flames.

also, if it’s at night and on snow make sure you get the color balance right and you throw some CC/lighting on areas around the fire where it would be illumnating the landscape

does this help?

[Author]

Posts