Bwana Kahawa

Right, seeing as I have to get something up by the stage deadline tomorrow, these'll have to do!

They're some of my initial sketches for the various sequences in the animation (drawn on a train hence the dodgy smudges everywhere! )

If I get time I'll add in a couple of others, but these are the basic ones I want to get done. Overall, the animation will follow the light through the day, opening at midday and finishing late evening.

I'm really struggling with this, as Maxwell has very limited supported for animation - it's non-existent apart from the option to queue renders.

I'm having to create a new object for each frame that I want a moving object to be in, export the lot, turn one on, save the Maxwell file as that frame, turn the object off and the next one on, resave as the next frame, repeat... And then add them all in to the queue, one by one.

So it's taken me two days just to set up a 6 second shot after I've exported to Maxwell... Plus the fact that this is pretty much the first animation I've done. And I'm rendering it all on a Mac Mini. So I'll just be happy if I have something that moves to show at the end of it all! Don't go getting your hopes up for anything rivalling ILM...

Attached Images

	AVC-1.jpeg (227.7 KB, 61 views)
	AVC-3.jpeg (157.4 KB, 30 views)
	AVC-4.jpeg (155.7 KB, 19 views)
	AVC-5-and-7.jpeg (124.3 KB, 20 views)
	AVC-6.jpeg (162.3 KB, 29 views)

AJLynn

Wait a minute. Maxwell... Mac Mini... Animation...

I'm a bit concerned about that.

amer abidi

one rough train ride that must've been!

couldn't make much of your sketches..but i gather it's all time lapse cordination for the time being?

one question though..what is the "ken borar" (or bonan..or boran..or..or ..) effect that you are talking about for the last shot?

__________________
"Well I figure it has to be bigger than smaller. Whats the point of designing a concept the size of a hot dog stand? "
J.G.

AJLynn

Oh, I missed the note about the Ken Burns Effect. I was going to use that as well. It's taking a still image and doing a pan-zoom on it to show areas of interest and imply motion. Ken Burns does it a lot, originally by moving the actual camera but now it's done in software. It a great way to work around render time problems when what you really want to do is show one scene but work with the area of foocus, because the only render you need is one shot at a high enough res that's always supersampled when you zoom. It can be very powerful depending on how you use it.

My advise would be not to use the built-in Ken Burnsifier in any Apple products for this, because they don't give you enough control, they're a bit random. Set it all up manually.

Bwana Kahawa

lol yeah my sketches and handwriting suck! I did it all as notes to myself rather than anything presentable, but still. Going for time lapse co-ordination as the main 'feature' to end the animation on, but I've yet to see how successful this will be given the noise I'm getting out of my renders so far.

Yeah, I've used it before on the one other animation I've attempted - used Final Cut's scale and move tools rather than a Ken Burns filter as such. Managed to get something vaguely decent out of it so hopefully it'll do the goods this time!

Apologies for the rant before - I'd been watching Maxwell crash my machine for a good two hours and was ready to throw in the towel! It's behaving itself today though. I've attached a screengrab for your amusement of what it did to my mac every time I tried to import a simple rectangle object...

Attached Images

crash.jpeg (282.7 KB, 30 views)

AJLynn

I admire your intestinal fortitude.

Adrian Bold

Crikey! that description of setting up your animation reminds me of the early days when i was using a software package that had zero animation ability, though it had a real-time engine:

To appease the client I had to buy a top-notch video camera, pci board, a gpu card that supported dual monitor, multiple hard-drives (when the largest capacity was some 15gb) firewire port; (a time when firewire was unheard of) setup the video camera as a second monitor, move through my scene using the mouse extremely slowly whilst recording the dualscreen output on the camera to miniDV tape, then once recorded, import all the tape(s) footage back to the computer as a video file (gigabytes!) into an early version of premiere, speed the footage up by 1000% to get it to a reasonable walking speed and allow for some post effect/title work to add polish.

Horhe

Hey Derek... sorry to hear about the maxwell problems. The sketches dont seem to convey much to me :P
Neat idea with the zooming in effects in order to produce an animation. Glad you didnt give up after all those issues with the renderer. Good luck! Looking forward to see more.

Cheers

__________________
Im starting to suspect that the world is a rendering...

Bwana Kahawa

Right, here's something a little better than my rubbish sketches for you all!

I've successfully rendered off the first opening sequence that pans round the building (the only sequence in the movie where the animation is created in the renderer - using moving cameras, moving elements, etc.). I now have to sort out a lot of issues frame-by-frame in Photoshop - the attached S1-Ftest jpg shows a very rough idea of what I'll be doing to each one.

I've also attached a few test renders from the time lapse sequence that will end the animation. I'll add more variation to each frame in Photoshop (people, boats, breaking waves, clouds, etc.) It gets dark pretty late as it's at the height of summer. Still need to make the lighting around the tower stronger so it stands out more in the darker shots.

The final image is a rough test of a Depth of Field animation showing the link between the centre and the tower. It'll open with the interior of the centre, and then change focus so that the viewer is looking at the tower beyond. Again, done frame by frame using the Lens Blur Photoshop filter and a ZDepth Map rendered in Maxwell.

One question I've got you might all be able to help me with. I want to populate the visitor centre in most shots with various people. As it's an animation basically produced frame by frame in Photoshop, my options are pretty limited. What I'm planning on doing (which I've done reasonably successfully before - http://www.lsiarchitects.co.uk/proje...thumb=6&text=2) is to add in video footage of people walking and talking (the set sold by Marlin Studios where each frame and it's alpha channel is a separate JPG).

However, it's a pretty time consuming process, so makes it difficult to fully populate the centre to the extent I'd like. So my options are:
1) Only have a very few people in the centre (two or three per sequence), all animated.
2) Have two or three animated and the rest still images (which might not be as obvious as it sounds - people reading signs, sitting down in the cafe, etc. might not be noticeably static, especially as the camera will be moving)
3) Have all still images, like the place is frozen in time.

Or, do I change my approach sequence by sequence? Any advice gratefully received!

Attached Images

	DOF-Test.jpeg (175.2 KB, 44 views)
	Time-Lapse-Test.jpeg (905.7 KB, 55 views)
	S1-Ftest.jpeg (136.0 KB, 39 views)

chutti

Man that looks so cool (the exteriors)...i'll go with the lighting between 22:00 to 23:00. which should not be too lite or too dark. wonderful site to be...love those stairs.

good luck...keep up the good work.