wow, I like what you got so far. looking forward to the final movie.
FXWars! THE OCEAN!: Colin Litster
RobertoOrtiz
#23
Nice entry.
Dont forget to post it on the voting thread when you are done.
FXWARS: THE OCEAN: Post your Entry!
-R
cogfilms
#24
Hi all,
Well I only just managed to get my entry in. The seconds were literally ticking away.
www.litster.nildram.co.uk/movies/colin-litster_FXWARS(634x270).mov
(QT-Sorenson 634x270 19.7MB)
In my effort to make the post in time I mucked up the image that I was trying to attach. Now the entry is made I can’t re-edit that post so here it is.
In actual fact that shot or title is used in the film although it’s over far too quickly. You know in the last hour of editing Premiere kept crashing and as a result I was unable to do too many previews. I even had to curtail several more titles I wanted to put in. But rules are the rules so if I do so it will be another movie which I will post here.
Cog
PS Well done to all the other entries I know how hard you will have worked.
cogfilms
#25
Since I didn’t give many details on the voting page I thought I should bring together the information here. Most of this is explained in earlier posts in this thread but time was only hinted at.
This has been an incredibly challenging exercise but very satisfying. All those that entered should be congratulated for the amount of time they all probably had to spend on it. Water is one of the most difficult things to simulate and I have learnt a great deal out of attempting it mostly to do with planning and time management.
All modelling and rendering was via the open source Blender 3D suite.
I also used PaintshopPro, Gimp, and Photoshop CS for some of the UV textures although the majority of materials, especially the water, were created with Blender procedural textures.
I used 3 PCs to spread the load of rendering. I had hoped to have a six PC linux render farm available but hardware failures and location difficulties became difficult in the time frame. I rendered to PNG sequential files that I had to convert to TGA in order to bring into Adobe Premiere v6.5 for editing. I could have used Blender for the edit but memory and speed constraints in current Blender version (solved in the latest CVS) made me nervous of using an untried new feature. Unfortunately Premiere became very flaky during the final edit which meat I couldnt see a preview whilst editing. The final render from premiere took over 30mins and the upload to the files location about 20mins literally within seconds of the deadline.
I spent many evenings and several weekends almost full-time on the project starting from 10th February. In the midst of this I had a conference to attend, that took up almost 5 days and several family events that subtracted from that available. That means that its difficult to give precise time spent but I would estimate about 164 hours.
60% of that was development of either models or materials.
40% was animating and rendering (although some of the rendering was started while the development and animating was progressing)
I unfortunately left most of the rendering far too late and as a result had to scrap some shots, that either had render errors, or were too complex to complete on time.
The glaring mistakes, and lessons, that I see in it now are:-
- Grammatical error in the opening quote:-
…THIS IS THE STORY OF THE OCEANS AND THE BRAVE BUOYS WHO DEFEND (IT)…
Should have been THEM.
Lesson learned:-
Dont change the words at the last moment when you have other more pressing problems occupying your attention.
- The first close-up of the Buoy looking at the ships has a normal problem with the neck mesh.
Lesson learned:-
Dont assume that remove doubles (vertices) will sort out a mesh in Blender. Sometimes it will produce odd faces that muck up normal mapping that not even CTRLN (force normals to outside will resolve.
Also if you test render do so at a resolution that will show such discrepancies. Leave more time to render at least 3 times longer than planned.
4. The shot of the mine surfacing has a texture map that scales too slowly just before the mine surfaces. That gives the impression that darkness descends on the mine.
Lessons learned:-
Monitor test animations on a good quality monitor. I used a LCD laptop for that shot and I just didnt see it.
I could have edited that out of the final movie but time ran out. Leave More Time to render.
5.In the same shot the ship textures, that are simple flat planes with a ship profile applied have a nasty black line at there top caused by not setting a switch in the blender texture buttons.
Lessons learned:-
LEAVE MORE TIME TO RENDER.
Planning is also vital because it should give a mechanism for estimating what can be achieved. I had 3 other shots that I would have loved to have included. One of these I called my complex shot.
It included a storm wave with both the mine and the buoy with; spray, mist, wake, splashes, and raytraced waves. I had 7 layers to makeup the final shot I was struggling to get the wake splashes looking right and it was just not worth fighting to fit it in when there were other effects to complete. So an overall lesson is to plan well and save wasted time.
Overall I am still quite pleased with the result. I believe it tells a story and makes a reasonable attempt at oceans using a 3D package that has none of these techniques built in. I will finish all the shots and the rest of the story for a short that I hope to enter into the Blender Conference Competition later this year. In anticipation here is a HERO render of the Buoy with toon shading that my final short will have.
Cog
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.