sprockets The Snowman is coming! Realistic head model by Dan Skelton Vintage character and mo-cap animation by Joe Williamsen Character animation exercise by Steve Shelton an Animated Puppet Parody by Mark R. Largent Sprite Explosion Effect with PRJ included from johnL3D New Radiosity render of 2004 animation with PRJ. Will Sutton's TAR knocks some heads!
sprockets
Recent Posts | Unread Content
Jump to content
Hash, Inc. - Animation:Master

2001 - A Space Odyssey - Modelling the Discovery


Tralfaz

Recommended Posts

It's been a while since I've done any real modelling in A:M. Stian's Cathedral and Eric's Nautilus and Millenium Falcon have inspired me to give it a go again (that and the carpal tunnel is doing pretty good right now).

 

I had modelled the Discovery from 2001 - A Space Odyssey in a poly modelling program a while ago, and thought it would be great to get back into modelling in A:M with. So, here goes nothing... :D

 

2001 is one of my all time favorite Sci-fi films. It even still looks great watching it today.

 

I first created an 'inner hull' sphere. It is relatively low in patches, as most of it won't even been seen. Next, I created a very high patch (over 40,000 patches) 'outer hull' sphere, slightly larger than the 'inner hull'. The high patch count would then allow me to essentially cut the raised panels as needed from the 'outer hull'. I cut the 'outer hull' in half since I will be copying and flipping one half to use for the other half. Then I started to reduce the patch count on each panel. The wireframe shows some of the panels that have been reduced and some that are still high in patches. After all the panels have been created, I will then work on the pod bay door openings, then add depth to the panels.

 

Have to thank Eric for his explanation on how he created the panels for the Millenium Falcon.

 

C&C always welcome.

 

Al

 

AO Render

 

test0.jpg

 

Wire Frame Render

 

test_wire0.jpg

 

Discovery (created in poly program)

 

disc_0003.jpg

Link to comment
Share on other sites

  • Replies 76
  • Created
  • Last Reply

Top Posters In This Topic

Thanks everyone for your kind words.

 

Definitely looking forward to your tutorial Eric.

 

In the meantime, I started working on the neck, just behind the command module. Things are pretty straight forward here when working with the panels. Here is what I have so far.

 

Al

 

test0.jpg

 

[edit]

 

Actually, when I was working on the neck, I found a way of reducing the patch count, which is making me rethink how I have done the panels so far. I was concerned that as I reduced the patches, I wasn't keeping the proper curved profile of the panel. I found that by selecting a line of CPs along the curve, then setting the translate pivot to the center of the sphere or the lathed object, I could then rotate the CPs, which in effect followed the curve as before. This way, I did not need to get into adjusting bias on the CPs, and kept things pretty smooth. I found that I would need a minimum of 3 CPs along a spline for a smooth curve when decreasing the patches. I also won't need to work with as high a patch count when starting the panel. Not sure if this is how you did yours Eric. My way is kind of tedious, and as mentioned before, am looking forward to your tutorial. Hopefully, your way is easier. Actually, I will have to go back and re-read yours to make sure I am not stealing your idea.

 

Anyway, this is how the neck will attach to the command module.

 

test1.jpg

 

[/edit]

Link to comment
Share on other sites

Thanks Stian and Steve.

 

Been doing some more work on the neck as well as reworking the panels on the command module. Modelling the opening for the pod bay was a bit of a challenge, but I think it doesn't look too bad so far. Once I get one pod bay done, I will copy and rotate it twice for the other two doors.

 

Anyway, time to go and get some shut eye...

 

Al

 

test0.jpg

 

test1.jpg

Link to comment
Share on other sites

Thank you Jaff and Eric.

 

Every now and then, I see an image of the Pod come up in the images at the top of the forum. It would be so cool to get my hands on that model as I do have plans for a little short animation. If not, I can always try and model one. I believe Vern did Dave and Hal at one point as well.

 

Been doing some more work on blocking out the panels on the command module. Also started doing some rough work on the pod bay doors. So far, the command module consists of 5,275 patches (a lot of them will still need to be thinned out), and 6,400 patches on the neck. Right now, I am going to model everything in small sub-sections and put them together later.

 

Al

 

test0.jpg

Link to comment
Share on other sites

Gotta' love having a laptop! Brought my laptop to work today and did some more work on the Discovery. Found I was starting to cut corners on the Command Module (which I don't want to do), so moved on to the containers along the spine of the ship. Will go back to the Command Module a bit later.

 

Al

 

test0.jpg

Link to comment
Share on other sites

Tralfaz,

As always, excellent beyond excellent. I'm just getting the hang of modeling, and this will probably sound like a stupid newbie question, (because it is) but; Is the entire mesh connected or are just sections modeled and placed?

Link to comment
Share on other sites

Thanks very much Mark.

 

And thanks Paul for putting me in with such an elite group as Stian and Eric. I don't consider myself there yet, but keep trying. They are both great modellers and keep providing the inspiration to push my skills farther and farther.

 

Thanks MJL. Right now, the Command Module is one model, the 'Neck' is another model, and the start of the 'Spine' is a third model. They are placed together in a Choreography for rendering purposes. As I work on each model, I make extensive use of grouping control points so that I can easily hide or lock them later as needed. It is also a nice way creating sub-sections within the model. For instance, in the Command Module, I have a group called Pod Bay Door 1 Upper, Pod Bay Door 1 Lower (and similar groups for the other two pod bay doors), a group for Inner Hull, Outer Hull, the different raised panels, etc. Say I decide to animate a pod bay door later on. Then all I need to do is create a bone for the upper door, select the upper door group and select Hide or Lock CPs. Then I would select the bone and then all the CPs that will be associated with the bone. I would then do that for each bone I need. Hope that helps.

 

Al

Link to comment
Share on other sites

Al:

 

You've really captured a sense of scale with that last shot you posted. That's hard to do considering the model doesn't even have any decals or textures. Consider it a tribute to your modeling skills :) I love it!

 

Eric

 

Thanks very much Eric.

 

Quick update. Did some more work on the cargo containers along the spine. Right now, I have 3 spine segments showing. Each segment is around 9,200 patches. It was almost double that, but removed some of the detail that wasn't really visible in order to keep the patch count down. It is great being able to assemble the different parts in the choreography. Makes things much, much easier to work on.

 

Al

 

Discovery1.jpg

 

Discovery2.jpg

Link to comment
Share on other sites

Thank you Eric for your excellent tutorial on using the Sweeper function in A:M. I re-did the ribbed openings around the pod bay doors and it looks so better now. I still need to tweek the door openings to round them out a bit better.

 

Discovery1.jpg

 

Also worked on the spine a bit more...

 

Discovery0.jpg

 

Thanks again...

Al

Link to comment
Share on other sites

That's what we're here for, to help each other out. Glad you could make sense of my tut. I saw some pics of the actual Discovery studio model, and your model looks 10x better than what was originally used - no kidding. Round out the bay doors better? How much rounder can they get, they look spot on - well, you are now officially AMOCD :D;)

 

Eric

Link to comment
Share on other sites

Thanks guys.

 

As for the pod bay openings, they are slightly out of round in a couple of places when you look at them really close. I think it is just a matter of adjusting the bias handles to smooth them out. AMOCD, hope my wife doesn't figure this out......

 

Been working my way further down the spine, and am now at the antenna array. I really love the AO renders that AM does.

 

Al

 

Discovery2.jpg

Link to comment
Share on other sites

That looks really good. Why not use displacement for the grooves? I had some models where displacement really paid off in render time when compared to high patch geometry.

 

Disp can often create acceptable details, but Tralfaz is going for hyper detail ... and any hyper detail affectionado naturally eschews "acceptable". Geometry always trumps displacement when you want clean lines and super detail, especially up close.

 

Plus, if you ever want to rapid prototype a model, displacement won't show up unless you jump through a bunch of hoops to turn it into geometry ... and it won't look as good.

Link to comment
Share on other sites

  • Hash Fellow

That's looking real sharp.

 

 

Why not use displacement for the grooves? I had some models where displacement really paid off in render time when compared to high patch geometry.

 

But did you really try the high patch geometry though?

 

I recall Greg Rostami showing a demonstration that high patch counts themselves weren't really a big hit. I think the real hit for High patch models may be that they tend to have many more edges and parts that need anti-aliasing.

 

The problem with displacement is that it shades unpredictably. The shading is really a trick rather than based on the actual shape produced; it takes the shading that an equivalent bump map would have and projects that onto the displacement shape.

 

I'm sure they have a use though.

Link to comment
Share on other sites

Thanks everyone for your comments.

 

As for modelling versus displacement maps, I haven't had a lot of success using them. I know it can be done, but it just wasn't working out. Also, I enjoy doing all the modelling, kind of a therapy for me. :D

 

Continued working on the antenna array. About all I have left to do with it is to model the mounting structure for the large dish. I've been grabbing some stills from a high-def video of the movie as a reference for trying to pick out all the little details. It won't be 100% accurate, but does give a pretty good representation of the Discovery.

 

Al

 

Discovery0.jpg

Link to comment
Share on other sites

Wow! Thanks for the photos Eric. They are amazingly clear and the detail just pops out. Much better than the images I have. The picture of the command module is the same as one that I have in the book '2001 - Filming the Future'. There is also a heads on view, but they aren't near the quality of yours. Where did you get the pictures?

 

Thanks...

Al

Link to comment
Share on other sites

I cleared some cobwebs, grabbed the trusty torch on the wall and went deep into the darkest recesses of my vault of resources called the internet. Once I was at the bottom of that dark and gloomy pit, I found these beauties in my quest to find a likeness of the EVA. It would take some work to get the exact website since I looked at so many of them and just grabbed what I thought might be interesting. If I stumble on it again, you will be the first to know ;) Oh, you are most certainly welcome, I'll see if I can dig up any more resources for your fine creation.

Link to comment
Share on other sites

LOL... that's pretty funny Eric. I know what you mean about searching through so many web sites and then not remembering where it was that I saw something! I do a lot of searching on my lunches at work, and have sent a ton of e-mails home with links to web sites. Unfortunately, most of my e-mails are called 'Links'. I need to be more specific in my subject line as now I have to search through my e-mails looking at the different links...

 

And thanks for the comments Paul and Steve.

 

Finished the antenna array for now. Will add more detail through textures later on. Here are three views in both AO and default lighting (with a star background).

 

Al

 

Discovery0.jpg Discovery_rt0.jpg

 

Discovery1.jpg Discovery_rt1.jpg

 

Discovery2.jpg Discovery_rt2.jpg

Link to comment
Share on other sites

I simply must know the total patch count. :-)

 

Also, from one high patch modeler to another, I'm exceedingly jealous.

Keep up the good work!

 

P.S. how do you maintain such well behaved curved mechanical surfaces?

Whenever I try to introduce lots of modeled details, the spherical mech surfaces invariably go wonky and no amount of bias adjustments can seem to fix them.

 

P.P.S I'm looking at your wireframe for the sphere module and some of those little panel details looks like they're a bit more than five point patches... however the max possible is five point.

I imagine I'm missing something in the translation.

Link to comment
Share on other sites

Thanks you everyone for your great comments.

 

Here is what I have so far for patch counts. I am modelling the Discovery as smaller models and combining it in a choreography Total patches so far is around 118,000 patches.

 

Command Module - 7,441 patches (Command module is not yet completed. It still has a lot of high patch sections to it that needs to be thinned down. At the same time, it still needs a lot of detail added to it (e.g. hatch for air lock).

 

Disc_wires_0.jpg Disc_AO_0.jpg

 

Cockpit - 1,062 patches (Also not done yet. Just want to have something visible inside the command module)

 

Disc_wires_1.jpg Disc_AO_1.jpg

 

Neck - 6,695 patches

 

Disc_wires_2.jpg Disc_AO_2.jpg

 

Cargo Containers - 9,245 patches each (There are 10 cargo containers along the spine. Each section is approximately the same patch count, so there are 92,450 patches in the cargo containers.)

 

Disc_wires_3.jpg Disc_AO_3.jpg

 

Antenna Module - 12,079 patches

 

Disc_wires_4.jpg Disc_AO_4.jpg

 

Antenna Array - 5,804 patches

 

Disc_wires_5.jpg Disc_AO_5.jpg

 

Still have to model the engines section. Lots of detail here with tons of pipes, panels, etc.

 

I try to keep to 4 point patches and use 3 and 5 point patches when only absolutely necessary.

 

Well, it is the Canadian Thanksgiving today, and I hear a turkey call me. Off to dinner I go. :D

 

Al

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...