Modelling and Rendering the TMP Enterprise

[Science Officer]

First let me get the torpedo bay stuff out of the way. Here are the current wires...

I'm happy with the front and side views of the bay. But the top view is wrong because it tapers too much at the back leaving little room for the light sources. The curve used for the top view is a 101m arc and it is widest where the green and red sections of the secondary hull meet. A little tweaking should solve this.

 

[Science Officer]

Hi AnyStar,

I couldn't figure out the hangar bay for ages. I was convinced that the shape of my hull was wrong. Images showing the top of the door always suggested it was circular and therefore by getting the radius right, it would automatically meet up with the rest of the hull.

Here's a quick dump of images...

http://img84.imageshack.us/img84/2613/stmpent07.jpg
http://img838.imageshack.us/img838/2623/stmpent22.jpg
http://img23.imageshack.us/img23/2796/constrefitandrewprobert.jpg

These first alerted me to what was going on at the back...

The back tapers - staring me in the face and I couldn't see it. To me it looks like the taper runs roughly vertically, dying out around halfway down the upper panel.

In my wires, I figure that there is a 90 degree segment that is circular. The side profile is a bit out, but shows that the bottom of the the segment needs to meet the hull about two polys below where the phasers are marked in purple.

So the bottom edge is roughly circular (or parbolic), straightens out and then bends in on a curve to meet the hull. Then a horizontal section is taken out of all of this to form the proper edge. So if the bottom edge is modelled as a parabola, and you take a horizontal section out of it, the proper edge would be more circular. Hope this makes sense!!

Now in modelling terms I haven't figured out how to do this in a tidy manner. The nice thing is that I've still got pylons and warp engines to model, before I tackle this.

Cheers,
S.O.

 

[AnyStar]

i tried out a new approach today, i sliced the hull off at the angle of the wall with the greeble, same angle as the final up-down gridline, and using that line just started extruding the endcap rotating from the flight deck height. i didnt think to take any caps in between stages, but i can probably do a quick hatchet job tomorrow and re-create it to post some wires. i'm actually kindof happy with the results, and overall i only spent about an hour on it. the way the fantail coninues out now is a much smoother line than the way's ive earlier tried to marry up a cylinder or something.

I'll try to get them posted tomorrow. all i have at the moment is a quick spot render to check smoothness, although it doesnt have that upper cap shown:

 

[Starship]

Are you looking for help about how to build that specific piece? If so, I believe that Madkoi (Dan Uyeno) has a few tuts with good approaches.
http://www.madkoifish .rr.nu/madfishway/
http://madkoifish.wordpress.com

 

[Science Officer]

Are you looking for help about how to build that specific piece? If so, I believe that Madkoi (Dan Uyeno) has a few tuts with good approaches.
http://www.madkoifish .rr.nu/madfishway/
http://madkoifish.wordpress.com

Thanks for the links Starship, I'll take a look at these to see if there is any useful information. I think AnyStar's post made me think about the shape a little more. Sometimes you can't put it together until you bung it in a post. I'm starting to think the bit above the hangar doors is a set of parabolas. When I'm ready I'll investigate it further.

For now the torpedo bay is my big concern. The width is bugging me a little (did I say I had a good feeling about this in my previous posts ). I know how wide it is at one point and my resulting curves look good. They suggest the bay is no wider than 10.28m. But it means that my template neck is 1m too wide at the light housing towards the rear. Some reference images do suggest the neck tapers slightly at the base and that might explain some of the descrepancy.

Interpreting reference images is so much fun!

Cheers,
S.O.

 

[Santaman]

I guess there's no way to get in touch with the people who build the actual physical model? They should have all the data you need...

 

[Science Officer]

I guess there's no way to get in touch with the people who build the actual physical model? They should have all the data you need...

Hi Santaman,

I wouldn't know where to start on that idea. But in some way, I'd rather not have the data. I like to think that when this is done, I figured most of it out for myself. That's the challenge I guess.

Anyway, I've figured out the torpedo bay problem. I was relying on one camera-mapped view. So I went to another, figured out some angles and corrected the bay width to 11.794 m. The template neck width, is nearly spot on. The error was due to cmv not being as accurate as I'd like - it's good for surfaces face onto the camera but errors grow as the surfaces become edge on.

So I'm back on track. I've now got two curves to combine to form the top and side surfaces of the bay.

Cheers,
S.O.

 

[AnyStar]

I guess there's no way to get in touch with the people who build the actual physical model? They should have all the data you need...

i cant find the photos now, but i was horrified about a year ago to see it being used as a centerpiece display at a wedding or something, i think it was even on the buffet table, perfectly in reach of everyones food'ed up fingers... the most elegant studio model in history to end up as a waiter in his retirement, i think i'll go cry now.

 

[Santaman]

Yeah.. saw that.. *sigh* someone needs to get her back to Olson, then offer him a gigantic sum to get her back to the TMP paintjob...

I guess there's no way to get in touch with the people who build the actual physical model? They should have all the data you need...

Hi Santaman,

I wouldn't know where to start on that idea. But in some way, I'd rather not have the data. I like to think that when this is done, I figured most of it out for myself. That's the challenge I guess.

Anyway, I've figured out the torpedo bay problem. I was relying on one camera-mapped view. So I went to another, figured out some angles and corrected the bay width to 11.794 m. The template neck width, is nearly spot on. The error was due to cmv not being as accurate as I'd like - it's good for surfaces face onto the camera but errors grow as the surfaces become edge on.

So I'm back on track. I've now got two curves to combine to form the top and side surfaces of the bay.

Cheers,
S.O.

well its your baby so.. in any case good luck.

 

[Starship]

i cant find the photos now, but i was horrified about a year ago to see it being used as a centerpiece display at a wedding or something, i think it was even on the buffet table, perfectly in reach of everyones food'ed up fingers... the most elegant studio model in history to end up as a waiter in his retirement, i think i'll go cry now.

Oh yes, it was very sad... end up as a waiter is an insult to the most elegant studio model in history, and for our memories too.

 

[AnyStar]

finally had a day to work on this again, here's how i shaped the bump


and after cutting the fillet, with an overhad shot of the model. its the best perspective match i can manage, even with the way the image distorts to the right, you can tell the model has more of a pinch - but every time i try to replicate that i get 'dimples" that just arent there on the real thing. so i think this is how im going to leave it (at least for the time being)


how do you think it compares to your cylinder approach?

 

[Science Officer]

finally had a day to work on this again, here's how i shaped the bump


and after cutting the fillet, with an overhad shot of the model. its the best perspective match i can manage, even with the way the image distorts to the right, you can tell the model has more of a pinch - but every time i try to replicate that i get 'dimples" that just arent there on the real thing. so i think this is how im going to leave it (at least for the time being)


how do you think it compares to your cylinder approach?

Well my cylinder appraoch was to get some idea of the shape and ultimately is going to fail. But it does allow you to figure out the dimensions of some of the decoration.

My later idea, which I touched upon in my previous posts, is to use a different shape such as a parabola. I've put togther a quick test below. The parabolas are angled to match the rest of the secondary hull mesh. The cylinder shows the shape formed on intersecting the parabolas, which is meant to represent that etched line around the spine. The rounded cube shows that small curve you get as you start on the diagonal down the side of the hull.

Now the result is "too pointy", but demonstrates the features on that part of the hull. So it probably boils down to the shape you use - if not a parabola, probably an ellipse.

Now looking at your attempt - I'd say it beats my cylinders because you have something concrete. But it looks like you have a join problem as the shading changes. I don;t know, I might run up a test to see what I can make of it.

But I must get back to my torpedo bay. I've got a problem what I can't pin down. Might post something on it.

Cheers,
S.O.

 

[Science Officer]

The torpedo bay is now starting to take shape. Well behind schedule on this because I'm having to change the vertical positioning away from where my (so far) trusty camera mapped background image would have it. However after a lot of head scratching and doubting this could possibly right, I checked it against some other CGI models and found they have the bay in pretty much the same position.

The neck and red section below the bay are going to be replaced. They're just there to help me out. The neck in particular is now definately the wrong shape with that etched curve way too high.

For the first time (ever) I'm going to exploit HyperNURBS to help get those beveled edges. Unlike other parts of the model, the amount of bevel changes as we move from the front (rounded) to the back (fairly sharp).

The ring is the protruding bits of my bridge airlock. Another interesting bit to tackle as I have a hi-res airlock to fit into a lo-res bay. But I have some ideas on that.

Cheers,
S.O.

 

[Saquist]

Why not just work a wire from from the orthos and then create sections to loft that fit the wire frame you've created?

 

[Science Officer]

Why not just work a wire from from the orthos and then create sections to loft that fit the wire frame you've created?

Hi Saquist,

Good to hear from you again - it has been a while. Sorry for the late reply as it's been a busy week at work. Concerning your idea - I didn't remotely consider that approach! I may have to try it out on some simple shapes first to see how C4D does it. But thanks for suggesting it.

Cheers,
S.O.

 

[Science Officer]

Hi Folks,

I've been taking yet another break from the Enterprise - thanks to a combination of work, a loft conversion and frustration and not making any progress. But I'm trying to crack on with it (again!).

A small digression first - Earth Vistas (http://www.nivnac.co.uk/). This website contains some nice weather balloon images that would prove handy for anyone looking to render a starship in Earth orbit. I've mapped the images onto a sphere and tested them against some shuttle-type shaders, with GI and blurred reflections.



Back to the Enterprise. I've started to look at the back of the Torpedo Bay. As usual I'm struggling to make realize the shape. I have a partial success in that it is smooth - however the spine would end up being distorted and way too wide. So I've got to find a way to narrow the spine, remove the distortion and bevel the edges.



If I can clear this hurdle, I can really push on and complete the torpedo bay.

Cheers,
S.O.