In-house 3D X Bomber...

[Bradster]

Oooh! Great idea, Andy. Let's compare them all!Well, the only suggestion I can think of (aside from starting on your textures! ) would be to "fatten up" the angled blocks sticking out of the midsection -- those blocks that LegTrax and MainBody fit into. I probably will need to do the same.Remember, though, that the camera's angle and zoom can be deceiving! Don't make any major changes until you're sure it's not just the camera playing tricks with your eye. For example, you already know that the profile is spot-on; so, I wouldn't go change the angle of the neck as you said -- not yet, at least.Actually, from that angle, it seems my model's "neck" is too fat. Oh well, that can be fixed later if necessary -- I'm determined to finish the majority of the textures before doing any more structural work.Ah! Here's an idea. Out of sheer curiosity, let's compare our fronts also. (Edited by Bradster at 4:08 am on Aug. 13, 2002)

[AndyThomas]

Hmm. Haven't got it with me for comparison just now, but looking at your model and indeed the real one I think the lower portion of my neck needs to thin down width wise. It's too uniform at the moment. I think I need to watch the series from the tapes a little to get a better idea of the forward profile.Looking at yours, I'd say something similar applies - your forward hull isn't as angled as the real model's and even mine could use a little more flaring out. BTW, I agree about my Dai X pods - they're more placeholders at the moments, I only got the general shape right last night and I realise they need to be a lot bulkier.Here's the closest I can get to a straight on comparison for you for now:There's something about the way my main engines are placed that's not quite right as well I think... Although it might be that the lower wing's not quite right in the right place. Hmm. Definitely need to step back and try and get a sense of it as a whole I think, I've been paying almost too much attention to the little details.Brad, any advice on camera settings? I've just managed to add a light source which makes things look sensible but as you say if the camera's not right it does distort things. I need to sit down with the manual really...Your textures look v. good - quite... stark. That's not quite accurately describing the impression, but... y'know! I suppose my question is, do screen grabbed textures tend to look a little artificial when you see the model actually moving? I suppose I'm thinking that because they're inevitably captured under certain lighting conditions, they'll never look quite the same as "natural" textures. Does that make any sense?!

[Bradster]

Heh, arranging the cameras can be both frustrating and fun at times. First of all, be sure you're working with an actual camera object and not the default "editor camera" in the view. This way you can arrange the camera and still have a different 3d view to rotate around the model when working.The distortion of the camera can be greatly used to your advantage if you tinker around with it enough. Here are two QuickTime clips of the same pass, but with different camera settings.http://brad.project-think.com/starfleet ... yby-01.mov (240 KB)http://brad.project-think.com/starfleet ... yby-02.mov (242 KB)The latter uses a much greater aperture width (but the same focal length settings) and is a bit closer to the model to compensate. This method gives the model a better sense of size, making it seem larger because of the excess distortion. It seems larger to the eye because of how we are used to seeing large object in cinema -- that because of their size the front edge is significantly larger than the rear edge. Hmmm, I'm not sure if I'm making much sense... so, here's a more specific example:See how the wing on the nearer side is larger and the wing on the far side seems so much smaller in the second shot? That is what I'm referring to.So, what's a good default camera setup? Well, a "normal" aperture width is typically about 36. You can zoom in with little distortion, but if you zoom out past the aperture width, you will encounter much more distortion (similar to that of a fisheye lens). The best way to fit an object into the camera's view is not to adjust the zoom, but rather move the camera itself closer to or farther away the subject. Here is where I have my camera set:As for the "real" textures, yes, you are quite right. It is very difficult to make them render well on the ship because the lighting on the original shots was not uniform.(Edited by Bradster at 9:11 pm on Aug. 13, 2002)

[Bradster]

Oh, and here's a third pass for reference that uses the aperture width of the latter pass, but is moved back a tad more:http://brad.project-think.com/starfleet ... yby-03.mov (355 KB)

[AndyThomas]

Heh - funnily enough I'd added a camera in the right way before reading those posts - not sure if it was overriding the editor's camera though. Haven't done much serious tinkering tonight, but you were interested in a front view - the right hand side as you look at it hasn't been adusted quite as much as the left. Oh, and that's another thing that's been annoying me - is it possible to copy an object and paste it as a mirrored version? At the moment I'm resorting to make copy, mirror, get rid of original. Which is fine up to a point, just wondering if there's an easier way to do it...

[Bradster]

Yes mirroring an object is actually fairly easy. I use symmetry for all of my duplicates. Here's how you do it:Choose this object:and then make it the parent of ythe object you want mirrored.

[AndyThomas]

D'oh!  I'm sure I've used that symmetry doohickey earlier on, I'm going to have to start keeping notes or something!*Goes off to experiment**Comes back*Ah, right. So unlike mirror, you can determine the line of symmetry by reference to the symmetry tranformation's axes. And whatever you do to one side, until you make it editable, gets reflected on the other. Ooooh, if only I'd asked about that sooner, would have made replicating my wings much much easier! A little reworking is in order I think...One thing I did notice though in replicating my lower intakes, the texture map gets applied as if it's a single object - which of course it is. How would you then seperate them e.g. to get port and starboard? Is that what the knife function will do? I can see why it takes so long to master these sort of apps....

[Bradster]

Quote Quote: from AndyThomas on 11:40 pm on Aug. 13, 2002One thing I did notice though in replicating my lower intakes, the texture map gets applied as if it's a single object - which of course it is. How would you then seperate them e.g. to get port and starboard? Is that what the knife function will do?Well, here is where it gets a tad complicated. I'll describe how I handle mirrored objects that need textures applied in different directions. This is probably not the best or most efficient method, but it's the only thing I've thought of.First, you need to get your model nearly complete, or at least finish the pieces that are to be mirrored. Select the symmetry object (with the other object as its child) and choose Current State to Object from the Functions menu. This creates a new object with both the original object and its mirror merged as a single mesh.From there, you can delete the old object and apply your textures to specific polygons on the new mesh. I do something a little different, though.I don't delete the original. Rather, I hide it and modify the new merged mesh to be the mirrored object. When the new mesh was created, C4D automatically selected all of the polygons of the mirrored parts. Select the newly created object, click the Use Polygons Tool. You will see that the mirrored parts are selected -- we want to keep these. Choose Invert from the Selection menu. Delete the selection. The polygons are gone, but the points are still there. So, with the object still selected, choose Optimize... from the Structure menu. Be sure Unused Points is selected and click OK. You now have two separate meshes -- one of the original and one of its mirror.You know what? Just watch this 504 KB QuickTime movie below and this will make a lot more sense. :biggrin:http://brad.project-think.com/starfleet ... orial.movI just prefer this method because keeping the objects separated makes moving them apart easier if I decide they need more distance between each other. Texturing the two is no easier or harder than using a single combined mesh.As for the knife, that's used for splitting individual polygons. I doubt it would be much help here.

[Slacker]

Lookin good boys, lookin very good.Just wondering what you plan to do about the actual characters, being as how living objects are notoriously hard to render in 3D?

[AndyThomas]

Ah, well if you have a further look around you'll see that Shane is having a crack at characters using a package called Poser. I've had a few "real life" things to take care of of late, but hopefully I'll be able to find some time to return to my model shortly...