[newman]

Speaking from (way too much) experience, working on more than one model at a time is a great way to spread yourself too thin and never finish anything. What's going to happen is you're going to reach a stage you dislike with one of them (typically UVW's for a lot of people) and then switch to the other one, then rinse and repeat. Then RL will take over, after several months it will be difficult to get back into it, and.. you get the picture. I know it's tempting to just go and grab every cool ship you can think of, but it's really a bad idea. Focusing on one at the time has always proven to be the best strategy during my work on Diaspora, and not just for me but really for everyone involved.

[rhettro]

Your N-1 looks good. If I might make a suggestion, though. Polygons that are the same smoothing group try to blend together, regardless of the angle between each other. Having a single line between two areas can be okay if the angle between them is low enough, but if your angle exceeds a certain amount, you get these weird shadows forming on your model.

That's great advice bobbtmann, thanks. If motivation moves me, I may try to fix it. My guess is that it's a shadow that more likely to show up in a high quality render than the game engine. Moving forward though, I'll being thinking more in terms of topology than approximating a ship's shape.

Speaking from (way too much) experience, working on more than one model at a time is a great way to spread yourself too thin and never finish anything.

Curses, newman has uncovered my true plan to never finish any model ever! Alas, this is also a good point.To quote Mythbusters, "failure is always an option." I make no promises. If finishing models was my motivation I'd definitely take your approach. But I'm not here to finish models, I'm here to create cool stuff, learn and fine tune my methods. To that end, a variety of models will fit my purposes and they aren't critical to any mod on this board.

[newman]

My guess is that it's a shadow that more likely to show up in a high quality render than the game engine.

Actually, it's the other way around. Things like these are visible in the engine, and can be quite easy to hide in a render. Renders have the advantage of not having players able to inspect the models in them from every angle and lighting scenario possible. I've seen quite a few horrible meshes that were used for pretty good renders

Curses, newman has uncovered my true plan to never finish any model ever! Alas, this is also a good point.To quote Mythbusters, "failure is always an option." I make no promises. If finishing models was my motivation I'd definitely take your approach. But I'm not here to finish models, I'm here to create cool stuff, learn and fine tune my methods. To that end, a variety of models will fit my purposes and they aren't critical to any mod on this board.

Fair enough

[rhettro]

I suppose I could alway purchase a professionally model mesh and copy the wings in to take care of the shadow problem. Here's a good one. http://www.turbosquid.com/3d-models/3ds-max-naboo-starfighter-star-wars/393912

[Thaeris]

By "good" do you mean "joke?"

No offense, but even if it were for a price, I'd not spend a penny on it. That's a really shoddy model.

[rhettro]

LOL

[Thaeris]

If I were to suggest a solution, it would be modeling by parts. Model the engine nacelles via a revolve operation, they are cylindrically uniform, after all. Then model the wings as an airfoil and extrude them out. I'd assume the foil is of constant section geometry, the scale is the only thing that varies. Merge that wing together with the fuselage, and then blend the rough parts, then merge the nacelles and the wing - and you may not even need to do that! In the end, flat quads and nice edge loops will do much to solving many of your lighting troubles.

[rhettro]

Actually, that's exactly how I constructed it Thaesis. All the quads are planar, I know because I went back with a tool that flattens them. They were already pretty flat to begin with. bobbtmann's suggestion is closer to the solution. The larger the angles between smoothing groups, the more likely shading errors are to occur. A narrow, uniform fillet between them aids in the transition. I could cut the polygons on my leading edge, but it will introduce triangles which may necessitate rebuilding the transition into the fuselage and the engine. Or maybe it'll be fine. I'll worry about it when it comes time to port it to PCS.

[Thaeris]

That's actually what I was noting when I said "blend." Manually adding an edge loop in a curved section is a LOT of work if you do it right, at least with the way I work. I assume higher-end programs may be able to manage this a bit more easily, but then, I've seen a lot of models made with very pricey programs that aren't worth a dime.

This is actually a project where tactful use of subdivisions may be worth the work - of course, that will mean carefully making the base model, which will subdivide properly (and splitting it where necessary), and then reducing it to proper levels of detail. Proper subdivision modeling is not a simple or short affair.

[bobbtmann]

Manually adding an edge loop in a curved section is a LOT of work if you do it right...

I don't think it would be that hard. Just select a ring along the leading edge of your wing, connect them, pull the new line forward a little bit. It might stretch your UV, but that shouldn't be too bad.

Also, I wouldn't worry about keeping your quads planar. It seems more trouble than it's worth, especially such rounded models.

[rhettro]

Well I may end up tweaking here and there, but I ain't rebuilding it. I agree it would be cool to construct it in a way to take advantage of subdivision. As currently constructed the subdivision modifier tends to take the bulk out of it while making it look slightly smoother.  If I ever get around to the J-Barge I might have to think about it.

[newman]

Well I may end up tweaking here and there, but I ain't rebuilding it. I agree it would be cool to construct it in a way to take advantage of subdivision. As currently constructed the subdivision modifier tends to take the bulk out of it while making it look slightly smoother.  If I ever get around to the J-Barge I might have to think about it.

Just apply what you've learned to the next model, then the next, and the next, and you'll be fine. Buying other meshes - even if they're good - wouldn't achieve the goal you're after, which is learning how to do it yourself. Sure you can see how those meshes were set up as an example but tbh it's not worth the money. Plus, that mesh on turbosquid isn't very good, and costs about 15$ more than it's worth

[rhettro]

I was hoping that the fact that I linked to such a bad mesh would clue people in that it was a joke. I would have expect a greater sense of humor from a board dedicated to grown men that like to pretend they're flying space ships.

[newman]

Well, when someone who's obviously a beginner at this (no offense intended) links to a bad mesh, the joke's not as obvious as you might think. Especially on the internet where you don't really know the vast majority of people you communicate with.

[rhettro]

All to true.

[mandobardanjusik]

hmmm, if he could model as fast as he defends himself with witty comebacks, he would be done already

[rhettro]

http://p3d.in/X0Jnz

Jumping on the P3D bandwagon. Still looks like a couple mesh issues to be resolved on the underside, but nothing major.

[Luis Dias]

The mesh in the bottom side seems irregular. There also seems like a weird triangle when the back of the wings meld with the rest of the ship body. Still it's almost perfect. The imperfections of the texture in the grey areas make it look uglier than it is.

[rhettro]

Yeah, I noticed that weird triangle. I'll fix it. Environmental reflection will pretty it up somewhat in game.

[Luis Dias]

Oh don't worry too much about the environmental reflection... just the per pixel lighting will improve it a lot (the vertex lighting is really a pain in the ass in low poly models).