Recruiting: AeroAssault

[RithiX]

Another update:
I started modeling the T1 mothership. Im planning on having them start small (but still kinda big) then increase as you advanced.


It's 57 polygons, as I am determined to keep my models low poly and effecient. I could probably take it down to 47-49 polygons if I tried removing some triangles on the bottomside but I am not that desperate.

Please leave feedback. This is my first 3D model, besides a few failed attempts at ships in anim8or. I rather like wings3D.

EDIT: Oh yeah, the mothership acts as a landing platform for planes to refuel and heal. Thought Id say that.

[MidKnight]

Feedback for your model: wings are not that aerodynamic. I know they're suppose to be landing pads, but imo it's a good idea to implement landingpadwings on a bigger, higher level commander

ALSO:the spring engine can handle lots of polygons. A classic example is Spherebot, a unit CA players make dozens of regularly. Spherebot has nearly 1500 polygons, and doesn't lag the game in the least, even if 30 are made. What the spring engine slows down with is usually effects and explosions. So feel free to use large numbers of polygons. For a unit the player is going to build only 1 of, it's fine to use even up to 5000 polies! What you want to avoid, though, is polygon waste. Use as few polygons as you can, but do not sacrifice quality for low polies!
otherwise, good first model!


As for my models, you can find the first one (along with Argh's reskin) HERE and the second one HERE

these and all my future models will be in the Art and Modeling forum

PS: you may want to add more design documents, that clarify the concepts of the game. My particular questions: are there *any* fixed structures? Will land move? If so, how? Do you plan on implementing "deployable" structures (ie: moves, stops, plants istelf into air and becomes immobile)?

also, a cool idea would be if morphs don't change how your commander looks, but instead adds stuff around the edges. (IE: with your current model, the morph wouldn't change what you have already made, but add giant wings and rocket launchers to the sides) so that after a few morphs, the commandr looks nothing like the starting commander, but if you looked closely, you could see the starting commander's pieces near the center.

[Saktoth]

START SMALL

Start with your basic economic model and a few units. Get your core gameplay concept up. THEN decide where to go. ONE faction first. If you have never made a game before your plans will change dramatically as you see the practicals of the implimentation.

Your design document shouldnt feature specifics of units and tech tree, it should feature what your goals are, your economic and gameplay model, your core balance and counter mechanics, what mechanisms you want to use, etc etc. People think games are just a bunch of units, they're not, the units designs are the last part you need to worry about.

It can help to start with free art assets to rough out your gameplay concept. See my post in the Libre thread for info on this. CA includes a lot of free content you can use to get a mod roughed out.

[RithiX]

I'm... a hardcore C++ programmer + GML and XNA. I will ponder your advice but Ive already planned this out beforehand. Now can somebody please suggest how the T2 commander can look. I'll either have to redo the T1 or somehow incorporate more into the T1 to make the T2. =S

[Saktoth]

If you've never made a game before, you will find your design changing as you progress. Thats all im saying. Start with a base you can actually run in spring, and work from there.

On your 'low poly, effecient model'- firstly, it doesnt use its polygons effeciently. You've got a lot of beveled edges there you dont really need given the complexity of the rest of the model. Secondly, the CA Spherebot has 1100 polygons. You see several hundred of these things on screen at once with no slowdown. Polygons are not that much of a bottleneck for spring.

You dont need to limit yourself to 50 polygons for your mothership, which will be a 1-off, unbuildable unit (i assume).

This is the Commander used in CA and XTA, by mr.D Look at the polgyons in the head alone. I dont know how many it is but its a lot.


OTA is 10 years old and even they have more than 50 polygons on their units.

[RithiX]

Landing pad: ~8 each
Body: 17
Back bit: ~19-22'ish

Just thought I say that.

Anyway, you are right. Im used to working with a 1500 poly pipeline (GML).

[KDR_11k]

I'm... a hardcore C++ programmer + GML and XNA. I will ponder your advice but Ive already planned this out beforehand. Now can somebody please suggest how the T2 commander can look. I'll either have to redo the T1 or somehow incorporate more into the T1 to make the T2. =S

Being able to implement does not mean being able to design. RTS design especially is very complicated and there are numerous decisions involved that aren't apparent just from looking at finished games. As a hardcore coder you've probably gone through the complete software design process including requirements analysis, rough and detail design, implementation and testing? Games need the same thing. A unit list is like a bunch of class diagrams, that's not a design.

[SpikedHelmet]

This is the Commander used in CA and XTA, by mr.D Look at the polgyons in the head alone. I dont know how many it is but its a lot.


OTA is 10 years old and even they have more than 50 polygons on their units.

Holy caydrhead.

[MidKnight]

One thing you should know: when it comes to modeling, your first revision is almost never very good. No matter how experienced a modeler you are, you will have to do more than one revision on most of your models before you finalize it. For example, Betta (my 4-winged aircraft) has had a total of 9 revisions, and many of Mr. D's amazing models are the result of many revisions, each improving over the last. So my advice is: try again!

[Hoi]

Well actually, I usually do it like I want it to be the first time

[MidKnight]

Well actually, I usually do it like I want it to be the first time

same, but then saktoth steamrolls it... so I've given up the idea of a hole-in-one model

[Squall815]

Hi RithiX

I 'm working on my mod too but I have some trouble to understand how lua works so I think is more useful to help you with 3d models than try to continue my own mod alone.

you can find what I have done here :
http://seb.charon.free.fr/spring/

Almost all models have less than 300 poly (by poly I mean quad)
except some model use as powerful units (screen shots).
As you can see I need to improve my textures but at lest I can do 3d models

Feel free to use any models (you can find every thing in stardust.sd7) bu t I not sure they can match with your idea.
So if you have some sketches I can try to make some models for you.

By the way I use some from different mods (lua scripts, explosion, resources) so I want to thank the mods E&E, BA and Sim base.

ps: please forgive my bad English.

[Pxtl]

START SMALL

That's what I was trying to say with my "start with L1". Start with the core gameplay - a handful of features, and then branch from there.

You could get a pretty good synthesis of common BA gameplay with only about 10 or so units (comm, solar, maker, mex, factory, LLT, radar, tank, conveh, and some kind of porcbreaker like a bertha or something). Yes, there's a lot more to BA than that, but most games come down to how you use those core assets.

So whittle down your idea to the very core kernel of the gameplay, and start with that. Future-proof it, of course - think long-term about how you would expand it out to L2 and L3, what special powers, etc. But start with something simple, small and achievable.

Did Argh ever reskin his old Nanoblobs units such that they no longer use that little team-colour-ball thing? That's also a good source of libre content.

[RithiX]

Those models are excellent!

[KDR_11k]

No matter how experienced a modeler you are, you will have to do more than one revision on most of your models before you finalize it.

Nope. That may be when you're learning how to model well but a veteran won't need more than one attempt (some games have two attempts on the main character because the first one is done early, before the tech is properly established and then at the end it's redone to really make the most of the tech). If you want to redo the design then do that in the concept stage, not once you've got the whole model done.

[MidKnight]

No matter how experienced a modeler you are, you will have to do more than one revision on most of your models before you finalize it.

Nope. That may be when you're learning how to model well but a veteran won't need more than one attempt (some games have two attempts on the main character because the first one is done early, before the tech is properly established and then at the end it's redone to really make the most of the tech). If you want to redo the design then do that in the concept stage, not once you've got the whole model done.

hmm... I don't know, maybe it only applies for ota remakes.
I was thinking of Mr. D's tremor/radar veh when I said that.

[Squall815]

I'm happy you like my models.

I start to make a model for a ship commander. It has 3 level and I think it could interesting to chose an orientation for the level 3. For example you can choose a level 3 with big engine which can increase the speed of your commander or choose a lvl 3 with new weapon which can provide a very powerful beam. The beam have the same nuke's power and works with stockpile.

(This model is not finish so there is no weapons yet)





And perhaps that would be nice to add new staff (weapon, landing pad, ..) without changing the level of the commander.



I 'm still noobs in low modelisation so do not hesitate to leave comments

[Squall815]

Any comments ??

Is my model ugly or do you wait for textures ?

Here are new screens shot.

The version T3 with engine


The version T3 with weapons (a)


An other version T3 with weapon (b) (but I think the version (a) is better)


There is roughly 2000 polys (quads and triangles) but if this units is unique it's not so much. What do you think ?

[smoth]

it is fine man, I have units way over that count.

[Pxtl]

Much better. I like the shape of it, and your polycount is probably about right for a major unit like that. You could probably even burn some more polys on the main fuselage to give it a bit more defined shape and sculpting if you have ideas on where to put them. However, you may want to re-think physically modelling the air intake fins - that kind of thing often results in a mess of z-fighting when viewed from afar - can't really say until I see it in-game. The guns might be too small (and plentiful) - I see where you're going, but a lesser number of cannons that are bigger lets the player better-see what's going on.

And the player will never see those belly-guns. That under-belly pipe thing looks kind of bolted on - it doesn't work with the shape of the rest of the model, and besides that modelling the interior of a gun barrel isn't generally done.

Still, a solid design.