Model Competition (high poly models of vehicles)

Optimus Prime · Post by **Optimus Prime** » 23 Oct 2006, 16:27

In this competition, you can show, what high (or not) detailed models of vehicles you can make.

Rules:

- you can choose any 3d programm you want
- the model must be untextured (you can put textured pics too, but they will not count)
- the model should have less than 30k polygons (in other words, the models should be made so that you can have a lot of them in game
- it must be 100% your own work
- it must be a vehicle (hovercrafts are allowed too)
- the contest ends at the 30th November with a vote (1 or 2 weeks

have fun! :)

KDR_11k · Post by **KDR_11k** » 23 Oct 2006, 16:57

Why is there a limit in a high-poly competition? That's fairly low for a highpoly model. Why aren't we allowed to use as many millions as we need? Consideration for some retarded 3d apps that can't handle more than 30k?

Optimus Prime · Post by **Optimus Prime** » 23 Oct 2006, 17:28

I know that this is not a really high poly competetion in comparison to really high poly models, but i thought it would be best, if it is possible, to put the final unit into spring. Spring engine is good, but i bet that a million poly model would slow the game if you build a lot of them.

KDR_11k · Post by **KDR_11k** » 23 Oct 2006, 18:40

30k will already slow Spring down unless your mod allows at most two units to be deployed per player.

Optimus Prime · Post by **Optimus Prime** » 23 Oct 2006, 20:25

no. A 16 player War Evo game runs without a performance loss and a normal unit has more than 3k polys (makes 16x3 = 48k polys). I bet that you can play with 10 or more mechs per player and the performance will still be fine. As 100 times said, its the pathfinding what is slowing the game most.
Sure, 30 k is much for spring and you wont be able to make an aa mod out of units of this kind, but because we want some nice units, we need a max limit which is much higher than the best looking unit we have in any (spring) mod at this time.
So lets try what can we do with such a limit.

LathanStanley · Post by **LathanStanley** » 23 Oct 2006, 20:35

30,000 will kill it.

Optimus Prime · Post by **Optimus Prime** » 23 Oct 2006, 21:30

wasnt caydr testing 100 of his 10k poly ships in spring a while a ago? When i remember right, he couldnt notice a big performance decrease.
Anyone has some test results?

LathanStanley · Post by **LathanStanley** » 23 Oct 2006, 21:32

make a 30,000 poly geosphere, wrap a 256x256 tex on it
and then watch frames fall....

lol

Peet · Post by **Peet** » 23 Oct 2006, 21:32

This has been discussed in its entirety at least twice.

Optimus Prime · Post by **Optimus Prime** » 23 Oct 2006, 21:59

never the less, here is my first version of my model Hercules.
It is still in alpha phase (cockpit for example is ugly and will be redone as other parts will).

polys: 18478 (can be reduced to 16k without detail decrease i think)

Snipawolf · Post by **Snipawolf** » 23 Oct 2006, 22:24

ACTUALLY, the object count contributes a good bit..

Das Bruce · Post by **Das Bruce** » 24 Oct 2006, 04:48

All the poly's in the world won't save you if your design is retarded.

rattle · Post by **rattle** » 24 Oct 2006, 04:55

That's Mickey Mouse on crack... well almost.

I could enter my... car. Just need to select all faces, hit S a couple of times et voila, 15k polygons.

Keithus · Post by **Keithus** » 24 Oct 2006, 06:43

I see alot of.......... Extruding and changing the scale of faces, which is good.........If it imporves visual QUALITY.

Zoombie · Post by **Zoombie** » 24 Oct 2006, 06:47

Those tires look to be over done. Maybe just have some simple treds? Also the miniguns seem to be placed...oddly. Maybe its the angle.

KDR_11k · Post by **KDR_11k** » 24 Oct 2006, 08:19

That looks like "hm, I need to drive up the polycount, let's extrude some random faces". Model functional detail, not just random extrusions! Pipes, storage boxes, a ladder to climb on the vehicle, an entrance hatch, something to feed ammo into the chainguns, something to connect the motor to the wheels, etc, etc. Stuff like the profile on the wheels is better done with a texture. And try looking at your model without hard edges because that's what it'd look like in a game. Unless you unweld all vertices and completely choke the graphics port bandwidth.

Also try some practical considerations, that cockpit looks like the pilot dies when the unit takes minimal fire from the front. Imagine what would happen if this thing got hit by a tank shell from its primary combat direction.

Optimus Prime · Post by **Optimus Prime** » 24 Oct 2006, 12:50

lol its so funny to see ppl acting like offended.
But i m ok with it, you all think its odd (for an alpha vers) - fine so i can wait and see you models which will be so much better and than we have a great contest.

KDR: some good points, some are not. Who says, that its not a robot? Ok sure, a robot wouldnt make such a design, but who cares? The cockpit is weak yes, but so are other parts (wheels) and therefore no sense in making the cockpit tanklike when the rest will get destroyed by a shot. Also i said that the cockpit looks ugly and will be redone. Its just the first step after some hours of work...

you too have time until november the 30th

update:

KDR_11k · Post by **KDR_11k** » 24 Oct 2006, 17:07

Thanks but I'd rather spend my time on something that makes sense like a model that can be used in a game or an actual highpoly model that could be used in a cinematic or for normalmap rendering.

caldera · Post by **caldera** » 24 Oct 2006, 18:21

just want to say that i think it looks great

rattle · Post by **rattle** » 24 Oct 2006, 19:57

You have to consider usefulness as well... let's wait for the finished product.

Spring RTS Engine

Model Competition (high poly models of vehicles)

Model Competition (high poly models of vehicles)