UGAC – Moving Juggernaut Vehicle Level - Rules

#41

Normally I would say quit-yer-bitchin’ but this time I feel like a mistake has been made, and I feel somewhat responsible. It is a difficult balance to make, if the specs are too high, people won’t finish on time.

Keep in mind it does not have to be the ENTIRE interrior and ENTIRE exterrior. You need to be creative on how to balance out your specs.

This is a challenge and it’s fun, but it is also for people’s portfolios. After revisiting the spec render, I feel that a 5000 tri’s for something like this is NOT competitive in today’s market. I have already PM’d DevilHacker urging him to edit the specs.

#42

I am just giving info. I have done my fair share of level design in the past, industry and user based.

I think putting normal maps with these specs is just laughable to be honest. It is like we want next gen but we will hinder you so much that you will not be able to make anything even close to UT2K4 quality what so ever. And than taunting us with pictures of Convoy and the other which are well passed the limit.

The idea is great and I think it is the best idea for game art contest ever done here imo. But the limits are really odd.

It would be hard to justify this to be portfolio worthy atm.

Btw just so I do not look like a noob I guess I have to say I made offical levels for UT2K4 while my stay at Epic Games.

#43

[color=white][color=white]I am sorry People. It is my mistake; I did not research the specs on the convoy vehicle enough, so I will fix it.
I am greatly increasing the specs, and am also adding a small amount to the time limit for any time lost, and to hopefully account for the greater degree of specs.

How do these new specs sound?
• 80,000 triangles LIMIT
• (6) 1024 x 1024 split for color (w/ alpha) maps
• (6) 1024 x 1024 split for grayscale bump, spec & normal maps

Edit: I will wait 30min.’s to hear the voice of the community, then make my decision.
To all contestants currently working, just know the specs will be a lot higher.

[/color][/color]

#44

weh weh weh woooow :eek:

yes finally I can really do this

#45

ladies & gentlemen, my nomination for understatement for the year. lol appreciate the more than generous boost, but just as i thought i was getting somewhere with what i had. oh well. i still have this evening to work. back to it!

-J

#46

I think that is a good number.

Some info that might help for level design and this contest.

Instancing objects like railing, doors, stairs, and any other detail that could be duplicated in the level design is a good way to do things. You save resources like this. This is the one of the reasons the poly limit can be really steep in next gen games. Plus it makes it easier just placing these decor objects and making the level.

Even though you have a huge amount of texture space, make sure to use it wisely. If your making a fps game for example you want to keep most of the detail of the textures in the playable areas but far away parts like ceilings can have lower res textures. So think of the playable area also. Reusing textures. Some wall textures can have a detail that can be placed as trims as well. Even though you have more room to work with you have more room for errors.

Edit: It would also be a good idea to split the polycount in 2 so 40,000 for the inside and 40,000 out.

#47

I think 80k and (6) 1024 textures is overkill. Not overkill for a Juggernaut, but overkill for a challenge for an individual to complete in 1 month.

Instancing would be awesome, but I was under the impression that not everyone here is familar with using game engines and wont be assembling their Juggernaut in a game engine. I dont even know where the specs should be placed for a contest like this, there are SOOO many variables.

#48

rofl - I have to agree :smiley:

Now I realize there’s a risk that this turns into a silly game of specification Pong, but isn’t that polycount a little over the top?

.

#49

What do you/everyone else think is an acceptable level?
I think that an
-Lower size level could be done in 20,000polys
-Mid size level could be done in 40-50 thousand polys
-Large sized level could be done in 50-80 thousand polys?
Question of the day is though, what should the size, and therefore the poly limit be?

[size=2]PS: Desp#2/Rog, That is some great advice that you gave there, do you mind if I quote you after the whole poly/texture limit mess is fixed?

[/size]

#50

ladies & gentlemen, my nomination for understatement for the year. [/QUOTE]
Haha.
Ya, I know…
What size limit would you like to see?
Express your opinion.

#51

Nah, I’m not sure really… I was mainly considering the fact that this is a fairly short challenge in terms of time to deadline, and to really make full use of 80K polys and associated textures strikes me as a somewhat daunting task (it would be silly to encourage needless tesselation so to speak).

But I will rest my case and let the more experienced game artists say stuff - I’m just a dabbler in this art :smiley:

.

#52

I have np with being quoted

The fact is this is how it is done for next gen and in some cases for some games last year.

Right now everyone must be saying… Wow this is alot of polys…

How about if I make a nice door that is 600 polys… repeated 8 times… It is already 4800 polys

Now think how many objects you really have to make. It will end up being small like around 10-20 objects to fill a place. Hell you can make like 2 trees and scale them different so it looks a bit diff and it will still be the same object in a game engine and you can repeat those like 10 times. You can think of instancing like that in any 3D program. You do not need a engine to work with instancing.

The poly count looks more daunting than it really is. The hard part is the textures imo.

#53

imo, i say stay where we’re at with 80k.
if you don’t use the 80k, you don’t use the 80k, that’s all. there’s no minimum, it’s just a ceiling. you know? just so long as it looks good & still feasible for a game.
now if you don’t mind me, i’m going to go play said ut convoy level for… um… ‘reference’. yeah. reference.

then more work!

-J

#54

You are totally right, but I just dont want some people instancing that door and thinking it is over. I guess if we put some thought into how it would be submitted, then there would be no problem with instancing!

#55

I think what ill_logic said is good also. Make it a range instead of being the amount everyone should shot for.

Level design has no 100% rules of how to do things hence why we are having such a hard time making sense of this. Working with ranges is best in a game level contest imo. At the end it is a balancing act.

#56

I, personally, think that next-gen levels are far too complex and time consuming to build by oneself.

:edit (I was looking at a post on the wrong page): 80,000 sound alright to me.

#57

um… ditto…

#58

very nice…cant wait.

-KDX-

#59

Ok. We been talking here behind the scenes here, and it sounds like these will be the specs.
[ul]
[li]15,000 tirs before instancing[/li][li]80,000 tirs after instancing[/li]li 1024 x 1024 color, alpha[/li]li 1024 x 1024 bump, spec, normal[/li][/ul]What does everyone think.
Hope to make this final by the end of the day.

#60

sounds good,though I was interested to see what people came up with using the original specs

do final submissions require any kind of in game shots?
I’d be interested in having a go but have no experience with porting a model to a game engine

80k sounds like plenty,40 for the exterior 40 for the interior…should be some good models coming out of this :slight_smile: