CTF-tanecris

[Kaal979]

Another news from thu front:
The uni corp managed to break the outer lines of the dark legion citadel tanecris.
But there is still the unknown and heavily defended center - the citadel itself ... .



Still inprogress with bot pathing.

[TexasGtar]

Ah yes a nice open field battle. Are those tanks and do they move and shoot?

[Kaal979]

This are armored transporters.
They had to stop there and throw out the infantry because of the ruins.
This really gonna be a huge battle with 64 bots in the ini file cheated.
The enemies spawn almost everywhere in the far lines so that they cant make serpents.

[TexasGtar]

Looks fun!

[RisingFury{TSU}]

Originally Posted by Kaal979 This are armored transporters. They had to stop there and throw out the infantry because of the ruins. This really gonna be a huge battle with 64 bots in the ini file cheated. The enemies spawn almost everywhere in the far lines so that they cant make serpents.

Don't even bother. Nobody will play cos it'll lag so bad.

[TexasGtar]

Originally Posted by RisingFury{TSU} Don't even bother. Nobody will play cos it'll lag so bad.

He needs to add a lot of hills to occlude the whole picture. But yeah that looks like about 3000 polys to me..lol

To Kaal
type in stat fps in your editor console or in the game console to see how many polygons are in view. You acan move around to see where your problem poly spikes are and try to build hills that hide a lot of the background polys.

There is no way to zone an open area so you cant gain performance from zoning this map. There are other tricks though.

Today's computers can handle 400-500 polys in view but ideally you only want 300 max in view from anywhere in your map if it's intent is to beplayed online.

i'd suggest 100 for really good gameplay online.

[RisingFury{TSU}]

It's not rendering that's the problem...... not by far. D3D8 and D3D9 and the OpenGl renders have the potential to render 1000's of polys...

Hell, if you look at an average UT character model, it's at 2200 polys.

So you're probably asking: What's the fuss about BSP terrain then?

Well, it's not rendering that's the problem... it's the collision detection. BSP is inherently slow because of the algorithm... and collision detection is handled by CPU completely. CPU's have gotten immensely powerful lately, but with multi-core CPUs having to restrict UT to one core, you can't expect much improvement.

So 300 polys is the topmost you'd wanna go to. And don't even dream about anything over 16 bots.

[Kaal979]

This is the so calle beta release.
The performance isnt the badest but the bots wont run for the red flag nor return with the blue one.
What causes this? Help please!

MEGAUPLOAD - The leading online storage and file delivery service

[TexasGtar]

Originally Posted by Kaal979 This is the so calle beta release. The performance isnt the badest but the bots wont run for the red flag nor return with the blue one. What causes this? Help please! MEGAUPLOAD - The leading online storage and file delivery service

Couldnt you just zip it up and put it here? Now i have to decipher thatr code which i tried 3 times??????????????? No clue even a human cant decipher their code thingy.

Now i have to download winrar. a lot of work for a 2mb file. winrar is for really really big files.

okay i got it brb

[TexasGtar]

Completely unplayable. I get over 2000 polys in some views. It's like a slide show. If you dont fix this it will never get played.

Aftwer looking at it in the editor i agree with Rising Fury.

Scrap this and read some tutorials on map making.

You obvious have creative talent and some working knowledge of the editor but without learning a little you wont be able to create anything worthwhile.

For example you have used a lot of spheres to make some of your things. Spheres are the worst frame rate killer there is . You can get away with it with some small extrapolations but not with high extrapolations like you have and so many of them.

You should really read about pathing your map. Right now the map path network will crash ut after being played for a small amount of time because of the amount of paths you have. There are over 39 path nodes that offer over 1000 possible routes. (in editor log) This means that bots try to figure the shortest route they have to figure 1000 routes for one node. It soon fills up memeory and will crash ut on the vbest of computers.

The theory of pathing a map isn't to place them equal distance in a large square pattern but to place them in a way to emulate a human player playing the map And though it iss good to have several routes it is not good to have over 1000 apples all connected to each other. If you look at your path network it is a big mess. Blue lines are good red lines are not good. Bots prefer the blue lines...maybe another reason why they don't grab the flag...plus the dificulty they are having figureing out where to go.

There is no way to fix this map in my opinion. I suggest you chalk it up to a grweat tool that you have used to gain knowledge and learned some things about how to make a map. However i see no way to save it.

It is just too much to fix. And no one will play it. So why waste anymore time on it. Of course it is up to you but that is my suggestion. Try to start another map....keeping in mind from the very first brush that a good map also has to be playable, not just cool to look at. I would also suggest you read some tutorials on optimizing a map before you start another one. And that bot pathing tutorial. I applaud you for attempting to path this because i have seen many a map without poaths at all which i can not stand people too lazy to path their maps. It is not a map without pathing. Sniper maps need pathing too.

But i hope i dont sound too discouraging because these are mistakes made by everyone starting out. You have undertaken a huge project in this map and made something very cool, but unfortunately you will be the only one who will try to play it.

I will be happy to try to help you on your next endeavor, but i don't think i can help on this map. Sorry.