Some wall issues
Some wall issues
G'morning, folks.
I found that Arma tends to cripple columns when they're stacked into each other. It always occurs from the 11th one (counting from the outside), getting worse quickly with more columns, and only when the outer walls come first in the code. Is this a known behaviour or can anyone verify this?
Also, when a height is specified and high_rim is zero, are walls supposed to be cut off at height 100?
I found that Arma tends to cripple columns when they're stacked into each other. It always occurs from the 11th one (counting from the outside), getting worse quickly with more columns, and only when the outer walls come first in the code. Is this a known behaviour or can anyone verify this?
Also, when a height is specified and high_rim is zero, are walls supposed to be cut off at height 100?
- Attachments
-
- z-0.aamap.rar
- The source code
- (688 Bytes) Downloaded 318 times
Re: Some wall issues
Well, it's expected that it'll break down at some point; I'll try to have a look at whether it can be pushed back a little. But you shouldn't do what you're doing anyway, it's a colossal waste of fillrate.
And yeah, height=0 gives the walls a default height.
And yeah, height=0 gives the walls a default height.
Re: Some wall issues
Nah, not height=0. walls above the limit 100 are clipped regardless of the attribute value, if the rim is set to low. Actually like
height>100 && high_rim==0 ? height=100 : height=height
And no breaking down when construction starts in the center, see the pic below.
Not a biggy, I was just curious if someone noticed this before.
height>100 && high_rim==0 ? height=100 : height=height
And no breaking down when construction starts in the center, see the pic below.
Not a biggy, I was just curious if someone noticed this before.
Re: Some wall issues
I'd say the issue is that the structure is too regular. New wall end points are constantly made right on top of connections between two existing end points (the diagonals of the squares). Slight distortions should be enough to make this work a lot better.
Re: Some wall issues
Sorry for not being able to help, but that tower looks just F***ING AWESOME in Armagetron
No, it's artZ-Man wrote:But you shouldn't do what you're doing anyway, it's a colossal waste of fillrate.
- Jonathan
- A Brave Victim
- Posts: 3391
- Joined: Thu Feb 03, 2005 12:50 am
- Location: Not really lurking anymore
Re: Some wall issues
Fixed.Z-Man wrote:But I shouldn't have made you so desperate that you would do this anyway. I'll go and implement proper 3D models now.
ˌɑrməˈɡɛˌtrɑn
Re: Some wall issues
I'm very, very impressed. I've wanted to do something just like this for a long time, but never got around to it (you can see how I hint at this idea in the AoT servers). I'm jealous you beat me to it, haha.vov wrote:it's art
Re: Some wall issues
Merci, guys. And you are exaggerating, sine :) Yes, your servers really inspire me a lot.
(Can anyone provide me a detailed isohypses map of the alps? Don't ask why.)
@Z-Man: Thanks, only... why does it work when stacked from the inside out? *confused* Well, not so important as long as there's a workaround.
Talking 'bout modelling, may I kindly ask the developers to consider these implementations:
<Wall height="x" bottom="y">
to make a wall start in the height y, ending at height x. Similar to zone_bottom, but for each wall segment.
(Alternative: <Hole/> as a child element of <Wall> with the attributes width, height and bottom)
Directly related would be something like
<Ceiling height="z">
<Corner x="x1" y="y1"/>
<Corner x="x2" y="y2"/>
</Ceiling>
for a rectangular ceiling (x1,y1 - x2,y2) in the height z (without collision detection of course). Actually, my little one would ask for this, he wants a tunnel.
And maybe some sort of
<Zone height="y">
to assign each zone an individual height, overriding zone_height.
Pretty please?
(Can anyone provide me a detailed isohypses map of the alps? Don't ask why.)
@Z-Man: Thanks, only... why does it work when stacked from the inside out? *confused* Well, not so important as long as there's a workaround.
Talking 'bout modelling, may I kindly ask the developers to consider these implementations:
<Wall height="x" bottom="y">
to make a wall start in the height y, ending at height x. Similar to zone_bottom, but for each wall segment.
(Alternative: <Hole/> as a child element of <Wall> with the attributes width, height and bottom)
Directly related would be something like
<Ceiling height="z">
<Corner x="x1" y="y1"/>
<Corner x="x2" y="y2"/>
</Ceiling>
for a rectangular ceiling (x1,y1 - x2,y2) in the height z (without collision detection of course). Actually, my little one would ask for this, he wants a tunnel.
And maybe some sort of
<Zone height="y">
to assign each zone an individual height, overriding zone_height.
Pretty please?
Re: Some wall issues
It's because the grid engine was never tested with regular patterns. It was designed for semi-random cycle walls. The better algorithms I have planned won't have these problems.
As Jonathan says, proper model decoration support would be a much better idea for things not relevant to gameplay. Don't get me wrong, it's cool what you're doing, but from an artistic standpoint, you're painting with shit and your bare hands on cave walls, and you're asking for more colors in your shit. Wouldn't you rather have a brush and a canvas and colors that don't smell?Wik wrote:Pretty please?
Re: Some wall issues
hahahaha, wow Z, that was unexpected. Still laughing as I type.
Re: Some wall issues
I take this as a "No".