Could you help me with this?

[Wired]

I have a question which is pretty hard to put into words since I don't know what the procedure is called but here goes.

 

When you are finished with the map there is one final step left, atleast it was in the quake 2 engine games, to take in order to make the map run faster and give whomever played it a higher fps.

 

I'm not sure but it might have been called 'zoning' the level or 'sectoring' it or whatever, basically you defined which area could be seen from another area or something like that. Well that's the best I can do, especially sine Engligh isn't my native language...

 

Is there anyone out there who might have an idea about what I'm talking about, and could you tell me how to do it?

 

------------------- --- -- - - -

 

Another question: What's up with the compile options. When I choose to do a FastVis(1/2) bsp it's completed in 2minutes and 20 seconds, which is fine. But whn I chose to do a FullVis nothing seems to happen and after a half an hour (28 minutes) it hasn't completed the build. What's up with that, should it taka so much longer to do a fullvis or is something wrong?

[Wired]

*KABuMp!!!*

[DaviJade]

Well I know that mappers should use the Caulk texture on all unseen surfaces. This will speed up the framerate and improve gameplay.

[Wired]

That's not what I was thinking about. What I mean is that you 'break' the level up in different blocks, then the engine calculates if it cas 'see' block(sector?) 1 from block(sector?) 3.

 

Say if I make two rooms and a twisty corridor in between, then I put a sector (I'm not sure if this is what it's called) around room one, another around the corridor and yet another around the last room. Now, if I stand in room one the engine make some calculations on weather or not it can see the other sector from the one I'm in. If it can then it draws whats in it and if it don't it doesn't draw it. Thus making the map run better since it don't draw everything at all times.

 

Crap... I really can't explain it better....

[Wired]

...

[Wired]

I've just put in a few area portals (the ones you use inside doors to limit what the engine can see.) but that's not what I was thinking about.

 

 

Here's another take on explaining what I'm thinkigg about:

 

Like in Vampire, when you make levels in that you really should use sectors, what you do when you use sectors is that you make every room into a new sector so if you have like 12 rooms connected with each other in a circular pattern (every room in the position like a number on a clock) and stand in room one (sector one) the engine doesnt draw room 3-9 (sector 3-9) because there is no way they can be seen from where you stand. Field of vision or whatever.

 

This means the engine doesn't have to draw the whole level all the time, it just draws what can be seen from the position you're currently at. I'm not sure you have to do this in a Quake 3 engined game, this is the first time I'm mapping for it, but the option was there in Vampire (and I guess other Q2 Engine games).

 

 

Anyone know anything about this?

[Darth Draugmahl]

I never did that in a quake2 map though I know what you mean from using UnrealEd. You create zones and when you are in that zone only the areas in that zones are drawn. I don't know how to do that in a q2 or q3 engine game. I have heard that if you make all brushs that do not touch the void as details rather than structural it will lower the r_speeds but I haven't had a need to test that theory personally

[Hubris]

Textures under system, "hint".

 

Be careful with it, you can screw up your compile times badly with over use.