--- Todd < temitchell at sympatico.ca > wrote: > Well it's big, I wouldn't want to see two of them this large anyway. Where should I put the full set of maps and scripts and stuff on Saturday? > Did you take your version of the maps in the last six weeks after all > the forest I added in the same area (but up and to the right...) and > after the changes to Brest (didn't see that on there). All my work has been on the release version, orignally 1.4 and the small world, but I've been using 1.5 since the day it was released. > I do think there should be a 'Great Forest' (names anyone?) to the east > of Scorn but four more forests of this size and the world would be one big > forest. There is an entrance to the "Dark Forest" near the Scorn / Navar road in that area. Or we could name if after the owner of the castle I built there B-). My idea is to have him be a big influence on the surrounding area as a powerful prince. He has some history with the nearby dragons, the King of Scorn, Navar, and Lythander. I am going with the "Elves look after their local forest" mythos, and the prince is an Elf. So basically, my large forest is an elven forest, with the prince being the head elf. On the other hand, his influence may not cover the entire forest, just the bit a few tiles either side of the deep lakes. > Did you change the elevation on the lakes to make them deep? I am As stated before, my script only changed the arches used, all else remained untouched. > starting to get cozy with the idea of having a single elevation value > per map (determined by finding the current average?) and then generating > the elevation of each tile on that map based on the arch and that map > elevation value, to generate the weather maps - I think this would be > better way to handle it than actually having the elevation values in the > arches. It would allow for a deep lake in a high mountain area or high > forests or deserts which is not the case currently. It wouldn't be hard > to do (he said expecting to be corrected...) I think, just changes to > the weather map genertation routine and the map headers, and it would > free up the map maker form having to change the elevation to make the > weather do the right thing... You worked with the weather code a bit, > what do you think of that? Elevation is used in more than the weather calculations if I'm not mistaken. LOS is oftem blocked by higher ground. Movement rates should also be modified slightly when going up or down hill (I do a lot of bush walking), dunno if they actually are though. Map makers changing the elevation is a problem, just look at the elevation around the road between Scorn and and Santo Dominion. Looks like when the orignal big world was generated and the first towns and roads where grafted onto it, who ever did that road lost all the elevation info. On the other hand, this could be seen as a general problem. Sometime in the future some other information could be attached to the squares on maps and the editors will have to be updated to not drop this information. I would rather solve the general problem, but thats just the way I am B-). Local mana level / god influence are the sorts of things that would have the same problem. A good example would be my elven forest, Lythander would have a greater influence in the forest, his altars would work quicker / better, his worshippers would have a little bit more luck, orcs and goblins would be scared to enter, etc. However the Dark Forest is inside of my large forest, with the Dragon City just on the edge, so a per map figure would not work. During the next bit of discussion, keep in mind that this is all from memory, my source code is at home and I am not. I can check my facts on Saturday. Also, this is more a bunch of random thoughts than a fully thought out response. As for the weather code, elevation is used in two ways. The elevation map that is initailly generated is an average over the area covered by that particular weather pixel, but his data is then cached. During the weather loop, temperature at any particular point on a map is calculated based on average temperature for the weather pixel and elevation at that square. It is the initial calculations that actually fix the granularity of the weather system, the current code is described by the author as a nightmare, it is not flexible. Note, average water / land ratio is calculated during the average elevation calculation, which is why it takes twice as long B-). So average elevation for particular maps would greatly speed up the initial elevation map generation (which does take a very long time) since it spends most of that time calculating that figure anyway. Going the other way, from an average per map to specific elevation per square as you would like to do will probably take longer. Doing that on the fly would probably be too big a performance hit. Caching could be done, but note that the inital elevation map is not currently recalculated when the maps change. Now, I want to chage the granularity of the weather system, making it flexible enough to handle any granularity. I have probably mentioned that I am in the process of pulling the weather code out of the server and running it on a cluster. I will need greater granularity. I suspect that your proposed change would mostly move the problems around, all for the sake of not having to be more carefull with editor code. I say we should make a general fix for the general problem of extra per square info in maps. On the other hand, I like your idea if it is applied to the problem of generating all the elevation in the first place, but I have not looked at the current elevation generation code. I doubt if the world generation program will get much use though. Having said that, the weather code implies that the big world could be mapped to one side of a sphere, leaving the other hemisphere currently uncharted... http://mobile.yahoo.com.au - Yahoo! Mobile - Check & compose your email via SMS on your Telstra or Vodafone mobile. _______________________________________________ crossfire-devel mailing list crossfire-devel at lists.real-time.com https://mailman.real-time.com/mailman/listinfo/crossfire-devel