Shared Map

[Magnus]

Yea I made a few errors in my last version that messed up bot play. I fixed it in this version I am about to post. The bots use the map MUCH better now. Lots of varried paths taken and reduced stupid decisions now like mass backflips into the lava in the midfeild.
I agree there is still some tweaking that could be done to deal with some balance issues on getting out of the base with the flag. Perhaps some obstructing objects to slow down getting into that lower path out of the flag room or maybe adjust that path so it is a bit longer to get from the flag room to your RG/fog tele/haste room.
I really like a13n's spiral stair/pillar/regen room, but the only areas getting used in that room are the upper path I reworkd and the JP I placed under it, the fog tele and the regen corner. The rest goes unused. So yea I would think it could use a rework.

Can't wait to see what you have once it is finished. Why dont you see what you can do to tweak a13n's spiral stair/pillar/regen room in your next version.

Also I made those ramp changes to the midfeild to deal with that quick crossing and plenty of cover path across the midfeild issue you mentioned. I think it works as well as looks really nice.

Anyway I was wondering where everyone has been.

Well lets get the team in here and have a look at this version. Plenty of changes to make the entire map and gameplay WAY better.
Also check out the .txt in the .zip and see if there are any issues or things that any of you can deal with or fix. Thanks.

magnus_ctfptm7

Well there it is. Check it out and lets hear from you guys.

Thanks again.

[obsidian]

I'll be spending the weekend looking over Wattro and your versions of the map and see what I can come up with.

[wattro]

bots do definitely use much more of the map which is great. but there is also a stutter in the game. i think this and the fact that everything is detail is leading to the z_malloc issue. let me try building in a caulk hull and see if that helps get rid of the issue.

[Magnus]

Yea I hate that stutter and the z_malloc crap.

It all started when I ran a compile to make a .aas for the map.
It had been taking about 25 seconds to create a .aas and only ran about 300,000 splits, but then this one time it took 1150 seconds and ran about 2,750,000 splits and it has been doing the same thing ever since.
I actually made a bunch of brushes detail after the first few times this problem happened. I hoped making many of the brushes detail would resolve the issue, but it didn't seem to make it better or worse.
I left the hull box around the map, any brushes with shaders, any brushes that were a func, and all trigger brushes structureal of course.
The last two times I ran a .aas compile it only took about 300 seconds and had about 1,400,000 splits. That was after I removed all of the Hint brushes, but the problem is still there.
Before this large number of splits and long compile time issue when creating the .aas there was no stutter and z_malloc problem.

I hope we can find the problem, because I am having the same issue with another of my CTF maps I built about a month ago and would love to know what to do to resolve it.

[wattro]

I left the hull box around the map, any brushes with shaders, any brushes that were a func, and all trigger brushes structureal of course.

i'm not sure why you are turning shaders/funcs/triggers into structural. the hull box is ok (i suppose), but not really preferred and usually leads to sloppy map construction (there were a lot of leaks to fix!). the only thing that needs to be structural is where you want to create bsp splits to help control how much of the level exists in each leaf-node.

essentially you are creating a huge box with some weird splits (wherever you have triggers, shaders, and funcs). this is probably why the hints (and cluster portals, if you have 'em) weren't doing very much.

The last two times I ran a .aas compile it only took about 300 seconds and had about 1,400,000 splits. That was after I removed all of the Hint brushes, but the problem is still there.

i'm not too worried about the # of splits, but it would be interesting to see what the reachability areas are (i'm guessing your bspc.log has at least one area with >2000 areas). anyways, i'm pretty sure that putting in a decent caulk hull will help with the problem. should even mean we can compile the bsp with lighting in reasonably fast times.

i'll put up a (hopefully!) optimized version in about an hour or so.

[Magnus]

i'm not sure why you are turning shaders/funcs/triggers into structural.

Sorry didn't mean to cause a misunderstanding. I was under the impression that brushes start out as structural and it is up to the mapper to decide to make some or all detail. My bad.

I never actually "made" any brushes structural; I just made many of them detail and left the rest as they were. So I guess in that case I never turned any of the brushes structural...lol. So any brushes that are not turned into "Detail Brushes" or Structural Brushes" would be refered to as what? "Normal Brushes"? Or just "Brushes"?

i'm not too worried about the # of splits, but it would be interesting to see what the reachability areas are (i'm guessing your bspc.log has at least one area with >2000 areas).

I bet you are right..lol (see below)

anyways, i'm pretty sure that putting in a decent caulk hull will help with the problem. should even mean we can compile the bsp with lighting in reasonably fast times.

"caulk hull"... you mean a big box around the map made of caulk or the actuall outer walls of the map it's self that create a barrier between the space inside the map the the big caulk box hull containing the entire map?
Because I have the big caulk box. So I assume you mean the walls and floor and celing and such of the map it's self.

Anyway here is a copy of my bspc compile:

nummapbrushsides = 39652
62 curve brushesWARNING: AAS_CreateCurveBrushes: no winding
85 curve brushesWARNING: AAS_CreateCurveBrushes: no winding
282 curve brushesWARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
726 curve brushesWARNING: AAS_CreateCurveBrushes: no winding
749 curve brushesWARNING: AAS_CreateCurveBrushes: no winding
850 curve brushesWARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
1256 curve brushes
1 threads
-------- Brush CSG ---------
7325 original brushes
6878 output brushes
-------- Brush BSP ---------
6878 brushes
62053 visible faces
0 nonvisible faces
72420 total sides
Win32 multi-threading
1 threads max
depth first bsp building
1303894splits
131267 KB of peak total bsp memory
BSP tree created in 1126 seconds
------- Prune Nodes --------
1091382 pruned nodes
---- Node Portalization ----
212513 nodes portalized
26176 tiny portals
39180 KB of portal memory
63406 KB of winding memory
------ FloodEntities -------
------- FillOutside --------
52295 solid leaves
18161 leaves filled
35801 inside leaves
35800 areas created
5780 face merges
1227AAS_TestSplitPlane: tried face plane as splitter
1295AAS_TestSplitPlane: tried face plane as splitter
1305AAS_TestSplitPlane: tried face plane as splitter
1531AAS_TestSplitPlane: tried face plane as splitter
2156area 40111 face 208767 flipped: front area 40111, back area 40112
AAS_CheckArea: area 40111 face 208767 is flipped
AAS_CheckArea: area 40111 center is -3062.414795 -1154.352539 176.119690
AAS_TestSplitPlane: tried face plane as splitter
2459
267 face merges
7291AAS_CheckArea: area 48010 face 112249 is flipped
AAS_CheckArea: area 48010 center is -433.105408 405.144104 236.008392
area 48010 face 112249 flipped: front area 48010, back area 0
18558AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.082306 404.500549 234.769592
area 59277 face 112249 flipped: front area 59277, back area 0
20637
41703 nodes pruned
17623 areas checked for shared face flipping
281Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
3660Warning: MergeWindings: front to back found twice
8003Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
8004Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
8005Warning: MergeWindings: front to back found twice
17976AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.088806 404.543488 235.043701
17977AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.096161 404.568329 235.302505
17978AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.104523 404.637115 235.125824
17979AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.103729 404.760864 235.235733
17980AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.090820 404.677338 235.539261
17981AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.086884 404.716705 236.020126
17982AAS_CheckArea: area 59277 face 112249 is flipped
AAS_CheckArea: area 59277 center is -433.074402 404.526794 236.060989
27484Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
31047AAS_CheckArea: area 54757 face 188962 is flipped
AAS_CheckArea: area 54757 center is 2168.070313 132.122467 63.229897
34801Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
37103Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
46812
53Warning: MergeWindings: front to back found twice
110MergeWindings: degenerate edge on winding 0.000000 0.000000 0.000000
192Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
Warning: MergeWindings: front to back found twice
230MergeWindings: degenerate edge on winding 0.000000 0.000000 0.000000
309MergeWindings: degenerate edge on winding 0.000000 0.000000 0.000000
482
0 ladder subdivisions
67085 edges melted
17623 areas provided with settings
allocated 23 MB and 622 KB and 220 bytes of AAS memory
17584 areas stored

AAS created in 1238 seconds
loading collision map...
CM_LoadMap( magnus_ctfptm7.bsp, 0 )
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
WARNING: CM_AddFacetBevels... invalid bevel
88 weapon jump areas
calculating reachability...
100.0
please wait while storing reachability...
found a trigger_push with velocity 73.001404 -73.001404 771.492065
found a trigger_push with velocity -73.001404 73.001404 771.492065
found a trigger_push with velocity -2311.232422 -0.001628 395.979248
found a trigger_push with velocity 50.977036 -50.977036 690.507080
found a trigger_push with velocity 500.472595 131.192902 724.432739
found a trigger_push with velocity -66.232727 66.232727 744.043030
found a trigger_push with velocity 0.000000 123.003647 744.043030
found a trigger_push with velocity -113.541824 0.000000 744.043030
found a trigger_push with velocity 0.000000 -86.366112 978.161499
found a trigger_push with velocity -500.472595 -131.192902 724.432739
found a trigger_push with velocity 2311.232422 0.001628 395.979248
found a trigger_push with velocity -50.977036 50.977036 690.507080
found a trigger_push with velocity 85.156372 -85.156372 744.043030
found a trigger_push with velocity 0.000000 -141.927277 744.043030
found a trigger_push with velocity 113.541824 0.000000 744.043030
found a trigger_push with velocity 0.000000 86.078697 981.427551
found a trigger_push with velocity -128.710953 425.733246 711.055542
found a trigger_push with velocity 128.710953 -425.733246 711.055542
trigger_teleport model = "*3"
trigger_multiple model = "*7"
trigger_teleport model = "*13"
trigger_teleport model = "*14"
trigger_teleport model = "*15"
trigger_multiple model = "*17"
trigger_teleport model = "*18"
trigger_teleport model = "*20"
calculating clusters...
0 forced portal areas
88 possible portal areas
78 removed portal areas
11 portals created
12 clusters created
cluster 1 has 4215 reachability areas
cluster 2 has 301 reachability areas
cluster 3 has 6 reachability areas
cluster 4 has 3 reachability areas
cluster 5 has 2 reachability areas
cluster 6 has 2 reachability areas
cluster 7 has 2 reachability areas
cluster 8 has 2 reachability areas
cluster 9 has 3 reachability areas
cluster 10 has 6 reachability areas
cluster 11 has 309 reachability areas
4851 total reachability areas
54017322 AAS memory/CPU usage (the lower the better)
writing magnus_ctfptm7.aas
numvertexes = 73448
numplanes = 27160
numedges = 164658
edgeindexsize = 447243
numfaces = 96056
faceindexsize = 165819
numareas = 17585
numareasettings = 17585
reachabilitysize = 23852
numnodes = 67013
numportals = 11
portalindexsize = 20
numclusters = 12
12018 walk
0 crouch
63 barrier jump
2755 jump
0 ladder
2830 walk off ledge
5542 swim
255 water jump
16 teleport
0 elevator
305 rocket jump
0 bfg jump
0 grapple hook
0 double jump
0 ramp jump
0 strafe jump
67 jump pad
0 func bob
freed 23 MB and 622 KB and 220 bytes of AAS memory
BSPC run time is 1312 seconds
Closed log bspc.log

i'll put up a (hopefully!) optimized version in about an hour or so.

Awesome!

If you run into any trouble perhaps my bspc log will help point out some of the issues.

[wattro]

Sorry didn't mean to cause a misunderstanding. I was under the impression that brushes start out as structural and it is up to the mapper to decide to make some or all detail. My bad.

I never actually "made" any brushes structural; I just made many of them detail and left the rest as they were. So I guess in that case I never turned any of the brushes structural...lol. So any brushes that are not turned into "Detail Brushes" or Structural Brushes" would be refered to as what? "Normal Brushes"? Or just "Brushes"?

"caulk hull"... you mean a big box around the map made of caulk or the actuall outer walls of the map it's self that create a barrier between the space inside the map the the big caulk box hull containing the entire map?
Because I have the big caulk box. So I assume you mean the walls and floor and celing and such of the map it's self.

afaik: i'm a bit out-of-date, but let's see how i do (and without pictures - but check the map file):

the big box around the map is the simplest of caulk hulls (should have said "structural hulls") that you can get. all brushes are structural by default. they can be either structural or detail. when brushes are structural, they will cause a bsp split. detail brushes will not.

basically, if everything inside of the big box is detail, then you will end up with zero bsp splits (i don't think that is entirely true - i think you will get splits starting at 0,0 and then at every 1024 units from there). the problem with this is that it makes for long compile times (esp. in regards to lighting) and can cause performance issues (eg: you might draw the entire level from anywhere else inside the level).

there are a few methods for building the structural hull of a level. the 1st and easiest, is the big box. a 2nd is to build everything detail and insert structural caulk brushes in order to encompass each of the halls/rooms/atria. a 3rd (the way i did) is to turn the outer walls into structural. a 4th way is to build the basic structural hull first, and then build the details. i prefer the 2nd and 4th as it tends to provide the best control over splits and i usually end up with planar splits. the 3rd method is ok, but often you will run into non-planar splits which isn't generally efficient.

anyways, the reason you use detail brushes is to limit the bsp from splitting unnecessarily or from creating inefficient cuts. once you have your structural hull, it becomes pretty easy to place hints and cluster portals as needed.

/wow that was terribly rusty of me...

i'll put up a (hopefully!) optimized version in about an hour or so.

optimized version

-map has a quick and dirty structural hull with hints thrown in (ps: 8 unit thick walls are hard to work with! 32 or 64, please)
-bsp compiles in about 150 seconds with lighting (i had to throw in ambient 10, which is still too dark - but at least you can see the lighting now)
-cluster portals are set up; aas compiles in about 1200 seconds
-no more bot stutter or memory allocation issues
-previously the largest cluster had 4000+ reachability areas, now the largest cluster has just under 900 reachability areas.

[Magnus]

Sweet. That for the info. Always nice to learn something and get better.

I understood that part of the point in detail was to clean up planes and thus areas and splits and such. i was also under the understanding that this is also what Hint brushes were for. Helping to encompas areas that had lots of brush faces especially if many of the brush faces are not on the same plane. Also I got the idea that detail and hint brushes help with fixing the max map brushed error you can get if you get it and help speed up compile times on bspc. I recently read read up on some info that said that over use of Hint brushes can cause more problems than it is worth making compiles more complicated and take more time.

See I am at a point in my mapping that I am really good at it on an atristic or invisioning level, but have a ways to go on the tech level and understanding all the things that a mapper should think about before even getting started.

Lucky for me I learn very fast as long as I can find a good and accurate sorce to learn from.

Anyway. Downloading it now and will play a bit after dinner.

[wattro]

yeah there is certainly no need to go overkill. it's pretty hard to cause performance issues on today's machines using quake 3, as long as you have a good idea of what you are doing.

i think plan b and djbob have the best tutorials on hinting/structural/bsp and check out cardigan's stuff on clusterportals/bots (i like the way you used the 'botdonotenter' - very nice - i wonder if it would be even more improvement if the non-essential sides of those brushes were 'nodraw'?).

hints tell the compiler where to make splits and work in conjuction with structural brushes. i don't remember the exact error message, but something to the effect of max_number_sides or something like that - meaning (i think) an overly complex volume. for instance if you turn all the little brick things floating in the lava to structural, you'll probably get this error.

[Magnus]

Wow nice. The dimness and lava light gives it an awesome ambiant.

Play is big fun without that studder as well.

I will check out those tuts you mentioned.

I could still stand to toy with bot use of the map. I think it would be cool to place a few torch/wall lamps and play with item placement and possibly rework the regen room, but other than that I really like it. WOOT!
With the advent of the above mentioned work I would actually be proud to call this an alpha, but that is up to the entire team as well and there may be things the rest of you might think of that needs to be done before we take that step.
What more can I say. As far as I am concerned we are really close to an alpha. We have some very nice versions now.

[obsidian]

Don't you people ever read all those sticky threads that we work so hard on putting together? Especially the archive thread, every good tutorial ever written on this forum is linked.

I wrote a hinting guide that also links directly to all relevant "caulk-hull" building method tutorials for all the info you will ever need. Read it here:

http://www.quake3world.com/forum/viewtopic.php?t=3620

At the top, check out the links. SPoG and Digibob's write ups are a little technical and on the theory side of things. The best practical guide was written by retired member Quakin' (AKA Q) which goes through the "caulk-hull" method. I learned much of what I know on part of what Quakin' introduced (hence my posts back then were somewhat noobish). RIP old friend (he's not actually dead, just figuratively in a level designer sense).

My tutorial sort of takes over from where Quakin' left off and resides more on optimizing hints and portals for a map properly prepared using caulk-hull.

Happy reading.

[Silicone_Milk]

I remember that thread obsidian. Excellent stuff.

Im feeling in a programming mood but I think I'll take a shot at this map and do some cleanup on it and throw in a few of my ideas this weekend.

[wattro]

Don't you people ever read all those sticky threads that we work so hard on putting together?

sorry, forgot about your's and quakin's tutorials. they are also

Im feeling in a programming mood but I think I'll take a shot at this map and do some cleanup on it and throw in a few of my ideas this weekend.

that would be

[Silicone_Milk]

I opened the map file up and a few things come to mind:

The walls are extremely thin.
The lava tunnel is a bunch of really skinny brushes... why weren't patches (in their cylindrical glory) used instead?

It also brought a smile to my face to see the different styles that were apparent in the method of stair building =)

Im removing a couple rooms, adding one or two in, and remaking the base.

Also, I'm going to garnish some Q2 love on this and have some industrial structure jutting out of the rock in halls and such ("supporting" the rock)

It's going to be a long day tomorrow.

EDIT: After going through the map for a test run I thought it would be neat to have a little cave area here...
Maybe stick a weapon or something down there and have it be an area people can jump to as a last resort to escape. Just a thought. I wont be adding it but if anybody likes that idea you should

[maz0r]

[...]
it would be good if we could get some of the ctf veterans around here to compare the versions posted. that would be a great guide for the next steps. any takers (maz0r, silicone, dichtfux, obsidian, and ix-ir - i'm looking at you!)?
[...]

Sorry, but I won't be able to check the current versions until next weekend. I was ill the past days (got the flu ) and will go on vacations tomorrow. So no time for mapping Good luck with your next efforts! I will contribute to this thread again when I'm back.

[Magnus]

Ok I reworked a13n's regen room to get rid of the areas that wee just not being used. It still feels a bit bla if you know what I mean. I just had a hard time coming up with something nicer for the back of the room. Not feeling too creative at the moment.

I adjusted the weights on a few botroams to more evenly split what path they take from midfeild to the opposing base.
I also done some botdonotenter work and botclip work to help guide the bots to better decision making and to help them avoid lava deaths.
Wattro done quite a bit of hint/portal work and things are a lot smoother.

BTW I tried this map in CPMA and it was not pretty. The way the bots straf jump and bunny hop all over the place they can cover getting from one base to the other in like 20 ceconds or less, but they do a LOT of falling into the lava pits in the midfeild and a LOT of cratering from bouncing around the flag room without getting any health/armor pickups.
They also seemd to have a few spots that they would get stuck in the middle of a path and a few of them would just gather in that spot running in place.
The seemed to use the map great in vanilla Q3A, Threewave, Q3:TA though.
So not sure what the issue is when playing in CPMA.

Anyway here is my latest version. The .map is in the .zip so those of you that want to use it as a starting point to add, remove or adjust anything.
Please read the magnus_ctfptm7.txt to see things that need to be dealt with and things that have been done that are not mentioned above.
Give it a play and let me know that you think so far and any ideas or feedback you may have.

magnus_ctfptm7

@Silicone_Milk:
I though about a cave entrance that you could drop down to as well but opening from the middle of the lava pit behind the MH pillar. So to that idea. Although I just hope it doesn't make things a bit overly complicated for a CTF.
I have an idea involving the midfeild lava pit cave idea and a LP. Possibly connecting where you are looking at placeing a cave entrance to where I was thinking of placing a cave entrance and useing a LP to get out of that area. I will get back with that later.
You will want to have a look at this version of the map as well as the other members latest versions before you get started because of the changes made to the midfeild MH bridge.

@obsidian:
All I can say is Oops. I should have thought to look into the stickies. Seems like a few of the links are dead and I am still getting used to finding my way to the archives and looking up good stuff there too.
Guess I just dorked out on that one.

@maz0r:
Sorry to hear that man. Hope you get well quick enough to really enjoy your vacation.

@all:
I am a bit concerned on who is still on the team as we have not heard from or had any real contributions/involvement from them in a while.
If you are on the team and have not checked in with the rest on the team in this thread in like the last 5 or 6 days then take a moment to review the latest versions or submit some work. At least stop in and let the rest of us know what you are up to so we understand your absence, but it better be a good excuse..hehe.
We have a few people that are really interested in being involved in the first Q3W CTF PTM event and it would be a shame for them to miss out because we thought we had a full team, but some folks had other things that needed their attention more or whatever.

Thanks.

[Silicone_Milk]

Im glad you told me to check out the other versions Magnus.
I'm in love with a room in wattro's version of the map. Has what I assume iron bars in the window and is very much non-organic in construction. I will be using this in my version =)

[obsidian]

I was just taunting you guys about the stickies, but yeah... some good reading in there.

You guys may want to start numbering your version numbers correctly since I'm really getting confused as to which one is the latest version. If you post a new version, ++1, please.

Primarily, I'm interested in condensing the map. A lot. It's still too large and confusing in all versions.

Interested in seeing what Silicone_Milk will come up with. I'm going through some of the others and see what I can do as well.

[wattro]

@Obsidian:
Taunt successful! I'll start with the versioning from here on out. The changes are often numerous and/or wholesale to be specific and I'm pretty chaos minded about how I operate around the map and change/fix stuff, which makes it difficult to track what I've done.

@Silicone:
Thank you! Always glad to hear comments like that, but dammit, i don't know which room you refer to! Actually, I believe you refer to the room with the 2 curved bridges (with the proto_grate3 texture)? Also, I like the idea of some industrial. The map could really use some proof-of-concept towards the theme & detail in the map. Can't wait to see what you got.

@Magnus:
Will check it out later in the evening once football is over. I'm not sure what has changed with CPMA bots, but interesting to note/look out for.

@All:
I've got a new version in progress. It continues to be a brainstorm, but the gist: modified Magnus' interpretation of Obsidian's midfield, killed a couple of the interconnecting paths between the different routes to/from each base, and reduced the overall distance of the map by 1024 units. I should have it up in a few hours. Hopefully it's a continued improvement. I also temporarily abandoned the sock-rock-wall that I was experimenting on. It turned out great, but the map isn't ready for it yet.

@maz0r:
Nothing like a good vacation to get over the flu!

[obsidian]

Eventually, you guys may want to read up on Q3Map2 alphaMod blending. This map is ideal for this kind of stuff. Just something for way later.

http://www.simonoc.com/pages/articles/terrain1_1.htm

[wattro]

way ahead of you

the big difficulty is the the rock-wall textures will likely be on x y and z planes and getting a texture projected and blended nicely to all 3 planes isn't the easiest thing in the world. definitely something to consider as it could be a somewhat painful process.

i also happened upon saberpeak which has a solution for this stuff, complementary to sock's ace potm tutorials.

ps: i'm hoping you can help with the technical aspects of this when we get there - i've never actually put alpha mod blending into practice but the tutorials make sense.

[a13n]

and will go on vacations tomorrow

Have a nice vacation!
But watch out for "terrorists".

@Magnus
I'm still alive.
Just got lost in the current maze.

[Magnus]

@a13n:
This was not really intended for you. Most of you have been popping in at least every two or three days with contributions and feedback.
The team started out with 4 of us and has grown to 6 with obsidian and wattro because they were really there helping all along anyway. There are obviosly others that would like to get in on this PTM event and I don't want the team size to get out of hand. So in order to make sure there is room for everyone we should know who is still onboard and who can afford the time to stay with this project.
maz0r and Silicone_Milk have actually contributed mapping and feedback and weren't even on the offical team. If they want it they deserve a spot IMO.
Unless everyone thinks it would be reasonable to have a much larger team than usual on a CTF PTM then we need to make some what may seem like harsh decisions and make some cuts.
I started this thread and this CTF PTM event to have fun and as long as the result is a awesome map that plays great, has good bot support and gets played a lot then I don't really care about team size.
Any opinions?

@obsidian:
Good point. How about we use lettering for versions with minor changes (i.e. ctfptm2b woul show a minor adjustment or change) and use numbering for versions with major changes (i.e ctfptm2b changed to ctfptm3 would show a major change has been made)
Or is that just getting too complicated with it and stick with the traditional (mapper_ctfptm1.1, mapper_ctfptm1.2 and so on) for each time we make some changes and post the changed version?
I think the latter would be best and should be posted with a description in the post of what changes were made.

Anyway I gotta go to muh jorb and do some work.

Later

[obsidian]

the big difficulty is the the rock-wall textures will likely be on x y and z planes and getting a texture projected and blended nicely to all 3 planes isn't the easiest thing in the world. definitely something to consider as it could be a somewhat painful process.

ps: i'm hoping you can help with the technical aspects of this when we get there - i've never actually put alpha mod blending into practice but the tutorials make sense.

There are actually a few alphaMod functions that will be useful here:

alphaMod dotproduct multiplies the surface normal vector with a user specified vector (typically the z axis) to yield a normalized value for vertex alpha. So this creates an automatic blending depending on the "steepness" of the surface. The texture does not necessarily have to be projected (tcGen ivector) so rock walls can retain their normal texture coordinates. This is a good place to start since it's automatic and requires little work.

alphaMod volume work like little flags that you place over vertex points to manually specify the vertex alpha value. This will probably be used for any fine tweaking of the dotproduct terrain as well as for any blends along walls, etc.

Lastly, I'm guessing that some well used decals and projected decals will come in handy here to really create some of those nice details like cracks, grunge, mossy bits, etc.


@Magnus, you can use whatever system you want. So long as people can tell what is the latest version.

[Silicone_Milk]

Well, I spent all day working on my version of the map. The midfield from Magnus' version and the "curved-bridge" room from Wattro's version of the map are pretty much the only original content that had been kept. Im still unhappy with the way the flow is on my version though so Im rebuilding the connecting areas between the NEW base design.

I will be playing with several versions of my base before I give you guys the map to look at. Currently the design of it actually makes you backtrack so Im going to do some gutting.

When you finally do see the .map file you'll also notice some whacky geometry from my t-junction phobia.

Ill be posting some shots of the current progress today.

1st Rough Draft of Base:



In between the faces Im going to slap in a long skinny window with greenish-yellow grungy glass that lets you see in to the tank and then Im filling it about halfway with lava and building an outdoor area with lava dripping in to the tank from a mystery source over the edge of the tank just to show some sky


EDIT: gray grid is my caulk texture. Dont worry I wont send it to you guys looking like that.
Also, Im redoing the base so its not symetrical because the rest of the level isnt. It just makes for an odd flow if the base is suddenly symmetric while everything else seems to be less planned/mirrored.