3×10: Competition 2: DefendVille

15th November 2013

Single Player First Person Shooter Maps and Mods for Half-Life 1, 2 and Episodes 1, 2 and 3

For this competition, you must create a scenario where you have to defend your outpost from a Combine attack.

The player should have approximately 10 minutes to prepare for the defence, including moving turrets, or other obstacles.

Ideally, the design would allow for players to try slightly different strategies on replays.

There are no limits to map(s) size or playtime. You can set the map in any environment that you want.

I would very much like you to start the map with this Radio Transmission. It can be from a radio or just played at the beginning of the map. The quality is lower than normal due to my having to use an older PC and I will try to supply a better quality one soon. If you prefer, you can have the content re-recorded by another voice actor if you prefer. You MUST NOT change the text though.

Judging Criteria

The winning map(s) will be the one that sets up the defence situation in the best way, paying attention to the atmosphere, the replayability and the general gameplay. If a map designes the outpost beautifully but is then boring to play, it won’t win.

Attention to detail is very important in this competition. I would much prefer smaller but highly polished maps than larger, less polished maps.

Ideally, what I would like to see is giving the player some defensive choices to allow them to have some control.

Beta Testing

Please contact Ade, the leader of The Beta Testers Collective, a free map beta testing project, to have your map tested. I fully understand that 10 days is very short but even one playthrough by somebody new is enough to highlight issues. She can be contacted here: [email protected].


Sponsor and Prizes

This competition is sponsored by SteamContests.Com who are kindly donating one $5 Steam game and I will also be adding another $5 Steam game. At this point I don’t know if the winner will be able to chose their games or if it will be a random selection.

Mapping Updates.

Please feel free to post updates to any work in progress maps. Add a comment and link to any images you want to.

Special Rules

Please ensure that your map works on Steampipe.

The moment you decide you will enter, look at the time and use the last three digits as your outpost name and filename. For example, the time as I write this is 16:58. Therefore, my entry would be Outpost 658 and my filename would be 658_pp_3x10.bsp. In the unlikely event there are duplicates, I will contact one of the entrants.


For more information on Steampipe, please see: Steampipe and the Current Issues, this FacePunch Mapping Thread and this Steam Community thread. You could also try a web search for specific issues.

In the 3×10 announcement, The_Blazer wrote “Specifically, the Source SDK tool is now entirely disabled, and Hammer needs to be launched from Steam > steamapps > common > Half-Life 2 > bin. It


  1. Christiaan Bakker (AniCator)

    Get ready to rumble!

    1. Christiaan Bakker (AniCator)

      Haven’t really got the time right now to give you all an elaborate post about my current progress.
      During day one I’ve mainly been doing things on paper.
      Today I’ve been experimenting a bunch. I streamed it as well on http://www.twitch.tv/anicator and I will probably stream more of it so if you’re interested you can follow that channel as well.

      I ended up scratching a bunch of ideas and coming up with a set of new ones as well. Shows you how much a project can evolve in its early stages. Haha.

      I created a few concept levels but I’ve started to settle on one of the initial designs. Here’s a few screenshots:

      This is all far from finished of course and who knows how unfinished it will be by the end of this competition. Ten days is a short amount of time and this is my first time participating in the PP competitions. Loving it so far though. Looking forward to seeing the updates and maps of others as well!

      1. Christiaan Bakker (AniCator)

        Question for Phillip:
        If we wish to use custom soundscripts and maybe even have closedcaption and VCD data, would it be possible to have those in the final product?
        For VCDs you would of course have to rebuild the scenes.image file which isn’t too problematic but I was wondering if there were any limitations on stuff like this.

        Of course I can pack part of this into the BSP but that would mostly be for data files. There are a bunch of files that aren’t used from inside a BSP as far as I know.

        1. To be honest, I can’t answer your question knowledgeably. I am not a modder, so don’t know how easy the things you ask are to do. If it is a case of putting files in the final product, then YES, but if I have to open Hammer or some other editing application then NO.

          I will do everything I can to help but know that I am very technically limited, both by my ability and my current PC, which is nearly ten years old.

          If other modders or you are willing to help then I will definitely try.

          1. JG

            To my knowledge, most contest entries in the past have avoided VCD because Phillip would have to rebuild scenes.image on his end when he makes the mod. I don’t believe scenes.image is something you can pack into the BSP because it is “global” data that applies to the entire mod.

            I know that as much as I’ve wanted to use Faceposer for some quick scenes, it just doesn’t work for me at all. The scenes.image is one obstacle – the other is that the program is completely screwed up following the Steampipe “upgrade.”

            As a potential workaround, you can use existing VCD files, for what that’s worth. You have to know their file names and manually enter them, because Hammer’s browser doesn’t find most of them. Most of Valve’s VCDs aren’t terribly generic, however, so you’ll probably need to experiment with a few to find what might work.

        2. JG

          Hey, AniCator,

          On a lark, do you actually have a Faceposer that works? For several months now, I’ve been able to load Faceposer, but I get completely textureless models that are oriented wrong. I haven’t been able to find out much about this issue – some people say it is specific to Vista and video cards, other say everyone has it.

          I’ve tried running Faceposer in the /bin folder of HL2, as well as the one in the Source SDK 2013 folder with the same result.

          I know Gambini logged a bug about something like this in the Valve Github, but it would appear they’ve gone back to giving us the silent treatment again.

    2. Christiaan Bakker (AniCator)

      Day Three | Progress Update

      I’ve been working on adding more buildings to the outpost/fort/complex, a blocked out hillside and I started working on adding back hostile AI units.

      Here’s a few screenshots:

      1. Christiaan, this looks like it’s going to be a lot of fun! I love your layout so far. Can’t wait to play it.

        Question for you:
        How are you getting that HDR effect? whenever I run my map it’s just the bland basic environment lighting (not the full bright, but not like yours either). It’s not a huge concern, but I’d like the map to look as nice as possible. Thanks!

        1. Christiaan Bakker (AniCator)

          For the exterior lighting I’m only using a light_environment.
          I do my interior lighting mainly with light_spot entities.

          I haven’t take an extensive look at your map but it seems like you’ve added a light_environment entity to your map without changing its settings. Which would mean it is pointed in a flat horizontal direction.

          Maybe you should take a look at the sky list on the Valve Developer Wiki.

          1. Thank you, Christiaan. I set the map to render HDR and it seems to work alright. I just need to add more lights. 🙂

            1. JG

              One thing to remember –

              When you build your map with HDR, the system actually creates two maps – one for non-HDR (bloom) and one for HDR. Both maps are packed into the same BSP file.

              When it comes time to build cubemaps, you need to make sure you do so on both maps using some console commands to swap between them. There’s a good section on the Valve Developer Community that talks about that.

              1. Thanks, JG. I’ll put that on my to do list. I’m excited to get the meat and potatoes out of the way and move on to the fancy graphics stuff.

                Not that scripting soldier movement is the worst thing in the world or anything. 😛

    3. Christiaan Bakker (AniCator)

      Day Four
      Today I only spent roughly 30 minutes on the map because I had to finish some other work first.

      I’ve worked on some dropship related errands and fixed some small issues that bothered me. Also started experimenting with some ideas I’ve wanted to start working on for a while now.

      Thought a lot about how I can make things more dynamic but it is on the other end also important to make sure the player can prepare properly for the attack instead of running into a surprise around every corner.

    4. Christiaan Bakker (AniCator)

      After not being able to work on the map for two days I’ve finally made some progress again.

      Started dressing up the level a bit with some non-dev colours and worked on some more I/O and spawning related things.


      My biggest worry is still the big picture. I’ve got some elements in place but this challenge is pretty difficult to get right.
      Making sure the player has meaningful defenses to setup during the ten minute pre-battle phase is actually really hard when you’re trying to keep it varied.
      Coordinating the events of the attack itself is also pretty challenging.
      I hope I make it to the end. Just a few more days to go.

      1. Looking good, Christiaan. Can’t wait to play around in it.

  2. Oh I can’t wait to see the results of this! 😀

    Aside: the judging criteria? Same as the ambushville one, or maybe needing to be tweaked a bit in words?

    I do so wish that I was any good at mapping. Not that I have any ideas for this specific one, but still it would be fun to eventually participate. >_> Me and Hammer, though, we have an understanding. I stay away from clicking on it to start it, and it doesn’t humble me with its complexity. 😀

  3. Defense… mmph, never had any creative ideas about a defensive scenario. Still worth giving it a shot though, I may come up with something eventually.

    Also, my AmbushVille map kinda sucked, so I should definitely avoid rushing things this time.

    If I find some inspiration and manage to sketch something on paper, I’ll start posting daily updates.

    I wish the best of luck for all other mappers entering the competition!

    1. Christiaan Bakker (AniCator)

      This is my first time participating. Where does one usually post these updates?

      1. There’s no official place to post them, but during AmbushVille the mappers who decided to post updates did it in the comments.

        If you’re participating, you should probably post updates in this comment thread.

        1. Christiaan Bakker (AniCator)

          Ah ok. Thanks. I’ll use my existing comment if I decided to post any updates. Not sure if I will though. :p

          1. You can’t use your existing comment. You will have to create a new one, which is fine.

  4. It’s a challenging theme, but I’m giving it a shot.

  5. Sound interesting, I was just thinking about a defence theme actually! I like the idea of tying the maps together using the transmission, but it does limit the theme a little bit. Is it okay to tell the player about the attack a different way? Similarly, can we still name our maps however we want to? I wouldn’t mind working “Outpost ###” into a larger title though.
    I have some free time now, so I will try to post map updates if I remember.

    1. I would prefer you told them via the transmission, but I am open to hearing your alternative. Before I decided, can you email the details and I’ll give you a final decision?

      1. No problem. I’m still considering different options and might end up having the transmission anyway, but I’ll email you once I’ve set on a course.

        1. I would prefer you email me before that, in case I say no. Actually, I’ll email you now.

  6. So, let’s start the “mapping diary” with Day 1.

    Today I did only paper sketches and idea gathering. For some reason I don’t have any particular scenario in mind that I like, so I surfed the internet for a good hour attempting to find something inspiring, unfortunately, no results.

    However, I came up with 3 different ideas and by tomorrow I should have picked one to work on.

    The first one is doable, visually interesting, but may result in poor gameplay because the setting is hard to adapt to game needs, and has very limited tactical choices. In few words, a campfest for you and a zergfest for the Combine.

    The second one may provide interesting gamepaly and cool visuals (as cool as my not-so-pro-skills allow, that is), but is very abstract and most likely impossible to place in a realistic environment.

    The third one could provide good varied gameplay, possibility for scripted events, and can easily be adapted to suit what I manage to finish in time for the deadline without breaking if I leave something out (which is very likely). However, the initial build of all the areas will be hard to do because of how complex it may turn out to be and may drag on for several days. It also uses a visually unappealing and very generic theme which would need to be spiced up by something cool – no idea what it could be, though.

    1. Now you are just teasing us with the different options.

    2. Day 2 Summary.

      Not much to say, really.

      I chose one of my initial ideas, and designed most of the map on paper. I also wrote down several ideas about Combine strategy, deploying, and use of special equipment. If they work, expect to see some new Combine technology employed on the battlefield (Well, not really new – more like modified versions of the stuff they already have).

      I haven’t opened the editor yet, which I’m kind of ashamed of, but designing a defensive scenario is proving to be harder than I expected. Tomorrow, however, I should start building the whole thing. I’ll start with terrain first, then I’ll add all the buildings and stuff.

      When designing the level on paper, I also noticed that keeping the player inside the game area without abusing combine forcefields (which sometimes don’t even make sense, for example, if they’re placed in areas that should belong to you) is very hard. I’d hate having to put an invisible wall triggering a super-cheesy “You can’t leave now, fight the enemy!” message on-screen when the player touches it. I’ll have to be creative in this respect.

      1. JG

        Agreed – one of the challenges with the Combine in a waved battle is that they always need a logical spawn point that the player can’t abuse. This is why Antlions have an edge in waved battles, such as the penultimate mine scene in Episode 2 – they can spawn out of the ground and unless it’s solid concrete, no one will complain.

        Unfortunately, the Combine don’t have too many justifications beyond the many flavors of forcefield, which Valve itself used like crazy. One thing I would suggest to you, though, is looking into the jump down system. Any enemy, including the Combine Soldiers and Hunters, can jump down a fairly significant distance given the right hint nodes. In that way, it’s easy to make a way for them to enter an arena that a reasonable player can’t get to.

        Because otherwise, yes, Combine have a twin problem in waved battles. They need a logical way to get in, but they don’t survive long once they are there. That’s why using things like Combine Dropships is good for effect, but kind of worthless in practice.

    3. End of day 3:

      For some reason finding motivation to work on this map is being much harder than it was for the AmbushVille competition. Dah, anyway.

      The paper gameplay is finished, now I just need to build everything in Hammer. I’m still uncertain whether there will be an object or “capture zone” to defend or if the goal will just be staying alive, although it will most likely be the latter.

      Today I started building the main area. I’m planning to have lots of things to help the player stay alive and offer varied tactical choiches, but that will slow me down in the mapping process. Because of this, I have devised a “fallback system” that should allow me to cut out parts of the map, in case I don’t reach the deadline with everthing 100% done, without affecting gameplay quality too much. Hopefully. Maybe. Or maybe not.

      I’m going to build from the main area and expand outwards. This will allow me to have at least a small area ready for playtesting (or more precisely, Combine Soldier testing) by tomorrow evening, and to tweak aspects of gameplay and other areas based on that.

    4. Late Day 4 summary.

      Day 4 (yestarday) was a problematic day for me in real life, so I had very little time to work on the map. Still, I have built the main features of the main area, and started wall design. That’s right, wall design. The walls are very large so finding proper details for them is a pain. Beams and trusses have already been added, pipes would make little sense, and there are only a few decals and overlays that fit the scale of those walls.

      Today I have decided to move ahead with the map and not waste too much time on the walls.

    5. Day 5+6 summary.

      I’m putting these two together because yesterday I completely forgot about writing the updates.

      The main area is pretty much finished. It should alredy bring solid gameplay on its own but it isn’t very big, and I really want the Combine to be able to push forward from the edge of the map towards the main area and claim territory if the player doesn’t actively keep them at bay. From tomorrow, I am going to expand outwards and build the capturable terrioty, then I’ll add spawners, nodegraph, and all that stuff.

      Right now, in the main area, here is not much cover and it could probably benefit from a central pillar or other large obstacle, but I’m not sure how to add something like that in the environment and still get it to make sense. I have chosen to add cover after playtesting because I don’t want to overdo it. If the player can hide behind stuff and block LOS from all sides, all the time, forcing soldiers to come at him at close range where he can slaughter them with his shotgun, there is no challange and no fun.

      The main area took awfully long to build. If I want this map to get anywhere, I’ll have to be faster when designing the outskirts of the map.

      1. Blazer, the Combine advancing sounds awesome. Are you using assault points for that?

        Can’t wait to play your map. I enjoyed ‘struggle” even though it has super duper hard (I’m bad at Half Life though). I’m excited to see how you’ve built this new area.

    6. Day 7 update.

      Pretty much no advancement in the gameplay or area design side of things, I mainly focused on lighting instead. It took quite a lot of time to tweak HDR settings for the skybox I’m using. Lighting for the main area also proved to be a difficult task because I’m using strong, focused spotlights to highlight certain important parts of the place, but I’m also using similar floodlights for the sake of proper visibility and realism to light up the rest of the area.

      Both focused spotlights and floodlights need to have a warm color, since that is going to strongly constrast with the dark world outside dominated by night and Combine cool, blue lights. I want to give the feeling that you’re under siege, that the darkness is pressing from all sides and that you must defend your little stronghold of warm light with teeth and nails.

      Unfortunately, this means that inside your “warm area” it is quite hard to create visual contrast to attract the player where needed.

      Lighting is decent now but I’ll definitely need to give it a second pass later on if I have time.

      Also, the whole Combine advancing thing may need to get scrapped. The deadline is quickly approaching and while the main area is almost done, I’m not sure I’ll be able to properly create and finish the outer territory.

    7. Day 8 summary.

      Today, I did some progress in designing the outer area. Seeing how long I’m taking, I’m only going to have a day or 2 to add all the NPC stuff, so the more complex territory claiming mechanic (which is not that elaborate or cool as it sounds, the Combine can simply prevent access to certain areas of the map using mobile force fields if you don’t stop them. They should also be able to teleport on claimed or partly claimed territory using mobile teleportation units.) may eventually get cut.

      Most buildings are finalized. The main area may still need some tweaking, but I don’t want to fix what isn’t broken so I’ll wait until I can create a proper large-scale battle to decide where to add or remove cover. There’s also a roof area that I’d really like to use, but I’m not sure if I’ll be able to create decent gameplay for it in time.

      Lighting has also been tweaked and I now have outdoor lighting that I like, and lighting psycology using yellows and blues is turning out nice.

      I feel like there is going to be a “big crunch” in this weekend for me, assuming I even finish the map. Tomorrow I’ll definitely have to complete the design for all buildings and areas, so on Sunday I can add all the NPCs and maybe custom textures.

      1. Given what you’ve said, I would agree with the decision to cut the more complicated aspects of the map. The territory concept did sound awesome, but maybe we’ll see it next time?

      2. JG

        Yeah, I started NPC stuff about 3 days ago and it’s been a real pain. Defense scenarios usually require more AI setup time than most.

        Actually, I’ve got to say, Hammer itself has been downright nasty this time. Not only is the Flag tab crash back again, but performance completely dies when I’m flying around with the Properties window open. It’s not enjoyable at all and I really wish they’d stop mucking with the tools, because neither issue happened during Endville and – lode and behold – there was a “new” version of Hammer between then and now.

        1. Christiaan Bakker (AniCator)

          I’m using the Source SDK 2013 Singleplayer Hammer and I haven’t had any issues with that.

          I already had a certain amount of AI stuff going on from day one pretty much. I scrapped most of the initial stuff and have currently got like one or two waves working. It’s not all that impressive but I hope to add more tomorrow.

          In the end this challenge is just really difficult to pull off, keep fun and graphically appealing at the same time.

  7. Hi Planet Phillip mappers. This will be the first mapping competition I’ve ever entered. I rolled Outpost 400!

    I’m going to try as hard as possible to meet the deadline, but I’m a total newb so we’ll see. Even if I can get spawns and wave timing working I’ll be happy. I want to try to finish maps for these competitions. Eventually I’ll get them done, and then eventually they’ll be good.

    Today I started by sketching a few concepts and working through what a defense map might need to be any fun. I typed out my thoughts. Please take a look and let me know what you think and if you have anything to add: https://docs.google.com/document/d/19kXJKJhN3Hny5AU4m7_921YMbGkbOqVDr2vbDr_2f44/pub

    The concept for this defense is protecting a power station located in a secluded seaside nook. I thought it would be just the kind of place the resistance might hide. The chapter will be called “Surge Protector.”

    Here is an imgur album showing the sketches and and map in Hammer:

    The initial idea was way to large, so I’ve trimmed it down quite a bit. There were going to be three spawn zones: a cliff soldiers would repel down, a dock area where a drop ship would land, and a driveway where armored cars would pull up. I ended up cutting the driveway completely and focusing on the other two main points of entry.

    The map begins overlooking the area. I’ll want to make it very clear immediately what needs to be defended.

    Once the combine spawn in, they will choose a path to follow into the main arena from their spawn zone. These choked routes are where the player can utilize turrets, mines, and barriers.

    The player can fall back to a secondary set of choke points that will be closer to together and tighter chokes, but more likely to be overrun by a persistent, large group of soldiers.

    Finally, there is a relatively small room leading up to the starting overlooking position that has four doors and a catwalk overhead. It is the last bastion and will be the easiest place to defend, but the player will lose all situational awareness of the outside. This area will be the closest to the ammunition stockpile and have the best cover.

    This still seems like a lot to me, so we’ll see how much more I can cut. Until then, I am going to finish the base bsp in dev textures, set up the assault waves, lay out some weapons, and pass the map off to the Beta Testers Collective. After that we’ll see what kind of changes are in order.

    Can’t wait to see what everyone else is working on!

    1. Welcome to PP. Thanks for the images and design detail. Sticking to a deadline is a very valuable skill and the biggest issue I hear about is mappers being too ambitious. If you can control yourself you are on your way to meeting the deadline.

      Good Luck and keep the updates coming.

    2. Thank you, Phillip!

      As I neared completion of the blocking for my map I decided to have a run around in-game before moving on to spawning. The map was WAY too big, so I have heeded your words to the Nth degree and cut the majority of the map.

      Blocking for the new, smaller map is now finished (though the level is just floating in a skybox void, the basic space is there). The current level consists only of the smaller room with four entrances, a courtyard, and two high ground fallbacks. Here are some images of the blocked level Day 2:


      I’ve found a great tutorial on dropships and spawning (http://www.nickurko.com/source-sdk-dropships.html) that has unlocked some major secrets for me, so the next step is planning out the assault timing and coreography, then placing some basic defensive items.

      If anyone would like to take a stroll though the map as it is now, please feel free to do so:

      There is one drop ship that will spawn, fly around, and deposit three combine soldiers in the courtyard.

      I would love to hear any feedback from anyone. Everyone’s opinion will be valuable. This is turning out to be a very fun learning experience.


      1. Christiaan Bakker (AniCator)

        I’ve played what you’ve got so far and it’s looking very interesting.
        Of course there isn’t much going on yet and I played it using mat_fullbright because the lighting makes it really difficult to see right now.

        What kind of defensive strategies/tools are thinking about implementing?

        1. Thank you, Christiaan.

          As my post below explains, I have cut the map you played completely. However I do intend to apply some of the same defensive strategies to the new, smaller map.

          The basics:
          – Turrets. There will be fewer of these than there are entry points.

          – Barrels and canisters will be scattered around. Some in suggestive positions, others ready to be moved.

          – Hopper Mines. There will be a small reserve of mines. It’s up to the player when to use these. I predict some “too good to use” behavior, though. We’ll see.

          – Lightweight physics props for shootin” at dudes with the Gravity Gun.

          I’d also like to try a few other things, but given the time who knows:
          – One time use physics traps.

          – Weighted containers that can block certain routes on the map. There will only be enough weighted objects to use two or so of these.

          The player will have plenty of room to move around in the map, so they won’t get trapped, though they will be pushed around by the soldier spawns. I have intentionally incorporated a figure eight and circular track in the map.

          I’m hoping to bring a lot of movement to the map, pushing the player to move around and take advantage of different defensive points and strategies depending on the group of enemies they are dealing with.

          1. JG

            Barrels in suggestive positions, huh? Can’t wait. 😉

    3. Well, today has been a day packed with learning and cleansing! Making maps is time consuming, especially for someone so new. I know it is reiterated over and over again, but it’s really true: everything takes ten times longer than you think.

      Anyway, I have *completely scrapped* the old map. Three days in to a ten day competition, this may seem like a horrible decision, but here is how I see it:

      First of all, something even as large as I had would simply not happen in the timeframe. Simple as that.

      Secondly, I started thinking about how defensive scenarios are presented in Half Life 2. With the exception of the final fight in which you protect the base from Striders using the Magnusson device, I am pretty sure every defensive scenario is a single room, and a small one at that.

      So I have crammed everything into one room with multiple access points scattered around. The challenge will have to be in the timing and location of enemy spawns.

      I have finished (again) the block brushwork and some of the scripting for how the soldiers will enter the building. I have completely cut dropships because they were giving me too much trouble, so this map is going to be simple, but that seems like the right way to go at this point.

      Already this competition has been a valuable learning experience, and I know that even if I don’t make the deadline with something respectable I will have had a lot of fun and will be better prepared next time.

      No pictures today, folks. I just want to get back to making these assaults work. I do have a few questions for all of you though:

      – HDR/bloom/etc does not seem to be working on my map. Any ideas why?

      – How do you get soldiers who spawn on foot to move to a point? info_target does not seem to serve that purpose.

      – How do I make enemies aware/unaware of the player?

      I’ll be researching all of these independently, but it would be nice to hear what you all have to say, seeing as how these questions are most likely trivial to you.

      Thanks in advance, and it’s been great seeing what everyone has posted. Good luck continuing on your maps!

      1. JG

        1. Are you building with HDR and have HDR enabled in your game’s options?

        2. The “Target Path Corner” approach on the NPCs themselves has never worked for me, but some have had success with it. What I generally do (assuming you’ve already placed a bunch of info_nodes) is use the aiscripted_schedule entity. Target NPC is the name of the NPC this applies to and Goal Entity is the name of the path_corner you want this NPC to run to. The Schedule to Run should be Run Goal Path and Interruptibility should be set how you see fit (if you want them to ALWAYS get there, use Death).

        If you enemies are being produced by an npc_template_maker, you’ll want to call the aiscripted_schedule using the StartSchedule input OnSpawnNPC. Otherwise, whenever else you’re doing to get the NPC’s attention (like a trigger), do the same thing.

        There are ways to make aiscripted_schedule a bit more flexible. The Target NPC can by a name shared by several NPCs (like “battle_a”) so you don’t need one of these for each NPC. Likewise, the radius allows you to grab all potential NPCs within a radius, so you could apply it to all nearby “battle_a” soldiers instead of everyone named battle_a.

        3. There are lots of ways to make enemies aware of the player. The most direct way is, again, using an aiscripted_schedule, setting the goal entity to “!player”, and setting the Schedule to run to Set enemy to Goal Entity AND Run to Goal Entity. Again, you use the StartSchedule input whenever you want them to do this.

        Of course, that gives the NPCs an omnipresent knowledge of the player, no matter where the player is. Probably the most effective “natural” method is simply assigning all the Combine NPCs in an area to the same “Squad Name.” That way, they will share information about where the player is and work better as a group. This is something you ought to do regardless of your desired awareness.

        1. Thank you so much, JG. My enemies are finally alive. 😀

          At this point I have a complete door breach sequence and enemies who enter with a goal in mind. With this bread and butter element unlocked I can really start to choreograph the map. I’m going to work as hard as possible tomorrow to finish the full assault sequence. Maybe I’ll have time to add some basic visuals as well.

    4. And so it continues…and wow, I have to say setting up AI schedules and linking timers together is only slightly less tedious than animation.

      But this is so amazingly fun I don’t care. Loading up a level after a half day’s work and seeing what works (and what doesn’t) is so rewarding.

      Thanks to this community I have managed to get my soldiers to spawn, move into position, destroy doors, and assault the base strategically. I’m sure the way I’m doing things is roundabout and likely easier/faster some other way, but it works and that’s all that matters to me for this particular competition.

      I’ve already learned more than I could have hoped, and today realized that I will at the very least have something to submit at the end.

      The map still needs a climax, but I’m working through the timing of the first couple of waves right now. Once everything is completely linked up I’ll be sending it off to the BTC. I’m really hoping to get that done tonight so the feedback can start flowing while I work on other stuff, but realistically it will probably be done in the early morning.

      I have some more questions today:

      – My turrets seem to be out of ammunition even though that box is unchecked. They just sit there and don’t even acquire targets as far as I can tell, and every Google search brings up something about the pop-out ground turrets or combine mounted turrets. Any ideas why they don’t work? They are set to resistance friendly.

      – Are the info_nodes really necessary? It seems so strange that everything has to be manually placed like that. It doesn’t take that long, so it’s not a big deal. I’m just wondering.

      _ I’m worried about the custom sound file for the broadcast working on other people’s computers. I’ve already converted it to .wav because it didn’t seem to play otherwise, but how will it play on other people’s machines if it’s stored locally? Is it packed into the BSP?

      Thanks for your friendly advice and help, everyone. Your updates are sounding good. So exciting!

      1. – I still need to put turrets into my map so will reply later for the turrets question after some testing.

        – info_nodes are one of the most important things needed by NPCs. Without them they basically won’t know how to move and will act really stupid. With them, they will be tactical and know the best path to move etc.

        – You should use the original .mp3 file. Hammer doesn’t autoplay .mp3 files but in game it will work. Also, the .mp3 file size will be smaller than the .wav which is why you should use the .mp3

        As for packing the file into the BSP. You need to do this for your map. Use Pakrat or similar. Just plug your map (.BSP) into pakrat and auto/scan it and it will embed the .mp3 into the .BSP which enables it to be used on anyones computer.

      2. JG

        Concerning the turret, you’re talking about the tripod turret you place on the floor, right? That’s an npc_turret_floor.

        For a turret that attacks the player, I have all the checkboxes under Flags disabled. Since yours is friendly to the player, checking Citizen Modified is correct, but you need to tell the turret what it is supposed to fire on.

        There’s an object called ai_relationship which does this. You give it the Subject (npc_turret_floor) and the target we want to alter its relationship with (such as npc_combine_s). Change the Disposition to Hate and make it Start Active. You also most likely want to set Reciprocal to Yes as well, so the Combine Soldiers will see it as a threat and try to knock it over. There’s also a priority number that should be something more than 0.

        You’ll want to repeat that for each class you want to fire on – so if you have Civil Protection units in your battle too, you need another ai_relationship with the Subject set to npc_metropolice.

        Just to confirm this, I looked at the Antlion defense in Episode 2 and they had the same setup with the ai_relationships. So I suspect that’s your problem.

      3. Thank you, George and JG.

        I think everything is working now. 😀

    5. Day 5 is over and all enemies are now spawning and moving the way I want them to when I want them to. There is a lot to be done still in terms of pacing and making objectives/assaults clear to the player. The whole scenario seems to be rushed and has no downtime, so that’s something I want to work on. Furthermore, there is only one decent punchy moment, and I would like there to be more of those.

      Here is one image of the map so far in-game and a bunch of reference:

      This album highlights my focus for the next few days: balance, pacing, and visuals.

      I’ve started texturing and lighting, and now I see what Jim was talking about! Next time my workflow is going to be significantly different. In any case, things are off to a good beginning. I’m really experimenting with the lights and textures, as I don’t quite know what everything does yet. The area is mainly lacking detail, so I’ll probably do a prop_static pass some time soon.

      As mentioned before, the first iteration of choreography has turned out to be too hectic. There’s no focus to the combat, and no transition from one assault to another, leaving the whole scenario to feel like one big flood of dudes. There just isn’t enough time to develop any strategy or do much besides run around frantically avoiding bullets.

      Anyhoo, it’s back to the map now. Thank you everyone for your help so far and I can’t wait to see more updates.

    6. Update Six


      Earlier in the morning I wrote up the choreography changes. They will be better paced and more focused. Part of those changes required the map to allow some more fluid gameplay on the player’s part, so I figured today would be a good day to begin the visual changes this map needs.

      As you can see in the image, the map is much more spacious than before. I did this for two reasons:
      – It was frustrating to be in such a small space no matter where you went on the map. The transparent textures and tiered platforms are my solution to this. The ground space the player can run on is roughly the same, but the environment feels more open.

      – Pinching the Combine into a death trap was way too easy before. The way the map will be laid out when I’m done will provide the Combine with slightly more routes and more visibility of the player throughout the map. I am hoping this will compliment a new feature I just got working. It’s a secret, but you might be able to guess what it is if you look at the image.

      Now I am rebuilding the space and doing a basic pass on textures and layered detail. Once that is finished I’ll be placing props (static, physics, and combat necessary). Then it’s back to the Combine, whose assault tactics will be treated with a major overhaul.

    7. Update 7

      Hey all,

      Here is an album showing the visuals as of now:

      (I’m working on toning down those env_glows.) It’s pretty much just prop placement at this point. I’ve got a very red scheme to make the Combine soldiers and entry points stand out as much as possible. I’ll be lighting important defendable areas as warm.


    8. Update 8

      – Visuals are as done as they’re going to get at the moment. I know many of the mistakes I’ve made, but there is not enough time to go back and fix that kind of thing right now. Lessons learned for next time.

      – I’ve opted for the lower end of 7-12 minutes because this map is so small. Ten minutes is way too long for what I’ve given the player to do. Hopefully seven or eight will be about right.

      – Spent way too much time on specific polish things that I shouldn’t have, like displaying the Chapter text and fading in. It looks nice, though, so there isn’t much to complain about other than the loss of time for scripting. Speaking of which:

      – NPCs are in place and ready to be told what to do. I have a new choreography plan that should feel much better than the hectic dance party that I made last time. Teams will now take cover behind props and attempt to gain higher ground or flank. The timing will be significantly different and each battle will feel like a focused part of the defense. Unfortunately it’s all going to end up being pretty predictable if played more than once, but what can ya do?

      – I still need to swap out the .wav I converted from the .mp3 Phillip provided. I didn’t know .mp3 was a better format, but thanks to George now I know better. I will switch back to the .mp3.

      – I need to pack it all together and compile for standard and HDR. I’ll try to have this done and get someone to test it before Monday so I know the map will work properly on everyone’s computer. Simply releasing a map is far more complicated than I thought it would be. Ah, well.

      That’s it for today. I’m going to try hard mctryharderson to get the combat scripted by tomorrow morning so I have a day and a half to test it and make any changes. I still need to write a paper due Monday, so that’s going to eat into mapping for a little bit.

      It’s been good. I’ll see you all at the finish line!

      1. dino_yoshi13

        You really don’t have to include any info_nodes, but if you don’t, NPCs don’t know anything about the geometry of the world.

        1. JG

          … which kind of means you need to have info_nodes. 🙂

  8. Hi Phillip, in the Judging Criteria, you talk about “ambush”, rather than “defence”, or am I missing something. I’m no mapper FYI, but I like the theme.

    1. That part appears to be copypaste from the previous 3×10 competition; you should probably ignore the word “ambush” and mentally replace it with “defence scenario” or something like that.

      1. Correct. I have updated it now.

    2. As has been said, that was a copy and paste piece of text that I forgot to update.

  9. My bad, I had really good comments on the ambushvile entries, but came way too late with chores and other stuff. Not that anyone cares but just apologizing ;(

  10. Mega Sean 45

    I can’t really make maps anymore. I got too much to do. But if anyone would like some map ideas, I can always help with that! I got alot. Just bringing it out there.

  11. Hec

    Wow it seems this recent competitions only are getting better and better!!!!!!!

    Hope I have plenty of time during my vacation to play them all!!

  12. wow,another “ville” already … I would love to have a go,however im confused with how the hammer editor works now
    I have the sdk base and 2006/2007 base installed,however do I also need the “source sdk base 2013 singleplayer” also ( for sp mod ) …
    and what is different with the source sdk base 2013 singleplayer ? …. was It released for a specific reason such as steampipe ?
    basically can anyone tell me how to start mapping again ( as the source sdk works but wont let me run any maps )
    as a novice ( and I mean novice with a capital N ) I wont be producing maps as good as miiggas and leon brinkmans ( my favourite 2 mappers/modders ) .. however with time who knows !

    1. Firstly, make sure that you have made this switch to SteamPipe. Go to the Steam Library where all your games are, and start up the game and play it. You just need to load any level, then you can quit. You may be advised that Steam needs to update your files, click OK and wait. It may take some time.

      Now, go to Steam > Steamapps > Common > Half Life 2 > bin, and in that “bin” folder you should find Hammer.exe. Launch it, and choose the game you wanto to map for (Half Life 2: Episode 2 in this case). There, you are ready to go.

      Note that now your game folder, where the map folder, along with materials, models etc… is located, is now under Steam > Steamapps > Common.

      Half Life 2 and its episodes can be accessed by going into the Half Life 2 folder, and then opening the subfolder of the game, which is normally named after the game’s shortened name (Counter Strike Global Offensive is csgo, Half Life 2 HL2, etc…). For example, Episode 2 is in Seam > Steamapps > Common > Half Life 2 > ep2.

      1. thank you so much for that advice … I found it and ive been messing around with some basic mapping … is it me or has the editor improved … it seems easier to use and one thing I noticed is how I can change what npc I want from the properties … so were I selected combine_s ,I could then select two other soldiers
        that said ( and I mentioned this in another post ) its ok to be able to map but if you have no imagination then its a waste of time … anyway can throw a simple map together but if theres no storyline its just a mish mash .
        anyway thanks for both the replys .. us ageing gamers have a lot to learn…. when I was in school we used the bbc micro and my first gaming computer was a amiga 1200 with a 20mb hard drive ( and that was big in those days )
        thanks again

    2. The Source base 2013 Singleplayer is for modmaking.

      Before the Steampipe update, it was impossible to make mods for Episode 2, wich uses the latest Orange Box engine. The EP2 mods was just HL2 mods with EP2 content. The Source Base 2013 is the latest and most stable Source Engine version.

      So, in short, if you want to make maps, you don’t need Source Base 2013. You just go to steam/steamapps/common/half-life2/bin and inside there should be hammer.exe. When you start it, it asks what game would you like to map for, just click Half-Life 2 Episode 2 and you are good to go.

      Hope you understood that.

  13. Greenman

    Looks like I won’t be entering this one, just no inspiration at the moment.

    Looking forward to release though.

    1. Pity, but you can’t force creativity.

      1. Greenman

        If the situation changes, I might still have a crack.

    2. Count me in. I have an Idea but when I open Hammer and start mapping, I’m just not having fun, and there is no point in doing it if its not fun.

    3. Mega Sean 45

      If you need some ideas, I can help! It would be an honor! 😀

  14. Its always interesting to me when people use dev textures…
    Unless you’re planning to create brand new textures for your map, there’s really no point. You may as well use, generic concrete or metal and apply NoDraw to the correct surfaces now, then tweak it later.

    Otherwise you have to go back through your whole map later and redo all the texture surfacing at a later date…

    Nice work so far though people!

    1. The reason people use dev textures is that it saves you time. First of all you can just block out the level, and if you change it later (which you probably will), you didn’t spend time picking and choosing textures.

      Once everything is done you can simply texture your map as you see fit a mass replace all the dev textures with nodraw using the options hammer gives you.

      1. Unq

        Heartily agreed, I’m a definite convert to blocking out maps with just dev textures. Too often in the past I’d waste time focused on shaping a detail to fit a particular texture, when in the end I’d throw it away or redo it anyway. Dev textures allow you to just throw the basics in quickly without worrying about details, get the gameplay down, and then go back with a couple passes to flesh things out. Of course you can use any texture you want as temporary but the dev textures stand out as placeholders.

        1. I don’t understand that.
          If you stick to 128 by 128 (which pretty much all wall/floor/ceiling textures adhere to) and use Alt/right click you should never have a problem.
          Interesting to hear how others work though. Do you NoDraw at the same time as using dev textures or do you go back later and do your NoDraw?

          1. I personally block out everything in a white dev texture, and once I’m happy with the layout, pass once to texture everything that is visible to the player and then just mass replace the dev texture with nodraw.

            When I started mapping alot of people and tutorials said to use nothing but nodraw and then go around and texture all the visible stuff, but the nodraw texture is horrible to look at and in alot of cases is difficult to understand the geometry. I found that dev/reflectivity_60 (/70) tiles well together, is pleasant to look at and works well by itself instead of the orange and black textures that most people use.

            1. After lots of different approaches I settled on using fairly standard size brushes and a few models with textures applied (and nodraw done) and use these to map out my area.
              Then once i’ve got the overal visual style and shape of the map down, I’ll start hollowing out rooms to create a playing route through the map.
              I have several general ideas for gameplay in the map, if the map shapes don’t fit well enough then I can always delete a few brushes and rebuild small areas but the overall look/feel of the map wont change.

              Heres some screenies of my current map.


              Its not playable yet. most brushes are 128 x 128 solid with textures applied…
              It also leaves me free to decide on the most exciting route for the player to take through the environment…

              1. dino_yoshi13

                Just to let you know, your link doesn’t work.

                1. Christiaan Bakker (AniCator)

                  Just remove the /home at the end and it will work again. :p

  15. I just submitted another daily update but it didn’t show up, however, when I tried to repost it, a message popped up telling me that I had already posted that comment. I still can’t see it, though. I’ll check back tommorow to see if it has appeared.

    1. You did exactly the right thing by posting this comment. This alerts me to the fact that you have one in the spam queue, which is in excess of 1000 per day – too many for me to check manually. Your update is now visible.

      1. I have a question regarding the use of custom material from other autors. I’d like to use one of these skyboxes: http://dl.dropboxusercontent.com/u/3779442/skybox_website/sunfolder/skybox_list_v2.html

        As you can read at the bottom of the page, all skyboxes are available for free as long as credited. I’ll obviously do that in my map.

        I’d also like to know if using content licensed under Creative Commons 3.0 By Attribution (http://creativecommons.org/licenses/by/3.0/) is allowed. Again, proper credits will obviously be there at the end of the map.

        Since they are available for free as long as credited, is the use of these assets allowed?

        1. Yes, you can use those sky boxes.

          1. What about Creative Commons-licensed content (such as music, textures, etc…)? Is that allowed?

            Sorry for being so fussy but I don’t want to break the rules.

            1. Yes, those are allowed. I added the rule because I didn’t want to have any issues with game developers contacting to remove the mod because it contained copyrighted material.

                1. Yes and No. Yes, copyright content has been included in *Ville mods (I didn’t know until after the release) and no, I haven’t been asked to remove a mod yet.

  16. These competitions are proving to be interesting. Let’s hope this one is as successful as ambushville. It seems like a similar formula.

    I have a map *almost* ready. But just like in my previous maps, completely random seemingly unsolvable problems occur. Just like with the APC causing crashes on many people’s computers in radiotower ambush, after testing of my map, sometimes the hopper mines cause crashes when they explode! Fantastic! Ah well, if I do manage to finish the map, I will still send it in and hope the random crash is not too prevalent!

    Atleast it has been fun although a heck of a lot of work, so, can’t give up now!

    Also, I have a question. Can I lower the amount of time until they attack from say 10-7 minutes to 5-6? because although the player spends about 4-5 minutes planting hopper mines and placing the turrets, I am not sure how else that big time gap could be spent.

    Regards, George.

    1. JG

      Well, tempting fate, I have an APC as well. I haven’t had anything unusual happen in terms of crashes. Perhaps we can compare setups.

      My npc_apcdriver is named “apcd” and has Vehicle to drive set to “apc” with all other settings left as default, including spawnflags (left at 0).

      My prop_vehicle_apc is obviously named “apc” and uses the World Model combine_apc.mdl and the Vehicle Script File scripts/vehicles/apc_npc.txt. It has Start locked set to Yes and everything else left to default.

      Scripting-wise, commands I give “apcd” include StartForward followed by GotoPathCorner (with the path_corner I want it to go to). On passing the end of that path, it gets an EnableFiring and Stop commands.

      The “apc” itself, on passing the same node, uses HandBrakeOn and then calls FireMissileAt a few times, with npc_bullseyes as targets.

      One discrepancy I’ve noticed is the HL2 prefab “apc_on_path” actually uses npc_vehicledriver and not npc_apcdriver. It also has the Target Path Corner initialized to a path_corner, which I didn’t do. I tend not to use the HL2 prefabs because I prefer to roll things myself, but if you are using “apc_on_path” as a basis, that could be an important difference.

      I’ve never had crashes with Hopper Mines, so I find that rather unusual.

      Also, I agree about the time requirement. I managed to stretch the setup time to 7 minutes, but 7 minutes is generally way too long for “setting things up.” I suspect if I had more time, I’d engineer some kind of interim objective into that, but realistically, it’s been enough of a chore getting the main battle done. I’m not sure what guided the 7 minute logic (let alone 10), because – my gosh – that’s an eternity in game time. I suspect we may end up with a lot of maps with annoying downtimes – which is kind of a weird thing, given that you’ll have to invest considerable time in each map just to get to the good part.

      1. Hi JG,
        Yeah I used the “apc_on_path” prefab, made by valve, built into the hammer tools. Obviously, I moved the path corners, so the apc would move into position instead of a straight line (even then, there were problems with the apc getting stuck on a 1 inch bit of wall). This is strange indeed that many people reported the APC crashing the game when it exploded.

        As for the hopper mine crash, it seems not to be the cause. After some more testing (and yes with vvis on normal), it seems random. Although the real reason I suspect is when the battle starts raging, many combine soldier entities are spawned in, and when many gibs and ragdolls are created, the engine hates this and crashes. I’ll see if I can reduce the amount of soldiers attacking at any one time.

        Either way, it seems to crash now every time the map is run (at a random time through the battle). So… without a fix, this map won’t see the light of day.

        Also, after reducing the wait time to about 320 seconds, or just over 5 minutes, the messing around with hopper mines and turrets plus exploring the map seems to be a fair amount of time for this map. While the 7 minute version was pretty boring. So, hopefully (if I manage to submit the map) this won’t disqualify me!

        I also thought about how the same audio will be heard in each map which could end up being boring. Plus that we hear the audio every time we test our map. :p

        1. I also thought about how the same audio will be heard in each map which could end up being boring.

          That can be the nature of having a set theme. I wanted to ensure a consistent as possible start for all maps. But yes, having the same audio on all the maps will be repetitive.

        2. Turns out I fell asleep at about 6am. Woke at 11am. Got some stuff done in real life and just got back (1:30pm), so, missed the deadline. Last night I seem to have fixed the random crashing error. I’ll see if I can submit this map in the next hour or so – obviously this means I am disqualified but it’s ok because a) I’m in it for feedback. and b)an extra map for the pack is always good.


          edit: and yeah my sleeping pattern is a bit skew-if.

          1. George,

            At the time of this message you have almost 9 hours left. The deadline is 11:59PM GMT tonight.

            1. ah! well, that’s great! I thought the deadline was 12noon!

              Thanks that’s really cool!

      2. I’m not sure what guided the 7 minute logic (let alone 10), because my gosh that’s an eternity in game time. I suspect we may end up with a lot of maps with annoying downtimes which is kind of a weird thing, given that you’ll have to invest considerable time in each map just to get to the good part.

        After no extensive scientific research or trials I plucked a number out of the air that seemed about right. Any less had the feeling of meaning that the maps might be too small to be fun.

        I won’t be timing each entry or basing the winner on that time.

      3. Speaking of APCs, they seem to be as light as a wooden crate… I can literally send them flying meters away with the gravity gun, is there any way to increase the model’s weight?

        Oh well, I guess I’ll have to remove either the gravity gun or the APCs. Assuming I can even get the whole thing finished by midnight…

        1. JG

          I’m not aware of a way to increase the weight of the APC in the map itself. However, the .txt file that it links to does contain numerous references to weight, which unfortunately appears to be identical to the jeep_test (which is used for the HL2 and Ep2 jalopies, I believe).

          Course, obviously, that’s a mod-wide setting and who knows what else it could affect.

    2. JG

      George, on a whim I decided to decompile Radio Tower Ambush to see if there was anything obvious about your APC crash there.

      This is my theory, and it could be completely wrong, but it’s possible you simply had too many particle effects going on. You have several func_dustclouds and func_precipitations over most of the level. You could be running close to whatever the particle limit is, such that when you do something that adds considerably more particles (like an APC explosion), it overwhelms some lower-end machines. It would also explain why the crash is unpredictable – particles themselves are unpredictable and the combination that causes the crash is not guaranteed.

      If you had something similar going on in your DefendVille map, hiding your particle brushes and trying the map without them might be a good test.

      1. George (not logged in)

        Yes that seems very plausible. I have done some minor extra work on the radio tower ambush already and decided to reduce the number of particles, the values were way too high.

        As for the defendville map, there are no func_dustclouds /precipitation entities and I do doubt I reached any sort of entity limit. It may have been some sort of entity error though (not a limit) but ive fixed the crash and have yet to get any other crashes. There was a map breaking error ‘too many t-junctions” cus the train tracks have t-junction brushes with all of the wooden boards along the metal lines (even though its func_detail) which took me a while to find (and the error sometimes was ignored by the compiler and other times stopped compiling consecutively). But in the end, just increasing the metal rail height so there were no t-junctions allowed it to compile. Strange but ey that’s Hammer and the compiler for you!

  17. Submitted. Balancing the game play in this entry took far more time than I expected or wanted. Lesson learned. Also, thanks to the BTC, they helped with that quite a bit. Now to play catch up in real life.

  18. Submitted my map earlier today. The theme was a challenge to pull off but I am fairly pleased with what I’ve made.

    1. A Migga map in a Ville comp. I am very excited!

  19. Shucks, what a failure (for me), I won’t be able to submit my map. There are just too many things that need adjusting, NPC scripting and navigation is at a too early stage, and there’s no way I’m going to be able to finish it in time, even if I worked all the way until 1 A.M. (we’re under GMT+1 here).

    I have learned my lesson though. I took too much time going OCD over the tiniest things, designing pretty buildings (which aren’t even that pretty actually) and making sure that my stairs and catwalks were perfectly aligned, and neglected actual gameplay until yesterday. Next time it will not have to be like this. At all.

    Motivation also played a role. I didn’t really get that excited about the theme. Defensive scenarios may be fun to play but they require so much complex gameplay design (which is something I certainly need to improve upon) and attention to it… If I don’t get excited or interested in a theme it’s pretty hard for me to find motivation, and I end up wasting lots of time and not working as I should.

    I may release the map later on as an independent level. Don’t ask me when, though. It may be in a week or in a year, or never I have no idea.

    EDIT: I wish the best of luck to all those mappers who submitted their level. May the best map win!

    1. JG

      There were many times I wanted to give up, simply because I didn’t feel it was ever going to reach the standard I set for myself. None of us are really in this for the prize, so what is lost if you don’t submit? As long as you went through the exercise and learned a thing or two – even if it’s what not to do – I’d call that a success. There’s no point in releasing something you’re not proud of. You walk away knowing what to do better next time.

  20. JG

    Submitted my map about an hour ago.

    This was a map that really came together in the last two days, after having to deal with a lot of troublesome entities and, surprisingly, enemy vulnerabilities and behaviors I simply didn’t anticipate. Fortunately, the space I built was flexible enough to support a number of different scenarios and once I settled into the composition I wanted to use, it was relatively straightforward to do. In the last two days. 🙂

    1. JG

      So while it is still fresh in my mind, here’s a little write-up about Outpost 905.

      The first day, I got to drawing some different map ideas for it. There were three basic ideas I was working from. The first was that we’re dealing with a Combine assault and I wanted them to be attacking something other than the player. The second was that because this would be a waved battle with Combine, it would be important to give them multiple ways to enter the area. And third, for the sake of adding something unique, let’s try using friendly Antlions and the pheromone ball.

      So the first draft of this went nowhere. I settled on having a bit of Kleiner’s rocket suspended from the ceiling in a warehouse that could be assailed from two sides, one side being the “parking lot” side and the other being the “railyard” side. Navigationally, this wasn’t terribly fun because running quickly from point to point in Half-Life 2 isn’t too entertaining on its own. There was only a day’s worth of work invested in this, however.

      The thing about 10 day competitions is that you need to find that initial spark really quickly. You don’t have too much time to sit around and experiment with different ideas, which kind of became a recurring problem with this map. Nevertheless, I scrapped that first draft, but kept the rocket and the warehouse premise.

      The next day, the second draft began. In this draft, I simplified the concept to a warehouse sitting at one end of a car junkyard that the Combine would have to navigate through after spawning along a road and hill at the other end. This was a fairly open concept – the cars could be easily moved around to create whatever formations I wanted – and there would be a couple supply huts out there so the player didn’t have to travel all the way back to the warehouse to get Hopper Mines and Turrets. This is also the point where I started using Rebels instead of Antlions. Rebels on mounted guns.

      I think because I had a fairly good idea of what the Combine Soldier waves would entail (these weren’t too problematic, fortunately), I decided to work on the more exotic entities first – namely the Combine APC and whatever the final “boss” was going to be. The APC, of course, is kind of a fragile entity in the sense that it all has to be “just right” in order for it to work. The APC’s role would be to lay suppression fire while the Combine Soldiers did the invading, as well as take out obstacles with its rockets. Two APCs quickly became one after it was discovered that the “Don’t fire rockets” flag never did anything.

      To combat the APC, the player had a mounted gun. Yet, here I ran into what would become another recurring problem – the gun wouldn’t damage the APC. It took me a while to realize that rather than an AR2 func_tank, you actually needed to use a func_tankairboatgun.

      Now, for the “boss,” I kept waffling between whether or not it would be a Strider or a Hunter-Chopper. I tried the Strider first, on the grounds that the Strider looked cool marching around the yard and it could conceivably attack the rocket with its cannon, unlike the Hunter-Chopper. Also, the idea of the mounted gun Rebel firing on it would look cool.

      Well, there’s that vulnerability problem again. Striders – and I presume “Synth” entities in general – are immune to the tankairboatgun. I really didn’t want to include an RPG because I felt it would trivialize the Combine invasion. And short of turning all the mounted guns into tankrockets (although I must remember that entity for another time because rapid firing rockets is amazing), that basically ruined that mechanic. The Rebels on the mounted guns were kind of worthless now. Plus, well, they were robbing from the need to set up your defenses anyway.

      This was with about 4 days left to go. Something drastic was needed because this just wasn’t working out.

      I ended up going back to the friendly Antlion idea and ditching the rebels completely. Antlions are in infinite supply, after all, and plus it was kind of cool to start the level with only a Crowbar, Gravity Gun, and the pheromone ball. There some something almost Fallout-esque about this scenario – being a lone, crazy guy with guns and a mutant insect. If I could make the Combine Scanners play clips from Enclave Radio, the feeling would be complete.

      Anyhow, this setup would immediately make the level different from most and create the need to find your other weapons, which ate into that 7 minute setup time. Plus, having a pet Antlion following you around gave the player something to do and interact with if they found themselves waiting. They also do a great job of distracting enemies and, between them and the rocket, keeping the fire off of the player. The last thing I wanted was 12 Combine all attacking the player.

      Now, if you have any experience with friendly Antlions, you might have noticed that they don’t really get along with Hopper Mines. This was a real letdown, because the Hopper Mines are these yucky, hardcoded entities. How would I keep the Antlions from tripping them and undoing all your work of placing them? Well, a hack, of course.

      I also ended up changing the boss. I like how the boss turned out. It’s a combination of a couple enemies that you don’t usually see together and you’re really going to prioritize how to deal with the them.

      By that point, I also formally added the dreaded seven minute wait and I was pleasantly surprised (with confidence renewed) after realizing how much of that time I ate up searching for weapons, setting up defenses and playing catch with my pet Antlion. There is a bit of a “story” that develops over the course of the seven minutes that sets up how the Combine finds you.

      So, in a (Citadel-sized) nutshell, that’s the development of Outpost 905.

  21. Despite aiming for both daily updates and beta testing, I fail yet again to manage either. Switching to a completely different concept almost halfway through didn’t help. Got my map in anyway, looking forward to seeing other people’s work.

  22. dino_yoshi13

    And it looks like I’m late…

    Well, I’ll see about getting the map out sometime soon. It’s probably one of the maps I’ve had more fun making. I’ve just been running into too many problems like the picture below to get it done on time. Oh well.


  23. Submitted about 5 minutes before the deadline! Whew, that’s a habit I’ve gotta kick. Well, this competition was fun to enter, albeit stressful. I learned a lot about combine combat ai (which I should have already learnt by now!) and I feel like I have learned how to increase my work-flow/speed so I am no longer spending too much time on one particular thing (although, I did spend about a day working on a rebel gate/defence thing which I ended up cutting!). So basically, these competitions are good motivation to learn more, because there is always more learning to do! Infact, simply reading the help comments given by other mappers here in the comments helped a great deal.

    BetaTestersCollective were a great help. Their feedback was invaluable in the last moments of making my map. I was surprised that they gave incredibly quick responses. I really do recommend the BTC, just shoot them an email and they will reply quickly, which is really good for these 10 day competitions.

    Next time I may plan on writing daily or every few day comments or screenshots on how my map is developing. However, I do not like leaving comments until I have released, because there is a high chance of not releasing due to the typical problems with game development, problems with steampipe and the hammer tools, and obviously, time management. These are all very difficult problems to wrestle with. Non-developers can’t underestimate the struggle of game development.

    Anyway, it has been fun. Good luck everyone! I can’t wait to play these submissions.

  24. I saw the theme posted on the first day, and decided to enter at 11:18AM, but I’m sorry to say Outpost 118 has not made it to the finish line.

    On the first day I came up with three possible settings I liked, listing the benefits and drawbacks of each. Ultimately I found the one with the most drawbacks also had the most readily available solutions to my concerns, and became my choice for feeling the most fresh and inspiring. It’s on the razor’s edge of believability, but the idea excites me, whereas the others were sort of derivative.

    Day two saw little work done, as I was feeling a bit under the weather, but I completed a few more tests and gained that much more confidence the idea was plausible. I stalled, however, on designing combat. Last time I had too many total soldiers, and too many alive at once. I didn’t want to make the same mistake again, but I didn’t want to overcompensate, either, and I wasn’t sure how to strike the right balance, so that got set aside for the third day.

    After that, it all went south. I didn’t recover from feeling ill right away, and was totally out of it for most of the ten days. I’d make a respectable amount of progress once in a while, but not enough to finish the map.

    My usual demons haunted me, too. I could never decide how many enemies there should be, at what times, or in what combinations. Despite having long ago learned the benefit of “don’t get it right, get it written”, I let my inner editor stop me from ever getting more than a handful of Combine in the map, and as of right now I’m still not entirely sure what I’m going to do.

    I also committed what I consider my gravest sin: though I knew several days ago that a polished version of my map wouldn’t be complete for the competition, I told myself I wanted to at least have an initial gameplay design finished by the deadline, to send along to the BTC for testing. No such luck there.

    In the end, though, I did learn a number of incredibly useful things:

    -Instances make absolutely everything a breeze (look up Metapyzik’s VMF Instance Inserter to use them with Half-Life 2 and its episodes), whether you’re interested in neat off-grid architecture, reusable prefab objects that can be easily updated en masse, or just keeping various parts of your map separate for ease of editing without constantly turning on and off VisGroups.

    -Apparently, even after the SteamPipe update and with all Source games” files verified, we might not all be running the same version of the engine. My desktop uses 2002278, while my laptop is stuck with 1923675, which it seems doesn’t support studiomodels” new scaling capability, or the SetModelScale input. I just so happen to use that feature, of course. Hopefully the latest version will be more common by the time I finish this thing.

    -NPCs look unbelievably awkward moving up or down slopes of more than about seven degrees. I had some at just over eleven and everyone’s legs started going wonky.

    There were countless other little bits and pieces, too. I’ve spent the past couple of hours working on a way to deliver some Combine defenses to the arena, and though I had to beat the I/O system into submission things are going very well, and should allow the design I’ve got to make sense.

    I’m still excited by the idea, so I plan to keep working on the map. Can’t wait to see what ideas everyone else came up with that I can shamelessly steal!

  25. Submitted earlier today.

    It sounds like a lot of people made maps, and even more importantly, it looks like everyone learned a lot. This should be a great map pack. Can’t wait.

  26. JG

    Actually, because Ade brought it up on Ambushville, could we not have the glowing PlanetPhillip barrels and crates this time? I’m fine with the PP logo on them as a form of watermarking, but the fact that they glow kind of ruins the atmosphere, for me at least.

  27. They glow? Iv’e never seen them glow.

    1. The oil drums, large wooden crates, and cargo container don’t glow, since only the textures are overridden, but the item crates have both a custom texture and custom material, the latter of which turns on self illumination. It’s not a glow in the sense of casting light, but the PlanetPhillip logo ends up with maxed out brightness.

      If you edit materials\models\items\itemcrate_sheet.vmt, changing the “$selfillum” property from 1 to 0, or if you simply delete or rename the file, you’ll see the default material properties applied to the custom PP texture, eliminating the effect. Whether Phillip wants to do that for the mod itself or not, I don’t know, but you can always modify it yourself if you’d like to.

      1. Thanks for that, I have removed the glow.

  28. JG

    For those who don’t follow Phillip’s Twitter feed, DefendVille received 10 entries and has entered its first round of testing. Phillip’s laptop has given out, so I’m not sure how he is playing them now.

    1. Thanks for the update, JG.

      1. JG

        Oh, don’t thank me too much. It seems there was a bit of a glitch in my submission, haha.

Leave a Reply

Comment Formatting Guide

Well formatted comments are much easier to read. Please copy and paste the HTML Tags to use in your comment

  • HEADER: <div class="fix"></div><div class="sbe3">TEXT HERE</div>
  • BOLD: <strong>TEXT HERE</strong>
  • ITALIC: <em>TEXT HERE</em>
  • SPOILER: <span class="spoiler">TEXT HERE</span>