This week I'm putting up the WTE a day early, as I don't expect to get around to it tomorrow. It'll simply last until monday (the 23d) again, as per usual. That said: this week's theme is once again inspired by Tatatatate, and it will be "Campaign". What does this mean? I want you guys to make a terrain that could be a level from the original StarCraft 2 campaigns; Wings of Liberty or Heart of the Swarm. I'm not saying you should try and copy an existing campaign map from memory (though you're free to do so if you want to try!), but rather that you make a map that could have been part of one of the original campaigns. Obviously, since the scale of this exercise is rather large, feel free to ignore some of the more specific details if needed, and focus on the lay-out. Perhaps you could also provide us with a description of the events taking place in your map?
[NOTE - These weren't made at Flash speed, these were made about two years ago, meant to be used as showcase and nothing more.]
I'm not usually one to quickly expose my golden eggs, especially when the time's not so appropriate, but since I'm not so much into that for quite a time, I feel it's best to let it out, using this opportunity.
I suppose you all remember I once started a Fanfic named 'Heart of the Swarm Alternate', meant to expose my personal view of what the original HotS campaign should've been instead of what we got, right? Well, I didn't actually manage to get past the Kaldir segment, though I managed to complete the Char one, mostly due to the fact I was working in my large-scale campaign I'm putting on hiatus for the time being. Well, sometimes I felt it was a bit difficulty to translate how I imagined the ideal surroundings for the said scenarios, battles and stuff, that's why I started terraining maps supposed to showcase directly the HotS-A story. And now I'll reveal what are nothing but PROTOTYPES for the said missions I imagined.
The first is, obviously, the assault on Warfield's fortress, which I called Harkonnen in the fic. I made this instantly out of my reaction to Warfield's defense on Char, which I didn't like, he was too cocky, the defenses were easy to overcome and the cop-out of 'the artifact was out' didn't glue for me. Here, he and the Dominion would still have the artifact in their possession and Kerrigan and Co. are supposed to take it away by having Za'gara assault an afar base belonging to a general Warfield was coerced into working with. The first phase would be setting up her forces at the same time the player would have to deal with air scouts whom, if they returned, would alert Warfield, who'd hold the Artifact, deploy it and a countdown would begin, zero obviously meaning game over. Otherwise, if all the scouts were killed before they could alert or if the player could reach and damage the artifact (Forcing the transport to pick it up and dust to orbit), everything would be fine. The second would be pushing through heavily defended bases and artillery/AA emplacements, minefields and an elite guard to destroy Warifeld's POC (Primary Operations Comand). All while stuck in Tier 1.5 with only Hydralisks, Zerglings and such, but here you'd be introduced to Overseers and a modified Contaminate Ability that'd shut down Bunkers, defenses and help detect both Spider and Widow Mines.
Second is a divergence/extension of the HotS story as to how the Protoss should've reacted to Kerrigan's return to the Sector, since they investigate their loss of communications with the Kaldir scientific expedition (Exposed in the Fanfic as a war band sent by Admiral Urun to slay the Broodmother Na'fash). After her return, the continuous assaults from the Shakuras Protoss forces Kerrigan to delay her plans for Korhal to head to Shakuras and stop Artanis' aggressions. Arriving at Shakuras, however, she discovers he was set up by Admiral Urun, whom staged a coup and turned out to be him calling the shots, including trying to coerce Horner and the Raiders into fighting Kerrigan lest the Protoss bombarded Haven because of the Purifier's destruction. Thus, Kerrigan allies with a Terran-Protoss Hybrid named Gestalt Zero into storming the Assembly, freeing Artanis and getting Urun out of command. Obviously the Protoss, misled into thinking the Zerg came to destroy them once and for all, ready a fierce defense of the Hierarchy (I tried to make something egg-shaped, but I realized a custom model would actually be needed for this, in both scale, design and such).
The third, finally, is Korhal itself, the final stages. Obviously, Kerrigan, willingly, plans to start the invasion at the very gates of Augustgrad, as to buy time for civilian evacuation and ordered the Broodmothers into blockading the planet against Dominion military reinforcements. However, this plan is disrupted by what's apparently the first Planetary Shielding System in history, and that can sustain, for days, direct fire and Sac bombardment. Yet, there is hope in a group of Spectres, led by Gabriel Tosh, whom joins Kerrigan's assault in shared interest of killing Mengsk. The first phase would be guiding a disguised Spectre into within heavily fortified Dominion bases to sabotage the Power links. The second would be setting up a base to directly assault and destroy the Shield generator itself. I originally wanted to go for a nostalgia Endor-Radar esque structure, but when this kitbash worked out, I said 'what the hell' and I thought this was a good job.
Other further works, if there were any, would be a stronger and much more lore-seeking view of the Zerus segment (NO PRIMAL ZERG, ESPECIALLY!), travelling through a Xel'Naga complex with a surprising plot twist and depicting both Skygeirr and Moros as more than just Hybrid making labs (Yeah, Moros wouldn't be a prison here), but also deeply involved with Zerg research, robotics based on Protoss research and, especially, human cloning and brainwashing.
These are not really applicable to the vanilla campaign, but as I will not likely ever have another excuse to post them on this site, I will post them here anyways, since the campaign they were being built for was still B&D and similar in style (but much larger scale and geared towards C-B+ iccup level players for tutorial level).
They are presented in varying stages of completion, mostly conceptual.
0x08 - Ruins of the Khalai Castleworld
This is a three-phase mission. You are given the liberty to chose your initial nexus location. You have about 2 minutes to make the decision before you are attacked. That means you will only have about a minute to get your defenses running. After that it transforms into a segmented defense against Armageddon tiers 1 to Nightmare 1. Your goal is to capture a neutral town before the enemy destroys it, which then triggers the Gardenship boss in addition to tier 2 waves. Your goal is to defend the Quasargate and construct the Convenient Stealth Shuttle which will require ample resources to begin building. That means you will also need ample map control.
When building the terrain I was immediately met with having no Protoss objects suitable to detail it with. I avoid using anything similar to actual player units in making maps because it's distracting and the gameplay is fast enough that even minor distractions could end a 50 minute mission instantly. Thus, the map was used mostly as an experiment to try different ways of weathering in types of supernatural erosion and damage into terrain in the most minimalistic way possible, hoping I'd be able to make suitable doodad assets later on.
Note that in a lot of maps like this one I intentionally made a lot of "high risk high reward" expansion locations or separated resources in seemingly unusual locations. Especially for lower econ maps (low econ for the campaign is 3-4 concurrent expansions are manageable in about 15 minutes) I felt placing resources in tactically unusual locations would lend well to the model I was aiming for in terms of map control.
0x03 -
Short and sweet here. I needed Russian structures and units to populate this 2v3 map. I didn't have those, so I didn't progress with it. This third map was the final tutorial of the campaign and was going to introduce you to both types of SDI - Artillery and Tactical. Nukes and other related devices were no longer tied to ghosts and functioned more like Supreme Commander, as did Artillery.
I experimented mostly with brushwork and cliff doodads here, later deciding I'd need to make custom cliff meshes for every single existing tileset, and the plethora of new tilesets I planned to add, as the game's default organic cliffs are extremely unsightly, and the few new models that were added in hots proved invaluable.
0x13 - Dead Zone Heart
Fringe of Eden's ruins. An experiment of color palettes. I believe I showed these images before. This was an end-game level map for Segment 1 of Apex F. Its difficulty would have required absolute mastery of not only the game's BW-style mechanics but also all of its new mechanics. You would face the Illuminati, Armageddon tier 3 nightmare, and a dragon boss 2v3. The map makes heavy use of islands and it was expected players would be baited into spamming air, so the various wizard towers you were going to face would make air play extremely difficult.
This map needed custom organic doodads and structures for the Forgotten race that occupied the other half of the ruins and acted as the Dragon's soldiers.
0x05 -
This map was the map that killed the campaign project. It was intended to be a 2v5 map that introduced you to the completely reworked Zurg. However, even at 2v2, the performance of the game's AI was simply too terrible. This map rarely ever managed above 14 fps with only a few hundred units in play and no fighting, all exclusively due to the AI. Over a year of time was invested into trying to make maps like this one playable. After a lot of empty promises and many dead-ends, that was that.
The map itself was intended to be the staging point of where the campaign started getting objectively "hard". You were in tight quarters with enemies at all sides using synergistic 300 food armies. From mass vipers and infestors with retooled Tactical AI to hordes upon hordes of ultralisks and various lurker types, this would be the first map you were given tier 1 artillery on. Gametime for the missions was expected to be 1-2 hours.
The use of destructible rocks helped expose how fundamentally broken (and terribly performing) the melee destructible wave code is. A simple alternative was to make the rocks neutral to players but hostile to computers and then delete the obs code from the wave AI, gaining back over 500ms in performance from that alone. This map also helped expose that the Spore Crawler Tactical AI is probably programmed by satan's deformed interns.
Although the terrain was early in, I experimented with waterless cliffing alongside watered cliffing to give the map a feeling of depth, particularly around the center, where it was insinuated the planet had been the target of an exterminatus in the past. No suitable doodad assets again held it back, but it's not like it would perform better with more detail.
0x14 - Eden
This map would have been impossible to make in a Blizzard engine I would discover. The scale of it alone is too much for sc2's AI to handle. A 5v3 against the Necropolis, this near-final mission had a colossal amount of planning put into it but none of it was ever actuated on.
Worth noting is the layout ideas I was planning to try out. I was really big on trying to get cliff advantage back into the game. I wanted the AI to prioritize cliffs and hold them, but it would be enough for the player to be encouraged to hold them against the various enemy heroes that would be sneaking around the choke points. This map was intended to encourage you to expand and hold locations.
The primary feature of the map was going to be the 5-phase Sethis boss fight, based on the old AO2 concepts and retooled. That boss alone would have been more work to create than most individual maps made for the game so far, if only for the sheer quantity of audio and visual assets he would have required. Unfortunately for me, while those assets would actually have not been that difficult, a 2fps boss fight would not have been enjoyable no matter what it looked or sounded like.
0x01 -
The first tutorial map of the campaign and also my primary testing map for the new AI we were building. Despite its extreme simplicity, performance was still very bad. I hand-painted the pathing on the map for every single cliff and everything. It didn't matter.
Like all tutorials of my projects, the primary function of the map gameplay-wise is to weed out casuals from players. Your first order of business is to survive the proxy bunker/tank rush that immediately happens as soon as the map starts. Even for experienced players I was hoping it would take a few tries to survive the first five minutes. From there the difficulty only scaled upwards. The map was a large FFA.
This map has a few of the new doodads I was porting. I eventually intended to replace all vanilla doodads. Until then, they didn't match very well. I think I may have also removed a bunch of doodads to simplify the pathing mesh. I kind of gave up on this one.
I showed some of the more complete maps earlier in time, so I won't respost them.
/e
Not sure if I ever posted these. They are from Apex A. So, 2011-2012 sort of era. Very old. Not very good. But ah, what the hell.
I've shown this map before a while back but it matches with the WTE since it was originally when I wanted to retell some of SC2. I've since changed it to fit with my new campaign idea.
It's one of Aiur's moons and the theme is supposed to show how mystical is is and then what happens when the Zerg dig in and destroy the land.
And to think I initially thought these were literally made within a day of this post, I would have fallen over and died, but then I read the text and discovered that was indeed not the case XD
Good stuff still! I always love me some cool looking melle stuff :)
Man I love campaign terrains! We've got some nice stuff here even though it was not built for the wte at all, it's still very very interesting to see these.
That Warfield's base is very well protected. Maybe a bit overkill on the amount of supply depot but other than that it looks awesome and would most likely be fun to play with a somewhat medium-hard AI.
About the protoss terrain, the first thing i though when opening the screenshot was "wow". I very like all these colors... and that huge temple of doodads. It makes me think of the old starcraft 1 Xel'Naga temple structure with some minor modifications. Very cool.
The last terrain is the one i hate the most out of these 3. The center mix of doodads looks really good, but the base around it feels awekward. The center is filled of pipes, electrical shit and all, and right when you get out of the radius of the thing, its flat and boring othe than default starcraft structures. It would needs alot more texture play, maybe a bunch of doodads passing between buildings, more electrical arcs, maybe computers idk, but not just flat mono-texture (with few roads).
I like your ideas. You should just believe and about the AI screwing your maps because it gets laggy? Just don't use pre-made AI, create the units and whatever else manually, issue your own orders to the comp, dont let him take decisions for you.
0x08: I wonder how protoss built walls with lava between cliffs... I think I would stick between either water or lava. And if you add lava, placing it in a way that it's not just there because i decided so. Try to make it more logical. Where is that lava coming from? Are there rocks? Are there volcano dark sand? Does water makes errosion so you could add the doodads that make the top of the cliff longer (something like cliff wide something... i dont remember the name of the doodad)
0x13: I like the fog with the tall cliffs, it feels really deep. The tall trees are meh... they are "ok" but I feel like something different could be better, I just dont know what.
0x14: The excessive cliffs and ramps terrains doesn't look very good at all. I personally never liked over-complicated cliffs and ramps maps. Like there is a game on the arcade about beeing a larva and eating stuff exploring and killing... and the terrains is very beautiful but oh gosh so many ramps and cliffs.
Noticeable other terrains are the creep. Its so creepy lol. It's like if diablo made a carpet out of his skin. It's visually interesting and even more disgusting than the normal creep xD I also like the very last picture with crystals around a small crater. I think you could have made it much much MUCH bigger or have like multiple small ones like holes in the ground and with the heigh tool just make a small mountains-ish. But its cool. I like it still.
That's cool. I like the mix of destroyed protoss building and Zerus big tree and tall grass. A problem with these tree is their auto-transparancy when something is behind like we can see on the right of the first picture. It's easy to fix, but it's still anoying. On the 2nd picture we see the river with small hills, well, maybe that could be a bit more rough instead of very polished rounded. Maybe more rocks, and rocks and rocks because rocks is rocks. I like the reflection in the water... mix of very beautiful transparent blue-ish water and bam green reflection just like if it was poisonned. I think the amount of plants that has light is a bit overkill... Maybe more work around the big trees, more diversification, more plant cluster... maybe the rest of a bridge made of rocks or temple things... So yeah, basicly thats most of it.
-
These are cool terrains. It warms my hearth to see so many cool stuff. It's very cool. We are all awesome :D
I like your ideas. You should just believe and about the AI screwing
your maps because it gets laggy? Just don't use pre-made AI, create the
units and whatever else manually, issue your own orders to the comp,
dont let him take decisions for you.
Then the gameplay would not be gameplay. The campaign relied critically on the AI functioning properly. The maps are basically staged melee scenarios. That was what the gameplay was built around. With no AI, there was no campaign. Plus, the workload of doing it manually to have anything even remotely resembling intelligence would have outscaled every other aspect, even the huge texture/sound rehash, by many fold...
0x08: I wonder how protoss built walls with lava between cliffs... I
think I would stick between either water or lava. And if you add lava,
placing it in a way that it's not just there because i decided so. Try
to make it more logical.
I was trying to make it appear like the lava should not be there. As in, it was a recent addition that occurred when a demon invasion killed off the Khalai. Hence the erosion. I should have made that more clear. But, yeah. It was kind of an experiment that didn't really go here or there.
0x13: I like the fog with the tall cliffs, it feels really deep. The
tall trees are meh... they are "ok" but I feel like something different
could be better, I just dont know what.
I hate sc2's trees. I had replacements lined up to be added into the game. I think it would have made the map really stand out. Until then it just felt... empty.
0x14: The excessive cliffs and ramps terrains doesn't look very good at
all. I personally never liked over-complicated cliffs and ramps maps.
Like there is a game on the arcade about beeing a larva and eating stuff
exploring and killing... and the terrains is very beautiful but oh gosh
so many ramps and cliffs.
0x05 - Yup, that was the goal!
0x14 - Keep in mind I zoomed out those, that terrain is actually quite large. That said, I would have needed to playtest the map to make the ultimate decision on how the cliffs were going to pan out. The Necropolis were going to have huge ground units and I couldn't be as reliant on choke points as I was in other maps.
Do you mean that even an "easy" computer would make it lag as well? Like, all basic actions like mining, it would be automatic, but sending attack waves and expanding (if you want them to) would be ordered with triggers. Or do you mean ally computers that you could control with the HotS new ai control buttons is laggy (which is totally true). I just can't believe there is no work around laggy computer. First thing would be not using multiple computers, but one big computer. From that you could work around making different color and if that base is destroyed say: "whatever comp has been destroyed" even though it's false. And if you absolutly need 2 or more comps like one toss one zerg... meh.. There has to be something.
I put 5 years of research into the game. Very long story short, there isn't something. The game is too hardcoded and Galaxy is too limited. To have an Ai that does what an AI needs to do in a matter remotely comparable to Brood War vanilla and perform at the same level Brood War vanilla did is just not possible. I tried a lot of the little tricks you mentioned. All things like the HoTS panel were totally removed from the galaxy. They don't exist. There is just hardcoded overhead and roadblocks at every turn. Triggers are nearly worthless for AI in this game unless you have a very linear map you are building around. None of my maps were like that, my goal was to avoid all of the innumerable pitfalls stagnating the vanilla campaigns like timed waves and lack of defense/expanding. I tried doing it through triggers for 2 years and it was just useless and inefficient.
The really bizarre thing is it isn't stuff like actions that are causing the performance loss. APM capping doesn't even make a difference. It's just overhead. Maybe threat evaluation. We'll never know. I've dropped the game a long time ago now. Every programmer and designer I've talked to about the game over the years encouraged me to jump ship, and so I did. But, hey, if something ever comes up, maybe I'll pick it up again. But given people like Jademus couldn't figure it out I doubt I ever will. It's Unreal 4 or bust for me.
I am working on a private game project in Unreal 4. As I am just beginning, there is not much to say other than that working with it has been quite enjoyable thus far.
Well no, but you people always complain about this when the day is like barely halfway. I tend to post the WTE's up between 12:00 and 20:00 my time, and it's currently 16:04 here.
Weekly Terraining Exercise #198
Hey hey lovely terraining folks!
This week I'm putting up the WTE a day early, as I don't expect to get around to it tomorrow. It'll simply last until monday (the 23d) again, as per usual. That said: this week's theme is once again inspired by Tatatatate, and it will be "Campaign". What does this mean? I want you guys to make a terrain that could be a level from the original StarCraft 2 campaigns; Wings of Liberty or Heart of the Swarm. I'm not saying you should try and copy an existing campaign map from memory (though you're free to do so if you want to try!), but rather that you make a map that could have been part of one of the original campaigns. Obviously, since the scale of this exercise is rather large, feel free to ignore some of the more specific details if needed, and focus on the lay-out. Perhaps you could also provide us with a description of the events taking place in your map?
Either way, good luck, and have fun!
The Global Information/Idea thread for Terraining Exercises can be found here: http://www.sc2mapster.com/forums/development/terrain/14658-terraining-exercises-global-information-idea-thread/
[NOTE - These weren't made at Flash speed, these were made about two years ago, meant to be used as showcase and nothing more.]
I'm not usually one to quickly expose my golden eggs, especially when the time's not so appropriate, but since I'm not so much into that for quite a time, I feel it's best to let it out, using this opportunity.
I suppose you all remember I once started a Fanfic named 'Heart of the Swarm Alternate', meant to expose my personal view of what the original HotS campaign should've been instead of what we got, right? Well, I didn't actually manage to get past the Kaldir segment, though I managed to complete the Char one, mostly due to the fact I was working in my large-scale campaign I'm putting on hiatus for the time being. Well, sometimes I felt it was a bit difficulty to translate how I imagined the ideal surroundings for the said scenarios, battles and stuff, that's why I started terraining maps supposed to showcase directly the HotS-A story. And now I'll reveal what are nothing but PROTOTYPES for the said missions I imagined.
The first is, obviously, the assault on Warfield's fortress, which I called Harkonnen in the fic. I made this instantly out of my reaction to Warfield's defense on Char, which I didn't like, he was too cocky, the defenses were easy to overcome and the cop-out of 'the artifact was out' didn't glue for me. Here, he and the Dominion would still have the artifact in their possession and Kerrigan and Co. are supposed to take it away by having Za'gara assault an afar base belonging to a general Warfield was coerced into working with. The first phase would be setting up her forces at the same time the player would have to deal with air scouts whom, if they returned, would alert Warfield, who'd hold the Artifact, deploy it and a countdown would begin, zero obviously meaning game over. Otherwise, if all the scouts were killed before they could alert or if the player could reach and damage the artifact (Forcing the transport to pick it up and dust to orbit), everything would be fine. The second would be pushing through heavily defended bases and artillery/AA emplacements, minefields and an elite guard to destroy Warifeld's POC (Primary Operations Comand). All while stuck in Tier 1.5 with only Hydralisks, Zerglings and such, but here you'd be introduced to Overseers and a modified Contaminate Ability that'd shut down Bunkers, defenses and help detect both Spider and Widow Mines.
Second is a divergence/extension of the HotS story as to how the Protoss should've reacted to Kerrigan's return to the Sector, since they investigate their loss of communications with the Kaldir scientific expedition (Exposed in the Fanfic as a war band sent by Admiral Urun to slay the Broodmother Na'fash). After her return, the continuous assaults from the Shakuras Protoss forces Kerrigan to delay her plans for Korhal to head to Shakuras and stop Artanis' aggressions. Arriving at Shakuras, however, she discovers he was set up by Admiral Urun, whom staged a coup and turned out to be him calling the shots, including trying to coerce Horner and the Raiders into fighting Kerrigan lest the Protoss bombarded Haven because of the Purifier's destruction. Thus, Kerrigan allies with a Terran-Protoss Hybrid named Gestalt Zero into storming the Assembly, freeing Artanis and getting Urun out of command. Obviously the Protoss, misled into thinking the Zerg came to destroy them once and for all, ready a fierce defense of the Hierarchy (I tried to make something egg-shaped, but I realized a custom model would actually be needed for this, in both scale, design and such).
The third, finally, is Korhal itself, the final stages. Obviously, Kerrigan, willingly, plans to start the invasion at the very gates of Augustgrad, as to buy time for civilian evacuation and ordered the Broodmothers into blockading the planet against Dominion military reinforcements. However, this plan is disrupted by what's apparently the first Planetary Shielding System in history, and that can sustain, for days, direct fire and Sac bombardment. Yet, there is hope in a group of Spectres, led by Gabriel Tosh, whom joins Kerrigan's assault in shared interest of killing Mengsk. The first phase would be guiding a disguised Spectre into within heavily fortified Dominion bases to sabotage the Power links. The second would be setting up a base to directly assault and destroy the Shield generator itself. I originally wanted to go for a nostalgia Endor-Radar esque structure, but when this kitbash worked out, I said 'what the hell' and I thought this was a good job.
Other further works, if there were any, would be a stronger and much more lore-seeking view of the Zerus segment (NO PRIMAL ZERG, ESPECIALLY!), travelling through a Xel'Naga complex with a surprising plot twist and depicting both Skygeirr and Moros as more than just Hybrid making labs (Yeah, Moros wouldn't be a prison here), but also deeply involved with Zerg research, robotics based on Protoss research and, especially, human cloning and brainwashing.
These are not really applicable to the vanilla campaign, but as I will not likely ever have another excuse to post them on this site, I will post them here anyways, since the campaign they were being built for was still B&D and similar in style (but much larger scale and geared towards C-B+ iccup level players for tutorial level).
They are presented in varying stages of completion, mostly conceptual.
0x08 - Ruins of the Khalai Castleworld
This is a three-phase mission. You are given the liberty to chose your initial nexus location. You have about 2 minutes to make the decision before you are attacked. That means you will only have about a minute to get your defenses running. After that it transforms into a segmented defense against Armageddon tiers 1 to Nightmare 1. Your goal is to capture a neutral town before the enemy destroys it, which then triggers the Gardenship boss in addition to tier 2 waves. Your goal is to defend the Quasargate and construct the Convenient Stealth Shuttle which will require ample resources to begin building. That means you will also need ample map control.
When building the terrain I was immediately met with having no Protoss objects suitable to detail it with. I avoid using anything similar to actual player units in making maps because it's distracting and the gameplay is fast enough that even minor distractions could end a 50 minute mission instantly. Thus, the map was used mostly as an experiment to try different ways of weathering in types of supernatural erosion and damage into terrain in the most minimalistic way possible, hoping I'd be able to make suitable doodad assets later on.
Note that in a lot of maps like this one I intentionally made a lot of "high risk high reward" expansion locations or separated resources in seemingly unusual locations. Especially for lower econ maps (low econ for the campaign is 3-4 concurrent expansions are manageable in about 15 minutes) I felt placing resources in tactically unusual locations would lend well to the model I was aiming for in terms of map control.
0x03 -
Short and sweet here. I needed Russian structures and units to populate this 2v3 map. I didn't have those, so I didn't progress with it. This third map was the final tutorial of the campaign and was going to introduce you to both types of SDI - Artillery and Tactical. Nukes and other related devices were no longer tied to ghosts and functioned more like Supreme Commander, as did Artillery.
I experimented mostly with brushwork and cliff doodads here, later deciding I'd need to make custom cliff meshes for every single existing tileset, and the plethora of new tilesets I planned to add, as the game's default organic cliffs are extremely unsightly, and the few new models that were added in hots proved invaluable.
0x13 - Dead Zone Heart
Fringe of Eden's ruins. An experiment of color palettes. I believe I showed these images before. This was an end-game level map for Segment 1 of Apex F. Its difficulty would have required absolute mastery of not only the game's BW-style mechanics but also all of its new mechanics. You would face the Illuminati, Armageddon tier 3 nightmare, and a dragon boss 2v3. The map makes heavy use of islands and it was expected players would be baited into spamming air, so the various wizard towers you were going to face would make air play extremely difficult.
This map needed custom organic doodads and structures for the Forgotten race that occupied the other half of the ruins and acted as the Dragon's soldiers.
0x05 -
This map was the map that killed the campaign project. It was intended to be a 2v5 map that introduced you to the completely reworked Zurg. However, even at 2v2, the performance of the game's AI was simply too terrible. This map rarely ever managed above 14 fps with only a few hundred units in play and no fighting, all exclusively due to the AI. Over a year of time was invested into trying to make maps like this one playable. After a lot of empty promises and many dead-ends, that was that.
The map itself was intended to be the staging point of where the campaign started getting objectively "hard". You were in tight quarters with enemies at all sides using synergistic 300 food armies. From mass vipers and infestors with retooled Tactical AI to hordes upon hordes of ultralisks and various lurker types, this would be the first map you were given tier 1 artillery on. Gametime for the missions was expected to be 1-2 hours.
The use of destructible rocks helped expose how fundamentally broken (and terribly performing) the melee destructible wave code is. A simple alternative was to make the rocks neutral to players but hostile to computers and then delete the obs code from the wave AI, gaining back over 500ms in performance from that alone. This map also helped expose that the Spore Crawler Tactical AI is probably programmed by satan's deformed interns.
Although the terrain was early in, I experimented with waterless cliffing alongside watered cliffing to give the map a feeling of depth, particularly around the center, where it was insinuated the planet had been the target of an exterminatus in the past. No suitable doodad assets again held it back, but it's not like it would perform better with more detail.
0x14 - Eden
This map would have been impossible to make in a Blizzard engine I would discover. The scale of it alone is too much for sc2's AI to handle. A 5v3 against the Necropolis, this near-final mission had a colossal amount of planning put into it but none of it was ever actuated on.
Worth noting is the layout ideas I was planning to try out. I was really big on trying to get cliff advantage back into the game. I wanted the AI to prioritize cliffs and hold them, but it would be enough for the player to be encouraged to hold them against the various enemy heroes that would be sneaking around the choke points. This map was intended to encourage you to expand and hold locations.
The primary feature of the map was going to be the 5-phase Sethis boss fight, based on the old AO2 concepts and retooled. That boss alone would have been more work to create than most individual maps made for the game so far, if only for the sheer quantity of audio and visual assets he would have required. Unfortunately for me, while those assets would actually have not been that difficult, a 2fps boss fight would not have been enjoyable no matter what it looked or sounded like.
0x01 -
The first tutorial map of the campaign and also my primary testing map for the new AI we were building. Despite its extreme simplicity, performance was still very bad. I hand-painted the pathing on the map for every single cliff and everything. It didn't matter.
Like all tutorials of my projects, the primary function of the map gameplay-wise is to weed out casuals from players. Your first order of business is to survive the proxy bunker/tank rush that immediately happens as soon as the map starts. Even for experienced players I was hoping it would take a few tries to survive the first five minutes. From there the difficulty only scaled upwards. The map was a large FFA.
This map has a few of the new doodads I was porting. I eventually intended to replace all vanilla doodads. Until then, they didn't match very well. I think I may have also removed a bunch of doodads to simplify the pathing mesh. I kind of gave up on this one.
I showed some of the more complete maps earlier in time, so I won't respost them.
/e
Not sure if I ever posted these. They are from Apex A. So, 2011-2012 sort of era. Very old. Not very good. But ah, what the hell.
@DeltaCadimus: Go
@IskatuMesk: Go
Both of those are breathtaking!
I've shown this map before a while back but it matches with the WTE since it was originally when I wanted to retell some of SC2. I've since changed it to fit with my new campaign idea.
It's one of Aiur's moons and the theme is supposed to show how mystical is is and then what happens when the Zerg dig in and destroy the land.
And to think I initially thought these were literally made within a day of this post, I would have fallen over and died, but then I read the text and discovered that was indeed not the case XD
Good stuff still! I always love me some cool looking melle stuff :)
Man I love campaign terrains! We've got some nice stuff here even though it was not built for the wte at all, it's still very very interesting to see these.
@DeltaCadimus: Go
That Warfield's base is very well protected. Maybe a bit overkill on the amount of supply depot but other than that it looks awesome and would most likely be fun to play with a somewhat medium-hard AI.
About the protoss terrain, the first thing i though when opening the screenshot was "wow". I very like all these colors... and that huge temple of doodads. It makes me think of the old starcraft 1 Xel'Naga temple structure with some minor modifications. Very cool.
The last terrain is the one i hate the most out of these 3. The center mix of doodads looks really good, but the base around it feels awekward. The center is filled of pipes, electrical shit and all, and right when you get out of the radius of the thing, its flat and boring othe than default starcraft structures. It would needs alot more texture play, maybe a bunch of doodads passing between buildings, more electrical arcs, maybe computers idk, but not just flat mono-texture (with few roads).
@IskatuMesk: Go
I like your ideas. You should just believe and about the AI screwing your maps because it gets laggy? Just don't use pre-made AI, create the units and whatever else manually, issue your own orders to the comp, dont let him take decisions for you.
0x08: I wonder how protoss built walls with lava between cliffs... I think I would stick between either water or lava. And if you add lava, placing it in a way that it's not just there because i decided so. Try to make it more logical. Where is that lava coming from? Are there rocks? Are there volcano dark sand? Does water makes errosion so you could add the doodads that make the top of the cliff longer (something like cliff wide something... i dont remember the name of the doodad)
0x13: I like the fog with the tall cliffs, it feels really deep. The tall trees are meh... they are "ok" but I feel like something different could be better, I just dont know what.
0x05: It made me felt like if i was looking at the old wasteland of starcraft 1. http://portforward.com/games/walkthroughs/Starcraft/Starcraft-small-229.jpg but with even more style with the deep cliffs again. Very cool.
0x14: The excessive cliffs and ramps terrains doesn't look very good at all. I personally never liked over-complicated cliffs and ramps maps. Like there is a game on the arcade about beeing a larva and eating stuff exploring and killing... and the terrains is very beautiful but oh gosh so many ramps and cliffs.
Noticeable other terrains are the creep. Its so creepy lol. It's like if diablo made a carpet out of his skin. It's visually interesting and even more disgusting than the normal creep xD I also like the very last picture with crystals around a small crater. I think you could have made it much much MUCH bigger or have like multiple small ones like holes in the ground and with the heigh tool just make a small mountains-ish. But its cool. I like it still.
@GhostNova91: Go
That's cool. I like the mix of destroyed protoss building and Zerus big tree and tall grass. A problem with these tree is their auto-transparancy when something is behind like we can see on the right of the first picture. It's easy to fix, but it's still anoying. On the 2nd picture we see the river with small hills, well, maybe that could be a bit more rough instead of very polished rounded. Maybe more rocks, and rocks and rocks because rocks is rocks. I like the reflection in the water... mix of very beautiful transparent blue-ish water and bam green reflection just like if it was poisonned. I think the amount of plants that has light is a bit overkill... Maybe more work around the big trees, more diversification, more plant cluster... maybe the rest of a bridge made of rocks or temple things... So yeah, basicly thats most of it.
-
These are cool terrains. It warms my hearth to see so many cool stuff. It's very cool. We are all awesome :D
Working on projects:
Then the gameplay would not be gameplay. The campaign relied critically on the AI functioning properly. The maps are basically staged melee scenarios. That was what the gameplay was built around. With no AI, there was no campaign. Plus, the workload of doing it manually to have anything even remotely resembling intelligence would have outscaled every other aspect, even the huge texture/sound rehash, by many fold...
I was trying to make it appear like the lava should not be there. As in, it was a recent addition that occurred when a demon invasion killed off the Khalai. Hence the erosion. I should have made that more clear. But, yeah. It was kind of an experiment that didn't really go here or there.
I hate sc2's trees. I had replacements lined up to be added into the game. I think it would have made the map really stand out. Until then it just felt... empty.
0x05 - Yup, that was the goal!
0x14 - Keep in mind I zoomed out those, that terrain is actually quite large. That said, I would have needed to playtest the map to make the ultimate decision on how the cliffs were going to pan out. The Necropolis were going to have huge ground units and I couldn't be as reliant on choke points as I was in other maps.
@IskatuMesk: Go
Do you mean that even an "easy" computer would make it lag as well? Like, all basic actions like mining, it would be automatic, but sending attack waves and expanding (if you want them to) would be ordered with triggers. Or do you mean ally computers that you could control with the HotS new ai control buttons is laggy (which is totally true). I just can't believe there is no work around laggy computer. First thing would be not using multiple computers, but one big computer. From that you could work around making different color and if that base is destroyed say: "whatever comp has been destroyed" even though it's false. And if you absolutly need 2 or more comps like one toss one zerg... meh.. There has to be something.
Working on projects:
@Scbroodsc2: Go
I put 5 years of research into the game. Very long story short, there isn't something. The game is too hardcoded and Galaxy is too limited. To have an Ai that does what an AI needs to do in a matter remotely comparable to Brood War vanilla and perform at the same level Brood War vanilla did is just not possible. I tried a lot of the little tricks you mentioned. All things like the HoTS panel were totally removed from the galaxy. They don't exist. There is just hardcoded overhead and roadblocks at every turn. Triggers are nearly worthless for AI in this game unless you have a very linear map you are building around. None of my maps were like that, my goal was to avoid all of the innumerable pitfalls stagnating the vanilla campaigns like timed waves and lack of defense/expanding. I tried doing it through triggers for 2 years and it was just useless and inefficient.
The really bizarre thing is it isn't stuff like actions that are causing the performance loss. APM capping doesn't even make a difference. It's just overhead. Maybe threat evaluation. We'll never know. I've dropped the game a long time ago now. Every programmer and designer I've talked to about the game over the years encouraged me to jump ship, and so I did. But, hey, if something ever comes up, maybe I'll pick it up again. But given people like Jademus couldn't figure it out I doubt I ever will. It's Unreal 4 or bust for me.
@IskatuMesk: Go
Well then... do you have any other projects?
Working on projects:
I am working on a private game project in Unreal 4. As I am just beginning, there is not much to say other than that working with it has been quite enjoyable thus far.
Though we've already had some good submissions, I'll extend this for a week due to laziness and a lack of worthy ideas for WTE #199.
@Mozared: Go
its cool Moz, take it easy.
Working on projects:
I dont have any Campaign experience :-( nothing to contribute! :-(
Marie T. Freeman If you're too busy to give your neighbor a helping hand, then you're just too darned busy. https://www.facebook.com/wargirlmaps.maps
Spread the love join DISCORD
https://discord.gg/Jtzt8Su
2nd week and still no new WTE
Moz is kill ripip
@ScrinKing: Go
Well no, but you people always complain about this when the day is like barely halfway. I tend to post the WTE's up between 12:00 and 20:00 my time, and it's currently 16:04 here.
Well it was almost Tuesday in my timezone