Mararthon: Yuge / YugePax

[Wrkncacnter]

Looks like you did everything right. Thanks President People!

There are buttons in the top bar of Weland for showing or hiding items. Make sure you have it set to show everything. Otherwise, not sure why your objects would be disappearing.

[irons]

guys

this shit is real

wrk is my hero (and is also irons)

jfoooooooooooooooooooooooooooooooo

[Wrkncacnter]

Yugemappers,

I'm putting out a last call for Yugeparts. Anyone still planning on mapping for version 1 of this scenario should get me their Yugeparts by June 30. I could maybe extend this a month, if I get the impression people would do a lot more mapping in July, but I haven't heard much from the couple people that expressed interest in mapping, so I'm going to assume that's not the case.

Besides, this scenario has already been in development for a whopping 2 months, and I think one more month is all anyone would need to create a mararthon total conversion. I'm planning for a release in the first half of July at this point.

Note that I will still accept yugeparts after release, if people like the maps and want to make them better. I am open to releasing updated versions with more map diversity over time.

[thedoctor45]

Yugemappers,

I'm putting out a last call for Yugeparts. Anyone still planning on mapping for version 1 of this scenario should get me their Yugeparts by June 30. I could maybe extend this a month, if I get the impression people would do a lot more mapping in July, but I haven't heard much from the couple people that expressed interest in mapping, so I'm going to assume that's not the case.

Besides, this scenario has already been in development for a whopping 2 months, and I think one more month is all anyone would need to create a mararthon total conversion. I'm planning for a release in the first half of July at this point.

Note that I will still accept yugeparts after release, if people like the maps and want to make them better. I am open to releasing updated versions with more map diversity over time.

The more I learn about this project the more I think it could come in handy for helping me complete or at least extend the singleplayer part of Halathon.

If you look at the original Halo level design you will notice a lot of repetitive interior architecture (corridors and rooms) throughout the entire singleplayer campaign - this is especially noticable during the first chapter of the game where you're basically just chasing through a series of similar looking corridors on board the Pillar of Autumn.

I've already made a few similar looking levels by utilizing Welands copy & paste plugin and I still have most of the map parts I used for compiling the levels sitting on my harddrive. The problem is I suck at level design and manually maerging each part into a map is still a tedious job, so I guess my question is would it be possible to adapt your automated level design technique to my Halathon map parts and actually make it look decent - they're already textured and lit btw so that's taken care of.

Please let me know what you think - I can provide the parts and a general overview of the original Halo leveldesign if neccessary.

[Wrkncacnter]

My level generation uses a small number of "template" areas, which are basically a rectangular area with known connections. If all of your stuff fits cleanly into a small number of templates like this, my framework could potentially be used. However, the way I'm doing it does not preserve any texturing or lighting, so even if you were to try to fit your stuff into my framework, you'd have to re-texture everything, so it's probably not what you want.

Also, the way the parts fit together is completely random, so if you're trying to replicate an actual halo level, it would be of minimal value.

I don't know exactly how the copy+paste plugin works, but I'd recommend checking out Obed. It allows you to multi-select polygons and drag them around as a unit. I believe it will also let you rotate things. So, you could copy+paste a room into your map with the plugin, and then drag it around/rotate it to where you want it to go using obed.

[thedoctor45]

My level generation uses a small number of "template" areas, which are basically a rectangular area with known connections. If all of your stuff fits cleanly into a small number of templates like this, my framework could potentially be used. However, the way I'm doing it does not preserve any texturing or lighting, so even if you were to try to fit your stuff into my framework, you'd have to re-texture everything, so it's probably not what you want.

Also, the way the parts fit together is completely random, so if you're trying to replicate an actual halo level, it would be of minimal value.

I don't know exactly how the copy+paste plugin works, but I'd recommend checking out Obed. It allows you to multi-select polygons and drag them around as a unit. I believe it will also let you rotate things. So, you could copy+paste a room into your map with the plugin, and then drag it around/rotate it to where you want it to go using obed.

Sounds like Obed is the better solution then, thanks for the input!

[Wrkncacnter]

Alternatively, if you're willing and able to write C#, I could potentially provide the weland plugin code I have, and you could possibly modify it to meet your needs. For example, preserving texturing isn't impossible, I'm just not doing it.

[thedoctor45]

Alternatively, if you're willing and able to write C#, I could potentially provide the weland plugin code I have, and you could possibly modify it to meet your needs. For example, preserving texturing isn't impossible, I'm just not doing it.

Not neccessarily willing and certainly not able lol...
I think I'll just try my luck with Obed instead of C#, but thanks for the suggestion.

[RyokoTK]

Some kindling for the pile.

group one.zip

(65.62 KiB) Downloaded 325 times

[RyokoTK]

Second batch.

group two.zip

(48.55 KiB) Downloaded 337 times

[RyokoTK]

More.

group three.zip

(170.94 KiB) Downloaded 325 times

[Wrkncacnter]

Thanks Ryoko. All of these are yugely helpful.

[President People]

A few questions:
• Is surface lighting ignored by the generator, like the textures are? What about lights defined for liquids or sound objects?
• Are Ambient/Random sounds kept, or does this only apply to sound objects?
• Will liquid types defined in a YugePart be carried over, or randomized for each map?

Lighting the maps by hand is fun, but if it can be ignored, it'll save a yuge chunk of time, allowing me to build more parts.


As for the item icons, I double-checked, and they definitely aren't hidden. It seems to be specific to objects that can't be rotated such as ammo, scenery, and sounds. As you zoom out, sometimes the items slowly become visible until you start zooming in again. Oddly, any objects with missing graphics, such as the Alien Weapon Ammo and the extra skulls are still visible as a small highlighted square.

Unless there's a quick fix for all that, I'll likely cross-post this into the Weland thread/SourceForge later.

[treellama]

Unless there's a quick fix for all that, I'll likely cross-post this into the Weland thread/SourceForge later.

Weland 1.4.3 fixes item drawing issues. What version are you using?

Also, SourceForge bad: https://github.com/treellama/weland

[Wrkncacnter]

Lights are ignored. The only light that is preserved is the tide light for liquids. I do not preserve lights for sound objects, either.

Only sound objects are kept, so no need to add ambient/random sounds.

The liquid types are carried over, although I may change jjaro to sewage in some circumstances, as jjaro will be... special.

Note that you have until June 30 to give me anything, which leaves you about a week and a half.

[Wrkncacnter]

The current yugepart counts are:

AmmoRooms - 20
Corners - 15
Ends - 8
HallwayIntersections - 23
Hallways - 13
Intersections - 16
LargeIntersections - 20
LongHalls - 19
MediumRooms - 30
Saves - 9
Separators - 17
Start Rooms - 8
Yuge Rooms - 24

We're in pretty good shape overall. If someone's still looking to make these, Intersections and Medium Rooms wouldn't be a bad choice. It seems like Medium Rooms get used more than anything else when generating these levels.

[RyokoTK]

group four.zip

(71.7 KiB) Downloaded 320 times

[Sankara]

I think I can shit out a couple parts by June 30th. I'll play around with it today.

[Sankara]

Just a quick question: it doesn't matter if all the maps I sent you are textured, does it? They're just textured with 1 texture, so every side is the same, just to test the flow. Ryoko's maps are also textured, so I don't think it's a yuge deal.

[RyokoTK]

Yeah I just hit Pave in Weland so I could walk around in the map real quick before I submitted it.

[Wrkncacnter]

Yeah, that's fine. My thing just ignores any existing textures, so it doesn't matter if it's textured or not.

[Sankara]

Hey, I've got another question about how your program works. I'm curious about enemy distribution per polygon. If I have a 1x1 WU polygon that I set to superglue, that would create one ranged monster I assume. But if I have a massive polygon set to glue/superglue, it'd be a little lackluster if the thing spawned one measly pfhor fighter. I'd just like to know if I need to break up some polygons to increase the enemy count.

I also don't want to overdo the enemies. I'm working on a map right now where I want a couple stacks of ranged units parsed along a staircase. I just want 1 per world unit. I attached a diagram of the map in Weland.


The red markers signify superglue polygons. The green signifies a glue polygon. The blue signify item triggers. I really don't have much of a concept how much these will produce, so that's why I'm asking. I was hoping for one enemy per red dot, but for the green dot it'd probably better to have more than one melee unit. So yeah, am I doing this right?

Here's the map itself if you want to take a look at it: http://s000.tinyupload.com/index.php?fi ... 9876444018

[Wrkncacnter]

You don't have that much control over the exact monster spawn locations. Since it's a Yuge Room, I know to spawn X number of monsters total, and it'll just pick points randomly in the marked polygons. So, if you have a bunch of polygons marked like that, sometimes certain polygons will have monsters and sometimes they won't. I think that tends to help the variety when you see the same room multiple times.

If you were to only have one tiny superglue polygon in a Yuge Room, you'd probably end up with a bunch of monsters spawning in that one tiny polygon. So far no one has done this, so it hasn't really been an issue.

Item triggers work the same way, and very few weapons/ammo actually spawn per room. So you'll probably end up with a tiny bit of ammo on the larger polygons and none on the smaller ones most of the time.

What you're doing looks fine, but wow, that's a lot of points. Remember, these things have to fit into a much larger map, and you might be getting to the point where there are too many polygons. I haven't run into the problem yet, but at some point the generated map will crash if there are too many points.

[Wrkncacnter]

If you want to see your room in action, here's something that'll get you an idea. It's just your room repeated 4 times, so it's not really the type of map my thing generally produces. Therefore, there's probably going to be an issue with ammo and stuff like that. But, you can see how things generally work.

I think it'll be fine, as there are other rooms with almost as many polygons.

[Sankara]

Thanks a lot for explaining it to me. After playing through that little demo you sent me I think I've gotten a decent idea of how this works.

Sorry for overdoing it on the detail in that yugepart. I just looked at the ceiling and thought it looked too plain—I couldn't help myself. But it's like 250 polygons or something, which could certainly overload the generator. I'll keep it lower in the future.

I'm giving you the 22 yugeparts I whipped up yesterday and today. The rest of them are much simpler geometrically and shouldn't give you any problems. A small portion of these are also variations on themselves in terms of enemy/item placement or small aesthetic details.

This pack contains 3 intersections (3 versions of 1 intersection), 5 unique medium rooms, 5 corners (1 unique and another with 4 variations), 4 hallways (2 unique and another with 2 variations), 1 longhall, 2 yugerooms (both unique), and 2 large intersections (both unique).

These maps all stay within the lines, tend to be below 100 polygons (with 3 exceptions including the MayanTemple, I think), and are free from errors in view-distance, bouncing walls, etc..

I think I did a good job playing with glue and item trigger settings based on my new understanding of your program. If you test it and think they're unbalanced I can modify them.

Anyway, I might put out another pack of these before the 30th. I'm pretty stoked to see to the finished product.

P.S. If anyone runs through these and happens to spot any mistakes, let me know before the 30th so I can fix them.