Alien Swarm as modding base

[Welsh Mullet]

A little, because while HL2 has the third person code built in (It's a command) AS has none of the view model code.

[Gary]

I exaggerated that a bit. Though it would be more difficult.

I assume there is more differences then just view model code, such as how the player entity is treated(sounds and physics).

[MrTwoVideoCards]

I exaggerated that a bit. Though it would be more difficult.

I assume there is more differences then just view model code, such as how the player entity is treated(sounds and physics).

A lot of that is entirely different now.

[stoopdapoop]

most things did seem different, but not alot different, I don't think it would be too much work to implement specific features.

[YokaI]

Maybe if importing content wasn't such a bitch still, people would still be interested in HL2 modding.

[Mr. Happy]

I wish UDK had a less plasticy renderer, though I'm looking forward to making maps for bullet storm.

Just don't bet against Valve, they break everything but they are also one of the smartest and most succesfull companies in the industry. Portal 2 and Ep3 could fix everything, If they simply just fixed all the bugs and made a few improvements to the pipeline at least. To bring the toolkit to modern day standards all that is really needed is a well thought overhaul to the tools and pipeline with a nice ide. All the modularity is in place, and a really good programmer could make something similar to vide or ducttape but fully integrated with all the other tools.

I wouldn't be suprised at in-engine map editor with some post ep3 engine since almost all of their tools are going in that direction.

On a side note, best engine toolset and pipeline integration goes to Love.

[Pingu]

... It would be great to have a lite AS source code skeleton, so that anyone could base their mods off (without having those AS related game mode entities and functions), so it would end up something like a regular Half-Life 2 mod but only on the AS engine...

That would make the life easier for those that decide to go over or start with AS. Right now there are too much Alien Swarm specific code in there. Many things are dependent on each other which creates high coupling so if you are changing it to a FPS game, like many want, you have to change quite a lot. The code is there but you still have to spend time to bring it up to a stable base that you can start adding stuff to.

The missing units from the hl2 universe could also be a problem. Npc_combine_s and other human like NPCs are a great use for many mods. Having just monsters to work with does not cut it so you would have to create your own models with animations and all that comes with it. So unless you are a full blown total conversion with skilled modelers and animators you are a bit limited here.

[atrblizzard]

An Open-Source mod would be a good idea. People could change the codes to their fits in order to get the native HL2 functions back. A subversion system would also be great for this idea.
Those Alien Swarm codes are dependent from one to the other, resulting major changes to be done in order to make it be as almost as HL2 was. Those AS codes can also be changed, moved, in order to fit our needs.
I'm also aware that there are many missing units and entities from the HL2 universe, but I was thinking, what if the source files from the regular Source SDK are copied into the AS project? The AS SDK still contains codes for HL2, found in #ifdef HL2_DLL, EPISODIC_DLL and so on. Not sure if adding those predefined preprocessor found in the regular HL2 SDK project would help much though. I'm guessing that the AS codes are sort of replacements on top of the old codes, but mostly the NPC entity source files should be fine, since I'm guessing they are "addons" to the mod.
There are also many changes between the two engines, such as codes split into separate files (making them more flexible and organised. Both codes needs to be examined and that way it would be easier to learn them. I'm afraid I can't help too much on this and I wish I could, but my C++ knowledge is not even near to basic. I only know bits of coding, and it's not enough for a big project like this.
If you have any ideas on how this could be achieved, please let us know. I'm afraid this would be the best solution in order to avoid those problems with the native Source SDK tools.

[ScarT]

Because it's a buttload more work than modding for hl2 or orangebox. For a start you have no materials, models, NPCs, or any assets at all to base yourself off.

The Gear team is currently in the process of converting City 17: Episode One (so pretty much all of the orangebox assets and entities) to the AS codebase seeing as Valve fucked the 2009 codebase for us.

I keep hearing that Valve broke your code base. Exactly what did Valve break that you aren't able to fix? I haven't run into any of this as far as I know.

[Gary]

Ugh... when I create a mod using the Swarm SDK, should more then 3/4th of the features and engine start out broken? Come on, at least give me a chance to break something. Half of the marines will crash the game if I choose them, all the menu strings are screwed, I can't load a map more then once without crashing, even trying to build a turret will crash the game entirely with no error dialog.

Oh and I tried doing what atrblizzard said, about trying to include the HL2 files(can't just copy&paste), I've tried many different things. It seems, for what ever reason, even some of the lower level functions are completely different. It just results in a endless page of errors. While you might be able to get a combine solider to half work, it takes hours to replace and fix all the errors generated because of great difference in dependencies.

[ScarT]

Welcome to the Source engine.

[Gary]

So, I would have to repair half the codebase just to have a stable engine?

[joe_rogers_11155]

you should know better IMO since i thought you had moved on to UDK. but i suppose you were just experimenting

[Gary]

Pretty much was just messing with the engine, wondered if I could make it into a FPS and throw up some neat scenes. But the codebase is so fucked it's not even worth the effort.

I'm gonna go install the June UDK once I can get it off my phone.

[dark0r]

Pretty much was just messing with the engine, wondered if I could make it into a FPS and throw up some neat scenes. But the codebase is so fucked it's not even worth the effort.

I'm gonna go install the June UDK once I can get it off my phone.

Stop being so lazy.

Saul Rennison says (11:37 AM)
Tony how hard is it to turn ASW to an FPS?
Tony Sergi says (11:38 AM)
its not
http://www.youtube.com/watch?v=PBZFPillw9E