Sprony
Sprony

Making Downfall: A Half-Life 2 Mod

downfall.png.db3978e107ba85801ea909c56d3bc09b.png

 

Maarten ‘Marnamai’ Frooninckx recently released the first chapter of his Half-Life 2 mod DownFall on Steam. It’s been some time since we’ve last seen such critically acclaimed single-player add-on for one of the most beloved series in gaming. In this article, Maarten and I will discuss the mod’s conception and take a look at its ongoing development.
 
Hello Maarten. Before we get started, could you please tell us a little bit about yourself?
 
I’m Maarten Frooninckx, better known as “Marnamai” on the internet. I have a background in art and I’m veteran Half-Life 2 mapper. I started out back in 2005, after having discovered that I could make my own Half-Life 2 maps with the Source SDK. Most of my initial work consists out of small test maps that I made for fun. I have been part of a few modding teams, but none of those projects ever reached completion. Is easy to become over-ambitious and underestimate the workload. That’s why a lot of mods fail. So in 2009 I decided to go solo, work on a smaller scope for DownFall and learn what I needed to get it done along the way. 

downfall_final1.thumb.jpg.0493b7b35340dec011605f2f6581f2b2.jpg

Watch out! We've got a sniper over here.

Wait a minute! You started DownFall in 2009? That’s one hell of a development cycle. Why did it take so long?
 
Yeah, it took me 8 years to release the first part on Steam. The original goal was to provide a fun expansion for Half-Life 2 which used the Episode 2 setting and learn as much as I could. Later this idea evolved into making DownFall a playable portfolio piece that I could hopefully use to land a job as a Level Designer. The entire project was a huge learning experience for me. The more I learned, the more critical I became. That’s why I’ve completely scrapped the entire project several times over. Actually, what you get to play now is the third version. Of course I haven’t worked on DownFall the entire time. I decided to take a 3D Modeling course so that I could make my own props. I also worked on various other side projects. It wasn’t until the end of 2015, when I started the latest iteration, that I made a serious effort to actually finish the project. 
 
DownFall is based on Episode 2, so it wouldn’t work without using Source 2013. Were you ever tempted to switch to a another version of Source?
 
I’m a sucker for the way Episode 2 utilizes the post-apocalyptic setting. I love all the old industrial buildings and big outside areas. I especially like the mines and caves. Plus, the gameplay is just spot on. It has a lot of appeal to me so no, I was never tempted. The main problem with using different versions of the Source engine is that you never know if everything keeps working and sorting that out would not only be a monumental task, but also require dedicated coders. Black Mesa’s customized Source engine would be the closest match, but the code they used isn’t publicly available, so I wouldn’t be able to implement the required changes I have planned for the gameplay in chapter 2. I’d also like to implement volumetric lighting which should result in some extra eye candy.

downfallgif.thumb.gif.b2e30035237b90738f1afa7568dc7e15.gif

Wait for it!

Speaking of using older technology. You’ve managed to get big open maps and beautiful vistas out of Source. What difficulties did you face?
 
Using the Source engine has a lot of benefits. There’s a vast library of assets available you can use and BSP is awesome for blocking out something fast and iterate on it. Plus, it has a big user base you can fall back to for questions or help. I know Source has been widely criticized for lacking proper documentation, but most of that has been taken care of by the community. You do face quite a few limitations, so over the years I’ve become real familiar with all the hacks to get things to work. Things like splitting levels if they are too big, how to optimize large outdoor levels, dealing with lightmap problems, hitting entity limits and how to handle leaks.
 
I really enjoyed the combat encounters in DownFall. You provided enough options and a lot of those felt refreshing to me. How did you go about setting these up?
 
My levels usually begin with an idea for a setting, location or gameplay encounter. I start doing research by collecting reference material and sketching out layouts or structures. During this process most of the ideas materialize. I’m very fond of creating encounters unique to a setting or environment. You could argue that the setting is only a layer of paint applied to the gameplay, but I prefer to blend setting and gameplay together. I strongly believe that it leads to more unique encounters that really complement each other. When I am in the early phases of a level, I make small prototype maps for gameplay scenario’s or scripted events. It allows me to test the viability and iterate fast. Once I am happy with those, I implement them into the block out (greybox) and test them as often as possible. During these tests, problem areas surface and I start the iterative process of adjusting the block out and gameplay setups to get them just right. 

sawmill.thumb.jpg.d7c242e1c161ce7a94c93d23a330290b.jpg

Planning is where it's at.

Another thing that I noticed is that the pacing is really on point. Care to elaborate on that?
 
I try to space out encounters to give the player some downtime in between combat. I fill that downtime with exploration, puzzles or story exposition. It is very important to have a good difficulty curve to keep things interesting. You start out slow, introduce the enemies and weapons that gradually increase the difficulty over time. You rely on gating to make sure the player knows the mechanics before being allowed to continue. It’s important to be fair but also challenge the player at the same time to provide a sense of progression. Some people have a natural flair for achieving good pacing, but in my case it came from years of playing games. You learn what works by observing, experimenting, testing and listening to feedback.
 
By now it’s obvious that you’ve learned a lot while making DownFall. Any do’s or don’ts you wish to share with our readers?

  • Show, don’t tell. A picture is worth a thousand words. Never have a player press a button without showing what it did. Players need the game to give them feedback on their actions.
  • Start testing early and test often. There’s nothing worse than finding a serious problem when you are late in development. Also, listen to feedback because you’ll discover solutions for problems and learn how to improve upon the existing designs.
  • Avoid over thinking solutions, the simplest solutions are often the best ones.
  • Have a group of trusted designers who you can share ideas with. Bring in other people to test your work for a fresh perspective. It can be hard to hear negative feedback about your work, but do not to get too defensive and try to figure out a way to extract useful data to help you improve.
  • Develop a personal discipline to work on your levels every day, not just when you feel like it and don’t be too ambitious. Aim for something you can actually achieve and take it from there.

downfall_final2.thumb.jpg.8184390daf8e22e136368a8f5b9875d7.jpg

Eat lead Combine!

Thank you for sharing that. I’m also wondering why you opted for a Steam release? Having a bigger audience is a given, but were there any other reasons?
 
Yeah, it has a huge install base and that means you can reach a bigger pool of potential players. Far more than any traditional modding site could reach. Of course, you still rely on those websites and social media to raise awareness and generate interest. But using Steam also allows you to use beta branches for playtesting and send out updates without having to worry if everybody is using the same version of the mod. Plus, you have the Steam community features which bring you much closer to your audience. 
 
Congratulations on the successful release. Before we finish this interview, I have to know, did you really do everything yourself?
 
Pretty much, yeah. All of the level design, additional models, animations, textures, Steam Greenlight campaign and the Steam release build are done by me. DownFall mostly uses default assets as I prefer to focus on the level design and only create assets when their role is pivotal in the design. I did bring in a Foley Artist, who’s is also a Composer, to make some new sounds and ambient tracks. For the next chapters I’m working together with two programmers, we have some new gameplay features planned and some bugged entities require fixing.

I’m very satisfied with the reception by the community. During the Greenlight campaign DownFall reached the third rank and was Greenlit in just over 12 days with a Yes/No ratio of 76%. The accumulated reviews resulted in a ‘very positive’ rating on the Steam Store with a 97% approval rating.

downfall_teaser.thumb.jpg.dfafc54b1c305d95e31e906f5d07ac23.jpg

A little teaser of things to come.

That’s an amazing job. Thank you for sharing this with us. I enjoyed DownFall, especially the funny touch on the end and I can’t wait to play more. To our readers, be sure to check out DownFall on Steam and leave a review. 

Vaya, AndyW, tomm and 28 others like this


Sign in to follow this  
Followers 0

Sign in to follow this  
Followers 0


User Feedback


I really enjoyed this mod, and Marnamai is a super knowledge guy. Very excited for Part 2 of Downfall

Share this comment


Link to comment
Share on other sites
Quote

You could argue that the setting is only a layer of paint applied to the gameplay, but I prefer to blend setting and gameplay together. I strongly believe that it leads to more unique encounters that really complement each other.

I find that extremely true and feels like it should be taken more into consideration from everyone.

Great talk, it really is a great HL2:EP2 mod (game ?).

Share this comment


Link to comment
Share on other sites

Really excited for part 2 :) best of luck @marnamai, again congratulations for the release, I'm happy you are getting all this marvellous reception 

Edited by blackdog

Share this comment


Link to comment
Share on other sites

Fantastic mod. Really inspiring and fun to see such devoted love for a franchise. I hope someone at Valve will see this and bring you in.

Share this comment


Link to comment
Share on other sites

Just played it, its very good

Almost all of my criticism would stem from stuff that is inherent to the source engine and not his design:

-Jumping off of sloped colliders of staticmeshes/props feels really clumsy / awkward

-Train wagons having ladders in the model but no trigger/entity to climb them. although he could have compensated by putting his own entity triggers on top.

-physics objects might fall in such a way that they block passage ways

-car getting SEMI stuck on a flat bsp wall etc ...

-this one is not valves fault though: there is one place were there are wooden boards that block a door that cant be destroyed even though almost every other door in the level has the same setup except destructible... inconsistency like this is a huge pet-peeve of mine.

 

Pacing and "show dont tell"-ness is top notch

I especial liked the moment were you:

1) knock away the combine energy ball

2) hear a bunch of chaotic sounds

3) the next window you run by shows you soldiers being droped of by a combine drop ship

4) and than the next window after that  a bunch of ant lions come running out of the tunnel

Perfect scene-geography / combat-scenario exposition / introduction.

 

 

Good job also felt good to get back into the half life universe for a little while ... I am however not as enthusiastic about the "episode 2-post-apocalyptic-canada-esque-outdoor" tileset graphics ... as the author clearly is  ... all I am saying is <3 city 17 !

 

marnamai likes this

Share this comment


Link to comment
Share on other sites
3 hours ago, Warby said:

-this one is not valves fault though: there is one place were there are wooden boards that block a door that cant be destroyed even though almost every other door in the level has the same setup except destructible... inconsistency like this is a huge pet-peeve of mine.

Oh yea, I know the door you're talking about, that one is a tough one in my opinion, I need the door as a shortcut, I want it visibly blocked so the player cannot open it, but I do want an npc to be able to destroy it. I used different wooden planks (bsp) as opposed to prop_physics models, but some players seem unable to differentiate between them.

 

Vaya likes this

Share this comment


Link to comment
Share on other sites


Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now