A.F.T.E.R. is a free open world fantasy RPG that takes place in a randomly generated world. It incorporates action adventure, city-building, dungeon crawling, and minor elements inspired mostly by the likes of Borderlands, Skyrim, and Starbound.
The game is currently being developed by a small team of three programmers with no experience. In addition to a large amount of content, the final product will leave a large amount of the game's functionality open for modders to remove and improve upon.
Yes, it's been a while. We may not have a lot of people following us anymore, but I owe it to all of you to give an update, especially after getting us this far.
First off, what happened in Fall 2014?
After summer ended last year, I didn't have a lot of time to work on the game, and following that it was a busy semester for everyone. That said, I did squeeze in a little bit of work here and there. So the game has had some changes, all of them good in some way or another.
The last time the terrain system changed, we got smooth terrain and a terrain system whose memory usage rivaled, no, it destroyed your web browser: lower memory usage, hands down. As we've found out over the year 2014, that point in particular has caused its own set of problems. As such, I've been going through and writing up plans for a new system combining the new and the old.
The terrain is the backbone of the game. It controls everything. I mean it too. At the basic level, it controls what gets loaded, where the physics simulation area is, when to center the game engine back around origin, who lives and who dies – okay, maybe not that last one, but you get the idea. It's the root of the game, and if it doesn't work well, it quickly becomes the root of all problems. That's funny, because the terrain has this variable named root and – yeah, never mind.
The point here is I've been putting it off because I don't want to deal with tearing out the backbone again. I can't quite pinpoint the cause of my laziness, but I think it's something like anxiety about needing to put in multiplayer when the terrain gets reworked.
Before, we mostly had some voices that myself and another person on the team voiced. Thanks to some wonderful and truly charitable volunteers that I am immensely thankful for, a good amount of the characters in the game now have voices. The quality varies, but you can probably expect the type of delivery between cheesy and terrible that is very close to my heart.
If you hate those kind of lines, go ahead and complain. Enough and I'll begrudgingly have VA's perform lines again while muttering to myself that your soul is dead.
Wait, you don't want to know more about what I did, and instead want to know what other team members are doing? Well shoot. FMF100 (Speck) is working on debuffs and cutscene backbone. Meanwhile, ChiefMasamune is writing more dialogue, adding unnecessary fun things to test the Lua implementation, and working on his own projects and life.
A lot of complaints were about speed. I've improved it. The rendering has come a long way in a year. It's not up to par with the big name engines, but with some of AMD's tools, I've been able to track down a bunch of the bottlenecks. Ironically, some of these bottlenecks still haven't been solved on AMD hardware. However, I'm happy to say the main “big-scene” bottleneck called animation has had its bandwidth halved and performance increased.
I came to the decision to fork the forward rendering and deferred rendering into two separate projects instead of trying to sync two systems that now have radically different behavior. Yes, the deferred rendering uses for the forward rendering system for a ton of things. However, it's become so specific to its usecase, it can be barely be considered the same thing anymore.
As such, the game has been moved entirely to the deferred rendering. In today's technology, this drops support for pretty much nothing. When this project was started, there were certain pieces of hardware that couldn't run a deferred rendering system. Over the last four years, with the advent of high-specification integrated graphics, and capabilities expanding ridiculously fast, the need to target old hardware has fallen by the wayside.
The system itself is in the works. Currently, it is a separate project, a multiplayer C++ framework with the goal of automatically sizing packets and syncing tables when needed. That said, I am eying RakNet, but the problem with that is licensing costs on consoles. The cost for licensing on consoles is so high it'll probably end up just being dropped as an idea.
You can tell that it's still early with this, but the framework works great over LAN.
I know I said we were going to put up a Kickstarter for our game last year, but the plans for that ultimately fell through. That said, this is still under discussion. The main point is that I personally take issue with using it for this project. Putting a game on Kickstarter is a sort of promise, and even with how much work and years poured into this game, I personally don't know if a promise is something I can commit to yet.
Single life forever.
I have said that when a Kickstarter is up for the game, that I'll say it. That's still true. I would plaster it all over the place, even try to talk to people in person. I'll make sure that everybody interested in AFTER will know. It's a massive commitment, and as a recently graduated nearly-jobless twenty-something year old, I can't afford that kind of commitment.
We're an extremely small team. It takes a while for things to get done. I realize that as time goes on, this kind of game is just going to get more and more dated. However, I'd rather it get done right than not at all.
More to come later, probably.
Latest tweets from @skarik_ehs
@MomoyoSeiso SATURDAY 1. massive winds knock things over 2. car's starter died 3. heavy rains 4. wind from earlier lead to no power/internet
Aug 30 2015
washington washington driving washington T.co
Aug 28 2015
jam jam jam jam jam yuri jam jam jam T.co
Aug 26 2015
@MomoyoSeiso Hella, I know that feel yo.
Aug 24 2015
Aug 24 2015