Sometimes when it comes to engine coding, it pays off to revisit some things that you previously tried(unsuccessfully).  Other times it pays off to throw the old code away, and start from scratch.  It almost always pays off to listen to what your client base tells you.  While we have been constantly pushing the graphical features of the engine, for years trying to catchup to AAA gaming engines, the efficiency was something that was still lacking.  I would not say that we ignored it, there were very substantial gains made in past revisions.  I recall more than doubling rendering speed around version 7.45, and 7.53 provided another large boost as well.  After the release of 7.60, and the Reloaded Edition, it was time to really bear down and make this sucker hum.  There was basically a two-pronged approach taken to tackle the issue.  First, the bsp rendering needed some major optimization, and Max Eliaser, who is somewhat mildly obsessed with optimizing and clever codework, ripped apart our existing system, and came up with something that provided some very impressive results.  Once a significant bottleneck, bsp rendering increased by 3 or 4 times over!

Another major bottleneck was the rendering of player models.  This is a problem that is common to a number of games, and it was certainly a problem for the CRX engine as well.  When we implemented skeletal animation in 2010, one of my future goals, aside from ragdoll physics, was GPU animation(or sometimes, IMO erroneously referred to as GPU “skinning”).  As time wore on, a couple of other engines, notably Qfusion and Cube 2 had implemented it with great results.  Not to be left behind, I personally took on this one myself using Lee Salzman’s IQM demo code as a base.  It was actually remarkably simple to implement, aside from a few hiccups which were quickly solved.  The results were astounding, as meshes rendered 3 times faster than before, when say a dozen or so of the same mesh were on the screen together.  The more meshes, the greater the performance gap.  In doing this, I revamped our VBO subsystem, and it also forced me to revamp the shadow system to use shadowmapping for all shadow types.

Despite the intense concentration of performance enhancements(and there were more than mentioned above, such as particle system, shaders, and general math cleanups), there were still some things added to improve the visual appearance such as cubemapping on meshes, soft vegetation shadows, high definition lightmaps, water shader improvements, and more.  The optimizations also meant that people who couldn’t previously run the game on higher settings now could, and even faster than they used to on the lowest settings.  All in all, 7.65 was literally a tour de force of a renderer rewrite.  Any of the old, idtech2/3 style rendering is now long, long gone.  We still have a number of things we are looking to improve, but for now we are ready to unleash this onto the public.

In the fairly recent past, I’ve lamented the slow decline of the fast-paced arena shooter games.  We’ve certainly taken steps to shake things up, such as the new Minderaser weapon that was introduced in 7.60, but I’ve come to a plateau when it comes to the gameplay aspects of Alien Arena.  I don’t expect there to be any major changes in how it plays in the future(though that could always change, you never know), and I don’t expect it toaatacticle_m suddenly regain it’s past popularity barring a major promotional campaign(always possible) or a resurgence in the popularity of these types of games.  With that in mind, for the last year or so, I’ve been considering a completely new game type set in the Alien Arena universe, one that is part arena shooter, part tactical shooter, and perhaps part rts game.  I’ve dubbed it “Alien Arena Tactical”, and have spent quite some time developing it’s concept and code base goals.  I’ve recently started coding the basic portions of it, and expect to have a beta released spring/summer of 2013.  It’s a large undertaking, but it’s something that I envision as being very exciting, and fun to play, even by those who prefer fast-paced arena shooters.  The movement and pace are slower, much like a tactical shooter, and while killing is your business(and it’s good!), it’s not the primary goal.  The game pits Martians vs Humans, and each team has a variety of classes to choose from, each with different abilities, and some that need to combine with one another to achieve a goal.  For example, a Martian Enforcer can plant a bomb in the enemy’s base, but needs a Martian Warrior to place a detonator in it.  Each base consists of multiple components that work together, and when one is destroyed, the others are affected.  For example destroying a power generator will cause the ammo depot to resort to backup power, and produce ammo more slowly.  In short, the game should be an all out war, set mostly in post-apocalyptic settings similar to Extermination, Annihilation, and Impact.  The game will be a little faster than the typical tactical shooter, but slower than traditional deathmatch, and will absolutely require strategy and teamwork to win.  This game will also serve as COR’s initial pay-to-play release.  Now don’t fear, there will be a demo version of the game in which you can play on demo servers, but you will need to purchase an account to play on the premium servers and maps.

Recently, there have been rumors of COR Entertainment reviving and revamping the Alteria franchise.  The rumors are in fact, true!  The game will be rebuilt from scratch using the CRX Gaming Engine, and there actually has been some progress made.  This will likely also be a commercial release, with possibly a shareware/demo version available initially.  This is a very major project, and while it will onalteria4_mly be sporadically worked on/updated until Alien Arena Tactical is released, I do expect this to come to fruition in the next couple of years.  The goal is to keep the original game play elements intact with only minor modifications and some improvements, but to create all new game content, though some characters/levels will certainly be based on the originals in concept and design.  It will be quite a different game from Alien Arena, though I do intend to explore co-op type game play modes.  I am very excited to recreate some old worlds using our new engine tech, and I think I will surely enjoy making something that is truly demonic and horrifying in a fantasy style setting.  Think of Diablo meets Lord of the Rings…only bloodier!

Well, that’s it for now, I will continue updating this blog as time permits, and I’m sure I’ll be showing off a number of things from our two exciting new games, as well as the engine tech that we are constantly pushing out.  Until then, get ready for Alien Arena 7.65, because it’s coming very soon!

Advertisements