You only have to click once, fool!

The Howling Abyss, the All Random All Mid (ARAM) map and game mode is, as of writing this, a fairly recent addition to League of Legends. Matches are usually pretty quick taking roughly 20 mins on average. If you like teamfighting and generating gold faster than a sloth picks its nose, then ARAM is the game mode for you. Having taken an extended break from LoL for just over a year and a half, I find myself mainly playing ARAM and little else these days.

Admittedly this return was spurred by a number of things, one of which was a desire to see what has changed in LoL, coming back to what honestly felt like a completely different game. Another was the fact that recent LANs demonstrated that it is still a fun game, provided you don't take it overly seriously and become obsessed (like I did for a time). Finally, and possibly most importantly, several friends of mine, both old and new, are playing it as well. These friends, one of which I have watched become incredibly competent at LoL in the last few months, are the real reason I bother to play LoL at all these days. If not for them LoL would still be sitting in a proverbially dusty closet even now. Which would be a shame because presently, LoL is a hell of a lot of fun.

Anyways, ARAM. This acronym based brunt of a terrible, repeated household mum joke is something that is bordering on being overplayed. I have made a frag video which, if anything, will give me reason to maybe ease off a little bit or even explore the other game modes I liked a bit more (Summoner's Rift, Dominion). Although, knowing me, this is probably an absurdly quixotic thing to expect.

As per normal with any video on this blog, I recommend watching it in 720p.

The video itself I am ... how should I say ... slightly disappointed with. I have been video editing a LOT lately due to my other work commitments so I maybe didn't quite find the energy to make this a serious production. While I am usually the first one to criticize my own work, I think maybe due to the nature of ARAM, the frequency of truly awesome, video worthy occurrences are incredibly rare. You are almost constantly teamfighting/poking, which while fun, is maybe not the best spectator sport for the average person with ocular sensory organs. For the most part, things that occur in ARAM are an absolute clusterfuck at best and discerning what is happening amidst the chaos is probably quite strenuous on first watch. The choice in music is a new experiment for me, but honestly something I have tired of having listened to 50+ times while editing. I also realise that most of the frags are performed by me which is probably a bit boring. Alas, that is the nature of the recorder also being the editor. Still, I hope the video portrays some of the amusing moments that have occured in ARAM and that will hopefully continue to.

ThreatBot

Like the game content I produce for personal or tutoring purposes, I have been incredibly reluctant to bring up my PhD research in any public place. Even a blog. However, as my PhD is (hopefully) coming to an end in the next few months, I do not see the harm in bringing it up in the form of a video I made for conferences and speeches.

My PhD is based around modeling certain player-like behaviour that is lacking in FPS bot AI and implementing it in games in a way that does not fundamentally change how typical AI systems (e.g. Finite State Machines) work. Within a plethora of possible player-like activities to improve on, such as my honours focus on a bot aiming system, I chose the somewhat vague and confusing idea of the 'threat' that players experience in games. For some reason, some people have difficulty understanding and swallowing what it is that I am talking about when I talk about threat in games. Maybe it depends on the type of games you play and how you think you play them. Hopefully this video shall aid in the digestion of the topic.

While this video talks primarily about the theory and design of the bot AI I have created, it should be noted that the entire game environment the AI exists in was designed from scratch too, as well as the default/base AI behaviour. This was not a game I just dumped my finalized ThreatBot AI into over an existing AI system, but developed from the ground up with attention to detail. This is a personal design choice, to see if a game can be developed from scratch to support the technology, but also to practice solo game development. Even now, with Australia's failing games economy, it is something I would love to get into.

The one percent

I have not posted anything in awhile. I am not sure if it has been because I am too busy IRL or if I am slowly tiring of even this meager means of anti-social narcissistic self expression (as opposed to social types such as Facebook). Part of me feels I don't really need to do this, but another part realises that its good to have something random to reflect on many months/years from now.

Nevertheless, I have a backlog of stuff to talk about. The first off the list is, again, Diablo III. Diablo III has copped a lot of flak from fans for many reasons in the past year. In my opinion, some of those reasons are legit (always online DRM, RMAH) and others are just people jumping on the jabberwocky bandwagon of wanting something to rip on. However, some reasons are ones that I will never understand, such as complaints about the game's story and not getting upgrades to your character consistently and immediately. I'm not saying they are not poor systems, I just could not give a hoot about them. Diablo games have always been about the gameplay for me, and Diablo III does not dissapoint on that front. At least in my opinion.

Anyways, enough of me trying to justify playing a game everybody hates. If I do any more of that I will have to start asking myself why I do it. A few months ago I decided to level a hardcore character. I am not sure what inspired me to do so, especially on an Australian connection (lag, disconnects etc). Maybe having leveled all 5 classes to 60 made me realise that there was only one area I had not tried. For anyone unaware, the Hardcore game mode in Diablo III is the same as the Hardcore difficulties in previous Diablo games. You get one life. When you are dead, you are dead.

I of course chose a Barbarian to play, the class I am most comfortable with and had the most experience playing. It is strange just how differently you approach everything in the game when you realise that death is permanent. I found myself gearing very differently than my previous Barbarian, focusing heavily on defensive and sustain gear at an early level and running abilities to buff my damage. Even with my knowledge of the game, I would find myself jumping into horrible situations and nearly dying at least once every 15 minutes. It was a much more interesting and tense play experience. As the hardcore AH is not the same as the softcore one (for example there is no real money option), the economy is completely different too. People understand that dying is very real and paying outrageous sums for gear is nonsensical. However, while stuff is usually cheaper, the supply of it significantly less (i.e. far less people playing hardcore). Taking into account the risk of farming of gold and items, the hardcore AH is much more balanced, but used sparingly due to the incredibly small amounts of gold people have. Upon reaching 60 and Inferno difficulty, amazingly my female Barbarian had over a million EHP and enough damage and sustain to take on the majority of the content.

Needless to say, Diablo was an absolute pushover. I think taking what I have learnt in softcore and my own personal preference for maintaining a balance between damage, sustain and tankiness was the exact criteria needed for finishing hardcore and killing Diablo. I will probably return to hardcore again in the expansion, Reaper of Souls, hopefully coming sooner rather than later.

 Note: The title of this post comes from the fact that less than 1 percent of Diablo III players have killed Diablo in hardcore Inferno. By comparison, only 6% have cleared Inferno in softcore. Amazing...

Pain is weakness leaving the body

Once again I have had the wonderful opportunity to face off against my colleagues in Unreal Tournament 3 at uni. This time with wires attached to read my sweat and pulse readings in multiple four player deathmatches in the map Deck (:D) with an armor and health giving catchup mod enabled. It was an interesting experience, especially as you have to sit rather contortedly and cannot move your hands as freely due to the sticky pads attached. Nevertheless, I enjoyed the test setup and hopefully one of my colleagues has gained some useful data. Unfortunately there still seems to be an apparent skill/experience divide, at least in UT3, which may have broke the mod on occasion (i.e. mod stacking 2-3 times more than it should), definitely noticeable at times. Plugging 5 rockets into someone and having them still standing was ... rather surprising.

Nevertheless, here is how one of the matches transpired. I would recommend watching it at 720p as UT3 looks god awful at low res:

Again I must apologise for the shoddiness of UT3's demo recording capability, especially client side. There is unfortunately not much I can do about the jumpiness and inaccuracy (aiming, audio repeating, some audio not working at all) of the replay as it seems to be a problem with the recording process. There are also no tools I know of that would allow me to play the file back more smoothly. My assumption is that it doesn't record scene-graph data anywhere near frequently enough, perhaps only 20 times per second. As a result you end up with very jittery playback.

Riding the Walrus

Diablo III has become a chill out game for me. Whether I am working or simply doing nothing, alt-tabbing in to kill some demons is a great way to ponder the universe and attempt to understand what it is I am doing regarding my PhD research. However, Diablo III is many things for many people. For some it was an utter disappointment, a narrative letdown (something I will never understand) that did not fill the shoes that Diablo II left. For others, it is a travesty of always online DRM (something I do understand), and should have offered an offline mode straight out of the box. Whatever the case may be, I feel it would be hard to argue against what Diablo III gets right. In my opinion that is its raw game play mechanics.

Never have I played a game where so much investment, research and physical execution and ability are required to make certain things work. Sure, you can slap a whole heap of spells on your toolbar and faceroll your way through the early game content merry as can be. But come late game content, you should be asking yourself both whether this is efficient and if you can possibly make this process better.

The most successful and efficient builds come in the form of ones that eliminate the need to run a primary resource generator. For the inexperienced, this would seem like heresy. "How on in the hell am I supposed to use ANY abilities if I don't have a way of generating resource?"one may say. "Easily!", would be my response. You just have to actually look at the spells/passives you are using and work out how they can best synergise with others. Observe:

In the video I am using two handed builds typically to show how they can work with the much more difficult gearing path compared to dual wielding (i.e. no bonus 15% attack speed). Usually this requires more crit chance or lifesteal to make up for the less frequent sustain you maintain.

Fight Night 2

Recently hosted another Fight Night ... night. This time with more games and different people. Overall it was a fun time (for most) and, although anything resembling skillful play was absent from all players (including myself), some entertaining moments did arise. Such an expectation is not something that I should really hold however, especially regarding the perspective the majority of gamers seem to have on fighting games. If you don't dislike them you will probably just regard them as fun party games for when people come round. Which, ultimately, is better than nothing.

Zerg Rushin

It has been quite awhile since I finished a RTS game's campaign. Hell, it has been awhile since I finished a game in general. Life at the moment is not permitting much time for games besides the occasional coop vs AI game in League of Legends, something that does get stale after awhile. Anyways, Starcraft II: Heart of the Swarm has, as of writing this, come out less than a week ago and I have already dedicated enough time to it to blitzkrieg through the campaign on normal and occasionally hard difficulties. My overall judgement? It's good. I'm not sure why it took them 2.5 years to develop but hey, this is Blizzard we are talking about. They were probably high on cocaine and wading through strippers for 90% of its development.

Nice view

What did I like? Well first of all I found the story to be quite interesting this time round (compared to Wings of Liberty) and not just because it was to do with my preferred Starcraft race. I loved the music, never really getting the space cowboy theme of the previous Starcraft games. While it is true I have a fascination with things that look positively deadly, I was honestly not expecting the amount of characterization and development given to the Zerg lieutenants Kerrigan encounters. They actually seemed like a cool group of ... monsters, some of which I wouldn't mind hanging out with (if that were possible). While the campaign is a little linear (strange complaint for an RTS I know), it is not boringly so as you do feel like you are amassing your Zerg army for a final showdown.

Fashion these days ...

It is obvious they have wanted to make the control of the Zerg a bit easier, at least for the campaign. A race that could be unsurprisingly confusing to control has had some of the multiplayer/original impairments lifted. Hatcheries of all kinds now generate up to 12 larvae without the need for a queen to keep ejaculating on them. Queens can now also roll with the swarm, unimpaired by non-creep terrain, healing units in the mass like bosses. The use of F2 to instantly select all offensive units is also a godsend. Similar to the WoL campaign, you can improve certain units (some unique to the campaign) to be more proficient at certain tasks. Some of these are incredibly well thought out, turning for example a new Zerg offensive siege unit, the Swarm Host, into the ultimate defensive unit en masse. Through the use of readily changeable mutations and one time only evolutions, I turned my Swarm Hosts into units whose locusts could attack both ground and air as well as tunnel to any location on the map that has creep. This kind of clever game design allowed me to make a small pack of Swarm Hosts that could move to defend any threat (including deployed siege tanks) anywhere I had a base in seconds. Brilliant!

In Soviet Russia ... rush Zergs you!

Unlike my Swarm Hosts however, most of my other units were mutated and evolved to serve two particular purposes: durability and/or numbers. Zerglings that were tougher and spawned in threes instantly, Roaches that took less damage and created roachlings from kills, Mutalisks and Hydralisks that could regenerate health or take a heavier beating. You get the idea. My dealings with Effective Hit Points in many games (particularly Diablo 3) has convinced me that it is the most efficient way to play. Surviving longer, perhaps indefinitely, results in the dealing of greater damage. If that damage can be split amongst tens or even hundreds of units that never die, well then you have a force to be reckoned with. Nay ... a swarm.

Gotta catch em all

However all of this would not be worth much if not for the abilities they have given Kerrigan. A decently thought out set of abilities can make base/unit building and combat quite interesting, depending on abilities chosen. You could potentially turn Kerrigan into either an autoattacking or ability using killing machine should you please. I however chose to base Kerrigan around keeping my swarm alive for unreasonable lengths of time (AoE heal), as well as providing decent damage when necessary. Her passives were also built around making an army as quickly as possible through increased efficiency in drones and overlords. For her ultimate ability I of course chose Drop Pods (easily the best of the three) for even more Zerg numbers.

Why should I ... care again? Geddit?

The end result was a Zerg swarm that could literally pop up in minutes and survive against even the most ridiculous encounters. Observe.

Ultrakill LAN 2013!!!

From the 11-13th of January, 2013, I hosted the third Ultrakill LAN I have ever hosted. While I personally was a little disappointed (for some reason), most of the participants reported an enjoyable time. It was also the most expensive, largest and undoubtedly the hottest LAN I have ever hosted at $160+ (not including electricity), 14 participants and 37 degrees celsius respectively. We even managed to overload one of our 15A circuits, blowing two fuses and needing to buy new one's mid LAN. Fun times.

The following video is only a small fraction of the games that were played, but probably were the most enjoyable overall. As usual, the perspective in the video is mainly mine. However there are plenty of instances of me getting killed by almost everyone else who attended.

 Because of the insane heat, I have decided I will not be hosting Ultrakill LANs during the summer anymore, moving them pretty well directly in the middle of winter. Depending on how things go, I may host one this coming winter, though probably on a smaller scale.

The value of teamwork

After trying several times in the past to get people to play the MvM Coaltown Halloween event map in TF2 and failing because of intense lack of interest (something I didn't and still don't get), I recently tried again at a few organised events. One was an afternoon of gaming with colleagues at uni in the GRID Lab. The other was at someone's minilan. During neither attempt was the team good enough to actually finish the map, which I found a little disappointing.

It is funny then that after just randomly scraping some people together in a couple of hours that I and a group of acquaintances managed to tear up the joint with reasonable ease, even after a start which was not even meant to be serious. It just goes to show the difference good players make when it comes to Coop gaming, when I guess 2-3 people aren't having to do all the work.

I have now come to the conclusion that the MvM event map is actually pretty easy with the right crew and as long as everyone does their job without dying much, fun and awesomeness can inevitably ensue. Speaking of fun and awesomeness:

In the video I make extensive use of the Black Box rocket launcher. Having been gifted a Beggars Bazooka, I can honestly say I am torn between two worlds in terms of using it. While the  Beggars Bazooka is undeniably the highest dps rocket launcher available to the soldier, and while getting to roughly the desired location on the map via huge chamber-chained rocket jumps, the Beggars loses some of the precision and timing I am used to with the Black Box. Not to mention the lack of life on hit, something I have pretty much adjusted my playstyle to make extensive use of.

Getting to a new tactical position quickly and easily or even jumping over incoming ordnance is something that can be done quite safely with the Black Box. After extensive use, I have come to realise that the sky is essentially a safe haven, allowing you to do maximum AoE damage from above and heal yourself with the Black Box's life on hit. It is as accurate as you are with the crosshair. There is also nothing better than rocket jumping from ammo pack to ammo pack, crit buffing teammates and dealing death below you in a precise manner.

The Beggars Bazooka on the other hand lacks that precision of aim (random spread), timely support and accuracy of mobility, but as mentioned, makes up for this in both sustained and burst damage potential. You can also fly, once you have figured out how it propels you. There is nothing better for traversing the entirety of the map, albeit in a rather reckless fashion, or for taking down giants of all kinds. While anything at range will give you trouble, anything in your medium range kill zone is going to be pounded by rockets, though not really from above. Standard rocket jumping is a bit awkward (different timing) and not as reliable to pull off in the thick of the fray.

Really I see uses for both rocket launchers. The Black Box I will use if we have heaps of damage (4 or more), little healing, but teammates that could benefit from timely, consistent crit buff support no matter where they are on the map. The Beggars Bazooka I will use if we severely lack damage, and I need to fill the role, possibly with medic/engineer support to make up for the lack of quick mobility and sustain. Ideally I would prefer to have both worlds, but you can't have everything.