Marsellus the Wise

i play hunt down against macgyver’s profit

[video mp4="http://starwarsccg.org/podcasts/tournament1game6.mp4" height=360 width=480]

i play trm against brother justs’ mistryl

[video mp4="http://www.starwarsccg.org/podcasts/tournament1game5.mp4" height=360 width=480]

despite what i say in the beginning, run time for this game is 1:31.

[video mp4="http://www.starwarsccg.org/podcasts/tournament1game3.mp4" height=360 width=480]

[video mp4="http://www.starwarsccg.org/podcasts/tournament1game2.mp4" height=360 width=480]

[video mp4="http://starwarsccg.org/podcasts/tournament1game1.mp4" height=360 width=480]

i’m going to tell a story. a story about a skill called sand shards. before i get into the story i need to give a good amount of background information. it is my hope that i can present this information in an interesting manner. if you are a fan of strategy games (as i suspect you are) i think you will find it interesting. this story took place almost four years ago, when i was playing a computer game called guild wars. guild wars is an mmorpg specifically balanced for player vs player (pvp) gameplay. unlike most other mmorpgs, you can start a pvp character at the maximum level with the best possible equipment. there is no need to spend hours leveling a character and waiting for the uber hammer of pwning to drop: the pvp is readily accessible with a minimal amount of time spent. so if your team is crushed you can be sure that it is because they are better than you, not because they have superior characters.

it also means that the game balancers primary concern is with skill interactions. before entering a match, each player must select 8 and only 8 skills for their character from either their primary or secondary profession. for example, a warrior/monk would most likely take something like 7 warrior skills and 1 monk skill. almost all competitive gameplay is done in 8 vs 8 format and there are no party restrictions. you can bring 8 monks if you feel like it, but doing so is a recpie for disaster. the game was balanced in such a way as to favor groups with a fairly even distribution of professions for maximum utility. in 2006 when this occurred, at any given time there were probably something like 5,000 people playing pvp. so if there are some skills that are significantly overpowered, you can be sure the player base would catch on fairly quickly.

so this story occurs during one of the beta release weekends for the game’s second expansion: nightfall. the development team knew that their internal testers could not possibly catch all broken interactions prior to release. so they would release certain aspects of the game to the public because 100,000 eyes catch things that even the most assiduous testers and developers could not. of particular importance here, one of the new things they were introducing was a character class known as the dervish, which had at its disposal a skill called sand shards.

at first glance, sand shards did not look like a very good skill. it was an enchantment skill that the dervish could cast on himself. when it was active, if the character failed to hit an attack, it would do a relatively low amount of damage to foes in the area. now, this was kind of cool, because of course the opposing team would spend a lot of effort in making sure the opposing melee characters were unable to hit all their attacks. disruption was a vital part of gameplay, an unmitigated melee character could do a severe amount of damage. so they would often cast spells that make the opposing melee characters blind, or allow their own characters to block attacks, and so a melee character would only ever do a fraction of his potential damage. and of course the team with those melee characters would try to make sure their characters did hit with attacks, and leaving them hindered so they could get the marginal sand shards benefit just didn’t seem worth it. sand shards was just not reliable enough or powerful enough to make the cut when you can only bring 8 skills per character. i, and most other top players, wrote it off with a glance and focused on the more obviously powerful skills.

but someone somewhere looked hard at this skill and realized an unintended interaction. if you could intentially make yourself miss rather than relying on your opponent to do it for you, it becomes a lot more consistent. if you could make yourself miss repeatedly at an accelerated rate, it quickly becomes truly powerful. if you could do that sustainably, then you have something broken. late on the friday night of the beta weekend, someone i know observed a korean dervish behaving quite oddly. he kept using the same attack over and over again, an attack that my friend knew was not even hitting. but strangely enough, he was somehow doing massive amounts of damage. he went to the opposing team’s base and killed their npcs at an astonishing rate. so my friend watched more, and slowly the pieces began to come together.

by saturday morning he had discovered and improved the broken interaction. the trick was to pair the dervish profession with the assassin. the assassin is unique in that his dagger attacks are based on combos. most of his attacks have to meet certain preconditions or they fail. for instance, many offhand attacks have to follow a lead attack. and then most dual attacks (attacks with both daggers) have to follow an offhand attack. these dual attacks were typically the most damaging, so the payoff generally lies at the end of the attack chain. the intent was to make their combos potent but prone to disruption; they do a great deal of damage but if you are able to prevent an early attack from hitting the entire combo falls apart. so if your first attack is blocked, you can go ahead and press the button for the next attack, but the attack will automatically miss and you will have wasted your energy.

but since sand shards rewards you for missing your attacks, so this is precisely what the sand shards dervish wants. one of the attack skills in particular, exhausting assault, would fail if not preceded by a lead attack. so the sand shardser brought this without even bothering to bring a lead attack: he wanted the skill to fail. and because exhausting assault is designed to interrupt an opposing skill, it has a shortened activation time of only 1/2 second. and it is a dual attack, which means that by missing with it, you’re not missing just one attack but two. consequently the sand shards damage is doubled upon each activation. to top it all off, if an assassin attack skill (such as exhausting assault) fails it recharges instantly and can be used again right away. so you attack someone with this. it takes 1/2 second and misses twice, sending out sand shards damage twice to all the foes in the area. that takes 1/2 second. then you can do the exact same thing right away because the skill has no downtime. so what you would do is go up to a group of enemies and just hit the same key over and over again to repeatedly activate your exhausting assault twice per second, sending out 4 pings of sand shards per second. now we’re talking about a lot of damage generated completely reliably at an extremely fast pace.

but matches are long. if you’re activating a skill every 1/2 second you have to be able to afford to do so, otherwise you will run out of gas and a good team will adapt to the quick burst of damage and take advantage of the later lack of pressure. and exhausting assault costs 5 energy and the player only gains energy back at a rate of 4/3 per second. and the maximum energy is 40. so you can go full blast with this combo for about 5 seconds, and then can do next to nothing. but. enter the final piece of the broken combo, the elite skill way of the assassin, another thoroughly mediocre skill. at the time way of the assassin was an enchantment skill the assassin could use to make all of his dual attacks free. now in a character with lead attacks, offhand attacks, and dual attacks, it is not worth wasting the your one elite skill slot on something that will improve at best half of your attacks. unless, of course, you plan on only using a single dual attack skill (for instance exhausting assault, the 1/2 second wonder) over and over and over, and are interested in doing so without expending ANY energy.

so there we had it, a character who could reliably and easily generate huge damage at unprecedented speed, and could do so for a prolonged period of time. now we just had to put it to good use. and of course the best way to put it to good use is to use multiple copies of it. and so we played four of the dervish/assassins and four characters whose job it was to keep them alive. to put into perspective how damaging this setup was, i’ll give some rough numbers (i don’t remember the exact figures). sand shards, on each miss, did about 25 damage. and because with each skill activation the dervish was actually missing twice, he did about 50 damage right there, per attack, to everyone in the area. a normal attack does something like 30-50 to a single target only and takes anywhere from 1.3-2 seconds, depending on the weapon type. a normal heal takes anywhere from 1/4 second to 1 second and heals a single target in a range from 80-200, depending on the heal. and the average maximum health is about 550. so four dervish/assassins using sand shards could do about 400 damage a second to multiple foes in the area. the opposing healers just had no way to keep up. within a couple seconds everyone around was dead, and it was short work to track down and eliminate those still standing.

once we had it all figured out we gathered together some people to abuse it like crazy. we chose a format of play that would best suit the brokenness of this build. we quickly realized that one of the obvious weaknesses was that we were not especially mobile, and relied on multiple opponents being in the area of our attack. so we chose a format that did not emphasize mobility and that had particular areas on the map of paramount importance. opposing teams would have to face our sand shards because if they could not hold that specific area, they would lose. we chose to play in the hall of heroes tournament.

i apologize in advance for making another digression, but the way this tournament works is fascinating. it does not promote skilled play quite as much as the highest level of the game, guild vs guild combat, but it is just such a great idea that i feel compelled to share. also, it is relevant to the story. the tournament is essentially a single-elimination tournament that has been running almost uninterrupted since the game was released in 2005. you enter in your team of 8, and face one or two opponents in each round, with varying maps and objectives. if you win, you move on to the next round. if you lose, you go back to the starting area. so does everyone else, so you expect the competition to get progressively better the further you advance into the tournament. after about 7 rounds you enter the final area, the hall of heroes. this is a three team king of the hill map and the team with their “ghostly hero” holding the hill at the end of the match is the winner, stays on and reaps the rewards while the losers get sent back to the starting area. each hall of heroes match features the returning champion and two new challengers who have worked their way through all the previous rounds. we chose this format for the exploitation of sand shards specifically because the final match was king of the hill. you can’t pick your spots in such a match. you can cover the entirety of the hill with sand shards brutality. it is either face the music or throw in the towel, neither of which is a particularly attractive option. and once you reap the rewards of winning the tournament, you go right back to the final match again and play more king of the hill. the format rewards teams that are particularly effective at king of the hill matches.

so on this day we entered the tournament and crushed our first few opponents. the first rounds are usually pretty easy but these guys were falling in a matter of seconds. in our fourth round we found that we were not the only ones to have discovered sand shards. but we were smarter than they were and had prepared for that eventuality. we had brought along a skill to counteract the effects of sand shards: healing seed. healing seed is an enchantment that a monk can cast on any character, and whenever that character takes damage, he and all his nearby allies are healed for 25 or so. a powerful skill that is balanced by being expensive, having a two second cast time (for monk skills, even one second is a long cast, two seconds is an eternity) and a recharge that lasts three times as long as its duration. even so, at the time most teams in this format brought two copies. we brought four, and coordinated our seeds so that one was always up. so as long as the seed was up on one character that was taking damage, they were all getting healed, and they were being healed for an amount equivalent to the damage. and so we defeated this team and another more sand shards teams en route to winning in the hall of heroes a few times.

the metaing had begun. we had identified the overpowered skill, and we had brought it. but we knew it would get out, so we also brought the counter to the overpowered skill. sand shards was the meta. and we were a step ahead of the meta. but the meta moves quickly. after a team is eliminated they can make adjustments and go right back in again. by the time you reach the final stages of the tournament your build may already be obsolete. and so it was with us. by the time we had won a couple matches in the hall of heroes everyone there was using sand shards, but we were the best and we had the counter to the meta, so we won. the meta adapted and soon everyone was using sand shards and healing seed and it became a giant slugfest. in a three way king of the hill match where there are 12 dervishes all on the hill each sand shardsing each other there is little room for error. your offense could wipe within a second if you slipped up, and that’s exactly what happened to our enemies as we won again.

but the meta adapted even further. many sand shards teams dropped one of their dervishes for a necromancer utilizing well of the profane. well of the profane is a necromancer skill that can only be cast if there is a dead person around. it exploits the corpse and creates (for a certain amount of time) a large area wherein all opposing foes lose all of their enchantments and can not have enchantments cast upon them. this goes right to the heart of the sand shards strategy. sand shards is an enchantment. healing seed is an enchantment. many healing skills are enchantments. the well of the profane totally negates both the offense and the defense of the opposing team. it was, as we say in swccg, the gg card.

so, upon being soundly defeated by the first team of this kind, we knew we had to bring a well of the profane necro of our own. in addition to the well, we brought a skill called consume corpse. now consume corpse essentially does nothing, except, well, consume a corpse. the key is that each corpse can only be exploited once, and well of the profane has a 3 second casting time and costs nearly all of your energy. consume corpse takes only a second. so if there’s only one corpse lying around and you try to cast the well, a smart opponent will use consume corpse and get to the corpse first. because the well needs a fresh corpse the well of the profane cast will fail. so against teams without a necromancer you can cast the well of the profane worry-free and destroy them because you have the “gg card” and they do not. but against teams also sporting a necro it becomes a battle of wits. both players are trying to get off the money skill and win it for their teams. but they know if they go for it prematurely they may very well lose it for their team. so the two necros cancel each other out and both teams are left to beat on each other until one of the necros can get the all-important well off, at which point his team wipes the other and often wins in a matter of seconds.

and so the meta continued to adjust. within a couple hours a new metagame was born, shaped, countered, adapted to. people found counters to the meta. then people found counters to the counters. and it went on. sand shards was the focal point, and everything shaped itself around that. who knows where it would have gone from there; the developers made an emergency update to the game and sand shards was fixed to only work with scythes, the normal weapon of choice for dervishes. the build we had been using only worked with daggers, the assassin weapon, so this change kept the skill in line with intent and curbed the abuse. and the guild wars population moved on to exploit other, somewhat less broken, skills for the remainder of the beta release weekend. and the developers gained a lot of information and so when it came time for the actual release they were able to put out a quality final product.

so let me know what you think. did you actually finish this whole story? i’m interested to hear responses