Wednesday, September 17, 2014

Dota 2 Next Level Meta: Support Brewmaster

Moving on to one of those heroes that I'm not very good with it, but like to play nevertheless. Brewmaster has some of the best quotes in the game - so, at least you can listen to them while failing as a support.

Hero analysis

My prediction for support Brewmaster was not that strong, primarily because he needs to be in melee range to be useful, and he has some serious mana pool issues. In theory his ultimate is just as strong without farm as it is with farm. However, he generally farms items that allow him to get his ultimate off more effectively - items like Blink Dagger or in some cases Black King Bar. Another reason why Brew is played as a core has also a lot to do with his ultimate. If he gets to level 6 early, he can dominate early game fights in a very convincing fashion. Supports' levels are often delayed compared to those of cores, unless you are fanatically leeching experience from a safe lane carry. Jungling supports can also get quite fast levels, but unfortunately Brew doesn't have a suitable skillset for that. In other words, you will be playing a melee lane support who has no stuns.

So what does he have? Well, his first skill Thunderclap is an aoe nuke and a slow and an attack slow - but it only hits around Brew, requiring him to be in melee range. His second skill has higher utility since at least it's ranged. The miss chance can be useful in the same way as Bane's Enfeeblement, to deny last hits from an enemy carry. However you need a substantial amount of mana regen to spam it. The slow is useful for chasing, allowing you to catch up and use Thunderclap. His passive is not all that effective for a support. It's kind of useful if you are facing a melee hero, as it allows you to trade hits more favorably. Most of Brew's effectiveness will still come out of his ultimate Primal Split, which gives a staggering amount of teamfight control and damage, especially in the early mid game. As a support you'll have to rely on walk-in ults at the time the skill itself is at its strongest. Obviously this is not very effective.

I think maxing Thunderclap first is still the way to go. Grabbing that one value point of Drunken Brawler should also be helpful, but beyond that you might want to max Drunken Haze second even though its scaling isn't that hot. That 75% miss chance at level 4 is still going to be very annoying for enemy carries to deal with, especially if they don't otherwise want to build either of the bars. The ranged slow is also going to be helpful whenever your ultimate isn't up - as a support you should expect to be fighting quite a bit more without it.

Items

I don't think there's much reason to deviate from the standard Brew build. Just go for that blink, even though its timing won't be very good unless you're somehow racking up kills. Getting an Orb of Venom after boots could be a good idea to improve your chasing potential. Buying wards is fine to some extent, and thanks to your ultimate carrying a gem is a lot safer than it would be on many other supports. Of course as a melee hero you need to bring a Quelling Blade to actually deward cliffs. If at all possible, just try and save for that blink. Grabbing arcanes at some point should be beneficial. If it's hard to get a blink, going arcanes first at least allows you to get more out of your other spells and still have mana for Primal Split. Beyond blink, just get whatever seems the most useful and attainable. Vlad is a good option with lots of utility as is Force Staff (even more mobility!) You could also build for a Hex, but actually getting there might be difficult. Other useful options include aghs and refresher, to get more out of your ultimate.

Game plan

Well, this here is the hard part. Brew is a rather useless lane support, and his ability to set up kills is nonexistent. If there is another support on your lane, or your carry can initiate, then Thunderclap is a decent followup nuke and slow. The same goes for ganking other lanes. You always need someone else to initiate - the best you can do is help them a bit with Drunken Haze. With 800 range you have a good chance of getting it off before the enemy realizes they're getting ganked, at least if you're smoked or it's night time. It might be possible to roam with, say, level 2 or 3 Brew this way, especially if you get that Orb of Venom. He's pretty bad at protecting a carry, and equally bad at harassing enemies or getting kills on a lane. So roaming is probably what you should do after leeching that second level either from the lane or from pulling. Beyond that it's all about getting those kills.

After finally getting to level 6, you'll be playing mostly like you would as a core who had his farm shut down hard. Because you have to rely on walking in to deliver your ultimate, you're very susceptible to silences and stuns. Two of the brewling are quite fast however, so you could just cast split a bit further away and then initiate with the water panda from invis. This does waste precious seconds, but it's better than dying without getting a split off. Using your ult to counter-initiate is another good way to get a lot of utility out of it. Just use it as soon as the enemy jumps in to screw up their plans. You should obviously avoid being the one in the receiving end of their initiation however.

Experiences

Based on one bot match, I feel like I was mostly right. Although his ultimate in itself is the same whether he is support or core, it is indeed slightly harder to deliver. There is however another factor involved in making him a less desirable support: it's generally ok for cores to rely on a long cooldown ultimate to be effective as they can farm while it's down; in the case of supports, they generally are required to be active even while their ultimate is on cooldown. It is these in-between periods where Brewmaster is very weak compared to other melee supports like Earthshaker or Treant Protector. While his two spells have a certain degree of utility, they are inferior to true support abilities. In particular, the only thing he can do from any sort of range is Drunken Haze, and while the slow is nice and the miss chance is significant, they don't do all that much to turn a fight around.

Having to walk in to use your only nuke really hurts, and it's only a slow. This puts Brew into rather significant danger and you typically can't ever use Thunderclap if you also want to use your ultimate. Another huge problem is his abysmal mana pool. A lot of time is wasted going back to base. In situations where the slow from Drunken Haze is enough of a gap closer, he can gank pretty effectively. If you manage to land a Thunderclap followed by a couple of poison attacks, most enemies should not be able to get away. However, Drunken Haze is not the best gap closer and generally requires a smoke in daytime to get close enough from a suitable angle. His burst damage is also quite weak with just one 300 damage nuke and a guaranteed double damage crit.

Support Brew is therefore a bit too greedy. While he can theoretically be very effective in teamfights, the in-between will be tough. His levels are also very likely to be late compared to a core Brew, and even more so his blink. I stand by my initial judgement, placing him in the poor category.

Tuesday, September 9, 2014

Tales of Xillia/Graces f

Getting lazy again and doing two games at once! Well, there's that, and the fact these are quite similar in many ways. It's a normal series thing, kind of like Final Fantasy before FFX and later games started messing around with things more. So yes, both of these games have the same fundamental game mechanics and core systems. The biggest differences can be found in character development, combat system details and various subsystems. Tales seems to be one of those series that are kind of reliable, but nothing truly amazing ever seems to come out of it. That's just my impression, and I might be terribly wrong. It does, however, definitely hold true for these two games I have played. One distinctive feature of the series is that you can play battles with up to four players. This makes it ideal for social RPG marathons because there is no need for taking turns holding the controller. Which is why I picked Xillia up initially in fact.

I don't feel like doing a complete analysis. I'll just go off on random tangents instead. Wherever that may lead.

1. Detour gaming

Rant time! This applies especially to Tales of Xillia, as the game is my source of inspiration for writing about this phenomenon. The point is still valid for many games, and even other works of fiction. Xillia is just particularly obnoxious. There's some minor spoilers about the game's structure, but nothing specific really. The game starts with Milla trying to destroy a device. She fails and has to retreat. Then we get to follow how she attempts to get back to the lab, to destroy the device once more. However, there's a but. One does not simply walk into... well, might as well call it Mordor. The next twenty or something hours of the game are basically spent running into yet another obstacle that ruins yet another way of reaching Mordor. When we're finally down to the very last option, the game still manages to throw several hours' worth of hoops for us to jump through.

In other words, the story does not actually proceed anywhere in the first twenty or more hours of the game. There is literally no progress in the plot. There's just an endless ball of side plot threads, all of which are incredibly pointless. They just have to get the player to visit every corner of the world, no matter how lame the excuses to do so are. The first half of Tales of Graces f feels a bit like the same in the sense that it too has the player run across the world map before the plot truly starts moving ahead, but at least it moves. Xillia just kind of spins in place. Ironically, once it actually gets somewhere, Xillia's plot is fairly good. Naturally once it gets to the good bits, it starts to move too quickly. It's kind of the same for Graces f, but at least it is somewhat more balanced with its pacing.

I stopped watching The X-Files because the series just throws bucketloads of fillers at the viewer. Same thing happens in anime a lot. Typically, not only does the main plot stand still for the duration of an entire episode, often interpersonal relations of the main cast do not develop either. Even if the individual filler episodes are great from time to time, it is still really annoying when the fiction does not grow. I love growth. These days I very rarely watch movies at all because they're hardly comparable to the massive arc of growth offered by extended works of fiction. I also expect this growth from such fiction, and get upset when it fails to deliver, choosing to unload filler after filler instead. That said, I haven't really found anyone who actually likes the kind of structure Tales of Xillia offers. So perhaps this is not just a personal quirk of mine at work in this rant.

The incredibly subtle parallel here is that I consider the hoops (subplots) in Xillia as something very similar to filler episodes. I think it's a fair comparison. In both cases, the main characters get diverted from their primary task into doing something less important (although technically it supports their main goal because it's an obstacle). The subplot often carries no weight beyond its own time frame either, which makes it a completely irrelevant event once it is over. It also often involves grinding down a horde of enemies that otherwise would have been left untouched. Yes, I do realize that by playing JRPGs I literally signed up for killing endless legions of enemies, but I'd rather encounter them while getting towards a worthy goal - otherwise it feels like I'm eradicating them just so that marketers can claim another +10 hours of gameplay in the box.

Games don't even need to necessarily have any stupid detours. Taking forever to get anywhere is just as frustrating, but it's a subject for another entry. Overall, I do admit my reaction to detour gaming these days is a lot stronger than it used to be in the past. The key factor here is time, or rather the ever-ironic relationship of time and money. Way back when I had all the time in the world, my ability to actually buy new games was rather limited. Back then a game had to offer a significant amount of gameplay hours to justify paying a full price. This made me a lot more tolerant. These days as I have a stable source of income and money to spend, the very fact of my employment kind of cuts into my gaming time. This has lead me to value my time all the more, and I have started to expect the same respect for my time from games. So, I feel a bit offended when they do not.

Perhaps it's a sign that I should finally ditch my long-lasting love for JRPGs. Or at least stop playing the titles that seem okayish, not great.

2. Real-time battle, almost there

The Tales franchise sports real-time battles, and they have been doing so for quite some time. At some point they switched to 3D battle fields, but to my knowledge the basics have stayed roughly the same throughout the series. As we have discussed in previous entries, real-time battle systems generally need to somehow control the flow of battle so that players cannot just mash buttons and hold enemies in stagger infinitely. In this regard Tales is a bit chaotic. Basic attacks generally don't offer much stagger and the stagger resistance of enemies didn't seem constant, or was dependent on their own state (e.g. they cannot be staggered mid-animation). I am not sure how exactly stagger worked in these two games - the system wasn't really all that transparent.

Of these two games, Graces f felt a bit more fast-paced, with more emphasis on tempo control. Although the games feel very similar on the surface, a number of differences contribute the difference in feel. For instance, Xillia controls ability spam through the equivalent of mana. Graces f does no such thing; instead it has a different point mechanic that controls the length of combos. It's quite similar to the system in Star Ocean 3. The end result is that while it's possible to hold an enemy staggered for the duration of an entire combo, at some point you'll simply run out of steam and have to wait a few seconds to regenerate combo capacity. This is the battle flow control mechanism that puts forces the player to play defense for a bit instead of brainlessly mashing combos in. In Xillia mana kind of acts as a similar mechanism: although repeatedly casting spells can hold enemies in place, you'll eventually run out of mana.

I did play most of Xillia as a caster so I cannot attest how it works for other character types. Overall I felt that the amount of control players can have over enemies is in similar levels; perhaps slightly stronger in Graces f. On the other hand, the amount of control players have over their allies is slightly higher in Xillia. In Graces f all other party members are almost completely autonomous; in Xillia, the player can choose to link their character with another character to have them work as a pair. Since we were playing with two players, we both had a link partner, whereas I'd imagine there would be more switching involved when playing solo because links between different characters provide different benefits and combination attacks. Both games have the option to bind a few abilities of non-controlled characters to buttons, allowing the player to invoke them when needed.

Overall I feel there's a bit of a redundancy problem in both games. They have a lot of depth, but fall short in utilizing it. With 3 AI controlled characters in the battle field, a lot of stuff is simply out of the player's hands. For instance, in Graces f you get quite significant bonus damage if you hit all of an enemy's weaknesses in one combo. There are so many different weaknesses that a single character can rarely do so, which means you have to hope that your allies have the sense to fill in the one's you cannot hit. It also requires a lot of memorization because attacks can have up to four different attributes, and the player has a combo tree of 16 attacks and another 4 attacks on top of that. You can always check the attributes in a pause menu, but it gets tedious. Xillia reduces the amount of available attacks drastically, and it also simplifies the weakness system to a more familiar element based one.

Another difference that may seem small but vastly influences the battle flow is free run. An interesting feature in Tales is that normally characters move in a 2D line in respect to their target (i.e. just near or far). Free run can be activated with a button, and it allows a character to move freely. In Graces f, free running depletes combo capacity, limiting it use heavily while in Xillia it's free. It follows rather naturally that you'll be using it a lot more, because it's the best way to avoid enemy attacks. In fact, since Graces f has a quick step dodge (also costs cc), there is very little use for free running. The end result is that Xillia involves a lot more running around and crossing distances is fairly quick. In Graces f evasions are done in a tighter space, and covering larger distances actually takes quite a bit of time. Xillia also has jumping which Graces f does not, and it seems very useless and mostly annoying. Jumps are slow, which makes accidental jumps quite dangerous.

I slightly preferred Graces f's gameplay, primarily because its controls are more stable. Accidental jumps aside, Xillia also has an annoying habit of characters not facing the right way after free running, which results in a lot of spells cast off-target. Neither of the games get to the same level of enjoyment as my long time favorite Star Ocean 3. Tales combat feels quite entertaining, and most of the time it's fast - a massively important factor in a game with hundreds of battles. Unfortunately enemies do not vary all that much, which makes stuff a bit repetitive. There are some alleviating factors however, and we're about to move on to them.


3. Character design musings

In both games, all characters have their own unique mechanics. I find this commendable, as opposed to games where all characters can be made into carbon copies of each other. These mechanics felt more emphasized in Xillia though, and were not that big of a factor in Graces f. There clearly had to be a reason why I chose to play just one character though, so let's look into this a bit. First of all, one problem I have with both games is that all character upgrades come in very small pieces from a system that has hundreds of nodes to activate (the systems are a bit different, but the rewards are not). A +3% conditional bonus here, a +2 stat bonus there etc. This means it's really hard to actually make any sense of a character's play style and advantages in just a glance, and it takes quite a lot of fights to grasp it in practice too. Arguably this could also been seen as a strength.

There is however something to be said about the actual impacts of different playstyles on characters. If you look at Dota 2, there's 108 different heroes and all of them have a distinct impact on how the game plays out. Meanwhile, in tales, it felt like I could have any party of four characters and the combat experience was no different - unless I would have purposefully avoided taking a healer with me, but that's just stupid. So, while each character may feel a bit different to play, the overall combat flow remains largely unchanged. I acknowledge that I did not particularly explore my options, because I largely had a "whatever works" attitude. The reason for this is the lack of variety in enemy design - the game very rarely forces its player to adapt. In Graces f I also didn't feel like memorizing a new combo tree too often so I stuck with the main character.

I don't often change characters just for variety. I only change when the game throws a curve ball my standard roster cannot handle, or if I feel combats are going too slowly. Sticking with the main character usually works well enough, because they're generally designed to be quite straightforward to play and still effective, and also the most well-rounded. I don't like playing healers or supports that cast buffs in real-time combat systems, because usually the AI is actually better at these roles (healing a low HP character or refreshing an expired buff are both very simple rules) - and it's usually more boring. In a cast of six characters there's typically one healer, one buffer and four others who usually do either physical or magical damage (or very rarely both but they're usually worse than focused characters). This leaves four feasible options to choose from.  

Because I don't trust AI in these games, I generally aim to play the character that is the most crucial for my strategy to work. Admittedly it's a bit unfair to make comparisons to Star Ocean 3 because I sank some three hundred hours into that game and experimented with everything. However, even on my first playthrough I felt more pressured to change my controlled character throughout the game. It would seem that for all their unique mechanics, characters in either of these two Tales games don't really have that different impact - some are just more tedious to play than others. It kinda comes down to the fact that there are just multiple ways of doing the exact same thing. If you compare to something I have hyped less, like Xenoblade, even that game had more pressing selection criteria for characters because they actually did different things.

Might be this is just my impression, and the system actually offers more than I got out of it. I also do commend the effort. It just really feels like - as long as you've got the basic requirements covered - there really isn't much of a difference between party configurations. It's also worthwhile to remember that it's not necessarily a problem with character design; it could just as well be a problem with enemy design. Ultimately it's always the game's challenge that drives players to adapt and make good use of their options. Certainly there are players out there who appreciate options just for the sake of having variety, but for me it feels like waste of interesting character design. In one regard Tales does fine: non-active party members still receive a lion's share of experience, so they at least do not fall behind.

Conclusion

Although there are a lot of individual minisystems in both games that are kind of curious, I will cut this post short here. Overall both games are decent JRPGs, but not really special in any regard. This really hurts them a lot, because mediocre JRPGs seem to be all we're getting these days - especially on home consoles. They're hurt by sloppy storytelling. All of their good ideas also kind of blend into a grey mix where nothing seems to really matter a whole lot. The end result is an experience that is kind of bland until you have put enough hours behind to develop some affection for the characters and their antics. This is actually another commendable thing about the series: it has a lot of (optional) banter between characters. Although it's quite cliched, it's highly entertaining and brings life to the otherwise mediocre characters. Some scenes - particularly in Xillia - are just downright absurd.

Overall, I might have loved this series a lot more had I played it like ten years ago. It was entertaining enough for these two games, but I'm not really dying to get more. Maybe for a full 4 player cooperative experience? Xillia 2 is out there, but it seems like yet another mediocre game, with just one benefit: recurring characters - but that's a topic for another day.

As a final note, maybe I should re-visit Star Ocean 3 and see if it's actually as good as I remember it to be.