Saturday, May 31, 2008

Being On The Path

Note: Sadly, because this post deals with the Season 4 PvP Ratings Requirement, I have the feeling that the comments will devolve into PvP vs PvE and miss my larger point.

The majority of Season 4 PvP gear will require the purchaser to have a personal rating of 1550 or higher. Even some of the gear obtainable through Battlegrounds will require Arena ratings. Many people, who thought PvP gear was too easily obtained for its quality, are happy about this, and many casual PvPers are unhappy.

It may surprise some of you--because I am in the camp which felt PvP gear was too easily obtained--but I believe that Rating Requirements are a bad idea.

The key is the concept of "Being On The Path" for endgame content. In nutshell, the number of people who reach the highest point of endgame is less important than the number of people who are working towards--and feel that they one day could achieve--that point.

For example, in PvE, the number of guilds in Sunwell does not matter. What really matters is the number of guilds who make it to Gruul and Magtheridon. Once a guild reaches Gruul and Magtheridon, they are "on the path" to Sunwell. Most of these guilds probably won't reach Sunwell before WotLK, but that doesn't matter. What matters is that they are steadily working on content, and feel that they could reach Sunwell eventually.

As long as a player is "on the path", everything is fine. The real crisis point are when people are unable to get on the path, or their progress on the path becomes completely blocked. I think it is less important to cater to the raiding guilds, and more important to help the guilds who are trying to become raiding guilds. To help the people who cannot get on the path at all.

This is why I feel Rating Requirements for Season 4 are a bad idea. If you are a casual PvPer, one who tries, but isn't really all that good, you've basically fallen off the path. And that is demoralizing.

In reality, of course, a casual PvPer is not likely to earn all the pieces of S4 before WotLK. But again, what is likely is not as important as what is possible. As long as you are on the path towards the end, that is the critical element.

If you look back at the last time I posted about PvP rewards, notice that I never outright denied a reward to a player. My suggestions made the process longer, but so long as a player kept at it, they would stay on the path towards Season 4.

The number of players who achieve the end is less important than the number of players who are on the path to the end. The number of players in Sunwell is less important than the number of players on the path to Sunwell. The number of players in Season 4 gear is less important than the number of players working towards Season 4 gear. With the advent of extensive Rating Requirements, the number of players on the path to Season 4 gear will drop drastically, and that is an unhealthy state for the PvP endgame.

Monday, May 26, 2008

Thoughts on Raid Schedules

I've noticed that there are very few 25-man raiding guilds that only raid for one or two days. There's probably a significant portion of the audience that would be happy in this type of guild, and it seems a little bit unusual that these guilds are so rare. I think that there are two reasons for the lack of 1 or 2-day raiding guilds: the nature of guild leaders; and the pressure to do more.

Guild leaders, by their very nature, tend to be a little more hardcore than normal. After all, starting and running a guild is a great deal of work and effort. As well, they tend to play a fair bit. People who are only online one or two evenings generally feel that they aren't able to devote enough time to the guild.

Because guild leaders play so much, they tend to shape the guild so that it matches their schedules. If they want to raid four days a week, that's what the guild works toward.

The other reason is that there is a constant pressure on a guild to do more. Every extra day you can add to the schedule makes you better off. An extra day of farming, or an extra day of boss attempts, sounds enormously appealing. A lot of times I've seen a guild start with small raid schedule, and then realize that, hey, if we add another raid day, we could farm Kara for badges, or have two days for boss attempts, and their raid schedule balloons.

Lately, I'm coming to the opinion that a non-hardcore raiding guild should ideally raid one less day than it is fully capable of. For example, if your guild absolutely cannot raid for five days, but you can for four, I think you should raid three days of the week. Having that extra slack can go a long ways towards combating fatigue and burnout.

I think the next major innovation in MMO game design will be an pressure or force that compels you to do less, and is one which is willing accepted by the player base. For example, you could always brute-force it, and say that you can only ever be saved to one raid instance at a time, but the player base would howl. I have no idea what this force could be, but I believe it will be key to the success of the MMO that dethrones WoW.

Friday, May 23, 2008

Fun and Raiding Video

Theoretically, this is still a WoW blog, so I may as well make a WoW post.

Apropos of our discussion on raiding and fun, comes this video of a Kil'jaeden attempt by <V A N Q U I S H>:

I like this video, because often raiding is a "quiet" sort of fun (except when a boss dies and Vent breaks out in cheers). You need to keep quiet in order to make sure that the raid leader can be heard. The surprise and outright delight in this raider's voice at the end is something that we all feel, but often don't give voice to, and thus is not really communicated to non-raiders.

It's sort of appropriate that this comes from a very hardcore guild, on the last encounter of the last raid in the game. It is heartening to see that even at that level, raiders still have fun.

Age of Conan: Odds and Ends

Tharok has challenged me to note some things that Age of Conan does better than WoW. I'm going to try to do so.


Umm, the water is really, really pretty. Seriously, the reflection, refraction, and diffraction is superb. Simply putting my toon in the water ups the graphics quality by an order of magnitude.

The use of positional effects such as knockbacks is very nice. My Priest of Mitra got this spell Repulse, which flings all enemies close to you backwards, and does a fair bit of damage. It's a lot of fun to use, and is a great spell for a priest type character.

The melee combat is more reactive than WoW's combat. Where WoW tends to fall into set routine, AoC combat tends to make you push buttons in a unique order each time.

Quests are a lot easier to do as quest objectives are marked on your map, so you can just follow the arrows.

I think the LFG system is better, but I haven't actually used it yet.

Well, that's done, and I can go back to pointing silly design design decisions. This is just going to be a quick list of stuff, not really in any significant detail.

Melee Combat. I personally find melee combat to be awkward, and not really that deep. Hit the guy on the side with no shields is not the height of strategy. Maybe it will get better if you can start identifying when a combo has been started and what moves are coming up.

Death Penalties. When you die, you essentially "respawn" at a graveyard. You have a penalty to your stats, which will go away if you reach the spot where you died (or after 30 minutes). Of course, if you die far away from a respawn point you have to run all the way back to your body, and there are normally respawned mobs in the way. So you have to kill these mobs all over again, only this time it's harder because of the penalty.

Quest tracking. As far as I can tell, you can only track one quest at a time. However, if you pick up a new quest, the tracking automatically switches to the new quest. So what ends up happening is that I'm following the arrow to a quest item and I see someone who has a new quest. I pick up the quest, look at the mini-map to find my bearings, and realize that I have to open the quest log and re-select the old quest I was working on. This behaviour is supremely annoying.

Chat Window. Here's the general rule of chat windows: Chat windows are for communicating with other players, the rest of the interface is for communicating with the game. 9 times out of 10, the player does not need to see an actual message from the game if the UI has been designed properly. AoC shows too many messages. As well, by default you can only see 3 or so lines in the window. The worst is when you sell loot after adventuring, and every transaction is posted. I'm not really sure it's even possible to carry on a conversation while doing something using the default chat.

Instancing. AoC doesn't really feel like a world. It feels like a collection of instances strung together. The instancing tech to balance loads and over-crowding is quite clever, but I'm not sure it was a good idea. I quite clearly remember my first griffon flight in WoW, and realizing that I could see other people and they could see me, that I was still part of the world. That was amazingly cool. In fact, I think Blizzard's decision to instance off the Blood Elf and Draenai starting areas was a mistake, and AoC is making that same mistake on a far larger scale.

Balanced areas. I was doing a green (low-level) quest that involved killing level 5 monsters at about level 8. However, to get to the correct area I had to get past a level 11 monster that killed me every time. I finally waited for a higher level to come by and kill it, and followed him through.

Graphics. I don't really like the graphics. Technically, they may be better than WoW (more polygons, more insert-CG-buzzword-here, etc.), but I find that Blizzard makes much better use of colours, contrast, and shapes. But I'm not really a fan of "realistic" graphics. I find that games that go for realism are not as "vibrant" as real life, and that more cartoony games are more likely to achieve that vibrancy. Your personal taste may differ.

Inability to switch characters. If you want to switch to a different character, you get to exit from the game and log back in again. I cannot fathom how Funcom decided this was a good idea.

Company logos. Yes, Funcom, when I start your game, I really want to see all 100 logos of every single company that had a hand in the game, including the pizza shop down the street, as well as the opening movie. Seriously, every time you launch the game, you get to sit through all the opening movies (or more accurately, hit Esc a whole bunch of times). Please, for the love of all that is holy, gaming companies need to stop doing this! Play all the movies whenever you make a new character, or the first time the game launches, but after that let us get to the actual game ASAP!

This isn't a negative for Funcom, but something that would have been really cool is if they had provided a set of "WoW default keybindings". Basically press L to bring up the Journal, B to bring up the inventory, C to bring up the character screen, P to bring up the Abilities/Skills and N to bring up the Feat trees. Just to make it slightly easier to steal WoW players. Similar to what Excel did to capture Lotus 1-2-3 users, by implementing all the Lotus "slash" commands.

All in all, AoC hasn't really grabbed me yet. There hasn't been anything that has really made me excited to play, and there are enough annoyances to make me not want to play. Admittedly, I haven't levelled very far yet, my highest character is my Priest of Mitra at about level 11. However, I have put in several hours on a bunch of different characters trying to find one that I liked. I predict that I will probably try to get my Priest through the starting area over the weekend, just to get my money's worth out of the game, and then probably stop playing it.

Thursday, May 22, 2008

Age of Conan: Melee Combat

Melee combat is the big selling point of Age of Conan. I find it fun and visceral in some ways, but also rather awkward.

Basically, there is no auto-attack. Combat is more like a console game, where you have three basic attacks (left, middle, right). Your enemy has three shields which they shift around. So basically, you try and hit them where they've left openings in their defenses.

This system is all well and good. Combat is bloody and fluid, and pretty fun with just the three basic attacks.

My problem comes with special abilities, called combos. To start a special attack, you press the hotkey for that attack, and then a specific sequence of basic attacks. That executes the attack.

For some reason, I find this really awkward on a keyboard. My basic attacks are bound to 1, 2, 3. I can use special attacks bound to 4 and 5 fairly easily. However, hitting specials bound to the other number keys causes me trouble. I generally have to take my fingers off the basic attacks in order to reach the specials, and that makes it hard for me to complete the specials, as I have to get my little, ring, and middle fingers back in place. I've been experimenting with using my other hand or the mouse to trigger specials, but then I can't use that hand to control movement. WTB one more hand!

Essentially, the basic attacks take up too much of "key space" accessible to one hand, at least with the default keybindings. I might be able to do something by remapping those keys to the middle of the keyboard. But right now I find that I'm skipping most of the combos and just hitting the basic attack buttons.

A very annoying note is that if you play an archer, your bound specials bar doesn't change when you switch weapons. If switching to a bow automatically swapped the specials available, it would be much easier. It's not so bad on a WoW hunter, as you can put range on one side, and melee on the other. The problem in AoC is that you use 1-3 in both situations. So it's like 1-6 for range, and 1-3 + 7-9 for melee.

It strikes me that this system would work superbly on a console controller or gamepad. You'd use your thumb to hit the basic attack buttons, and you could queue up specials using the right shoulder pad buttons.

Though to be honest, if I had designed this system I would have designed it differently. I would have allowed the player to replace the basic attacks with new attacks, but still keep the whole directional combat with the shields. For example, if you got a Poisoned Strike, you could replace your Middle Basic attack an Middle Poisoned Strike. Then whenever you made an middle attack, you would make a poisoned strike. (You could bind the same Strike to multiple directions.)

If you combined this with abilities that had cooldowns, it would get pretty interesting. If you have a Haymaker with a 10s cooldown on your Right attack, you'd make attacks to draw shields away from the right side, then Haymaker the opponent for a ton of damage. But now you can't use your Right attack for 10s, and a smart opponent can balance shields to protect the Middle and Left sides.

It would keep the twitch and visceral nature of combat. You could still have combos, but now the components of the combo could change. (I'd also remove having to start the combo by hitting a new button.) If Left, Left, Right triggers a Sweep, you could set it up so that Left and Right had abilities of your choice. It would be a little more strategic, and would be a lot easier to hit keys with one hand.

Wednesday, May 21, 2008

Age of Conan: Graphics, Priest of Mitra

Continuing on, the graphics in this game are very nice. There are lots of options in character creation. Of course, the game aims for the "realistic" side of things, and thus tends to be mostly greens and browns. Seriously, I don't understand why everyone is so scared of bright, primary colors.

The water in particular, is superb. The way reflections interact with the surface is nothing short of spectacular. AoC requires a pretty hefty machine, but it uses that machine well.

Animations are very well done for the most part. The only quibble I have is that the female running animations seems a little odd. I don't think the way the elbows are positioned is quite right.

I'm skipping to the second character I made, as I want to talk about the first in more detail later. My second character was a Priest of Mitra, which is pretty much as pure a healer as you will get. You start off with a Smite spell which is pretty neat. It does a lot of damage early on, and it was pretty amusing to be one-shotting enemies as they get lifted up into the air and then dropped on the ground. Then at level 5 or so, it settled down into a more normal routine of needing to blast the mob several times.

There were some very nice touches for a priest class. They had a health buff which automatically hits everyone in your group and lasts an hour. The first healing spell I got was an Heal-Over-Time that was placed on the entire group. Are the days of Whack-A-Mole over?

The only weird part was that healing seems to be really weak. I was dishing out and taking about 20-40 damage per second. Yet my heal spell healed a grand total of 2 (!!) health every second (per person). I was trying it out and wondering if it was missing a zero or something, as it literally seemed pointless to cast. I'm not sure if healing changes later on, or if this is intended, but it seemed a little unusual.

I got the first city and started doing the destiny questline (sort of like a solo instance of the world with specific quests). That's a pretty neat idea and use of zoning. It's interesting to see more games move through time in the same zone. I didn't get very far, only to about level 7.

I do like the initial quests, though, as they are fairly meaningful, and not just "kill 10 rats." I died a few times, though it was mostly my fault for attacking multiple enemies.

So far though, I like the Priest of Mitra class. I'll probably try some of the other ones, but this class looks to be a solid choice for a healer who is used to other games.

Age of Conan: Installation, Races, Classes

I picked up Age of Conan last night on a whim, and thought I'd write about my first impressions. First off, this is probably going to sound unnecessarily negative. AoC does a lot of stuff well. For example, the servers were up, running and pretty solid, which, sadly, is an accomplishment for an MMO. I really only glitched out once (I died once and the dialog to resurrect didn't come up), so technically the game--so far--is smooth. My issues are almost entirely design-driven, which I suppose is progress for the MMO industry.

A general warning, the technical requirements are pretty steep. I wouldn't pick AoC up if you have less than the recommended specs.

I'll start with the install process. When we talk about Blizzard games, we often throw around the word "polish". Whenever someone asks for a concrete example of polish, I point to the installation process for World of Warcraft. When you install WoW, the installer plays a noticeable sound (an orcish drum sound, I believe) when it is ready for the next CD. This allows you to do something else while the game is installing, and still be alerted when it is time for the next CD. Then when you insert the next CD, the install process automatically picks up from where it left off. (My guess is the autorun of the new CD communicates with the installer.)

That is polish. Taking the time and effort to clean up the little details that other people would leave undone. AoC's install, in contrast, worked fine. But you had to manually check if the installer was ready for the next DVD, and you had to click the OK button after putting in the DVD. It's a small thing, but it really symbolizes the difference between WoW and AoC, between Blizzard and the rest of the gaming industry. AoC works, and is serviceable and fun, but WoW is polished.

On to races and classes. There are three races: civilized good guys, who are white; noble barbarians, who are also white; and evil bad guys, who are black. Gotta love the casual racism of the gaming industry. These racial distinctions are enforced in the game. If you roll a Cimmerian, you cannot pick a non-white skin colour. If you roll a Stygian, you cannot pick a non-dark skin colour.

And the good/evil divide extends to classes. The black people have a single neutral class (the Archer), and all the other classes are outright evil (Demonologist, necromancer, Herald of some elder demon god, priest of Set). Meanwhile the white people have mostly good or neutral classes (there is one evil class, the Dark Templar).

Of course, Funcom probably justifies being racist by claiming to be "true to Conan's lore." Yeah, well, Howard was pretty damn sexist, but I don't see AoC imposing any but the most cosmetic differences between males and females. There aren't any stat differences between male and female characters, and either gender can be any class. But then I guess Funcom is European, and therefore, "Racism good! Sexism forbidden!"

Funcom will probably react to this by trotting out some token black NPC who they claim is a good guy ("See, not all black people are totally evil"). But actions speak louder than words. In Age of Conan, if you want to play a good, heroic class, you make a white character. If you want to play an evil class, you make a black character.

Monday, May 19, 2008

Lack of Faith

Lately, I've been reading the forums and discussions about the proposed changes to raiding. It seems to me that there is a significant--and vocal--segment of raiders who lack faith in 25-man raiding as a fun and attractive activity.

Basically, their arguments are that Blizzard should continue with 25-man raiding because 25-man raiders are somehow the most deserving of WoW players. All other activities must have significantly lower rewards so that people who wish to progress have no choice other than 25-mans.

It seems that deep down, in their heart of hearts, they fear that most people do not like raiding. That other people just raid for gear. And since they like raiding, they try to promote raiding by loudly denigrating all other choices, and seek to have Blizzard stack the deck in favour of raiding. Basically, they seem to think that if raiding had to compete with other activities on a level playing field, it would lose out.

This lack of faith hurts the raiding cause far more than it helps. Blizzard should not make 25-man raids because 25-man raiders are somehow "more deserving". Blizzard should make 25-man raids because raiding is fun, and a large portion of the audience enjoys it! Blizzard should make more battlegrounds because battlegrounds are fun! Blizzard should make more 5-mans, because dungeons are fun! Blizzard should make more reputation grinds, because grinding rep is...wait, no, no it isn't.

The raiding community--especially the high end--does a really poor job conveying the fact that raiding is fun to non-raiders. We do a lot of complaining, of calling people "bads", of indulging ourselves in trying to prove we are better than "casuals". But these types of behavior hurt raiding far more than they help.

Imagine you are a new 70, thinking about raiding. You take a look at the Raid & Dungeons forums on the WoW forums. Do you honestly want to end up in guild with people like that? Or your realm forums. I'm sure you've seen a newish guild post about their latest triumph, and get torn to shreds by some high-end raider sneering, "Months behind". Can raiding really be fun if raiders act like this?

The thing is that raiding is fun. Working together as a team to kill a dragon is a thrill. Watching a raid move through the complex dance that is a boss fight is an experience. Even wiping is fun (to a degree) as you slowly get better and better, as you get closer and closer to victory. Raiding is more than capable of holding its own against the other options in this game.

The problem with raiding is getting new people to try it, trying to convey to new people that yes, this is worth doing, that raiding is fun in and of itself, and the loot is just icing on the cake. If we raiders want Blizzard to keep making 25-man raids, the best path is to get more people raiding.

This is why the 10-man raid progression is worth having. The single largest barrier to getting new people raiding is not gear or skill. Gear improves with time, and skills can be learned and taught. The major barrier is time. It is getting people used to the idea of blocking out one or two evenings a week for WoW. Getting someone to say, "I raid on Thursdays" is 95% of the battle of creating a new raider. 10-man raids will do this, will allow people to become raiders in the safety and comfort of a small guild. From there, it's just a small step to the 25-mans. 10-mans also offer a safety net, so that if 25-mans don't work out, you still have options to progress, to see new content.

People will raid if they believe that raiding is fun. Emphasising that raiding is an enjoyable activity, and nurturing and being helpful to the new raiding guilds, will do far more for the cause of 25-man raids than trying to argue that raiders are somehow "more deserving" of Blizzard's time than other players.

Thursday, May 15, 2008

WotLK 10/25-Man Raids

Raiding is undergoing some pretty interesting changes in WotLK. The main points that I've seen are:
  • Each raid instance will have a 10-man version and a 25-man version.
  • 25-man raids will drop loot one tier higher than 10-mans.
  • No attunements or keys for any raids.

I must confess that when I first saw this, my immediate thought was, "Wow, that's going to create a lot of work for me." I maintain the Guild Progress thread on the Skywall forums, which lists the raiding guilds, their website, and what bosses they've killed, ordered roughly by how far each guild has gotten in endgame.

With two separate progressions, how would this thread be ordered? Should I have two lists, one for each raid size? Should guilds be classed as 25-man guilds or 10-man guilds, or listed in both lists? I expect the 25-man guilds to do the 10-man instances, and is it really worth cluttering up the list with the same Nexus-25 and Nexus-10 bosses?

Of course, these concerns are pretty trivial. In pretty much all other respects, these changes to raiding are an unambiguous win. Content creation is expensive, and I think that this will allow a lot more people to see the raid dungeons.

I think that 25-mans are pretty clearly pitched as the hardcore option, and probably will be tuned very tightly. The 10-mans will still be complex, but will definitely be tuned lower than the equivalent 25-man. Arthas-10 will be easier than Arthas-25. The loot structure is set up for that. Remember that for raiding, the quality of loot should depend on the difficulty of the encounter. Better loot equals higher difficulty.

Various other people are debating if 10-man encounters can be as epic as 25-man encounters. I'm not really sure I know that answer. Epic has a lot to do with lore and the archetypes of the adversaries, not just the complexity or difficulty of the fight. But really, the difference in "epicness" doesn't matter:

Is it better for Arthas to be an epic 25-man fight you never see, or for Arthas to be a less epic 10-man that you get to participate in?

I think Blizzard's plan strikes the best balance. As long as Arthas-10 is harder then the 10-man bosses that come before him, that there is an actual difficulty progression in 10-man raiding, I think everything will be fine.

The interesting questions will be how guilds react to the changed situation. Will the 25-man guilds do the 10-man raids? Will they recruit from the 10-man guilds? What will happen when someone burns out of 25-man raiding? Will they quit the game, or will they see 10-man raiding as a good alternative (huge wins for Blizzard if people take this path)? Will "softcore" guilds--who work on 25-mans at a slower pace--continue to exist, or will the only 25-man guilds be ones that raid 5 or more days in the week?

Exciting times ahead.

Wednesday, May 14, 2008

Lil Timmy and the White Kitten

The very first non-combat pet I bought was a White Kitten, which is sold by Lil Timmy, who occasionally wanders in Stormwind. Coriel was only level 12 or 13 at the time. This was over two years ago, way back when I was still new, before I knew about such options as Instant Quest Text.

I remember running across Lil Timmy, and clicking on him, and reading his story about how his family's cat had kittens and they couldn't afford to keep them. So he had to sell them, or they would have to be drowned. But the kitten cost about 60 silver, which at level 12 was pretty much all the money I had.

I've never been a big non-combat pet collector, as they tend to take up too much inventory space for my tastes. I think I spent several minutes deciding whether or not I should spend all my money on something that had no obvious value. But what sort of paladin can pass up a kitten in distress? So I spent the silver, got the kitten, and it's been in my inventory ever since. That kitten has been all over Azeroth, and has gone to the Molten Core, Blackwing Lair, Outlands, and many other places where kittens were not meant to go.

Yesterday, I ran across Lil Timmy again. This time when I clicked on him, there was no story, as I was taken straight to the Vendor screen. It seemed unusual at the time, but I chalked it up to not correctly remembering an event that happened two years ago.

However, later I realized what had happened. Several patches ago, there was a change to NPCs. If an NPC only had one option, for example a windrider asking if you wanted to fly somewhere, the game would automatically open that option, rather than making you click through a screen of flavour text. Lil Timmy was one such vendor. The only option you could pick after the flavor text was the option to buy the White Kitten, which opened up the Vendor screen. So like all the other NPCs, the flavour text got skipped.

I really liked the change when it was first implemented, and I still do. 99% of the time, it's an extra annoyance to click through an NPC's initial screen, especially when it is an NPC that you interact with again and again, and when you haven't found out about Instant Quest Text yet. The windriders were the worst for this.

But almost all changes are trade-offs, which can have unexpected ramifications, and Lil Timmy falls in that 1%. Without his story text, he's just a random boy selling a somewhat rare pet. And the game is slightly diminished for that.

Tuesday, May 13, 2008

WotLK Raid Makeup

Raids in Wrath of the Lich King need to be balanced around 20% healers. There are many reasons for this:

1. The introduction of Deathknights as a DPS/Tanking class means that there is extra competition for the non-healing slots.

2. Blizzard has done a really good job of making non-healing specs raid-viable, and that means even more competition for the non-healing slots. Let's say you want to bring an Enhancement and an Elemental Shaman. You need at least 2 Resto Shamans to fill out the healing quota, so you're looking at 4 shamans in total. That means you are taking one slot away from a pure DPS class, and that's unfair to them.

3. The rest of the game is balanced around 20% healers, so bring raiding in line makes balancing the population easier.

4. Healers have always been the hardest role to recruit for. Requiring less healers and more DPS will make more players happy.

Take a look at this proposed WotLK raid makeup:

Group 1 - Prot Warrior, Prot Paladin, Feral Druid, Deathknight, Deathknight
Group 2 - Rogue, Rogue, DPS Warrior, Enhance Shaman, Ret Paladin
Group 3 - Balance Druid, Mage, Mage, Elemental Shaman, Shadow Priest
Group 4 - Warlock, Warlock, Warlock/Mage, Survival Hunter, Hunter
Group 5 - Disc Priest, Holy Priest, Holy Paladin, Resto Shaman, Resto Druid

To me, that looks like a gorgeous raid makeup. Every class has 2 or 3 slots. No class takes more than its share, and no class is completely left out. There's lots of different specs, and lots of synergy and options available.

However, this setup is completely useless under today's raiding conditions. There simply aren't enough healers. But if raids are balanced around 20% healing, raid compositions like this can be viable, and I think that would make raiding healthier and more enjoyable.

Monday, May 12, 2008

Divine Intervention Idea

Divine Intervention is a very weird spell. The spell itself is a terrible, terrible mechanic. You can only cast it on another person, who must be a resser AND in a position where she won't draw aggro when the fight ends. Compare that to the Warlock soulstone, or Shaman self-res. And yet the flavour of Divine Intervention--the paladin sacrificing herself to save her companion--is so strong, that the spell somehow works.

In reality, the major use for Divine Intervention is to help the paladin avoid a repair bill. It would be nice to have DI do something useful, while still retaining its flavour.

Tharok and I were talking about it, and came up with the following:

Divine Intervention
Instant Cast, Reagent: Symbol of Divinity, 100 yards range, 60 min cooldown
The paladin sacrifices herself to return all other dead group members to life with X health and Y mana. This spell can be cast in combat.

So it's a combat res, but the paladin has to die. It's possible to get up to a 4 for 1 trade, but you can only resurrect people in your group. You'd probably need to combine this with preventing groups from being changed in combat, and maybe give druids a real resurrection spell as well.

Edit: Actually, Divine Intervention is a really interesting spell from a design point of view. The key element of the spell is that the paladin has to die, to sacrifice herself. However, the paladin is often the healer or the tank, and usually when the tank or healer dies, the group wipes soon after. So you need an effect which makes a large impact, and can lead to the group defeating the fight, but which overcomes the extremely large drawback of losing the tank/healer.

Friday, May 09, 2008

A 2-cycle Example

Perhaps an example will help illustrate the difference between 2-cycle DPS and normal DPS.

You are a caster with 1000 mana. The next fight is 90 seconds long. You have the following two spells:

Burn Spell - 1.5s cast, 500 damage, 100 mana cost
Efficient Spell - 1.5s cast, 100 damage, 10 mana cost

What combination of spells produces the maximum damage?

Rather than actually using an algorithm, we'll just list out all the possibilities:

The most damage comes when you cast 4 Burn spells and 56 Efficient spells. Now let's make a slight change, and add 20 mana to your mana pool, for a total of 1020 mana.

This time, maximum damage comes from casting 5 Burn spells and only 52 Efficient spells. You even spend 4.5s doing nothing! The extra mana, a Longevity stat, allows you to do more damage by spending more time in burn mode. Note that this system is also very sensitive to fight length. If the fight was longer or shorter, that would have produced a different result. You could also increase damage by adding DPS stats, making each individual spell hit a bit harder.

Now, contrast this system to the process used by my Destruction (Shadow) Warlock. I don't have two spells, I only have Shadow Bolt. Adding 20 mana doesn't really do a whole lot. I spend all of my time and effort making my Shadow Bolt hit harder (spell damage, crit rating) and more often (hit rating, haste). Longevity stats are far less useful than DPS stats.

Of course, optimizing those DPS stats is pretty complex. But it's a completely different process than balancing Longevity and DPS stats.

Thursday, May 08, 2008

Tension in Gearing: Arcane Mages

In the comments to the previous post, Kadaan makes a very strong case that Arcane Mages have to balance between Damage and Longevity stats, and have very similar tension in gearing as healers and tanks.

So why are Arcane Mages special? Arcane mages possess a spammable, high dps, high mana-cost, unsustainable nuke in Arcane Blast. Because Arcane Blast cannot be spammed for the entire fight, an Arcane mage has to use what Elitist Jerks calls a "2-cycle theorem of spell selection," where they alternate "between a mana intensive burn mode and a mana efficient mode." The burn mode consists of spamming Arcane Blasts and the efficient mode is usually something like Frostbolts or Scorch.

Getting the correct ratio of burn mode time to efficient mode time to maximize damage done is non-trivial, even before you take upgrading gear into account. It's a Linear Programming problem, and can be solved using the simplex algorithm for a given initial gear set and fight length. Under this setup, Longevity stats are valuable because they allow you to increase the amount of time in burn mode, which increases total damage done, even if those stats do not increase the DPS of either mode directly!

In a lot of ways, it's very similar to paladin healing mechanics, in that we have a burn mode, Holy Light, and an efficient mode, Flash of Light. The major difference is that the spell paladins use is dictated more by incoming damage in the short-term future, rather than attempting to maximize total raw healing done.

However, Arcane Mages are very different from most other DPS classes in the game. Most of the DPS classes use the simple DPS algorithm that I've outlined before. They have a single, sustainable nuke (Sinster Strike, Shadow Bolt, Fireball, etc.) that they rely on, while juggling cooldowns for certain higher-dps abilities and lower-dps enablers. I'm not sure I know of any other class that uses a 2-cycle spell rotation. Perhaps Balance druids with Starfire and Wrath.

(Note for the random DPS who seem to think I am putting them down: the simple algorithm is only "simple" in comparison to the 2-cycle algorithm. Please do not interpret this as me insulting you.)

Now, should more DPS classes have a setup such that they have to use a 2-cycle rotation, and thus increase tension when gearing? Maybe. However, one issue is that a burn mode can be quite devastating in PvP, where burst damage is more important than total damage done. Arcane Blast keeps this in check by making the mage take time to ramp up to the high-dps setting.

Wednesday, May 07, 2008

Tension in Gearing

When it comes to choosing gear, tanks and healers seem to differ from DPS in a significant way. Tanks and healers are trying to reach a balance between two points, while DPS is only intent on maximizing one point.

Tanks need to choose between Survivability and Threat. There's no need to put out 50% more threat than the nearest DPS, and there's no need to have so much health that you never drop below 75% health. Similarly, healers need to choose between Healing Throughput and Mana Longevity. You need to put out enough healing to keep people alive, but you also need enough mana to last the entire fight.

DPS, on the other hand, really optimizes for only one thing, DPS. When looking at gear, pretty much the only thing that matters is which gear results in higher damage.

Part of the reason for the difference is that, past a certain point, Threat, Survivability, Healing Throughput and Mana Longevity stops being useful. But higher DPS is *always* useful. There's no tension, no balance you need to maintain as DPS.

The ramifications for this show up in a lot of different places. Tank and Healer builds often make trade-offs. DPS builds, on the other hand, are pretty much the maximum damage builds possible. Tanks and Healers are more likely to use different colour gems because each color enhances a different aspect. On the other hand, DPS absolutely hate blue gems, because blue gems rarely contribute to increasing damage.

I'm not really sure which is a better system. DPS is simpler and straightforward, and makes gear evaluation a lot easier. As well, you can minimize the amount of gear you carry. Tank and healer gear can get more complex, and you can try for a balance that is very specific to the needs of a specific fight.

The DPS class that comes the closest to needing a balance are mages, and there is a lot of complaints among mages for needing that balance. Mages complain that Spirit is useless, about needing to use Mage Armor over Molten Armor, etc. Mage builds rarely emphasize the trade-off, and usually sacrifice longevity for DPS.

Monday, May 05, 2008

The Dark Secret of Paladin Levelling

The dark secret of paladin levelling is that the whole Seal/Judgement system--though it looks shiny and complex on the outside--is pretty much irrelevant. 95% of the time, you're better off casting or judging Righteousness (or Command if Retribution), rather than trying to debuff the mob with a special Judgement.

To see this, take a look at Judgement of the Crusader. It increases your damage from subsequent Holy sources, which is usually Seal/Judgement of Righteousness. But instead of judging Crusader, you could have started by judging Righteousness. Your subsequent hits are smaller, but you do a chunk of damage up front. Each method takes pretty much the same time to kill a normal mob.

But judging Righteousness also has the advantage of using less mana, as you can judge the seal you already have running instead of casting a new Seal. For example, if you are fighting two mobs, you can go:

1. Cast Seal of Righteousness.
2. Judge Righteousness on mob 1.
3. Cast Seal of Righteousness.
4. Kill mob 1.
5. Judge Righteousness on mob 2.
6. Cast Seal of Righteousness.
7. Kill mob 2.

That's 3 Seals and 2 Judgements. Now if you debuffed each mob with Crusader first, you end up with:

1. Cast Seal of the Crusader.
2. Judge Crusader on mob 1.
3. Cast Seal of Righteousness.
4. Kill mob 1.
5. Cast Seal of the Crusader.
6. Judge Crusader on mob 2.
7. Cast Seal of Righteousness.
8. Kill mob 2.

That's 4 Seals and 2 Judgements. If you want to debuff mobs, you essentially cast an extra Seal for each mob. And given that it's pretty much the same damage either way, you save mana by using Righteousness all the time. That increases your uptime, allows you to run Blessing of Might instead of Wisdom, which lets you kill even faster, and generally makes you more durable.

That's not to say that the Judgement debuffs are useless, just that their use is very specialized. Judgement of the Crusader is good for fighting much higher level mobs or elites, which take a long time to kill. Light and Wisdom are great in groups. Justice is invaluable to keep your target from running off and aggroing other mobs.

But 95% of the time, you're better off ignoring the whole Seal/Judgement system and sticking with Righteousness when levelling.

Sunday, May 04, 2008

The Failure of Team-based Ratings

Patch 2.4.2 and Season 4 have some big changes coming for PvP. The most important change, in my opinion, is the following:
- If a character’s personal rating is more than 150 points below the team rating, they will earn points based on their personal rating instead of the team rating.
- If the average personal rating of the players queuing for a game is more than 150 points below the team’s rating, the team will be queued against an opponent matching or similar to the average personal rating.

This is Blizzard throwing in the towel on team-based ratings. Which, quite frankly, is needed.

The point of a rating system, as I have pointed out before, is to measure the skill* of a player. But in a world where you can easily join, leave, form and dissolve teams, the rating system quickly bears no resemblance to reality. Team-based ratings are simply too vulnerable to exploitation. Point-selling by high-end teams is endemic.

In the ideal scenario, a player's rating should quickly settle around its true value, and after that, change fairly slowly. You should not be able to reset it. Wild swings after the system becomes settled is a sign of a flawed system.

Team-based ratings were a bad idea to start with. Hopefully in WotLK, Blizzard will implement a proper personal rating and reward system for PvP.

Once ratings settle, there are a lot of interesting things Blizzard could do. For example, you could have a Tournament each Saturday. Teams would be divided by rating bands, and you could give the winner in each band enough Arena Points for a bonus item. Under the current system, if you did this, you'd probably end up with some full Gladiator team that just started a new team in your bracket, and they would romp all over everyone.

I would take a look at games like Magic Online, which--even though they have a rating system--don't rely solely on ratings to hand out rewards.

*Where "skill" means the ability to defeat the other team, and may include gear.