Monday, March 15, 2010

Glyphs

Glyphs are terrible.

First of all, just like in my gems post, I should point out that I am very aware of the fact that glyphs are fantastic, especially considering the alternative to glyphs is not having glyphs.

Glyphs were a cool idea that worked out pretty badly. I have at least four gripes about glyphs, which I may, of course, add to in the future as I think about more problems with glyphs. They are:
  1. The level of your inscription skill bares no relation to glyph power
  2. Glyphs are mostly uninteresting and trend towards being more uninteresting all the time
  3. Many glyphs seem to have been abandoned as useless
  4. There are *really* not enough minor glyphs
First of all, the inscription skill is completely bizarre. The reason you skill up leatherworking is to make better leather items. You skill up jewelcrafting to cut high grades of gems. You skill up inscription to make equally powerful glyphs. While a few glyphs apply to class skills that are only attained at high level, and as such can't be used until high level, a lot of the "high level" glyphs are just as useful at level 15 as they are at level 80. Glyph of Fireball, a spell that mages get a level 1, requires 385 skill to create. This is not because they didn't want you to have it at low level. The glyph's item level is 6 and it requires level 15 (the level you get your first glyph slot) to use. It's just that if you want to use it at level 15, you'll need to get it from someone who is at least level 65 who makes it out of herbs found in Northrend.

On the other hand, Glyph of Pestilence only requires 75 skill to make despite requiring level 55 to use (since there are no death knights below level 55). Glyph of Power Word: Shield, the most important glyph by far for discipline priests, required 80 skill to use, so there is no correlation between skill to make and usefulness either.

Basically glyphs are assorted randomly at different skill levels and herb requirements. They had to do this to make inscription fit their skilling up paradigm, but it makes no sense whatsoever. If I started playing this game today, made a mage, and decided that inscriptions sounded cool, I would hit level 15, get my skill to 75, and my profession would give me the incredible Glyphs of Fire Ward, Frost Ward and Arcane Blast. So two spells I never cast and one I won't learn for 49 levels. Inscription is designed entirely with the idea that you have max skill in it. Until then the profession doesn't make sense.

When glyphs first came out, the idea seemed to be that you could modify your spells to give yourself different options. A spell might be faster but cost more, or do less damage but more healing, or have some other kiss/curse type effect. There were also those glyphs aimed at changing the way you play. In practice, half of the glyphs just gave spells a percentage boost and had no trade-off. So, of course, people use the ones that make the things do they flat out better rather then using the ones that change things around. Also in practice these "interesting" glyphs often made the spells just plain bad. So many of the trade-off glyphs started to vanish to be replaced by simple percentage boost glyphs. Many glyphs could just ultimately be replaced by "+2% damage" and no one would notice. It's a lot of work and game mechanics to accomplish this.

But the fact that glyphs have that kind of effect makes many of the glyphs pretty close to useless. If there are glyphs that give you 2%+ damage then you aren't going to take glyphs that don't directly add to damage. If there are glyphs that give you 2% more healing, or 2% more damage mitigation, then you are going to skip utility glyphs in favour of these as well. As a result, many utility glyphs seem to have just been abandoned completely as useless. Glyph of Mass Dispel reduces the mana cost of Mass Dispel by 35%. According to Armory Data Mining no one uses this glyph at all. Even if they are just rounding down the very small number of people who use it, that would still mean that more discipline priests are using Glyph of Divine Spirit than Glyph of Mass Dispel.

And who can blame them. When I was running ToC and Ulduar my spreadsheet told me that Glyph of Mass Dispel would be worth about 4 Mp5. With my ICC logs it would be worth zero. I haven't cast Mass Dispel in a while. Even if they upped the 35% to 100% it would still see little use, though some people might swap to it for particular fights and it would probably see use in PvP. Glyph of Barbaric Insults is similar. So are Feint, Entangling Roots, Scorch, and several other glyphs from every class. These glyphs aren't good, it seems like they can't be good, so they've just been left to die. But some of these glyphs are really pretty easy to rehabilitate. Up the damage percentage on Glyph of Scorch so that a Scorch mage does almost as much damage as a Fireball mage and you'd see at least a few people using it. Up the effect of Glyph of Rapid Fire so that it's overall contribution to your damage is somewhat similar to other hunter glyphs and people would use that too. Others, like Mass Dispel, might be rehabilitated best by making them into minor glyphs.

Which would help a lot with the last problem I'll talk about today with glyphs. There aren't enough minor glyphs, and the homogenization in their use is rather incredible. About 95% of priests use Glyph of Levitate. A similar number of druids use Glyph of Unburdened Rebirth. For every class there are the minor glyphs people use and the minor glyphs people don't use. Some degree of homogenization is inevitable with only six glyphs to choose from and three slots to fill, but even in quality of life improvements there is room for balance. It if fairly obvious that people value getting rid of reagent costs far above duration increases for self-cast spells. Just like the unused major glyphs, I get the feeling that minor glyphs are just seen as unimportant, and that the ones people don't use have just been left to die. Minor glyphs aren't terribly important, but if there is anywhere in the game where it would be nice to see variability in behaviour, this would be it. They want to see options in talent builds, sure, but that is very hard to do when talents are tied to your performance in your raiding role. Variability in a part of the game that should be appealing to people's personal taste for quality of life improvements shouldn't be as hard as that.

So what to do about these problems? The first one seems absolutely impossible to solve without significant new tech that is unlikely to be implemented just for one profession. I assume that when Cataclysm comes out there will be a new Greater Heal glyph that you'll need to be level 80 to create while many glyphs that can be made at level 5 will continue to be useless to characters under level 60. The other option is to put in lots more glyphs and actually make the low level ones worse than the high level ones. I don't see this happening.

The solution to uninteresting glyphs is to just nix them all. If something simply says, "Increase the effect of something you'd cast anyway X by 5%" then replace it with something better. Things that increase the effect of spells by 20% are meaningful as they probably rearrange the priority of your casts, but small percentage to commonly used spells really take away the meaning and intent of glyphs. Of course this would be a ridiculous amount of design work, and they would probably rather introduce new glyphs than rework old ones, but if people are using "X does 5% more" type glyphs now than unless the new glyphs are "X does 6% more" I don't see that people will use them.

I already said that a big part of the solution to glyphs no one uses might be to make some of them minor glyphs. I don't think this is appropriate for all of them, but it would sure work for Mass Dispel and Barbaric Insults. But another part of that solution is rebalancing the minor glyphs that no one takes, reducing the attractiveness of reagent-replacement glyphs by selling Light Feathers and Fish Oil at reagent vendors, and making at least six more minor glyphs for each class. If you only get a quarter rather than a half of the glyphs then which ones you take will inevitably be less of a foregone conclusion.

Of these solutions I think that the last one, including more minor glyphs, will probably come with Cataclysm. I have little hope that any of the others would ever be done. Two years from now we will still most likely be able to look upon Glyph of Feint and weep.

Friday, March 12, 2010

Brute Force

Q. How do you feel about the limited attempts mechanic that has been use in Trial of the Crusader and Icecrown Citadel?
A. We're not crazy about how they worked out. They were designed in essence to save players from themselves. In the past, some people would make 400+ attempts on new bosses. That's not healthy and at some point you're not measuring skill but tenacity. Yet, that's not what happened. Guilds just made alts to get around the limitations, and things like disconnects and accidents make losing an attempt really frustrating. We like gating because if nothing else it lets the community focus on more than just the final boss in the zone. If we do limited attempts again it would probably be limited to optional bosses like Algalon.

This is a quote from a recent developer chat. I'm glad they don't like limited attempts because they are probably the worst idea that has ever been introduced in WoW. What I'm really not glad about is that they are denigrating practice with their comment, "At some point you're not measuring skill but tenacity."

I've seen this a lot recently on various forums and fan sites. Most people seem to have jumped on the anti-practice bandwagon. The words "brute force" are often used when people mean to say "practice." It gives it an ugly tone, but it doesn't change the underlying facts.

Yes, top guilds, in order to earn their world-first positions, put in a lot of hours in the form of long continuous play sessions. This seems rather extreme, and I'm sure that the developer is right that it isn't terribly healthy, but they do it because it is what makes them the best. But the key is that it actually does make them the best, it doesn't just mean they lucked into that position.

Practice actually makes you better at things. Malcolm Gladwell suggests that to be a real master at something you need 10,000 hours of practice. Do you have 10,000 hours of practice playing WoW? If you type /played that would be 416 days and 16 hours. The vast majority of players don't have nearly that many hours played, but I would wager heavily that the majority of the players in world-first competing guilds do. They practice and they get good.

And they practice new fights as well as practicing their characters. The Lich King is not roulette. If it takes 10 or 20 or 40 hours for a top guild to get through a fight it isn't because they had a 1 in 200 chance at winning each attempt and they finally hit it. They had basically no chance of winning their first attempt, and probably a pretty decent chance of winning on their winning attempt. They practiced and got better. This is how WoW works and it is also how everything else works.

There is tenacity in being the best at anything. Olympic athletes need a lot of tenacity to get to where they are, but should we insist that you can only compete in Olympic events if you promise not to spend more than 8 hours a week training for them? Eleven million people play WoW, I would bet that is significantly more than the number of people who have ever been in a four-man bobsled. If you have that many people doing something then some of them will have the tenacity to be the best and others won't.

There is no such thing as brute forcing your way through new content by putting in lots of hours. There is practicing and getting better at it until you can do it.

But for those who think that it is not a measure of skill but only a measure of their ability to dedicate the time when the boss come out, you are wrong about that too. Just like the top-tier athletes, you need to get in a lot of practice to be the best, but that doesn't mean that anyone who puts the practice in will be the best. There is natural talent and ability that factors in. Odds are if you practiced as much as the best in the world you would still not be the best.

Thursday, March 11, 2010

Gems

Gems are terrible.

We all know that gems are great. Not only should you socket up all the gear that has sockets, but you should normally choose gear with sockets over gear without sockets because it ends up being substantially more powerful. That's not what I'm talking about. I mean that the concept of gems, while it sounds cool, just isn't working, and I think this is just going to get worse in Cataclysm.

A recent blue post confirmed that the "cost" of sockets in an item's budget is taken out of the most attractive stat on that item. Thus, tanking gear with sockets should have less stamina than tanking gear without, plate damage gear with sockets less strength, leather with sockets less agility and so on. But of course they really have more of these things. They have more because that's what you socket them for. This is why we prefer gear with sockets to gear without, because we can maximize the one stat we want.

And so the problem is really that there is one stat we want. Almost every class and spec has a stat that is better than the other stats by a fairly significant margin. Even if they have two stats that are about equally good there is always the plateau of high stats followed by the drop off to stats that are worth two-thirds, half or even less than half as much towards their main function.

Suppose you have one stat that is good and every other stat is worth at most two-thirds of that stat1, and that you can get a stat that is two-thirds as good in both of the colours that don't have your main stat.

Aside from heads that have metas, the typical pattern for socket bonuses is that matching a single socket gives a +4 bonus, matching two gives a +6 and matching three gives a +8. Epic gems are worth 20 stat points. How extreme do circumstances have to be for you to gem outside your best stat?

A single socket item is only worth matching the socket if you get your best stat out of it. So if your best stat is strength, match a non-red socket only if the socket bonus is strength.

A two socket item you should match only if one of the sockets is the same colour of your main stat. So if your best stat is strength and the item has a yellow and blue socket, you should just gem strength even if the socket bonus is strength. If one socket is red to begin with then matching the other is good as long as the socket bonus is one of those two-thirds stats.

A three socket item should be matched only if either one of the sockets is the same colour of your main stat and the socket bonus is your main stat or two of the sockets are the same colour of your main stat. So match gems if two of the sockets are red and the socket bonus stat is useful or if one of the sockets is red and the socket bonus stat is strength.

In any other case just gem strength.

And this example is designed to be very favourable to socket bonuses and to variety. In reality most dps wouldn't value any blue stat at even close to half of their main stat and most tanks wouldn't value any stat at half of stamina2 let alone two-thirds. This means people gem all stamina, they gem all strength, they gem all armor penetration. They use other colours of gems to match sockets only to turn on their meta-gem.

Sockets on dps plate might as well just be replaced with twenty strength and we'll be done with it. Sockets are little more than a very complicated system to let people trade in small amounts of a stat they really want for small amounts of hit to get as close as possible to the hit cap without going under.

With a number of the changes in cataclysm - such as clearly defining and accepting the idea that each class/spec has a best stat and the coming rating changes - I think we have every reason to believe that the difference between your best and second best stat are going to get even more extreme (with the possible exception of tanks). This will mean more gemming for your best stat no matter what type behaviour.

In order to redeem gems stats need to have a relatively equal value. One stat will always be better, but it should not be half again as good as other stats. Without that, gems will slip more and more into just being passive stat bonuses by a different name. Gear with gems lets people trade gold for item points.



1. On an item point basis. For example, this would mean that if your best stat was Strength then either haste or crit would be two-thirds as good and a blue stat would be two-thirds as good. If your best stat was stamina then a red and a yellow stat would be as good as stamina point for point, since stamina only costs two-thirds of other stats in terms of item budget. Similarly, if your best stat is spell power than a yellow or blue stat would be more than two-thirds as good as spell power. You can see from these examples that this is an idealized situation that is making stats seem more balanced than they are. The reality of stat imbalance is much more extreme.

2. Valuing a stat at half a stamina item point wise would mean valuing it at three quarters of a stamina point for point wise.

Monday, March 8, 2010

Spell Power Coefficients & The Invisible Hand of Balance

This installment on spell power coefficients argues that the hidden nature of spell coefficients makes spell effects hard to understand for players. The argument is essentially this:
  1. There is no in-game source of information to distinguish spells with larger or smaller than normal coefficients
  2. Inconsistent wording and design of talents makes it hard to know how those talents affects spells
  3. Therefore the only way to access information about how your spells work is by calculating it from data, making the information non-transparent
One could argue that this isn't necessarily a bad thing, but I think that a) there is a good case to be made that transparency of the effect your gear has on your character is a good thing and b) WoW developers have clearly come in favour of transparency, for example, their justification for removal Armor Penetration Rating as a stat. When announcing that Armor Penetration Rating would be removed they used two arguments. One was that it was just a straight dps increase that happened in the same way as attack power increases dps and therefore was unnecessary. The other was that the calculation of it was too complicated and lacked transparency. If a lack of transparency is bad with Armor Penetration Rating, it is bad for spell coefficients as well.

Spell tooltips are not so kind as to include the coefficient of the spells. For the most part spells follow the cast time divided by three and a half rule, but some spells have a higher or lower coefficient and there isn't a way to know which ones or by how much. For example, Lightning Bolt and Starfire have typical coefficients. Arcane Barrage and Vampiric Touch have unusually high coefficients. Shadow Word: Pain and Mind Flay have slightly lower than typical coefficients. In the case of Vampiric Touch the coefficient is actually double the expected value of 100%.

There is no way to know this by looking at the spells. When a shadow priest hits level 50 and takes Vampiric Touch, she also learns a new rank of Shadow Word: Pain that deals 606 damage over 18 seconds. The base level Vampiric Touch deals 450 damage over 15 seconds. That's 101 vs. 90 damage per tick. Shadow Word: Pain is the harder hitting spell. A level 50 character typically does not have a lot of spell power. An heirloom staff, chest and shoulder total 73 spell power at that level. Depending on how dedicated a player is to gear it is easy to imagine having 150 spell power, but more commonly spell power would be closer to 100, or more like 50 for those who aren't handing off gear from higher level characters. Shadow Word: Pain gets 18.3% of your spell power per tick while Vampiric Touch gets 40%. So the first-time player who is just going through in their questing greens will probably find that in reality SW:P deals about 110 damage per tick and Vampiric Touch mysteriously deals pretty much the same. A player who is fairly tricked out for levelling - we aren't talking twink gear but enchanted heirlooms and craftables - is going to do about 128 per tick with SW:P and 150 with Vampiric Touch.

By the time we are level 80 and wearing good gear we already know that, as I said last time, base spell damage numbers are largely just window dressing for talents, glyphs and other effects of the spells. In this case, however, there is nothing in the tooltip for either spell, in any talent or glyph, that would indicate that VT is the harder hitter of the two. Unless they strip naked, any level 80 character is going to get more damage per tick with VT, but nothing in the game explains why.

On the subject of those talents that determine the real power of the spells, they themselves are far from transparent in their effects. There are two different things a talent can mean when it says it increases the spell power coefficient of a spell. They even appear in the same talent tree. Here is Empowered Touch from the druid restoration tree:

Your Healing Touch spell gains an additional 40% and your Nourish spell gains an additional 20% of your bonus healing effects.

And here is Empowered Rejuvenation from further down the tree:

The bonus healing effects of your healing over time spells is increased by 20%.

The difference is this: If the coefficient of a spell were 50% and I said to increase it by 20%, what would you think I meant? Would you think I meant to add 50% and 20% together to get 70% or to increase the 50% by 20% of 50% to get 60%? In the first case, they mean the former, in the second case they mean the latter. Of course the wording of the two talents is different, maybe we are supposed to know how they work from the different phrasing.

Here are some other wordings that appear on actual talents: "Increases the benefit from spell power gained by X by Y%"; "Increases the effect of X by an amount equal of Y% of your spell power"; "The spell power bonus to X is increased by Y%"; "Your X gains an additional Y% of your bonus damage effects"; "Your X gains an additional Y% of your bonus spell damage."

This confusion over what a percentage increase to a coefficient means even appears in two different ways of increasing the coefficient of a single thing. It's not spell power, but a Death Knight has two ways of increasing the amount of her Strength that her ghoul gains, one "Increases the contribution your ghouls get from your strength by Y%" and with the other "Your ghoul receives an additional Y% of your strength." To save you from looking it up, the former is the talent, and it means it increases the base 70% of your strength you share with your ghoul by Y% (Y = 20) of the 70% per point (or 60% of the 70% meaning 42% for all three points, giving a total contribution of 112%) while the latter is the glyph and means a flat add of Y% (Y = 40) of your strength to the ghoul's strength (so 110% of your strength is added for just the glyph without the talent). The talent is applied and then the glyph is applied so that the total contribution if 152%. If the glyph were applied first - and while this may seem counter-intuitive there is nothing in the game that says it isn't - you would instead add 176% of your strength to your ghoul's strength. Both things also affect stamina, but the base contribution from stamina is only 30%, so the final total is 88%. The talent and glyph have exactly parallel wording on Strength and Stamina, but the talent increases the ghouls' strength slightly more than the glyph does while it increases the ghoul's stamina by less than half as much as the glyph does.

There is no consistent wording or way to tell which is which. Moreover, even if you can tell which is which, since you don't know the spell power coefficient of spells to begin with, you don't know what half of these talents actually do.

This means that a person who wants to know how their spells work needs to go to a website and read formulas. This is precisely the behaviour that they want to prevent by getting rid of Armor Penetration Rating. Armor Penetration Rating isn't actually any harder to understand than spell power. For someone who isn't coding a spreadsheet or simulation tool, they both work the same way: having more is better because it makes you do more damage. In fact, the actual effect of Armor Penetration is to understand. There is one, admittedly complicated, formula. Any class that uses spell power usually needs at least three, sometimes more, equally complicated formulas that are no more transparent or available within the game.

Compounding this problem is that fact that spell power coefficients are used as balance tools. Back in BC mages paid the infamous coefficient tax on Fireball and Frostbolt. If they invested in the talents that lowered their cast time, these spells suffered a 5% penalty to their spell power coefficient. This was documented in a set of patch notes but was not indicated within the game. When shadow priests weren't doing enough damage in Wrath one of the fixes was to increase the coefficient of Vampiric Touch to the incredible 200% I mentioned earlier. This certainly upped Shadow Priest damage but it also took a spell that reads like it is meant to do weak damage (you cast it to get replenishment, not for it's punch) and turned it into one of the highest damage per cast spells in the game. At the same time they left Shadow Word: Pain with a sub-normal coefficient and subsequently made it the only shadow priest DoT unaffected by haste; the reasoning on that would really interest me.

I know in my last post I specifically advocated for varying spell coefficients, so it may seem strange that I am now complaining about them varying spell coefficients. But I am not suggesting the remedy to this problem is making spell coefficients uniform. Rather, I think spell coefficients should vary in a way that is consistent with and predictable (at least in a ballpark fashion) from the base spell. Hard hitting spells should have larger coefficients while weaker spells have smaller ones. That way looking at the base spells would give you an idea of the spells' comparative power. Of course this would involve retooling the power of the base spells (Mind Blast should not do triple the dps of Mind Flay) as well. Talents that modified one spell and not another would clearly change this balance, but presumably you know which talents you took and understand that taking a talent that makes Frostbolt more powerful but doesn't affect Fireball will increase the power of Frostbolt relative to Fireball.

Once this is done, all talents that modify spell power coefficients should be worded in one way and mean the same thing.

Furthermore, spell tooltips should show the total damage of the spell given your current gear as well as your current talents. Rather than my mind flay tooltip indicating that it does around 600 damage, it should probably say it does around 6000 damage, since that's the actual truth. Melee abilities show damage numbers based on current gear, it's time for spells to join them.

Tuesday, March 2, 2010

Spell Power Coefficients & The Irrelevance of Base Spells

For casters damage in the end game is largely governed by the invisible hand of spell power coefficients, and these coefficients have a number of major problems. In this first post on what is wrong with spell power coefficients, I'll talk about how spell power coefficients overwhelm and trivialize base spells.

Here is a sketch of the argument
  1. Most of your damage comes from spell power rather than base spells
  2. The relative strength of base spells rarely affects the relative spell power coefficients
  3. Talents and multipliers that affect spell power are what truly determine spell damage
  4. As a result it is the form rather than the content of base spells that determine their effectiveness
Spell Power coefficients come from an old standard default value of casting time divided by three and a half, or duration divided by fifteen for DoTs. AoE spells are penalized as are spells that have other effects (such as a snare). Basically, you should be able to add 2/7 of your spell power to the damage you do every second if you stand and chain cast nukes based on this formula.

At higher level the amount of a spell's damage derived from spell power overtakes the base spell damage dramatically. To use the example of Arcane Blast, the base damage of this spell is 1185 to 1377 for an average of 1281. The coefficient, with the increase from the Arcane Empowerment talent is 80.43%. For a well geared mage with 4000 spell power, this means the spell power component of this spell is 3217. Both the base damage and the spell power component are multiplied by various effects, but the ratio between them doesn't change. The pre-multipler damage of the spell is 4498, 28.4% of that is the base spell, and the remainder comes from spell power. Now 28.4% of your damage is certainly relevant, and you don't want to lose that, but certainly the substantial majority of your damage comes from spell power, not from the base spell.

While there are a few counterexamples, most spells follow the default coefficient as described above. So lets compare two other mage spells to consider their function and how spell power tends to wash out the design of the base spell. Levelling up as a mage, most players would be familiar with Fire Blast as a high damage quick burst they can use to finish an enemy off, or to pump in extra damage at higher mana cost. Comparing the base spell to a fire mage's main nuke, Fireball, Fire Blast is exactly that: quick high damage at a higher cost. The base damage of Fire Blast is 1010, it costs 21% base mana and it is an instant. The base damage of Fireball is also 1010, it costs 19% base mana and takes 3.5 seconds to cast. Since instants use up 1.5 seconds of your time, Fire Blast deals 673 dps while Fireball deals 288. The dps of Fire Blast on the base spell is two and and a third times that of Fireball. And when you are levelling the way you attack your enemies reflects this. You cast Fireballs on them and use Fire Blast when you can to kill them faster. If Fire Blast didn't have a cooldown you'd just spam it, and you'd be pretty great until you got up a few levels and spell power (and talents) started shifting the balance. Why wouldn't you pay 10% more mana per damage to do 133% more damage per time?

The talents that boost Fireball are more numerous and powerful than those that boost Fire Blast, so it is no wonder Fire Blast falls behind. But let's look at the untalented spells to get an idea of what spell power is doing to them. With 4000 spell power, both spells gain 1143 dps. That means Fire Blast deals 1815 dps while Fireball deals 1431. Now Fire Blast only deals 27% more damage than Fireball per time instead of 133% more. Would you pay 10% more mana for that? Of course you would, and you would continue casting Fire Blast, all other things being equal (which in reality they are obviously not) through to very high levels of spell power. But of course the total damage of the Fire Blast is 2724 while Fireball does 5010, so the mana per damage on Fire Blast is now 96% higher. Mana is hardly a thing, so mages would still do that, but if mana were a thing they'd really have to think twice. Spell power radically alters the relationship between these spells. The spell power coefficient increases the dps of each spell equally rather than increasing the dps of the spell with higher dps by more.

The problem is also very evident for shadow priests and in the relationship between Mind Blast and Mind Flay. If we had no spell power, Mind Flay would deal 196 damage per second while Mind Blast would deal 680 dps. Because of the 8 second cooldown, we can get 8 Mind Flay ticks for each Mind Blast. That means 272.4 dps overall from those two spells. By taking Improved Mind Blast we can drop the cooldown to 5.5 seconds. Idealizing this so that we can actually put 5.5 ticks between the Mind Blasts (which would be possible with the right level of haste, I'm just abstracting away the granularity) the dps would become 299.7. That's a 10% dps increase for five talent points. I'd take those talent points. The reality is that spell power inflates the damage of both spells to very high levels and makes the percentage difference between them very small. My actual average non-crit Mind Flay tick was 2584 in a recent raid while my average non-crit Mind Blast was 4847. Instead of Mind Blast dealing over three and a half times the damage of Mind Flay per second cast, it deals only about 25% more per second cast. Where does that leave the Improved Mind Blast talent? My own spreadsheets and Simulation Craft agree that the talent ends up giving you around 6 to 8 dps per point, probably closer to a .1% dps increase per point than a .2% increase per point, and that's in an ideal situation. If you are moving around, dotting multiple creatures, or doing other things with the time that Mind Blast is cooling down, it could be even less or nothing at all. Most shadow priests still take this talent, maybe in part because people will take talents that increase their dps in favour of talents that give some other kind of utility no matter how small the dps increase. But I think the main reason why people take this talent is because it intuitively seems like a useful thing. Looking at the two spells, Mind Blast does big damage and Mind Flay does small damage. You want to spend more of your time doing big damage, so the talent is appealing. In reality because the majority of the damage you do comes from spell power, both do medium damage. All spells do medium damage.

But of course some spells are much better to cast than others. Which of these spells would you rather have: A spell called Fireball that deals 1010 damage or a spell called Fireball that deals 0 damage but also deals 50% more damage. The second spell looks pretty stupid because 50% of 0 is still 0. In reality, with 4k spell power, every mage would prefer the second spell, and it might be enough to make mages all raid as fire instead of arcane.

So what do we look for to determine which spells are best? a) talents that increase the output of that particular spell and not other spells - so that the spell becomes relatively more powerful; and b) elements of the base spell that increase the base spells damage - spell that deal increased percentage damage under certain circumstances, effectively giving them a much higher spell power coefficient. It is the form of the spell, not the content that determines its worth.

What could be done to remedy this situation? More variance in spell power coefficients to bring them closer into line with the intended power of the spell is the answer. Of course that is going to take a bit of thinking. I don't think Fire Blast doing triple the dps of Fireball is a good model to go with, or similarly, Mind Blast doing triple the dps of Mind Flay. This would undoubtedly create burst damage problems in PvP. It would be nice, though, if reading spells gave us an idea of what they did and in what circumstances they were useful. Talents can obviously alter this, but some of the base spell should manage to shine through the talents and not be washed out under spell power coefficients that account for the vast majority of our damage.