Category Archives: Science!
Expansion Iteration: A visualization
I’ll be brief.
Blizzard Watch just posted a great analysis of the iteration times for all the past expansions. I took that data and visualized it.
Blue is the gap between the release of the previous expansion and the announce of the new one; red is basically the time between that announce and the start of beta; and the orange is the length of the beta to release.
Crunch it however you want, but the big takeaway for me is the damned heroic effort in getting MoP out in less time than all but one previous expansion, and that’s INCLUDING a total refactoring of The Jade Forest.
In Defense of Damage Meters
It is all the rage these days to disavow any regard whatsoever for damage meters. One is expected to denounce their use, purge them from one’s system, deny them access to your chat window, and make fun of those using them.
Well, I say, nuts to that!
If you are DPS, this is your instrument
Neil Armstrong did not land on the moon by looking out the window ((He looked AT the window, which had guide markers on it!)), he used instruments – and Buzz Aldrin calling out other instrument readings. Lindbergh didn’t even have a front window, he flew across the Atlantic on instruments. There are automobile races where the participants don’t even depart at the same time – they completely use instruments to determine who won.
In short, a reliable instrument is worth any number of other observations.
And a damage meter is the DPS role’s instrument of measurement.
You need to know if you are performing properly
The DPS role is dependent on its numbers, whether you take them subjectively or absolutely is irrelevant. But of the two, an absolute reference is much better than a relative one. Numbers are absolute. You can feed them into spreadsheets, save them off, compare them to each other. You can make multiple passes and chart your progress or lack thereof. Your damage meter is your friend. If you were doing 20K last week on a particular boss, and only 18K this week, you have something to look in to before you’re the cause of an enrage-timer wipe in the future.
"Well, fine", you say, "turn it on for your target dummies, I got no problem with that, but using them in a live encounter is bad!" To which I say, pfah! Target dummies give you a baseline, but they don’t take anything into account that you get from a live boss. You won’t see all the group buffs, or group procs, or even be able to use your execute abilities such as Kill Shot or Decimate. You might as well just sit there with autoshot, the approximation will have the same level of accuracy (and much less variability!).
No, a live boss (or live trash, if that is your interest) is the only way to truly gauge your performance in a raid setting. And since things vary depending on raid make-up, procs, and the like, you will need multiple samples.
Well, you don’t need to run it for everyone
Yes, you do.
You are not a single unit. You are part of a team. And how you perform relative to the rest of the team is important, if for no other reason than that of self-preservation. For if you’re performing in line with the guys at Elitist Jerks, but behind that of your guild (what, you think that EJ is infallible? Lol.). You may be in danger of being sat without realizing it. Because if you’re part of a serious raiding guild, I guarantee that your Raid Leader is watching your performance.
The more you know …
But that’s what World of Logs is for!
It is indeed, and in my opinion it is a far more accurate instrument than Omen or Skada, provided all members contribute logs (if it’s just you, then it’s on par with the other two, not better). But you probably won’t have WoL for all of your Heroics, trash runs, and so forth. You need all the things. Else your dataset is incomplete.
A damage meter is always there.
People use them badly!
They do indeed. Jerks spam chat with them all the time. But not you, right?
And the damage meters don’t do that automatically, so if yours does, it’s totally your fault. You are misusing the instrument. Stop it.
What idiots do with damage meters is not my concern, and it is not the fault of the damage meter. Get over it.
I don’t need a damage meter to know how well I’m doing
Yes you do. You will always do better with solid statistics than you will with a "gut feeling".
But if you just want to use the Force, have I got a game for you.
It might also be that you’re a PvPer and see no need. I contend that you don’t even belong in this conversation. Fire up All Healers Must Die and go do that honorable thing you do.
There are valid performance issues.
Yes, there are. But not for me, and not for most people that I know of. If, however, you are one of those people, and cannot afford a computer made after 2001, then by all means don’t run with one, because for certain it does suck CPU cycles.
You already have problems and damage meters are the least of them, but, whatever. ((I also recommend 800×600 resolution.))
Don’t shoot yourself in the foot.
In general, however, a damage meter is a valuable and useful tool for DPS self-improvement. Feel free to sneer at the idiots spamming party chat, and feel free to kick people that get hung up over somebody else’s DPS in a PUG ((We did that last night, and when asked why, the reason given was "Insufficient Beardiness." The fact that the mage in question was so quickly votekicked speaks for itself, however. He was an insufferable bore.)). But don’t blame the instrument for these things.
After all, both Tommy Dorsey and myself play the same musical instrument. But nobody has ever proposed that the Trombone be banned because of me. ((Well, my neighbors. And maybe my mom.))
Your damage meter is your friend
If you’re serious about self-improvement in a raiding environment, you need to use your damage meter to its fullest to provide nice, juicy data from which you can draw useful conclusions, and then apply those conclusions in such a way as to improve your performance (or detect bad decisions of that sort).
This is my damage meter. There are many like it, but this one is mine.
My damage meter is my best friend. It is my life. I must master it as I must master my life.
My damage meter, without me, is useless. Without my damage meter, I am ineffectual. I must use my damage meter wisely. I must DPS better than the boss that is trying to kill me. I must kill him before he kills me. I shall.
My damage meter and myself know that what counts in this raid is not the DPS we do, our meter dumps to raid chat, or the noise we make. We know that it is the overall damage that counts. We will do massive damage.
My damage meter is human, even as I, because it is our life. Thus, I will learn it as a brother. I will learn its weaknesses, its strength, its range, its triggers, its filters and its scope. I will keep my damage meter prepped and ready, even as I am prepped and ready. We will become part of each other. We will.
Before the Light, I swear this creed. My damage meter and myself are the defenders of my world. We are the masters of our enemy. We are the saviors of my life.
So be it, until victory is ours and there is no enemy, but peace!
Taking the mystery out of herb pigment yields
One thing that WoWHead could do ((They have an addon that automatically harvests data for them. If they tracked this, they could provide some great statistics.)), but does not, is report on herb milling yields. So the best you can usually do is comb through comments and hope they’re accurate.
The problem is that without accurate yield information, the enterprising Scribe cannot accurately forcast what her needs will be from day to day, and that makes for a sloppy business model. Whether you are buying off the auction house and need to know how many stacks of X to buy, or harvesting in the wild and just want to know when you can safely stop, it’s kind of a big deal.
Not only that, but different herbs yield different results. Logically, they seem to group thusly:
- Lower tier – Cinderbloom, Stormvine
- Middle tier – Heartblossom
- Top tier – Twilight Jasmine, Whiptail, Azshara’s Veil
The assumption is that the higher the tier, the more valuable it is for alchemists, therefore the better the pigment yields. But there was no solid evidence either way.
To get that evidence, then, I set out on a several-week study. Many high-end herbs were sacraficed in the cause of science – or at least, solid statistical data.
Here is the fruit of my labor.
You can see the raw data in this spreadsheet.
[table id=1 /]
When you look at this you immediately see that there are, in reality, only two tiers. Moreover, they don’t fall out the way we first expected. Azshara’s Veil, a high-end herb for Alchemists, is actually lower-tier in yield. And the two tiers are incredibly consistent within themselves. Basically:
[table id=2 /]
Simply put, I gathered herbs as needed and ground them up, counting the yields per stack and recording each one. For each herb I acquired 100 samples. For Cinderbloom, I acquired 200 to check for consistency (the first 100 is not there, don’t look for it).
And we conclude
So, what lessons have we learned?
Lesson 1: Tier 2 might be worth a premium
At a 12-10 advantage, Tier 2 can deliver Ashen Pigment at a 20% improved rate. If the premium for a Tier 2 herb is less than 20% over a Tier 1, you’re better off buying the Tier 2.
Better, Tier 2 offers a 2:1 advantage for Burning Ember. If you’re making decks, the gains will likely far outweigh the cost of the premium.
Lesson 2: In Tier 1, it doesn’t pay to be fussy
No matter what you buy, all Tier 1 herbs will deliver approximately 10 Pigments and one Ember. So, shop smart and don’t get too worried about “brand name” on these.
Lesson 3: If you’re not interested in Embers, Tier 1 is probably the best buy
As 4.3 hits the downloaders, the products derived from Burning Embers (T11 products) are becoming less interesting and not much in demand. A glyph proprietor will find them to be less attractive than Pigments.
True, that 20% boost to Pigment yields is a boon, but if the cost of Tier 2 outstrips Tier 1 by 20% or more, and all you care about is pigment, go cheap. It’s just math.
On the other hand, if 4.3 brings out some new gear to make with the embers (do they ever?), then Tier 2 will be very valuable to you should you choose to exploit that market.
Lesson 4: If you can get Tier 2 for about the same as Tier 1, jump on it
Despite its higher price, there is no denying the superior quality of the Tier 2 herbs. When opportunity knocks, don’t hesitate to reap the benefits of a fickle market.
On my server, Whiptail routinely falls below any Tier 2 herbs. On the other hand, Heartblossom rarely falls below Tier 2 herbs in price. Knowing what I know, I will always buy in the case of the former, and never buy Heartblossom unless it is a real bargain.
Lesson 5: Within the two tiers, things are incredibly consistent
You will hear anecdotal evidence, nearly religious-fervor sales pitches, and just plain assumptions that have never been born out. But this you can take to the bank. These are statistics, taken from live samples. 100 samples of each, which I believe to be statistically meaningful ((In that, if you plug in a big number into any one cell in that spreadsheet, the other 99 keep it pretty well in check. Freaky yields do happen, but it all averages out.)) and which have yielded remarkably consistent numbers.
Lesson 6: Unless you’re an alchemist, Azshara’s Veil isn’t worth it.
The scrawny herb population, awkward movement mechanics, and hordes of ninjas and cheaty druids make Vashj’ir a wretched experience, doubly so for farming herbs. Alchemists have no choice, they have to go by brand and no substitutions.
Stop throwing things, Flora!
Lesson 7: When farming, don’t let herb type be the deciding factor
Unless you’re hot for those Tier 2 yields, you’re better off choosing based on comfort, enjoyment, population, and presence of asshats. On my server, if I need a LOT of herbs, I’ll hit Deepholme, which as about twice as many nodes as any other zone at level 85. For my daily needs, I usually go to Uldum and get some tasty Tier 2s in there as well. I almost never go to Twilight Highlands because the cheaty druids pick it as clean as possible.
Whether you are farming or buying, knowing the yields for real gives you an advantage when it comes time to do the deed.
While it is true that price per stack will vary from realm to realm, the yield will be consistent across all realms. The former is based on player needs. The latter is simple game mechanics.
This late in the expansion, knowledge is power. You need every advantage you can get in a competitive market.
How many zones, again?
UPDATE – Darthregis pointed out a math error in the table. I’ve updated the post and table.
The actual count for Cata is 11 zones (previously stated: 14). You can read the comments for my conclusions on how this happened (hint: anatomical limitations)
Hey, I’m a frostamancer, not a mathamancer. If I was meant to count past 10, I would have more hands.
Regardless of the lower number, Cata is still perfectly competitive with the previous expansions. Nice to have a margin for error, in this case, since error did occur.
It’s seems to be popular amongst the blogeratti to say - in a dismissive fashion – that Cataclysm is in some way inferior to other expansions because it doesn’t have very many zones.
I did some quick poking around. Let me share this table with you.
Compare to BC’s 10 and WotLK’s 13 zones and suddenly Cata doesn’t really fall so flat. So let’s drop that argument.
One might object that three of the zones for Cata are all part of one mega-zone named Vashj’ir, but even so, that would change Cata from
14 11 to 12 9 zones. Not a huge difference, and not, say, half of the previous two.
I realize that some may consider Wintergrasp in WotLK, and I’m OK with that. Traditionally, PvP zones aren’t counted in such surveys. However, since you can visit and do things in it in between battles, it would be reasonable to accept WG, especially in light of Tol Barad showing up under Cata’s header (though it has a non-PvP half, which is what I counted). So, feel free to add WG, or take TB out. Doesn’t really alter the tally drastically and does nothing to support the argument that Cata is deficient in zone count – (added) at least not in the way that most people characterize it.