Tuesday, May 26, 2009

Lessons from Flame Leviathan Hard Mode

Here are a few things we learned while attempting Flame Leviathan (FL) with several towers up the other night (our best attempt: 25% with two towers up).

- The Freya tower is probably the hardest one since it introduces adds (one big one, then medium ones and then small ones). These adds change the dynamic of the fight radically so I suggest destroying this tower first.

- The Hodir tower slows down everyone and occasionally ice tombs players, who need to be freed by rockets quickly.

- The Thorim tower creates a beam that damages people around it and also increases the damage that Leviathan deals.

- Only Mimiron's tower beam goes along a predictable path, the other two seem to follow one person for a little while, they stop and then cause damage.

- Because of everything that's happening with additional towers, you can't expect to do the "kite clockwise" strategy. The fight becomes a lot more chaotic and while it's recommended to try to kite along a clockwise/counter-clockwise path, kiters need to do whatever they can in order to dodge beams as well.

- As a consequence of the previous point, motorcycles need to adjust very quickly by trying to predict where the kiter is going and then lay an oil slick on FL's path. Lighting up these slicks with the rockets is more important then ever.

- Flame Vents need to be interrupted at all costs, and because of the high enery cost of Electroshock, Siege Engine drivers should use the same strategy as we did during Malygos: wait for your energy to pool over 60% before you Ram, so that you can cast an Electroshock immediately if FL starts casting Flame Vents. Interrupting seems a bit flaky: ofentimes, it seems to go off (you see the blue arc) but doesn't interrupt.

- Passengers of Demolishers need to pick up Pyrite in-between launches. So far, we have given priority to launching them in order to maximize Overload times, but we probably need to balance this more with 2+ towers. The cycle should look like "Pick up 5 pyrites, get in position to be launched, launch, Overload, picked up by motorcycle, get back into a Demolisher, pick up pyrite, etc...". In short: don't get launched before you have reloaded 5 Pyrites (note: I haven't verified this in the field, it's just something we should try).

- If the Hodir tower is up, everybody gets slowed down, which means that everyone needs to start backing off earlier during Overload or they will quickly be caught up by FL.

Wednesday, May 6, 2009

worldoflogs.com

A newcomer in the combat log analysis web site: worldoflogs.com

Sample reports:

* Guild calendar
* Overall report
* Specific boss fight

The uploader works in real time (you let it run while you're raiding) and as you can see, the reports show more sophisticated graphs than wws and wmo: they show deaths, Blood Lust/Heroism times, etc...

Need to play with it more, but it certainly looks interesting.

Tuesday, April 7, 2009

As some of you know, I'm a big believer in "money over all". I think money is the best way to get mats and that even farming is usually a wealth loss since unless you get lucky, you will be much better off running a few dailies and buying the mats you wanted to farm off the AH.

With Ulduar just around the corner, I decided to move one step further in this direction and I started selling pretty much everything of value that I own.

The trend started with obvious items such as Frozen Orbs or Dragon's Eyes, but soon, I was selling various food, green and blue gems, leathers, scales, silks, eternals, etc... If my main is not guaranteed to need it very soon (which is pretty much limited to a couple of blue gems), off to the AH it goes. And hell, it can be argued that even if it's something I might need in two-three weeks, I might still be better off selling it now and buying it from the AH when the need arises.

I have a couple of reasons to back this up this decision:

  • Ulduar will make most items drop in prices at the AH. Note that I said "most" (see caveat below).
  • There is little point in immobilizing hundreds of gold in assets in the expectations that I might need it later. I'd rather sell all my stuff for 500g and have to buy some back for 100g in a few weeks.

I'm also planning on spending pretty much all my Emblems of Heroism to buy Frozen Orbs since these sell for 80g pretty reliably. I'm even considering converting a few Emblems of Valor for that, although I still occasionally see people looking to buy the bracers for 1500g+, so it might be worth holding on to at least 60 of those.

My big question is: what items do you think will actually go up in price when Ulduar comes out?

Thursday, April 2, 2009

Powerleveling Jewelcrafting

I'm almost done powerleveling Jewelcrafting and since I already received a few questions, I thought I'd share my notes with everyone.

Blizzard really did a good job with this profession in WOTLK.

One thing I like about Jewelcrafting is that you don't need to max it out to start deriving benefits from it, even if you raid. The target is 375, which is when you unlock the daily quest and when you can cut the Dragon's Eye gems, which only Jewelcrafters can use.

These gems are prismatic, meaning they can fit in any slot, and they are also more powerful than the regular gems (e.g. +27 strength instead of +16). You can't use more than three on your person, though, but it's still going to be useful for people who are using blue fillers in order to satisfy a metagem requirement.

In my case, I had two purple gems +12 sta/+8 str, which means that once I used of my three Dragon's Eyes, I ended up with +59 str (19 + 19 + 11), which is nothing to sneeze at.

Here are the guides I used:

WOW Wiki
WOW Guide Online
WOW professions

It cost me about 2000g to get to 420, so it wasn't too bad. Make sure you use two guides when you powerlevel because they all have holes at some point, or they give you instructions that are sometimes more expensive than their alternative. AuctionLite and Postal are lifesaver add-ons for intensive powerleveling since they save you a lot of clicks.

Leveling slows down after 420. You can get 5 more points by creating Prisms, which have a 24 hour cooldowns, but other than that, you'll need to create meta gems to max out. It's not that expensive if you can get an alchemist to transmute the gems for you, and each recipe costs five tokens, so that's five days of dailies.

Unfortunately, just like all the other professions, there doesn't seem to be much point in leveling Jewelcrafting past 440. Hopefully this will change with Ulduar, although I haven't read anything about this yet.

In the next post, I'll cover 420-450.

Wednesday, February 25, 2009

PTR 3.1 early impressions

First of all, check out the gorgeous 2mn preview of Ulduar.

I'm not sure if I'm more impressed with the movie or with the content of the page:
<?xml-stylesheet type="text/xsl" href="layout/ulduar.xsl"?>

<page lang="en_us">
<ulduar/>
</page>
How awesome is that?

As for the PTR, the client keeps crashing so it's barely usable right now, but the short time I spent there confirmed to me that dual specs are fantastic. The UI is very intuitive too, and you get a full set of new toolbars for each spec (I was afraid they would limit the change to the main bar, not sure why).

I was a bit disappointed to find out that talent points are still spent immediately, so you need to plan ahead and not make a mistake or you'll have to pay the respec fee to start again. I seem to recall that Blizzard wanted to do the same thing as the Talented add-on: you define your spec at your own pace, move points around until you're satisfied and then you say "Ok" and the spec is applied.

Oh, and instant spec change == epic win.

I haven't tried the gear manager yet but it seems to be a bit primitive compared to the existing add-ons (and it doesn't appear to automatically switch when you change your spec). I suspect that ItemRack/Outfitter will still be useful, especially for their abilities to define gear sets based on other sets.

I noticed a few changes in the DK talent trees that were not documented anywhere, need to spend more time playing with them.

Monday, February 23, 2009

Calculating a gear score by scraping the Armory

I am very impressed by the content of the Armory pages: they're 100% XML. No HTML, no CSS, no Javascript. An XSL spreadsheet does all the magic and a request just returns your full model in XML.

As an experiment, I wrote a quick Ruby script that calculates a gear score for a character. Here is a sample output:
Valorous Bonescythe Helmet:213
Icy Blast Amulet:213
Valorous Bonescythe Pauldrons:213
Valorous Bonescythe Breastplate:213
Belt of the Tortured:213
Valorous Bonescythe Legplates:213
Boots of Captain Ellis:213
Thrusting Bands:213
Dislocating Handguards:213
Strong-Handed Ring:213
Ring of Invincibility:213
Mirror of Truth:200
Fury of the Five Flights:213
Hammerhead Sharkskin Cloak:213
Webbed Death:213
Sinister Revenge:226
Spinning Fate:213
=====
Gear score for Rozz: 3595
Reference: Blue (Superior):2992 Heroic:3096 Naxx10:3200 Naxx25 (Epic):3408 Max:3616
The script is about forty lines of Ruby and you can download it here.

Thursday, February 19, 2009


Here are the dps charts from our latest Sarth3D kill last night.

Top DPS. I haven't seen that since the early days in Black Temple on my Rogue, and Rogue DPS has been declining steadily ever since.

I still find it disturbing that a Death Knight that has been raiding for basically two weeks can top the dps charts like this, even though Sarth3D is obviously very AOE intensive.

Wednesday, February 18, 2009

Confessions of a botter

No, not me, him.

This guy makes very good points (and his responses to haters in the comments are worth their weight in gold), some of which hit near home for me.

I have run out of things to do in WoW besides raiding, and that's even after I switched mains (my DK now has almost reached the same level of gear that my Rogue had).

I'm playing Left4Dead when I'm not raiding now, but honestly, I would rather continue spending my videogame time in WoW, and that would probably be possible if I had more than two 80 characters, but I'd rather eat my own eye than leveling one more toon in this game.

Thursday, February 5, 2009

Twilight Vanquisher

We finally killed Sartharion + 3 drakes last night.

The final kill was actually one of our sloppiest attempts of the night. We had killed Shadron a couple of times in our previous wipes with most of the raid still alive, but for some reason, we lost one healer and quite a few dps to enraged drakes and void zones during the Shadron battle. Still, with Shadron dead, the damage on our Main Tank decreased considerably and the remaining healers were enough to keep him (and the adds tanks) up.

I was one of the unfortunate people to die early, which was a bit frustrating since I'm usually one of the last people standing on our attempts, but this allowed me to witness the rest of the fight as the rest of the dps slowly whittled away Sartharion's health while keeping the add population under control and dutifully entering each portal as it spawned.

Looking back at our history, it took us over 120 attempts and it cost the guild about 20,000g in repair. By far the most difficult fight we ever faced.

Sunday, January 25, 2009

DPS ranks per class for Wrath of the Lich King

I came across this very interesting chart on wowmeteronline.com that lists all the DPS by all the classes on the boss fights for Wrath of the Lich King (Naxx and Obsidian Sanctum). It's a good way to get a quick reference for how well you are performing on a boss-by-boss basis, but I thought I could derive more information by calculating scores for all the classes.

Here is the current standing:







































ClassScore (lower is better)
Mage36
Warlock56
Death Knight60
Hunter60
Warrior72
Shaman97
Priest107
Paladin119
Rogue123
Druid150




You can find the full details in this spreadsheet.

Mages, Warlocks, Death Knights and Hunters have a commanding lead in all-around DPS, and it's especially surprising to see Death Knights in the top 3 since they are a melee class (most of the fights in Naxxramas and the Obsidian Sanctum are very mobile and therefore more suitable for ranged dps, which doesn't need to run back in melee range before resuming DPS).

My biggest dissapointment is to see Rogues so far down. For one of the few single-spec pure DPS classes in the game, it's depressing to see that they have so little to contribute overall. Not only do they provide close to zero raid wise utility buff, but their DPS can only be noticed on very specific fights, and it's not even that great. Another surprise for me was to see Priests so far down as well. I have noticed very impressive DPS from shadow priests these past weeks, but I guess they don't get a chance to express themselves fully in these fights.

Druids are the worst off but even though I have found Cat Form dps to be average overall, I can't really explain their very low score on this chart. I'm guessing the charts I got the numbers from don't really take specs into account, so it might be that most of these were not DPS specced.

Friday, January 23, 2009

Malygos Power Spark buff and dots

I have been wondering for a while if the damage buff we gain when standing in the Power Spark aura applies to dots or not, so I decided to find out for sure. 

Short answer: it does. 

I used wowwebstats.com to find out. The filter is unit="rozz" and spell="rupture" and you can see the result here.

In case the URL is not working, I pasted the trace below. In short, my initial Ruptures tick in the 700-900 range. Then they jump to 1628 (I believe we had two sparks stacked then), then drop to 1345 (probably down to one spark) and back to 900. 

I also tried wowmeteronline.com but 1) the event log browsing is broken and 2) it reports DPS that are half(!) those of wowwebstats.com. Has anyone else tried to update the same report to both sites to compare? 

The bottom line is that you can keep using your regular rotation while standing in a spark aura, unlike some other fights such as Loatheb (which only increases crits, so it doesn't make sense to use dots on this fight while you have the spore buff). 

-- 
Rozz 
your friendly and theorycraft obsessive rogue 



04:35'29.762 Malygos #1 suffers 965 Physical damage from Rozz Rupture. 
04:35'31.589 Malygos #1 suffers 743 Physical damage from Rozz Rupture. 
04:35'33.549 Malygos #1 suffers 743 Physical damage from Rozz Rupture. 
04:35'35.549 Malygos #1 suffers 743 Physical damage from Rozz Rupture. 
04:35'37.557 Malygos #1 suffers 742 Physical damage from Rozz Rupture. 
04:35'39.553 Malygos #1 suffers 743 Physical damage from Rozz Rupture. 
04:35'41.525 Malygos #1 suffers 743 Physical damage from Rozz Rupture. 
04:35'53.228 Malygos #1 is afflicted by Rupture. 
04:35'55.269 Malygos #1 suffers 941 Physical damage from Rozz Rupture. 
04:35'57.302 Malygos #1 suffers 941 Physical damage from Rozz Rupture. 
04:35'59.293 Malygos #1 suffers 941 Physical damage from Rozz Rupture. 
04:36'01.300 Malygos #1 suffers 940 Physical damage from Rozz Rupture. 
04:36'03.287 Malygos #1 suffers 941 Physical damage from Rozz Rupture. 
04:36'05.241 Malygos #1 suffers 941 Physical damage from Rozz Rupture. 
04:36'07.240 Malygos #1 suffers 723 Physical damage from Rozz Rupture. 
04:36'09.256 Malygos #1 suffers 724 Physical damage from Rozz Rupture. 
04:36'27.374 Malygos #1 is afflicted by Rupture. 
04:36'29.436 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'31.401 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'33.358 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'35.336 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'37.317 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'39.359 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'41.353 Malygos #1 suffers 1628 Physical damage from Rozz Rupture. 
04:36'43.383 Malygos #1 suffers 1627 Physical damage from Rozz Rupture. 
04:37'15.141 Malygos #1 is afflicted by Rupture. 
04:37'17.169 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'19.167 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'21.149 Malygos #1 suffers 1034 Physical damage from Rozz Rupture. 
04:37'23.175 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'25.207 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'27.231 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'29.143 Malygos #1 suffers 1345 Physical damage from Rozz Rupture. 
04:37'33.986 Malygos #1 is afflicted by Rupture. 
04:37'36.072 Malygos #1 suffers 928 Physical damage from Rozz Rupture.

Tuesday, January 6, 2009

Sartharion + 1 Drake

It was a pretty good week for us: not only did we kill Malygos in 10 man but we also had our first kill of Heroic Sartharion + 1 Drake (Vesperon, the health one). I give Heroic Malygos less than a week to live.

The Sartharion fight was a lot of fun and deliciously chaotic and so far, it qualifies as the supreme test of "don't stand in shinies". Our initial attempts made all the expected mistakes one by one: being overrun by adds, people dying from void zones, people being wiped by lava walls, tail swiped while running away from a wall, and of course, any combination of the above ("running away from a wall and then running into a void zone" or "tail swiped and then swept by a wall", you get the idea).

We debated a few times the strategy once inside a portal. The questions were:
  • Get all the DPS in or only melee and range stays out to deal with adds?
  • Go all out or try to time the kill so we come out of the portal right after a lava wall?
What ended up working for us was "All DPS inside" and "Go all out and be careful when you come out". There is still a little bit of risk associated with this last part but being hit by a lava wall doesn't have dramatic consequences and with our DPS, the Acolyte inside the portal usually goes down in less than fifteen seconds.

As for the rest, we tank Sartharion in the upper left part of the island and the Drake in the lower right corner.

It took us about one hour and a half to get the kill.

Saturday, January 3, 2009

Dead Malygos

Here is the strategy we used to kill Malygos (10 man) the other night. I'm mostly focusing on the little details that we learned during all the wipes it took us to get there, please make sure you read the basic strategy for the full details on the encounter.

Phase 1

We have someone call where the Power Sparks are coming from and the main tank reacts by moving Malygos away from it ("Next spark north west", "Turning him south"). This way, the party is positioned between Malygos and the incoming spark. Additionally, the main tank will pull Malygos slightly toward the outer rim of the platform in order to maximize the odds that the Sparks will all die in the center.

Note that on our kill, we didn't do a perfect job at it: the sparks were not optimally stacked and I think a couple even reached him. We still managed to kill him before the enrage, so there is a little bit of leeway here, but obviously, stacking the sparks makes our job much easier in phase 3. We also try to decide when it's best to call a Blood Lust (typically when we have two sparks stacked).

Phase 2

We call an order for melee to climb on the platforms. Someone puts a skull/hunter mark on the closest Nexus Lord that lands and we all focus fire on him. Move to the next dome when it appears and keep track of the Deep Breath timer to make sure you won't be out in the open when it hits.

In order to avoid a situation where the tank and his healers are standing in two different domes and unable to get together, healers need to stay with the tank, even if he's not moving to the next dome right away.

Whenever a Nexus Lord dies, a melee gets on the platform according to the order decided. It is very important to let the melee go first, because a melee staying on the ground during this phase will be useless, which will make us lose precious time for phase 3. I'm not quite sure if there is a point in a range getting on a platform once all the melee is up in the air since the ranged person will probably do more damage from the ground(?).

Phase 3

We experimented with the "one healer per group" approach but it didn't work well in 10-man, so we ended up having raid healers. Everybody groups together at the south and we have 2 or 3 designated healers. We should be able to run with 2 eventually, but we'll start with 3 and see if we hit the enrage timer with people still alive, which would be a huge progress. If this happens, we'll just use one less healer.

The healers stack up 5 HOT's on themselves (3 3 3 3 3) and then cast an AOE heal (4). Remember to save some energy to be able to cast Flame Shield (5) in case you get targetted with Surge of Power, so you might want to insert a delay in your cast sequence to let the energy fill up (3 3 3 3 3 [pause] 4 3 3 3 3 3 [pause] etc...).

Same idea for DPS: 1 1 [pause] 2 1 1 [pause] etc...

Occasionally, Malygos will cast an Arcane Pulse, which causes a big AOE splash damage for anyone staying around it. This is the cue for the entire raid to move clockwise. We will all move 30-40 yards on our left and resume our cast rotations until the next Arcane Pulse, which will cause us to move again.

It is absolutely mandatory for everyone to have DBM installed, and not BigWigs. BigWigs will *not* notify you when you get targetted with Surge of Power, which means certain death. Again, we'll try to have a few people dedicated to calling who is about to be targetted so they can cast their Flame Shield, but don't rely on it: you should be able to manage this entire phase just by paying attention to what's happening around you.