Wednesday, 11 July 2012

A Recent Delve into Solo Top Nunu


With me finally settling into a work routine after uni I've found myself able to play more games and actually write a new article inspired by a recent game.

It was a fairly basic normal draft game with me planning to try out Jayce solo top as I had just got him however that all changed when the enemy team picked one of the stronger top laners: Yorick. Naturally with me never having played Jayce at all and being fairly out of practice in general I didn't really want to lane against Yorick especially if he was a good one, after a brief moment of thought I locked in as this fellow:


uhm... I meant these guys:


The league's first mounted champion (none of that Sejauni stuff) makes a very strong tanky AP top, especially since they cut the cooldown on his consume and the number of attacks needed to charge his passive to attempt to make him better at jungling.

Nunu's strength as a top laner comes from three factors about him:

 First off we have consume the ability that gives Nunu stupidly good sustain. A minion only nuke that a) guarantees a last hit and more importantly b) Heals Nunu for 125/180/235/290/345 pretty impressive for a 60 mana skill, though not quite as impressive as its 1.0 AP scaling. Rushing a RoA on top lane Nunu makes you pretty much invincible due to the health and AP.

 Next we have Nunu's harassment, anyone who has played against a support Nunu will know how much this skill hurts with little AP not to mention if he has a RoA or AP runes, AP Top Nunu will max this skill as quickly as possible as it is a 6 second cd spammable nuke with again a 1.0 AP ratio and a hefty slow on top of it which stops a typical melee solotop from being able to hit you back.

 Finally the other key component of Solotop Nunu, his passive, for every 5 autoattacks nunu's next spell will be completely free, this allows you to spam iceblasts in-between trades and last hits, keeping your opponent at bay as they will mostly have melee harassment and back off after a few iceblasts.

The trick to top lane Nunu is to be a bully, a massive bully to the other person in lane. Your inherent tankiness allows you to run in and autoattack the enemy champions in order to get your passive charged for more harassment. 

As mentioned earlier I picked Nunu for a Yorick matchup and I strongly recommend anyone who sees the other team pick Yorick do the same, Yorick's ghouls not only give you something to eat with consume but also give you a nice target to autoattack allowing you to charge up your passive without pushing the lane too far.

Well that's a brief overview of Solotop Nunu, I wont bore you with build details as RoA+Defensive items is pretty standard on him and works very nicely solotop. Have a good one with your Nunu games.

It's good to be back,

Thalric





Thursday, 19 April 2012

Stacking Items


Inspired by the  double stack Garen build on Reddit I've decided to do a post on When people talk about stacking items we generally mean Sword of the Occult, Mejai's Soulstealer and Leviathan due to them getting stacks purely on champion kills (2 stacks for a kill, 1 for an assist) however we'll also include Warmog's Armor and Bloodthirster as they also incorporate stacking elements.

With all but Warmog's a proportion of the stacks are lost when you die, this is to create a risk vs reward mechanic with these items rewarding better play with snowballing and punishing weak play with items that are sub par to others of the same price. Due to this mechanic they are usually frowned upon in competitive games. With the Sword, Soulstealer and Leviathan 1/3rd of your stacks are lost upon death and with Bloodthirster 1/2 of them are. Due to the typically large amounts of champion kills on Dominion all these items are banned there. Okay now the basics are out of the way let's look at these items in a little more detail:

 Sword of the Occult, built from a Long Sword and a hefty 954gold for a measly +10 attack damage Sword of the Occult starts off in a pretty bad place, however a decent AD carry or tanky dps character who can get kills and hold onto them could use this item well as you get +5 damage per stack. Compared to other early game AD items it requires at least 5 stacks on it to just be cost effective however Sword of the Occult does have a fairly big perk, it generates more AD than any other flat AD item in the game, even more than a fully stacked Bloodthirster. Of course getting it to that level requires either a great team on your side or stupid plays by the enemy but anything can happen in games. Also upon reaching 20 champion kills or assists you get the benefit of a nice +15% move speed which helps you flee to hang onto your stacks.

 Mejai's Soulstealer is the stacking item for AP champions built from an Amplifying Tome and  800 gold, it grants you 20 ability power straight away (so no better than the Tome it was built from) it grants 8 Ability Power per stack and also requires about 5 stacks on it to just scrape by on cost-effectiveness but unlike Sword of the Occult it is not the best AP item in the game as even at 20 stacks (Deathcap and fully charged Archangel's both beat it AP wise) Though at 20 stacks you do gain a nice 15% cooldown reduction but on champions that need it you can easily hit the 40% cap without it. However on champions with global abilities, namely Karthus and Soraka, it can be very easy to get charges on it due to Karthus's ultimate damaging an entire team can easily finish off injured enemies, at the least could grant him 5 assists each victorious team fight you team has and Soraka can use her ultimate to heal and remember if you heal one person who takes part in a kill you get an assist. This allows them to get stacks from a safe distance without risk of losing them, however getting Mejai's on them is still seen as suboptimal as a lot of items can benefit the characters more.

 Last of the champion kill stacking items and the one that seems to be the least popular is Leviathan, built from a Ruby Crystal and 800 gold it gives you the ruby crystal's bonus health straight away and then grants you 32 health per stack you acquire. With 20 stacks you will gain 820 health from Leviathan which is 100 less than an unstacked Warmog's. However at 20 stacks you do take 15% less damage resulting in you being a fair bit tankier from it. The problem is that, apart from this % reduction at 20 stacks, Leviathan only gives health and generally flat health without something else is a poor choice unless you already have a large amount of MR and Armor and even then Warmogs gives a lot more health so is generally a better choice for an item slot than Leviathan.

Okay now let's move onto some of the items that actually see quite a lot of play:


 Excluding the rarely picked Sword of the Occult, The Bloodthirster is the highest Attack Damage item in the game. Costing a total of 3000gold Bloodthirster's base stats are +60 Attack Damage and +12% life steal and you gain +1 AD and 0.2% lifesteal for any minion or champion kill you get, to a maximum of 40 kills, fully stacked that's +100AD and +20% life steal which is pretty impressive on any AD character. You lose 1/2 your stacks upon death but they are easy to reclaim off minion waves and once you're fully stacked all that life steal makes it quite hard to die. Another note is that Bloodthirster's passive isn't unique so you can stack them for even more AD and Lifesteal but that makes you a big target and lose a lot of power when you die.

 Last but by certainly no means least you have Warmog's Armor. This pretty expensive, 3000 gold in total, item grants you a huge chunk of health straight off giving you 920 health and 30 health regen, however you gain +3.5 health and  +0.1 health regen per minion kill and +35 health and +1 health regen for each champion kill or assist. It takes a maximum of 100 minion kills or 10 champion kills/assists to reach it's full potential of +1270 Health and +40 health regen. You can't lose these stacks through death so ,despite the potentially large amount of time needed to charge it up in a game with few kills, so it's a solid defensive item for health. However health alone without other defenses is fairly useless stat though most characters who would buy this generally get a large amount of armor and mr, usually in the case of an atma's for armor and to let Warmog's also give you a large boost to your offense.

Well that about wraps it up for the stacking items in the game hopefully I'll have a less point out the obvious boring article next time, I just felt like I should spend a moment to talk about them.



Wednesday, 18 April 2012

Hecarim Patch Notes Commentary


Okay with the Hecarim patch finally out I feel like doing a commentary on some of the changes in there. I wont do anything on Hecarim himself until I've had a good chance to pick him up and play.

Okay let's start with Champion Changes:


Ignoring what is a small tweak to Annie the changes start with Akali base armour and health both increased by a small amount to make her slightly tankier early game and some bug fixes on both her shroud and ult. Her shroud notably now should always give assists when you hide in it and slow people so Akali players should be getting more assists round the board and should be pretty happy with these small tweaks to make her slightly better.

From then it's pretty much all bug fixes until Kayle where her generally lackluster heal is getting some changes with the amount healed and movement speed increase both buffed fairly substantially at low levels, the movement speed duration increased to 3 seconds and the mana cost increased slightly with the greatest difference seen at low levels. These changes should help both solo top Kayle's sustain and support Kayle's healing potential and general utility. Here I was thinking riot hated all heals and sustain...

Janna is getting a couple of tweaks, her tornado is being changed so you actually have to charge it to gain the same knockup that you got instantly pre-patch. This will make Janna players plan ahead more as it will be slightly less useful for chasing and instant CC. Her ult's mana cost is being reduced by 50 across the board which is a good buff however the radius is being reduced to 875. Overall a bit of a rocky patch for Janna with some buffs and some nerfs.

From Janna it's all bug fixes until Orianna where her command: attack's mana cost being changed to a flat 50 at all levels, this should further encourage maxing command attack first and will give her substantially stronger laning as a result of this spammable, powerful, low mana cost poke. After Orianna, Sejauni gets some cooldown reduction on her ult and a small mana cost reduction on her Arctic assault. This shouldn't affect Sejuani's playing too much but it is still a nice little buff to a rarely seen champion.

Swain's getting mana cost reduction on his ult and is getting a slight improvement to his godawful auto-attack animation, as a result he should be able to stay as a giant cr- CAWCAWCAWCAWCAWCAWCAW for longer and be a little better at last hitting.

One of the major things from the patch preview was the Teemo changes, though only a bugfix to toxic shot it means that suddenly on hit Teemo should be doing noticeably more damage along with that we have a small bugfix to his ultimate and Move Quick's cooldown has been reduced meaning the little yordle/rabbit/badger will slip out of your grasp more often. More small bug fixes for lots of champions until we get to Zilean who is getting a mana cost reduction on his Time Warp and Chronoshift which should make support Zilean a little more appealing to people.

Well that about sums it up for champions let's move onto...

Item Changes:


I've already discussed the changes to Atma's Impaler below so that's already been dealt with. Doran's Ring and Blade are both dropping down from +100health to +80 making stacking them slightly less viable for winning your lane as you're slightly less tanky but apart from that they still give decent all round stats for early game but aren't quite as great value as they were before. A small buff to Hexdrinker making the shield last longer, this combined with the release of Maw not too long ago should make Hexdrinker a more popular item. Finally on major item changes the cost of wit's end is going up by 150 gold, though a nontrivial increase in price I think wit's end is a good enough item for it to still be bought on many champions and the price rise wont affect it's popularity or usefulness too much.

'General' Changes:
A couple of HUD upgrades like showing your health and mana regen on the bars and displaying damage above your enemies when you hit them are quite nice and well received but the major change this patch is the changes to champion kills. You now gain substantially more xp for slaying an enemy of a higher level than you and get reduced xp for killing lower level ones. As mentioned in the patch preview this is especially good for supports as they can tend to drop behind and getting more xp for their assists will mean when teamfights are in full swing they'll be able to easily catch up. This also helps losing teams fight back and possibly pull a win out of a loss, as happened on my team just before writing this article. Combo'd with this is the fact that the maximum gold for stopping a killing spree has been boosted to 600 making taking down strong opponents even more rewarding. I'm fully in favor of this change as it helps counter snowballing teams and gives more late-gate oriented teams a better chance in competitive play.

I'll throw the rune changes into general changes, minor MR rune nerfs will probably effect tanky dps the most as they tend to run MR blues and will make them more vulnerable to mages early game. This being combo'd with buffs to both magic pen and AP runes meaning mages should be a little bit more dangerous to most people early game.

Well apart from adding in a couple of bots to Co-op vs AI that about sums up this patch. Quite a lot of changes in there and in my opinion pretty good ones at that. Now we wait for the Varus patch which hopefully shouldn't be delayed as long as this one.


Saturday, 14 April 2012

My Little Mumu

Today I'm going to be talking about this little chap:
Amumu the Sad Mummy hasn't been getting the same attention he had when I first decided to take the short lived plunge into ranked in Season One. He used to be a very common first pick or instant ban and if he got into a game it was generally a win for his team if the Amumu player was any decent. Nowadays the undead midget doesn't seem to have any love in anyone's eyes but I'm here to analyse if he is still a strong pick or if the community ignoring him is because he is genuinely sub-par.

Looking out over his patch history it is immediately clear that Amumu hasn't been touched much by Riot since his release, a few minor changes such as Curse of the Sad Mummy being instant damage instead of over the duration and the couple of buffs to him should suggest that he became stronger than when he was always banned.

Before we look in further depth as to why Amumu isn't considered the powerhouse he once was let's have a look at his skills:

His passive:Cursed touch reduces the magic resist of anyone he autoattacks by a decent amount for 4 seconds. This is a pretty great skill for 1v1s as most of his damage is magic and helps your team's mage burst down people in teamfights.

Bandage Toss is a pretty great skill: useful for initiation, chasing and CC'ing during teamfights its only downside is its fairly long cooldown (8 seconds at level 5) With long range, a one second stun, decent base damage and a 1.0 AP ratio it is a very useful skill.

 Despair is a great ability for taking jungle buffs, clearing camps and providing constant damage during teamfights, costing only 8 mana a second it is fairly easy to constantly have this up for teamfights and even all the time if you have blue buff. This skill gives a decent dps AoE toggle that deals a percentage of the target's max health per second in addition to base damage. The percentage damage has a low AP scaling.

 Tantrum provides a passive reduction to physical damage taken, making Amumu inherently more tanky by just leveling this skill, it also has an active AoE effect that deals decent damage with good AP scaling (0.5AP ratio). This skill is a strong jungling skill due to the fact each time he is hit the cooldown is reduced by 0.5 seconds allowing Amumu to spam this skill as he clears camps.

 Curse of the Sad Mummy... This skill is a large AoE nuke (with a 1.0 AP ratio) plus root stopping all enemy auto-attacks for 2 seconds. A successfully timed Curse of the Sad Mummy will win a teamfight for you as it effectively immobilizes the enemy team allowing your carry and mage to lay down a lot of damage. The only problem with this skill is ensuring it is off cooldown for every teamfight due to it having a relatively long cooldown.

Due to the AP ratios on Amumu are fairly good it can be tempting for many to build him as a straight mage...don't. Though it could be very effective against teams not prepared for it, against teams with good communication and skills he will be squashed every time going into a team fight. All his skills either have short range or get you close up to the enemy resulting in you being in the middle of the enemy team for the majority of teamfights. I'm not saying don't ever buy AP on Amumu ever, if you already have enough survivability from your items and the enemy team has little magic resist then by all means grab an AP item and go to town.

So overall he still has a strong CC kit that will throw out enough AoE damage to punish anyone who ignores him and his ultimate will secure won teamfights. His poor laning presence is mitigated by the fact he has really strong jungle clears. So... what's wrong with him? One of the answers is that Riot making the jungler easier which had the resulting of bringing out a number of new strong junglers to play. Junglers who's clears are faster, safer and have better invasion and 1v1ing. Though I'm still of the belief that Amumu's clear time is good enough to warrant him being a strong candidate for the jungler role and his ganks ,both pre and post level six, along with his teamfight presence should keep him at the pinnacle of top champions. 

Ah well...another possibly 'sleeper OP' champion I guess.

Friday, 13 April 2012

So... Twisted Treeline


Twisted Treeline has always been void of love from Riot since I've been playing with all patches being for the balance state of Summoner's Rift with Riot avoiding Treeline like a plague. However it was with this, the April fools announcement for 'League of Legends 3D' where Riot ignored one of the maps entirely in favour of Dominion and Rift. Despite Dominion seemingly being about as popular as Treeline it is riot's more favoured map for some reason and naturally Rift gets the most attention due to it being the most played.

However in my opinion Treeline is the most fun map, I'm sure many would disagree, the action is more fast paced than Rift but keeps the same basic structure, the matches are shorter so you can get easily one and a half to two matches in the time it would take you to play one Rift game. However some people claim it is hideously unbalanced with some champions who are only average on Rift being powerhouses on Treeline. Examples of some of the more powerful champions include Jax, Dr. Mundo and Yorick.

 Strong champions on Treeline seem to be relatively tanky, have a good early game, maneuverability, good CC or strong abilities to 1v1. A couple of the above features on a champion tend to make them perform well on Treeline though as I said in my earlier article about the meta it is still more player skill that matters than champion selection. Though a small meta-game of two people bot lane with a solo lane top has developed and I can't say that I like it (even though more often than not I conform to it) the benefits of running a dedicated jungler with all the Treeline buffs (more on jungling below) to leave two solo lanes to get all the experience and gold. Even if you don't run a dedicated jungler it makes more sense to run two laners in top as their are more buffs in top jungle than two bot.

Speaking of the Jungle and buffs on Treeline there are more than on Rift and I will briefly run through each one:

Grez the Lizard Lord- Your standard redbuff (slow and dot) giver except actually named as opposed to Rift's Lizard Elder Grez is also slightly tougher and hits harder.
Rabid Wolf- Grants a buff that gives both attack speed and cooldown reduction however this buff lasts only one minute so you need to be careful when you pick it up not to waste it
Ghast- A slightly bigger wraith that gives a large movespeed increase (30%) like Rabid Wolf's buff this only lasts one minute so you've got to make good use of it.
Ebonmaw the Terror of Zaun- This is Treeline's equivelant of Baron, though significantly easier to take, Ebonmaw grants a large amount of global gold and a strong buff that increases all damage delt by a percentage equal to your character level.


All these buffs are very useful and it is customary if not running a jungler to have your top laner take either ghast or wolf, along with some of the smaller camps frequently whilst laning. If you are running a jungler top lane is very easy to gank anyone with a gapcloser that goes through walls can immidiately jump into the lane from almost anywhere a long the lane whilst the entrances to the lane from the jungle are all close to the turrets resulting in a hard to escape gank if caught in the middle of the lane. Taking smite to secure and steal big buffs like Ebonmaw gives the jungler another use except strong ganks and more experience, even if not running a jungler then I would still recommend trying to fit smite on one of your team members just to ensure you get those buffs. A problem with jungling on Treeline is that any blue dependant jungler will not be able to clear very quickly and it is quite a lot harder than Rift's jungle since the change making small camps give health back was not applied to Treeline (suprise suprise). However if you lane until you have a razor or wriggles your jungle clears should be fast and easy. I will be experimenting with a variety of different dedicated junglers on Treeline in the near future and hopefully documenting which ones performed well.

 People tend to think Flash essential on Treeline due to all the walls that can get in the way and it helps gank, escape ganks, and chase. On the champions without built in maneuverability it pretty much is essential. Speaking of Ranked Treeline a 'meta' (using the phrase loosely at best) seems to consist of grab a variety of tanky champions or champions deemed 'OP' at Treeline send two bot, one top, ward occasionally and hope for the best. Though I'll admit my steps into ranked Treeline this season has been few and fairly poor with about an even w/l ratio last season we saw almost the same but luckily it can be countered with well chosen picks and bans (not going into detail due to my last article on the meta) and once again there is no need to conform to the metagame and play how everyone else does. Once my ranked team was quite easily beaten by a team running a dedicated jungler and a ranged AD carry, something that did not in any way conform to the metagame of Treeline and remember that due to the lack of Treeline players compared to Rift lots of options haven't been fully explored so many strong strategies may await for your team to uncover.


Well I think that's everything I wanted to say about Twisted Treeline, now back to Rift stuff...




Thursday, 12 April 2012

Evolution of the Metagame



On the occasions that I step into ranked nowadays there are several common bans that I see pretty much every match:
Shaco- His very strong early game ganks and strength of jungling.
Morgana- Her massive laning presence, in built survivability and CC shield, and powerful teamfight ult
Kassadin- Shuts down the majority of mids and kills squishy champions in seconds
Shen- Powerful solotop bruiser, able to teleport to allied champions and turn any fight in your favor.
Cassiopeia- Pretty much the same reasons as Morgana

These four I will commonly see immediately banned in every single ranked match without the team captains thinking about any other options. All four of them have decent counters and their presence can be negated fairly  easily with the right team composition, captains should be focusing their bans on what would prevent their strategy and what champions their team specifically struggles against as opposed to what the community deems 'op' currently.

As most of you know the current metagame is generally bruiser solotop, mage mid, and carry/support bot lane, it's been proven to work time and time again in tournament play and is a go to decent strategy to use. However when I first came to have an interest in competitive play the apparent metagame was quite different. The first 'competitive' match I watched was Noxus vs Ionia, though not a true tournament game it still shows what the playstyle was like back in 2010. Mage top, AD carry mid, tank/support bot. This would truly horrify anyone playing ranked nowadays as clearly our modern playstyle is the best right?  Well Judging by the play by Fnatic in Dreamhack running a double mage setup and more recently by Dignitas in IPL4, no. Choosing three champions defined as 'support champions' by riot to allow Kog'Maw to super-carry and shred the enemy team.

With various team composition proven to be viable why do people feel the need to be shoehorned into a rigid metagame? I suppose the mantra of 'If it aint broke, don't fix it' could be the mentality of many players however I believe that the majority of characters in League of Legends are viable ranked picks, at least for starting out in ranked, and that skill level matters more than getting the better team. AoE teams, Poke Teams, Triple Support, 'Meta'-Teams, are all viable and doable in ranked and it is more important to use your bans to stop any specific threats that your planned team composition can't deal with or that specific players hate playing against. Team cohesion, planning and skill are all far more important than conforming to the meta.

So a few little thoughts on the current meta, I'm still no LoL expert so feel free to completely disregard everything I say if I'm wrong.

Now we wait for the meta to change again...

The Deal With Supports



Support is probably the least wanted, played and valued role in normal and low ELO. People want to play the carry or initiator and see sitting in lane babysitting as pointless. However it is one of the most essential roles in a team with wards giving map awareness, auras items giving subtle but strong boosts to the entire team and heals, shields and CC enabling your carry to get fed and farmed without a problem.

The thing with supports is that they're hard to screw up but even harder to master. Most of them, in my opinion, have fairly low skill floors but pretty high skill ceilings and getting good at supporting is one of the ways you can almost guarantee a win. Of course some are slightly simpler than others, as difficulty level varies for different champions, and another positive point of playing support is even if you aren't the best player it is hard for you to cause a team to lose. Naturally a mediocre player who has awful positioning and feeds can cause a loss for the team but someone slightly more skilled (but not by much) can still ensure the team has a good match which wouldn't happen if they played the carry or initiator. All a decent support needs is to ward, use skills on teammates and buy aura items to help their team.

Speaking of Wards they are one of the most essential items in the game, but warding isn't just the supports job, it is everyone's. Whilst this is clear early game with each lane needing to see the area around it so  your top and mid will be buying wards this falls off a lot late game when map awareness is equally as important for big map objectives and for the jungle to avoid being caught and to allow you to catch enemy team members in compromising positions. A common excuse for many not to buy wards is 'I didn't have the money after buying my items' wards should be a priority for all players, you only need to pick up one (two or three for supports) each time you go back and if you're last hitting properly or buying gold items then it should never be a problem. Ensuring you always have decent map awareness is essential for any team.

Okay now we've sorted that out let's talk about one of my favorite supports, Janna. Janna, in my opinion is the best and most well rounded support character, despite lacking an in lane heal like most supports she makes up for it with a strong shield, godly CC and that ultimate. First of all her passive is incredible, sure it only works whilst she is alive but it affects the entire team and more move speed, for chasing and escaping, even just 3% more (which helps a lot in early game) makes a huge difference. Then we move on to that Q an easily spammable knock up in a line, this skill shouldn't be used to harass in lane, unless in the build up to a gank, due to the relatively high cooldown and manacosts. This can be used to stop escapes, counter initiation or generally mess up enemy teamfights. Her W gives a ghost like effect, further boosting her movement speed and stopping unit collision. It's active deals decent magic damage and is a long slow, 48% for four seconds at 5th level. It's a good skill once more for ensuring successful ganks and chasing but is also a decent tool for harassment in lane as, unlike her Q,  it is purely single target so doesn't disrupt the creep wave for the carry. Her E is a pretty incredible shield. Fairly strong shield strength (with a .9 AP ratio) which also gives the champion bonus attack damage, this varies from 14 at first level to 50 at 5th. Maxing this and throwing it on your carry to harass in lane can result in a lot of damage getting through to the enemy team. Finally we have her ultimate, usage of her ultimate is what divides good Janna's from great ones. It can be used to initiate, save your team, counter initiate, lock someone to a wall and more. Unfortunately it can also be used to mess up your own teams positioning in a teamfight, blasting their carry's away and their initiators into you or accidentally helping the enemy chase and get kills (See Here)

Well that's Janna sorted out and a few small thoughts on supporting in general, so get out there pick up a support and pray your lane partner can actually last hit.


Wednesday, 11 April 2012

All this fuss about the Atma's Nerf


In League of Legends there is an upcoming Atma's Impaler nerf taking the items Attack Damage scaling off Hp down from 2% to 1.5%. Naturally this being occasionally one of perhaps two damage items on a tanky dps or bruiser it has caused great rage across all the forums. I'm here to know if this rage is justified or if it is just the community over-reacting like usual.

First of all let's analyse old Atma's and see what it gives us: 45 Armor and18% Crit chance but we can buy those items normally as Atma's doesn't improve upon the base items in that respects. The important part is 825 gold for 2% of your hp to damage.

But how much hp do you need for that to pay off? According to League wiki it's 1125 hp which is a minuscule   amount compared to the hp of those who should be building it. With the change from 2% to 1.5% you will need slightly more hp to make it worth while and you only gain 15AD per 1000Hp...as opposed to 20. All in all the change shouldn't affect the majority of people building it who are also stacking frozen mallet, warmog's, sunfire cape etc except in a possibly beneficial way to discourage the stagnation of 'metagolem' builds and a further shift away from tanky dps being king.

So the rage on the forums is pretty much unfounded, sure you wont be a nigh-unkillable damage dealing one man army (as much) anymore but it is still a solid defence and damage item for anyone building it.

Now we just wait until the next minor change sets off the community...