Combat Ratings at level 85 (Cataclysm) - Page 2

Borodin · 07/31/08, 12:28 PM

Aha that makes sense. Curse my lack of Beta access.

Nezralix · 07/31/08, 12:32 PM

If these are accurate, then I suppose they make some sense. The 60->70 conversions probably weren't steep enough, given the 75% average increase in gear level. *If* they were happy with the level 60 conversion rates (as opposed to the level 70 conversion rates), and *if* 70 to 80 sees another 75% average increase in gear level, then you might assume roughly a 1.75*1.75 = 3.06 multiplier of the level 60 conversion rates (i.e. 14 critical strike rating per 1% crit converts to 14*3.06 = 42.8 critical strike rating per 1% crit). That is, if we don't consider the level 70 values at all.

And that appears to be approximately what we're seeing in these numbers (well, these indicate about a 3.2x multiplier).

So if the falloff is steeper than it was from 60 to 70, then it could well be because they're adjusting for a 60->70 transition that they didn't feel was steep enough, without actually nerfing the level 70 rates.

♦ Praetorian · 07/31/08, 12:39 PM

Fantastic post, thanks. Hopefully you can get a mod-author beta key soon.

Lord BEEF · 07/31/08, 12:58 PM

This actually makes the new gems at level 80 slightly weaker than the ones we have at level 70.

That is to say, a 10 hit rating gem on a level 70 provides more hit than a 20 hit rating gem provides for a level 80. Very slightly so, but it's still there.

• Hamlet · 07/31/08, 1:03 PM

Originally Posted by Lord BEEF

This actually makes the new gems at level 80 slightly weaker than the ones we have at level 70.

That is to say, a 10 hit rating gem on a level 70 provides more hit than a 20 hit rating gem provides for a level 80. Very slightly so, but it's still there.

Well, it's within an integer step of the same value (i.e. the amount of HR required to be exactly equal to the hit you currently get at 70 is between 20 and 21). I think that's all that's going on here.

kalbear · 07/31/08, 1:05 PM

Sorry for being ignorant here, but why do we believe that all stats scale exactly the same way as haste does?

• Vulajin · 07/31/08, 1:07 PM

Originally Posted by kalbear

Sorry for being ignorant here, but why do we believe that all stats scale exactly the same way as haste does?

Because they all do from 1-70.

• Hamlet · 07/31/08, 1:08 PM

Originally Posted by kalbear

Sorry for being ignorant here, but why do we believe that all stats scale exactly the same way as haste does?

It's consistent with what they did from 60->70. All ratings conversions were given nice integer values at level 60, which were generally concordant with the pre-rating itemization costs of the underlying stats. Then all the coefficients were multiplied by the same value (41/26) to arrive the level 70 versions.

kalbear · 07/31/08, 1:10 PM

Okay. So assuming that they've not changed that this will work. I've seen many people test this for haste; could we have some testing for some of the other conversion rates to make sure they correlate?

Bisbus · 07/31/08, 1:14 PM

Am I crazy or will a fire mage only need 33 more hit rating at 80 than at 70? Assuming we're all using the melee scale now. If it's the spell scale I come up with two less rating needed.

9% miss rate minus 3% from talents= 6% needed x 32.8=196.8

With the increased ilvl values this leaves a ton of room for other stats.

Akka · 07/31/08, 1:15 PM

Originally Posted by Nezralix

If these are accurate, then I suppose they make some sense. The 60->70 conversions probably weren't steep enough, given the 75% average increase in gear level.

0_o

I don't know what's enough for you... If anything, they were far TOO steep.
The problem is in the extreme inflation in gear level, not the conversions.
And the problem is that they seem to follow an even more extreme road for WotLK.

• Vulajin · 07/31/08, 1:23 PM

Originally Posted by kalbear

Okay. So assuming that they've not changed that this will work. I've seen many people test this for haste; could we have some testing for some of the other conversion rates to make sure they correlate?

Tested just now with hit, crit, haste, spell hit, spell crit, spell haste, and expertise using the formula for level 75, and all the values matched.

Whitetooth · 07/31/08, 1:28 PM

Originally Posted by kalbear

Sorry for being ignorant here, but why do we believe that all stats scale exactly the same way as haste does?

Although I only posted the haste table to make this post simple, I actually have data from level 1 to 100 for every other rating. I did check the formula with everything and they match.

Erdluf · 07/31/08, 1:47 PM

Originally Posted by Bisbus

Am I crazy or will a fire mage only need 33 more hit rating at 80 than at 70? Assuming we're all using the melee scale now. If it's the spell scale I come up with two less rating needed.

9% miss rate minus 3% from talents= 6% needed x 32.8=196.8

With the increased ilvl values this leaves a ton of room for other stats.

Saying that +Hit applies to both melee hit and spell hit is not the same thing as saying that spell-hit will use the same table as melee hit.

Has anyone done the testing to see what table caster's are using? The test is trivial. Give yourself 9% spell hit (including talents) at 70 and attack a 73 mob. Is your miss rate 0% or 8% (or something entirely different)?

• frmorrison · 07/31/08, 1:49 PM

People have tested spell hit, it is 9%. That means there is no more hidden 1% resist chance.

Lord BEEF · 07/31/08, 1:53 PM

Originally Posted by Bisbus

Am I crazy or will a fire mage only need 33 more hit rating at 80 than at 70? Assuming we're all using the melee scale now. If it's the spell scale I come up with two less rating needed.

9% miss rate minus 3% from talents= 6% needed x 32.8=196.8

With the increased ilvl values this leaves a ton of room for other stats.

If you're grouped with an elemental shaman and a moonkin, you need zero hit rating on the entirety of your gear. This goes for most (all?) casters actually.

Roywyn · 07/31/08, 2:08 PM

Originally Posted by Lord BEEF

This actually makes the new gems at level 80 slightly weaker than the ones we have at level 70.
That is to say, a 10 hit rating gem on a level 70 provides more hit than a 20 hit rating gem provides for a level 80. Very slightly so, but it's still there.

Maybe we get a second set of higher level level epic gems after 1-2 raid tiers, just a thought.
[Edit]: We didn't get our current BC epic gems until the BT/Hyjal patch, for example.

Also, can you extract the armour value/damage reduction for level 80 (DPSing) and level 83 (tanking bosses)?

• Hamlet · 07/31/08, 2:16 PM

Originally Posted by Roywyn

Maybe we get a second set of higher level level epic gems after 1-2 raid tiers, just a thought.

But why would we expect them to? The original point of ratings was basically to prevent untenable perpetual increasing of %-based stats like crit and hit. Consistent with that, we'd expect gems to provide the same % of each in the next expansion as they do in the current one.

kalbear · 07/31/08, 2:21 PM

Tested just now with hit, crit, haste, spell hit, spell crit, spell haste, and expertise using the formula for level 75, and all the values matched.

Although I only posted the haste table to make this post simple, I actually have data from level 1 to 100 for every other rating. I did check the formula with everything and they match.

Thank you both.

That's going to be very odd given what we've seen come out for itemization so far. Given items like Band of the Kirin Tor, we're looking at something that barely gives comparable results to most of the basic rep rewards out now. I guess that's similar to vanilla->TBC, but it goes somewhat in the face of their supposed goal to make the gear transitions less jerky relative to TBC gear.

It's also going to be much harder for warriors and paladins to hit the defense rating cap without their talents providing defense skill.

Roywyn · 07/31/08, 2:52 PM

Originally Posted by kalbear

Thank you both.

That's going to be very odd given what we've seen come out for itemization so far. Given items like Band of the Kirin Tor, we're looking at something that barely gives comparable results to most of the basic rep rewards out now. I guess that's similar to vanilla->TBC, but it goes somewhat in the face of their supposed goal to make the gear transitions less jerky relative to TBC gear.

It's also going to be much harder for warriors and paladins to hit the defense rating cap without their talents providing defense skill.

Kel'thuzad gave ilvl 92 loot like [Frostfire Ring] and ilvl 89 weapons.
The first raid instance Karazhan gave ilvl 115 loot, a jump of 28 ilvls or 25%.

Kil'jaeden gives ilvl 164 loot like [Crux of the Apocalypse].
The newly added ring Band of the Kirin Tor is ilvl 200.
That's a jump of 36 ilvls or 22%.

If that ring is representative for the first 10-man raids, it's a similar relative jump like in Vanilla => BC after the 2.1 overhaul.
There will be a gear reset, level 80 blue items do have the budget to replace sunwell gear.

Yes, items that randomly pick 3 stats out of sta/int/spi/mp5/dmg/haste/crit/hit with random amounts of them will suck. But well designed socketed gear can outscale Kil'jaeden gear.
Weapons are already there, Weapons - Items - World of Warcraft has a level 78 dagger with more DPS than the Kil'jaeden drop.

Aravan · 07/31/08, 2:57 PM

Does anyone know the rough numbers on the Agility to Dodge ratio?

At level 60 and 70, when you obtain 1% crit from agility, you also obtain 2% dodge from agility.

60: 29 agi = 1% crit = 2% dodge
70: 40 agi = 1% crit = 2% dodge

Does anyone know if these ratios still hold at 80?

Whitetooth · 07/31/08, 3:16 PM

Originally Posted by Roywyn

Maybe we get a second set of higher level level epic gems after 1-2 raid tiers, just a thought.
[Edit]: We didn't get our current BC epic gems until the BT/Hyjal patch, for example.

Also, can you extract the armour value/damage reduction for level 80 (DPSing) and level 83 (tanking bosses)?

The formula for armor damage reduction hasn't changed.

To get 75% reduction against level 83 bosses, 49905 armor is required.

function GetArmorReduction(armor, attackerLevel)
	local levelModifier = attackerLevel;
	if ( levelModifier > 59 ) then
		levelModifier = levelModifier + (4.5 * (levelModifier-59));
	end
	local temp = 0.1*armor/(8.5*levelModifier + 40);
	temp = temp/(1+temp);

	if ( temp > 0.75 ) then
		return 75;
	end

	if ( temp < 0 ) then
		return 0;
	end

	return format("%.2f", (temp*100));
end

kalbear · 07/31/08, 4:12 PM

If that ring is representative for the first 10-man raids, it's a similar relative jump like in Vanilla => BC after the 2.1 overhaul.
There will be a gear reset, level 80 blue items do have the budget to replace sunwell gear.

They'll certainly replace it, but that's not what I was talking about. What I was saying is that the gear you'll have at the start of 80 (quest reward blues, heroic purples, random rep rewards) is going to be less powerful in comparison to the similar gear available to lvl 70s as you compare what the stats give you and how much they give you. At least if this ring is any comparison. That's entirely due to how these stats scale compared to before. Because the stats are scaling at a faster rate, in order to reach parity you would need the values on the items to scale at a faster rate. They're not, however, so relative to TBC we'll find that we won't be in the same place, power wise. Of course it really depends on the encounters whether this matters, but the benchmarks for classes won't be the same.

Arindelest · 07/31/08, 5:06 PM

Originally Posted by kalbear

They'll certainly replace it, but that's not what I was talking about. What I was saying is that the gear you'll have at the start of 80 (quest reward blues, heroic purples, random rep rewards) is going to be less powerful in comparison to the similar gear available to lvl 70s as you compare what the stats give you and how much they give you. At least if this ring is any comparison. That's entirely due to how these stats scale compared to before. Because the stats are scaling at a faster rate, in order to reach parity you would need the values on the items to scale at a faster rate. They're not, however, so relative to TBC we'll find that we won't be in the same place, power wise. Of course it really depends on the encounters whether this matters, but the benchmarks for classes won't be the same.

While this may be true, Weapon DPS, Spell Power and Attack Power are continuously increasing. So even if you have less other stats (in %-based terms) at 80 then at 70, you have a ton more attack power/spell power.

This isn't any different than 60->70, compare [Bonescythe Bracers] T3 bracers to [Bracers of Maliciousness] Kara bracers. The former has ~2% crit and 26 attack power @ 60, the latter has ~1% crit and 50 attack power @ 70. %-based stats go down, attack power goes up, and you make those %-based stats back up as you go through the content. Why? If it didn't work that way, each expansion would give you "more" crit or whatever, until you eventually ran into a hard cap (i.e., you can only have so much crit, 100%-dodge/parry/miss/resist etc.). And if they refused to scale the %-based stats for fear of running into that cap, then they'd be shoehorned into itemizing into only a few stats, again dramatically reducing the power of items.

Latas · 07/31/08, 5:33 PM

To give a comparison using my 73 lock in beta.
Here are my stats with my percentages. For haste i calculated using spell casting times since the tooltip on the character screen does it by whole percentage points.

Stat - Rating - Percentage - Rating per %
Hit - 189 - 12.03% - 15.7107
Haste - 193 - 9.649% - 20.0021
Crit - 340 - 10.669% - 31.868
For crit i took out the amount given by int and the warlock constant of 1.701 .

Here are the calculations for int.
Int - 515 - 6.90% crit added - 74.6377 int for 1% crit (if its a direct conversion)

Hope these values help some in determining the new formula that was pushed last patch.