LightCC

Hello,

I put together a spreadsheet for spell DPS for warlocks quite a while ago. I recently have been updating it, and found several issues. I've been reading/mining the forums to try to understand all the issues and get things right, but I've still got some questions.

So I'd like to outline my process on a few items, and ask a few questions before spending hours finding out with the parser (as I already have to get where I'm at so far!) - in case someone already has done the work.

[ Edits in green with found answers ]

Most of the answers here so far come from the 'Maximizing DPS as a Raid Wizard sticky in the Wizard forum.

I. BASIC CAST-TIMING CALCULATIONS
(Cast-time, Recovery, Re-use)

1. At time 0, you begin to cast a spell. This takes the base spell time reduced by a modifier, however, the modifier is not exactly what is shown in the persona window, there is a conversion that has to happen. The actual cast-time of the spell at any given time is shown in the tooltip for that spell.

1a. Does anyone have a formula or a table that converts persona cast-time reductions into actual percentage reduction in cast time?

1b. What are the soft and hard caps for cast time reductions?

There is no 'soft cap' per se. The hard cap is 100% which equates to a cast-time reduction of 50% (you cast twice as fast).

The curve is a curve of diminishing returns, with the actual drop in remaining cast time falling off most notably around 50% of persona casttime reduction. Here are a few points of reference:

Persona Cast Haste % / Actual Cast % / % Reduction over last 20%
20% / 83% / 20% (1% reduction/1% persona cast haste)
40% / 71% / 18% (0.9% reduction/1% persona cast haste)
60% / 63% / 13% (0.65% reduction/1% persona cast haste)
80% / 55% / 13% (0.65% reduction/1% persona cast haste)
100% / 50% / 10% (0.5% reduction/1% persona cast haste)

2. At the end of the spell casting, both the recovery timer for all spells and the reuse timer for that spell begin.

2a. Recovery is 0.5 seconds, modified by any recovery reductions, from things like Time Compression. At the end of recovery you can cast another spell.

2b. Are recovery reductions similar to cast-time in that the actual reduction is different than the stated reduction? Are there hard and soft caps? Or is a 40% reduction from TC just simply a straight reduction to 0.3 s from 0.5 s?

Recovery works just like Cast reduction/Spell Haste. See 1b above. The hard cap is 100% at which point your recovery is 50% of normal (0.25 s instead of 0.5 s)

2c. What happens when you have TC and FC both up? TC is 40% reduction, FC is 50%. Is your recovery down by 90% to just 0.05s? Is it 50% of the 0.3 s that is normal with TC? Or is there some other soft/hard cap in place?

See 2b above. TC of 40% and FC of 50% recovery reduction would net a 52% reduction in recovery time, to a net of about 0.26 s.

3. Reuse starts when the spell gets done casting. Reuse is like cast-time in that the actual reduction is less than what is shown in persona.

3a. Like cast-time, are the hard and soft caps?

See 1b above. Exactly like Cast-time, hard cap of 100% = 50% reuse.

3b. Does anyone have a conversion formula or table for it?

See 1b above.


DAMAGE CALCULATIONS
(INT, Crit, SD/BoE)

There is an excellent post with the basic formula: http://www.eq2flames.com/warlocks/22...lockery-3.html

The post from Dehah outlines how to calculate spell damage - but still a few things I don't understand clearly:

4. Start out with the base min and max damage, and modify it by INT. The only way to modify the base damage is by increasing the quality of the spell (app, adept, master, etc.)

4a. The soft cap is Spell level (not character level) *12 + 20? The hard cap is Spell Level *15 + 20?

There is no soft-cap per se. It is a continuous curve of diminishing returns.

4b. Does anyone have the function or table that explains how the INT calculation works?

The INT function varies the base spell damage in a range of 60% max damage at 0 INT up to 100% max modified damage at the hard cap of (Spell Level) * 15 + 20. So being at the INT hard cap for a spell will increase the damage by 67% over what the damage would be with all your gear broken.

Here are a few points of reference for a level 80 spell:

% of INT CAP / INT / % of max dmg / Act. Dmg Incr. over last level

0% / 0 / 60% / 0%
20% / 244 / 73.4% / 22% (11 INT for +1% damage)
40% / 488 / 84.4% / 15% (16 INT for +1% damage)
60% / 732 / 92.4% / 9.5% (26 INT for +1% damage)
80% / 976 / 97.8% / 5.8% (42 INT for +1% damage)
100% / 1220 / 100% / 2.3% (108 INT for +1% damage)

So, we see that over the first 244 INT every 11 INT on average gives us a 1% increase in damage. Going up to 488 INT, it takes an average of 16 INT for each 1%, up to 732 it average 26 INT, up to 976 INT it takes 42 INT, and up to the 1220 INT cap it takes a whopping 108 INT on average to increase spell damage by an additional 1%.

5. Determine if it crits and the crit damage.

5a. Whether it crits is a straight percentage equal to the persona crit chance. I've seen claims that it is lower based on mob level, etc., but I think this is wrong. If you look in your parses it will show a total crit level less than persona, but if you look closer you will see this includes attacks from CAs (like ambidextrous casting), melee attacks, and attacks from item procs, pets and other buffs that use melee crit instead of spell crit.

5b. The crit calculation is tricky because if the crit value is less than max-base-damage + 1, the game bumps the damage up to that. Otherwise, my understanding is that it is simply 130% of normal damage. However, Looking at Flames of Velious, I had a 21% increase on crits in RE2 the other night... Any explanation? Or am I looking at something wrong (the crit average was pretty much dead on for Distortion however)

Still not sure on the Flames of Velious issue... need to do some in-game testing.

I think I understand this 'reduction' of crit damage. It has to do with the fact that my SD is not capped for flames of Velious (it's only around 500 right now). Crit does not apply to SD, so if base damage is roughly 1200, crit damage will be roughly 1560. Add SD of 500 and you get normal damage of 1700 and crit of 2060 - a 21% increase. If SD was capped with crits, you would see the full 30% increase (i.e. crits do raise the SD cap by 30% - if you have that much available).

So, this means you cannot set up a spreadsheet based on raw tooltip spell damage at a given INT/SD. Instead, you have to take off all SD (and INT preferably) to read base damage, and then calculate crits and SD bonuses in the spreadsheet itself.

5c. Crit calculations apply to dot ticks just like DDs. However, some spells may not be affected by crits. What are these spells? Pets, I think are melee crit (Netherbeast/broodlings). So are some AA abilities. Is anything else abnormal on crit damage or not based on spell crit %?

Still looking for more info. I think examining some parses will resolve this easily enough now that crit % is in the parser.

6. Add the SD (+Spell Damage or BoE) damage. The cap is 50% of the total normal or critted damage. So if you hit min damage, the SD cap is much lower than if you hit for max crit. The SD is calculated at this point, but it is not added until after base damage modifications (like choker) are added in (so crit increases SD cap, but base damage mod does not)

6a. The cap includes total damage for a DOT. Is it calculated on both the initial tick and each following tick like a separate DD? Or is it front-loaded partially or completely? (I suppose this will be easy to test)

According to the Wizard DPS post, all SD is added to the initial tick.

6b. The SD cap is reduced by 1/3 (so it is 33.3% instead of 50%) on aoes. Is this all aoes (green and blue)? When the epic clicky is on for single-target mode, is the SD reduction still in place?

The actual reduction is in the SD that is added to the spell damage, it is reduced to only 1/3 of the normal amount. I'm still not sure if the SD cap is reduced to 1/3 of normal as well or not.

This affects warlocks a *whole* lot more than Wizards with all our aoes. Hopefully the epic clicky to change Apoc and Absolution to single-target changes this SD penalty to the normal single-target behavior as well

7. Base damage mod. This is a straight percentage. It is multiplied by the normal/crit damage, and then the SD damage which is already calculated is added to the result.

7a. If you have multiple mods, I assume they are additive, not multiplicative (i.e. two 10% mods would be simply +20%, not 1.10*1.10 = 1.21 or a 21% increase)

Appears to be correct (additive)

8. Finally there is some kind of calculation that can reduce the damage based on the mob level and mob magic mitigations. My understanding of this is that being higher level than the mob and debuffing the mob properly will simply reduce the mob's mitigation effect - in other words it keeps your damage from being lowered, it never actually increases your damage.

Any confirmation of this?


OTHER CONSIDERATIONS


Other calculations at this point are simply to determine the casting efficiency based on cast-time, based on reuse, and based on power.

One could calculate the effect of TC and focused casting (anything that reduces recovery will make spells with shorter casting times have greater gains in cast-time efficiency).

One could also factor in the effect of Tandem, PoM, Aria and other percentage based buffs/procs (faster casting is higher efficiency with these).

In calculating procs that are "x times per minute", does the game factor in your recovery time? If not, then longer casting times will actually make those proc more (and recovery time reduction will make the difference smaller).

Yes, the procs are based on Cast-time + recovery time. So a proc that triggers 1 time/min with a 2.5 cast time spell would trigger (2.5 s + 0.5 s) / 60 = 5% of the time.

The more interesting information is that the proc rate is based on unmodified cast and recovery time. So if you have hard cap 100% cast time and 100% recovery increases, then your proc rate will actually be double the listed rate (as you will cast your spells twice as fast but the percentage chance to proc will be the same).

That makes me finally understand why proc items are more valuable than the raw dps you calculate from the item description would say.


Anything else that affects dps I'm not thinking of? There's lag, effect of 'instant-cast' spells (like freehand sorcery - can't queue the next spell so there is some 'user' lag), etc...

I suppose you could talk about spell resists, that's going to be per mob... The resist rate of Aura being higher (it's subjugation instead of disruption) may be a worth factor to consider as well.

SPELL SKILLS

Base spell skills at level 80 are 400 (level *5). The hard cap is 520 (level *6.5). Raising disruption/subjugation from 400 to 520 is supposed to make them 20% harder to resist.

If this means that outright resists drop from say 5% to just 4% for a particular mob, then I can see why no one notices the difference in spell levels - but I'm not sure this is what it means.

DEBUFFS

Regarding Debuffs, tests on the Training wall in KJ showed an increase of about 6.5% when the wall was debuffed vs. elemental by about 1000. Of course, this is a single point in the caster lv vs. mob lv and Mob mitigation curve matrix..

It would take some dualling along with mentoring to get better test points on debuffing, and then an idea of what the mobs mitigation levels are starting at (I suppose average resist rates should tell us this)


Comments, help, concerns all welcome. If I missed some good posts, please help me out... I did search, I promise!

LightCC

Ok, there is a wizard sticky post that is a guide to Wizard DPS that answers many of my questions. I would highly recommend any warlock read through that if you haven't - probably 75% of it applies to warlocks (not most of the spell specific section, but the theory even of that does).

Here it is: http://www.eq2flames.com/wizards/119...ard-guide.html

I'll edit my original post to add some of the answers from there when I get a chance...

[edit: Original post now edited with some updates and answers.]

LightCC

Findings/progress to date:

[Edits in Green, updates as of 8/12/08]

1. Base DMG does affect SD cap.

No it doesn't... it only affect SD cap on the tooltip, actual damage numbers prove that base damage does not affect SD Cap (but crits do).

Spreadsheet is working to where I can input INT, +SD, base dmg mod, Crit%, Casttime % mod, Recast/reuse % mod, recovery % mod, all based on persona values, and it will calculate DPS efficiency of the spells. You can also adjust individual spell dmg, cast, reuse mods, etc. for AAs. It also shows what the tooltip damage should be based on current inputs, and that is now accurate for all spells that have tooltip damage displayed (except for low INTs, under 30-40% or so of cap there is seems to be some errors in the curve). Still need to go grind some time with the parser to verify actual average damages.

Based on the tooltip damage of the spells, in order to calculate everything correctly, I have concluded that the base damage mod actually does increase the SD cap. This is true at least for tooltip damage (still need to verify actual damage with a parser). It doesn't increase the actual SD that is added to the spell damage, unless the spell is capped, then it does so indirectly by raising the cap.

This is true of AA base damage (WIS line), equipment (MC cloak with 2%), and individual spell mods (like the +10% AAs for Dissolve and Absolution). Calculated any other way, the tooltip damage doesn't come out right in non-capped vs. capped SD situations (easy to see on spells that are capped at the low damage but not at the high damage).

This includes ALL base damage mods, including damage mods from AAs (like the 10% damage on dissolve from warlock AAs), WIS line for sorcerers, and equipment base damage mods (choker/MC cloak)

2. Aura of Void

This spell has an SD cap that is calculated based on just one trigger, but the total SD is spread out over all three triggers. Effectively, this lowers the SD for this spell by 1/3. Pretty crappy. This makes it just an average efficiency spell that is capped at very low SD (around 130-180 depending on stats). This is pretty crappy, because to get that average dps, you still have to make sure that all 3 triggers go off, if only 2 do, you are already down into dissolve-level crap dps.

Apparently this is considered a reactive, which means they cut the +SD down to 1/3 of normal value (cap to 1/3 as well).

This is because any hostile spell from any member of group/raid will set off one of the 3 triggers, I think. Stupid reasoning in my opinion, but that's the way it is. On the other hand, this does mean that in a raid all three triggers go off almost instantaneously (no worries about casting this close to the end of the battle unless rigth at the last second).


3. Neatherealm and Gift of Bertoxxulous

These spells get no benefit at all from SD. Zero. Sure they are AoE (or can be), and sure you can't calculate how many triggers you will get from Gift, but no +SD at all? For Netherrealm you *can* calculate total dot damage for single target, assuming all trigger (and if you equalize the SD across all ticks then missed triggers just lower the total SD damage).

Gift I understand the issue because you don't know how many triggers you will have, so there is no way to figure a total damage for capping. Surely a compromise could be reached (assume 10 triggers to calculate cap and then equalize? Remove SD if more than 10 triggers are reached? something like that). And, reduce the final SD to 1/3 like other AoEs too.

By the way, does anyone know if Netherrealm works like the description and requires a poison or disease spell to trigger? Or like the effect text and just requires any hostile spell?

Also, does Gift of Bertoxxulous require poison damage like the description, or any successful hostile spell like the effect text reads?

Need to get the complete list of spells that trigger each one of these. But a lot of debuffs like aura, root, nullify, and vacuum field do trigger gift. Netherealm the triggers are a bit less important since you just need to hit it once every 3.3 or 5 s...

However, FC reduces the duration of other dots... I wonder if Netherealm triggers faster during FC as well... hmmm...

4. Miscellaneous

Spell damage doesn't always read correctly when looking at a book/scroll on the broker... By the way, the difference in base spell damage from adept III to Master 1 is exactly 14.3% every time - of course this doesn't mean the tooltip damage will change exactly this amount (your SD benefit may vary)


5. Dissolve

Dissolve is still a crappy spell. It's DD with lower DPS efficiency than every spell but Radiation. This includes all other AoE spells, even if just on one mob. The only time to use it is right at the end of a battle and there are no other DD spells to cast (and the mob is about to die so dots are actually less dps).

This is bad because noobs playing warlocks naturally think this should be one of their main damage spells. No wonder people who haven't run the numbers think warlocks are so bad single target - if you are casting this every time you can you are dropping your dps by at least 1/3, maybe even 1/2.

Small possibility that dissolve moves up over absolution and the blue aoes in dps when coupled with percentage based procs, due to it's shorter cast time. Still haven't finished working procs into my calculations.

6. Swarm pet spells...

I still need to figure out the base damage and mods for the swarm pets spells. Does anyone have this already? Does anyone know if their damage is calculated based on the casters +spell damage/int/spell crit%? Or something else (like melee crit%, etc)? Probably going to just have to grind the parser for these.

Crit on swarm pets seems to be 1% and ignore all caster attributes. I am assuming I will find damage is the same way.

Still need to parse to find the base damage, but if it is unaffected by caster stats, then it will be easy to figure - just pull out of any raid. However, swarm pets may get the benefit of group buffs...

Ok, need to use a few on some individual parses with different equipment to see if there is any damage bonus, then can compare to raid damage, should be easy to figure.

7. Mount/Warg bonuses

Warg/mount bonuses for +SD do not show up in the persona window, but do show up in tooltip damage.

Reconfirmed in the field when testing Dissolve.


8. Preliminary numbers for value of SD, crit, base dmg, etc.

I've set up a preliminary spell rotation for a 1 min fight, and it's very easy now to calculate the change in dps/damage for that one min fight when changing cast time, recovery, INT, etc. Here are some preliminary numbers with a particular setup.

This is based on approximate numbers for a grouped Warlock with a little INT buff, in mostly legendary and T1/T2 raid equipment, not including the dps effect of procs. Note that cast time and reuse reductions on equipment don't show up in the persona window, but they do apply, so you need to add them up separately...

INT = 1000
+SD = 500 (persona, no warg bonuses)
Crit = 50% (persona)
Cast time = 21% (total persona + hostile cast time red. on equipment)
Reuse = 18% (total persona + hostile reuse red. on equipment)
Recovery = 0%
Base Dmg mod = 6% (WIS line + 2% from cloak)
Note: Base dmg mod for individual spell AAs are on top of this.

At these settings, damage is affected as follows:
+1% base dmg = 0.8% dmg
+1% cast = 0.657% dmg, (1% cast = 0.82% base dmg).
+1% crit = 0.285% dmg, (2.3% crit = 1% cast, 2.8% crit = 1% base dmg)
+20 SD = 0.361% dmg, (16 SD = 1% crit, 44 SD = 1% base dmg)
1000 INT to 1100 INT = 0.662% dmg (43 INT = 1% crit)
+10% recovery = 0.212% dmg (about 40% recovery = 1% base dmg)

or

0.8% dps = 1% base dmg = 1.2% cast time = 2.8% crit = 44 SD = 40% recovery = 120 INT

Of course, all this logic changes depending on your stats at the time (raid buffs, TC, focused casting, etc.).

One thing I failed to consider so far is power usage. +crit and +SD give you more damage with no additional power requirements. But cast haste, recovery, and reuse gain dps by allowing you to cast more spells in the same amount of time, which means an equally higher percentage of power useage.

If this means you are running out of power and having to cast hearts/shards, Void Contract, Mana trickle etc. in combat, then you are losing dps and need to take this into consideration, and put more value on crit/+SD.


9. Reuse

There is no easy way to account for reuse - you need to figure out a new cast order based on a particular reuse% in order to compare differences. My current working guestimate on reuse is that it's about equal to crit% in value, or 2-3% = 1% base dmg. Probably closer to cast% for solo/group where it will make your roots/stuns/daze come up more often. Yes, for heroic encounters I count radiation as a stun, virtually it's only use....

Part of the problem we have with reuse is that two of our main dps spells, Vacuum Chamber and Acid, are DoTs that have a longer duration than reuse (even with AAs). Since reuse only reduces the recast timer and doesn't lower the duration, it allows us to cast these spells more often, but will cut off more and more ticks at the end of the duration if you do recast them faster, lowering the dps.

Broodlings, Arma and Neatherealm have such high DPS that reuse on these really helps dps... also improves reuse on FC and even deity abilities.

Additionally, on trash mobs, the better the reuse the more often you'll have your big hitters up at the start of fights. So I'm growing fond of reuse, just still not sure how to put a number to it - very situational.

Comments, concerns, tips, and data all welcome.

Naeca

Gift is any hostile spell, boss.

Very nice post though

__________________

LightCC

Thank you on both counts.

That makes it one of our best spells for dps (roughly 20 spells should hit instead of roughly 10 using about 30% persona cast haste). Even better with TC, better gear and, of course, during FC. As long as you get in the full duration (45 s with AAs...)

TheOtherDude

Did you actually test Gift? I had always assumed it was any hostile spell, but I went to test it on solo mobs in Jarsath, and Encase and Absolution didn't proc it (but Acid did).

Also, Upheaval, Encase, and Velius didn't proc Netherealm, so it looks like that description is correct as well.

EDIT: Interestingly, Vacuum Field and Nullify proc Gift, even though there's no way to tell those spells are poison based from the description. Concussive doesn't proc it. Even more strange, Aura does proc it, which means the "debuff" it applies is poison based, even though the damage it deals is disease.

Xede

pretty sure they trigger from poison spells

Pinski

The tooltip is wrong. +Base Damage does NOT affect your spell damage cap, nor do crits.

__________________

LightCC

TheOtherDude,

Thanks for the info... Netherealm is less important to me - single target acid will constantly be on them, and multiple mobs cataclysm will, but with AAs it ticks fast enough I need to see if I'm missing some ticks, especially with upheval not setting it off...

Gift I just need to make a list now of what triggers it. I guess that means I can actually assign dps to nullify and vacuum field... when run after gift.


Pinski,


I know this is true of base dmg on single-damage spells like Flames of Velious (I've seen the tests and tested it myself)...

But all the things I've read so far indicate that Crit does affect +SD. I thought I tested that... hmmm. Well, it will be easy enough to test with Flames of Velious.

Also has anyone tested this with any spells that have a damage range? Some preliminary tests I've run with dissolve make me think that base dmg does affect SD in this case - which would simply mean they programmed a special case for single damage spells and there is a bug in that case only.

But - some of my data is coming in outside of the range of either calculation - I think this has to do with mob mitigation, but not sure.

Anyhow... has anyone done the proper testing with anything other than flames of velious?

LightCC

Pinski,

Just finished some testing that proves (if my calculations are right), that base damage does not, but crits indeed do, modify the SD cap.

Here's the data if you wish to check it out.

Test spell. Warlock spell Dissolve (level 71).
Dissolve Base Damage = 862-1600 at 668 INT
Base Damage at Test INT = 879-1632 at 736 INT (calculated)
Base Damage Modification = 14% (10% from dissolve AA, 4% from WIS AAs)

Spell Damage Mod = 1375
Calculated SD Cap for Tooltip, low = 879 * 1.14 * 50% = 501
Calculated SD Cap for Tooltip, high = 1632 * 1.14 * 50% = 930.5
Both caps are under actual SD, so,
Displayed Tooltip Damage = 1503-2791,
which matches actual tooltip damage in game

Now, if we leave Base Dmg Mod out of the SD Cap calc:

SD Cap, Norm, low = 879 * 50% = 439.5
SD Cap, Norm, high = 1632 * 50% = 816

And calculate Damage ranges with the above SD Caps and base mod:
Actual Damage, non-crit, low = 879 *1.14 + 439.5 = 1442
Actual Damage, non-crit, high = 1632 *1.14 + 816 = 2677

This means min crit will be 2678, now lets determine what max crit should be.

If we assume that Crit DOES NOT affect SD Cap, as you propose, the max crit will be:
Max Crit, no SD Cap increase: 1632 * 1.14 * 1.3 + 816 = 3235

But if we assume that Crit DOES affect the SD Cap, that means the cap will be:
SD Cap, Crit, high = 1632 * 1.3 * 50% = 1061
so the max Crit damage will be:
Max Crit, SD Cap increased 30%: 1632 *1.14*1.3 + 1061 = 3480

Here's the data:
All attacks on level 69-71 mobs in KP (satharians)
level 80 warlock
INT 736, SD 1375, crit 32%
215 hits, 31.6% crit (68 out of 215)
Min non-crit damage: 1443
Max non-crit damage: 2675
Min crit damage: 2678
% of crits that were min damage: 42.6% (29 out of 68 crits)
Max crit damage: 3470
# of crits over 3235: 11

Unless I did something wrong here, this is proof that Base Dmg does not affect SD Cap, but that Crits either do affect SD Cap, or that calculated SD damage is multiplied by 1.3 (+30%) whenever a crit occurs. So now we need to test whether the SD Cap is actually raised with a crit, or whether the no-crit SD value is simply multiplied by 1.3 when you do crit. The difference is that if only the SD Cap is raised when you crit - then you have to have a high enough SD to take advantage of it...

So, to test, I have to lower my SD to under the cap for crit. The theoretical non-crit cap is 817 with 736 INT, so I choose this value to start with. After playing with my equipment, I can get exactly 817 +SD by my INT turns up being 744 at this level of SD, so my damage for dissolve bumps up just slightly... just taking the numbers from my spreadsheet and not actually showing all the calculations a second time:

INT of 744, SD of 817,
Tooltip Damage is 1507-2682
Actual Damage is 1445-2682
Actual Crit Damage is 2683-3241 if Critting only raises the cap by 30%
Actual Crit Damage is 2683-3486 if Critting multiples the non-crit SD value (capped or not) by 30%

Test data:
Level 69-71 mobs, Satharians in KP
Level 80 warlock
INT 744, SD 817, Crit 52%
Dissolve, 88 crits of 168 hits out of 168 attacks
Of Crit attacks, 32 of 88 are minimum crit value of 2683
Highest crits are 3132, 3139, 3164, 3190, 3214, 3224

So we come within 0.5% of our value that shows us that Crit only raises the SD cap - if you don't have enough SD you will not take advantage of the extra SD cap from crits. Since the other case would increase the non-minimum crit range by over 44%, when we are within 2% of the value based on this same range... I'd say this is definitive (we have over 50 values in this range).


The only thing that is off here... is that theoretically I should be at minimum crit 50% of the time I crit, while I'm actually at minimum crit only 42.6% of the time on the first test, and only 36.4% of the timeon the second test. Granted these samples only have 68 crits and 88 crits, so the sample is not huge, but maybe something else is going on here... But all my other numbers are within 0.5% of their theoretical values, very definitively showing that crits do, in fact, increase the SD Cap of the a spell, and if you don't have that additional SD, you won't take advantage of the higher SD cap whenever you crit.

It also proves that Base damage does not modify SD at all... either actual SD damage or the SD cap, as everyone already believed.