Casting speed vs spell crit/boe - Page 3

dakkota · 11-30-2007, 04:33 PM

Quote:

Originally Posted by Slippery

1% Base Damage = 2% Casting Speed = 3.3% Spell Crit = ~39 BoE

I equipped the mastercrafted +2% increse to hostile spell cloak...perhaps excluding ice nova and fission, did no where near what this suggests it should? Or when comparing the 2h's - one give +4% crit, +50 BoE and a 1079 2.0x a min proc. The other gives +200 BoE and +2% cast speed and +2% reuse timers. According to this theoretical formula, the 2nd 2h listed equates to a straight +(~)20% crit? Just doesn't seem like equivalent translations to me. /shrug. Maybe im not paying attention to the math but it seems to suggest such figures.

I've also noticed when swapping cast speed gear during a solo buff setup and a raid setup (TC + allegro), adding the gear giving the cast speed bonuses, doesn't have as high of impact always without these cast speed buffs. It's very minimal but so are the adjustments to cast speed from these items in the first place.

Pinski · 11-30-2007, 05:45 PM

Quote:

Originally Posted by Vexus

hrmm I must be missing something then.

If I were to say an 8k Ice nova (or Bolt of Ice if you wish) at 4.0 sec (base cast time and average damage)

8000/4.0 seconds = 2000 dps (Base)
1% = 3.96 seconds
8000/3.96 seconds = 2020 dps (101% Base)
2% = 3.92 seconds
8000/3.92 seconds = 2041 dps (102% Base)

now the only thing I can think is that if it's not a 1% faster recast, you're lowering efficency of what you can cast

if I went with base recast, 45 seconds

Total dps of the spell would be Damage/(cast + recast + recovery)
0% casting:
8000/(4.0 + 45 + 1) = 160 dps (Base)

1% casting:
8000/(3.96 + 45 + 1) = 160.13 dps (100.08% Base)

2% casting:
8000/(3.92 + 45 + 1) = 160.26 dps (100.16% Base)

1% recast:
8000/(4 + 44.55 + 1) = 161.45 dps (100.9% Base)

2% recast:
8000/(4 + 44.1 + 1) = 162.93 dps (101.8% Base)

1% casting + 1% recast:
8000/(3.96 + 44.55 + 1) = 161.58 dps (100.99% Base)

2% casting + 2% recast:
8000/(3.92 + 44.1 + 1) = 163.2 dps (102% Base)

Is my math wrong? I don't think either of those models are completely accurate, due to the fact that you can cast more inbetween, but I'm not sure what to go with

Please note, you are failing to include recovery time in your burst DPS calculation. You are going from 4.5s on Ice Nova to 4.46, which means you are going from 1777.78 Burst to 1793.72 Burst, while a 1% Damage Increase will give you 1795.56 Burst.

For sustained your calculations are also wrong, as Recovery time is never factored into sustained DPS since recast starts as soon as the spell finishes casting along with the recovery time. That being it goes:

Time
0.0s Cast Starts
4.0s Cast Ends, Recovery Begins, Recast Begins
4.5s Recovery Ends
49s Recast Ends

So, yah. Learn to do math right.

Vexus · 11-30-2007, 07:57 PM

Quote:

Originally Posted by Pinski

Please note, you are failing to include recovery time in your burst DPS calculation. You are going from 4.5s on Ice Nova to 4.46, which means you are going from 1777.78 Burst to 1793.72 Burst, while a 1% Damage Increase will give you 1795.56 Burst.

For sustained your calculations are also wrong, as Recovery time is never factored into sustained DPS since recast starts as soon as the spell finishes casting along with the recovery time. That being it goes:

Time
0.0s Cast Starts
4.0s Cast Ends, Recovery Begins, Recast Begins
4.5s Recovery Ends
49s Recast Ends

So, yah. Learn to do math right.

ok, yes I fucked up on the recovery times, but the numbers are still extremely close to what they were before. you pointed out the mistakes but didn't really answer my question, are these correct for comparing casting and recast, and how does it relate to crit/BoE?

So if I redo them :
8000/4.5 seconds = 1778 dps (Base)
1% = 3.96 seconds
8000/4.46 seconds = 1794 dps (100.9% Base)
2% = 3.92 seconds
8000/4.42 seconds = 1810 dps (101.8% Base)

that makes a small difference, 1% casting = 0.9% dps

Total dps of the spell would be Damage/(cast + recast + recovery)
0% casting:
8000/(4.0 + 45) = 163.27 dps (Base)

1% casting:
8000/(3.96 + 45) = 163.4 dps (100.08% Base)

2% casting:
8000/(3.92 + 45) = 163.53 dps (100.16% Base)

1% recast:
8000/(4 + 44.55) = 164.78 dps (100.9% Base)

2% recast:
8000/(4 + 44.1) = 166.32 dps (101.9% Base)

1% casting + 1% recast:
8000/(3.96 + 44.55) = 164.91 dps (101% Base)

2% casting + 2% recast:
8000/(3.92 + 44.1) = 166.6 dps (102% Base)

Are these now the way you'd compare them? and how does that stack up vs crit/BoE?

Slippery · 12-04-2007, 06:05 AM

I haven't come back and respond to this yet because I haven't felt like running the numbers again to give the "correct" result. However, the inconsistency in my results and your results are because my results are based off of my current stats in a raid (66ish casting speed, 73ish crit, 4 base damage). The first part of your post where you divide damage by cast + recovery is correct. When it comes to figuring recast, this is a grey area. You can run perfect numbers but they will not necessarily be true. The reason behind this is the recast only matters if you actually have the ability to cast the spell. Generally a spell is going to become ready to cast while you are casting a different spell, so that extra half a second you gained from the recast reduction might just be going to waste. That isn't to say it will never help, but quite often it won't.

shadowflame · 12-04-2007, 08:20 AM

Basically Slippery lost track of time playing LoN again

gte386v · 12-04-2007, 11:27 AM

It’s really hard to quantify and compare casting speed haste to base damage increase to crit change to BOE for the reason that casting speed and BOE suffer from diminishing returns. Because of this, it matters what you’re starting point is...

Using a vastly overcomplicated spreadsheet, warlock spells (mostly T7 masters as I’ve been too lazy to get the T8 numbers yet), and 37% casting haste, 17% recast haste, 0% recast haste, 35% crit change, 750 BOE, 4 proc items, and 4% increase proc chance I see the following gains:

Calculated DPS: 2179.8
1% Increase in Base Damage: 2196.6 (0.77% gain – doesn’t increase BOE gain or proc damage so less than 1%)
1% Increase in Casting Haste: 2188.0 (0.38% gain)
1% Increase in Recast Haste: 2183.6 (0.18% gain)
1% Increase in Recover Haste: 2183.5 (0.17% gain)
1% Increase in Crit Change: 2186.3 (0.30% gain)
25 Increase in BOE: 2190.6 (0.50% gain)

To conform to the format in previous posts 1% Base Increase ~ 2% Cast Haste ~ 4% Recast Haste ~ 4% Recovery Haste ~ 2.5% Crit Chance ~ 39 BOE.

If anyone cares to see the spreadsheet and update it for wizard spells, it’s available here. The inputs should be fairly intuitive with the calculations being the messy part. Please let me know if you find any mistakes in the formulas.

Lauraliane · 12-04-2007, 12:19 PM

Quote:

Originally Posted by gte386v

1% Increase in Base Damage: 2196.6 (0.77% gain – doesn’t increase BOE gain or proc damage so less than 1%).

Increase in Base damage, also increase the amount of BoE that will affect your spell, as the cap is 50% of base dmg.

The goal shouldn't be to find the most important stat between all of them, but to determine the optimal cap for each of them. At which value BoE lose its interest ? Spell Haste ? Spell Recovery ?
if we can get those values, especially BoE which depends on a lot of others factors, then you just have to gear yourself by getting as closer to those cap as you can.

gte386v · 12-04-2007, 12:27 PM

Quote:

Originally Posted by Lauraliane

Increase in Base damage, also increase the amount of BoE that will affect your spell, as the cap is 50% of base dmg.

If a spell does 3000 damage, then the BOE cap is 1500. If you are below 1500 BOE, increased base damage doesn't give you anymore BOE gain. Now if you had 1525 BOE, and you got a 5% base increase to 3150, the new BOE cap is 1575 so there is a little gain. However, outside of a couple of spells, I don't see warlocks and wizards capping BOE and thus base damage increase doesn't help BOE. Regardless, that is taken into account in the spreadsheet.

Slippery · 12-04-2007, 02:12 PM

Quote:

Originally Posted by gte386v

It’s really hard to quantify and compare casting speed haste to base damage increase to crit change to BOE for the reason that casting speed and BOE suffer from diminishing returns. Because of this, it matters what you’re starting point is...

There is no diminishing returns and there is nothing complicated about it. All of these give quite linear benefits. You could graph from 1% casting speed to 100% casting speed from base dps and you will get a very straight line as too how much dps you gain. From the base dps 1% base damage will give you 1% more dps, 33% base damage will give you 33% more dps. To see the general and get a rough guideline of these you calculate from base, and it will never deviate far enough from that baseline to change the outcome. Base Damage > Casting Speed > Crit With BoE depending on amount. The only reason the numbers will EVER appear different is because YOU wanted to use a different baseline that isn't actually a baseline.

Base damage does effect procs.

In the end, none of this even fucking remotely matters. You can do math and bitch and moan numbers all day long, it doesn't mean shit without practical experience. Their is far more to consider that will increase your dps in other ways than just casting in a certain order, and sometimes even just casting the best spell is up is not that right choice.

gte386v · 12-04-2007, 02:31 PM

Quote:

Originally Posted by Slippery

There is no diminishing returns and there is nothing complicated about it. All of these give quite linear benefits. You could graph from 1% casting speed to 100% casting speed from base dps and you will get a very straight line as too how much dps you gain. From the base dps 1% base damage will give you 1% more dps, 33% base damage will give you 33% more dps. To see the general and get a rough guideline of these you calculate from base, and it will never deviate far enough from that baseline to change the outcome. Base Damage > Casting Speed > Crit With BoE depending on amount. The only reason the numbers will EVER appear different is because YOU wanted to use a different baseline that isn't actually a baseline.

Casting time with 0% haste: 3 seconds
Casting time with 50% haste: (3) / (1+.5) = 2 seconds (1 seconds change)
Casting time with 100% haste: (3) / (1+1) = 1.5 seconds (1.5 second change)

First 50% nets 1 sec less. Second 50% nets 0.5 seconds less. Non-Linear

The graph is attached...

Quote:

Originally Posted by Slippery

Base damage does effect procs.

I'll take your word on this and will test it as well.

Quote:

Originally Posted by Slippery

In the end, none of this even fucking remotely matters. You can do math and bitch and moan numbers all day long, it doesn't mean shit without practical experience. Their is far more to consider that will increase your dps in other ways than just casting in a certain order, and sometimes even just casting the best spell is up is not that right choice.

Didn't say this isn't the case but you'd be surprised how well the models work. Daray had a post a while back comparing BOE to crits and his practical results and theoretical results were less than 5% different.