On the topic of disintegrate, I've noticed something odd happening with my Stone of Jordan. (+15 Arc power, -5 Arc Power Cost for Disintegrate). Namely the spell can be channeled for a REALLY long time(almost 25x longer) with my .9 swing speed weapon with the ring on. So I did some testing at different speeds.
Below is the data:
There is a big jump in how long I can stand in one place and cast between the 0.9 and 0.92 swing speeds and also the 0.95 and 0.97 swing speeds.
So my question to the community is, should the reduced cost of the spell from 20 down to 15 arcane power make it last this long? (up from 7 seconds to 173 seconds)
Also has anyone figured out how spells like this(channeled spells) use their resource costs? Because my best guess isn't anywhere in the ball park of what numbers I'm seeing in game.
The equation I used for estimating casting time is:
0 = (Max AP - Min AP) + (Regen Rate * Time) - (Swing Speed * Cost *Time)
0 = (115-6) + (10*T) - (0.9*15*T)
T=31.14 in this case
Don't forget you can also get the bonus Arcane Power/second regen from the Enchantress.
Simplify it.
Normally, it costs 20AP per second, and you are regaining 10AP/second. Every second, you lose 10AP, so you should be able to hold it for 10 seconds normally.
If you have 1.00 Attacks per second, and disintegrate costs 15AP per second, and you are regaining 10AP/second, you are only burning 5AP every second. Then with a base of 115 AP, you'll be able to keep it going for 23 seconds. Your ring with -5 Cost, and +15 Max increases the duration by 13 seconds. If you were to have the Enchantress bonus of 0.5AP/second, you'd only be burning 4.5AP every second, meaning you could keep it up for 25.5 seconds.
If the ticks are related to attack speed as you suggest, you'll be able to keep it active for a very long time.
Don't forget you can also get the bonus Arcane Power/second regen from the Enchantress.
If you have 1.00 Attacks per second, and disintegrate costs 15AP per second, and you are regaining 10AP/second, you are only burning 5AP every second. Then with a base of 115 AP, you'll be able to keep it going for 23 seconds. Your ring with -5 Cost, and +15 Max increases the duration by 13 seconds. If you were to have the Enchantress bonus of 0.5AP/second, you'd only be burning 4.5AP every second, meaning you could keep it up for 25.5 seconds.
The thing is I don't have the enchantress nor do I have astral presence in that data collection. Also, the spell stops casting when I reach 6 AP (not 0).
When I tested a 1 swing/sec weapon I lasted for 42 seconds, not the 23 you predicted nor the 21.8 I predicted. But the mind blowing number is the 173 second cast duration with my 0.9 swing speed mace.
In the OP I did use a simple formula, though it may not have been written out as such, either way it predicts only a 31.14 second cast time for the 0.9 swing speed. 5.5x less than that I see in the game.
The equation:
0 = (Max Arc - Min Arc) + (Regen Rate * Time) - (Swing Speed * Cost *Time)
0 = (115-6) + (10*T) - (0.9*15*T)
T=31.14 in this case
It is because of the way that the game data doesn't match the obvious simple equation that I am posting here. My thinking is that there must be some kind of time-averaging formula that the game uses to calculate the cost of each 'tick' of the beam, since the draining of your AP happens near continuously, not in discrete chunks as is assumed in our simplified equations.
I just spent 100k on a -5 disintegrate cost helm to test this out and I was indeed able to reproduce your results. The behavior seems very odd and I don't really have a reasonable explanation for it. I'm pretty sure it's due to the cost reducing item though. If I remove it the duration seems to be quite close to the expected value.
I tried to see how much arcane power was charged for casting disintegrate if it was only cast for an 'instant' (i.e. a quick tap of the button) to try and estimate how much each 'tick' of the beam cost to try and relate it to the swing speed....but I couldn't get my results to be consistent.
Below is the data:
There is a big jump in how long I can stand in one place and cast between the 0.9 and 0.92 swing speeds and also the 0.95 and 0.97 swing speeds.
So my question to the community is, should the reduced cost of the spell from 20 down to 15 arcane power make it last this long? (up from 7 seconds to 173 seconds)
Also has anyone figured out how spells like this(channeled spells) use their resource costs? Because my best guess isn't anywhere in the ball park of what numbers I'm seeing in game.
The equation I used for estimating casting time is:
0 = (Max AP - Min AP) + (Regen Rate * Time) - (Swing Speed * Cost *Time)
0 = (115-6) + (10*T) - (0.9*15*T)
T=31.14 in this case
Simplify it.
Normally, it costs 20AP per second, and you are regaining 10AP/second. Every second, you lose 10AP, so you should be able to hold it for 10 seconds normally.
If you have 1.00 Attacks per second, and disintegrate costs 15AP per second, and you are regaining 10AP/second, you are only burning 5AP every second. Then with a base of 115 AP, you'll be able to keep it going for 23 seconds. Your ring with -5 Cost, and +15 Max increases the duration by 13 seconds. If you were to have the Enchantress bonus of 0.5AP/second, you'd only be burning 4.5AP every second, meaning you could keep it up for 25.5 seconds.
If the ticks are related to attack speed as you suggest, you'll be able to keep it active for a very long time.
The thing is I don't have the enchantress nor do I have astral presence in that data collection. Also, the spell stops casting when I reach 6 AP (not 0).
When I tested a 1 swing/sec weapon I lasted for 42 seconds, not the 23 you predicted nor the 21.8 I predicted. But the mind blowing number is the 173 second cast duration with my 0.9 swing speed mace.
In the OP I did use a simple formula, though it may not have been written out as such, either way it predicts only a 31.14 second cast time for the 0.9 swing speed. 5.5x less than that I see in the game.
The equation:
0 = (Max Arc - Min Arc) + (Regen Rate * Time) - (Swing Speed * Cost *Time)
0 = (115-6) + (10*T) - (0.9*15*T)
T=31.14 in this case
It is because of the way that the game data doesn't match the obvious simple equation that I am posting here. My thinking is that there must be some kind of time-averaging formula that the game uses to calculate the cost of each 'tick' of the beam, since the draining of your AP happens near continuously, not in discrete chunks as is assumed in our simplified equations.
Any suggestions?