| Author |
Thread Statistics | Show CCP posts - 0 post(s) |
Entity
X-Factor Industries
Synthetic Existence
   |
Posted - 2009.12.31 15:59:00 -
[1]
Side note: Absolute peak recharge occurs at 25% shield (or cap, it's the same formula), not 30%.
_
Got Item? | EVE API? | Cache? |
Entity
X-Factor Industries
Synthetic Existence
|
Posted - 2009.12.31 16:45:00 -
[2]
Edited by: Entity on 31/12/2009 16:45:49
Perhaps it feels like it's 30%, but it's really at 25% give or take 0.1%
F:\tank>peak.py
7.10589169958 peak recharge at 24.90%
At 30% it is 7.03891192716
This script uses EVE's charged attribute curve (as used for shields and capacitor, and possibly heat), and records the deltas at 0.0001% intervals. Figures are for an unfitted Navy Caracal at max skills
As you can see the difference is way too small to be noticed, so at 30% you have virtually the same delta as at 25%, but the mathematical peak is most definitely at 25%.
_
Got Item? | EVE API? | Cache? |
Entity
X-Factor Industries
Synthetic Existence
|
Posted - 2010.01.01 00:04:00 -
[3]
 Originally by: Liang Nuren
Originally by: Entity
Edited by: Entity on 31/12/2009 16:45:49
Perhaps it feels like it's 30%, but it's really at 25% give or take 0.1%
F:\tank>peak.py
7.10589169958 peak recharge at 24.90%
At 30% it is 7.03891192716
This script uses EVE's charged attribute curve (as used for shields and capacitor, and possibly heat), and records the deltas at 0.0001% intervals. Figures are for an unfitted Navy Caracal at max skills
As you can see the difference is way too small to be noticed, so at 30% you have virtually the same delta as at 25%, but the mathematical peak is most definitely at 25%.
Source?
-Liang
http://pastie.org/762889
based on the formula here:
http://go-dl1.eve-files.com/media/corp/Corvax/Capacitor01.gif
yes, it's really also this formula for shield recharge or I will eat my virtual hat ;)
_
Got Item? | EVE API? | Cache? |
Entity
X-Factor Industries
Synthetic Existence
|
Posted - 2010.01.01 12:06:00 -
[4]
Originally by: Pottsey
But if you have a very large hitpoint buffer as most passive tanks do you should be able to tank at 25% or 27% all the way up to 29%. But this is not happening in game or in testing.
If peak HP regen is 100 and someone does 100dps but a single hit only knocks you down 1% or less due to a large buffer you should hold at 28% not fail and die once the shields drop below 30%.
More testing is needed.
Alright. I'll design a test to (dis)prove the 25%. I know just the thing to do.
Might take a while though because it's rather complicated stuff :P
_
Got Item? | EVE API? | Cache? |
Entity
X-Factor Industries
Synthetic Existence
|
Posted - 2010.01.01 16:45:00 -
[5]
Alright, here is my proof.
Test conducted as follows:
1. Sit in space with an arbitrary ship.
2. Someone else strips your shield using whatever weapons.
3. Record the shield value as shown on the HUD (absolute readout) whenever it changes and record the time of each change (see bottom of post).
4. Normalize data points to a 100hp shield with 100s recharge.
5. Convert the data points to deltas since previous point (both time and shield) and normalize the shield hp deltas to hp-gained-per-second (using Δhp * 1/Δt)
6. Spit out a table with the observed normalized recharge in hp/sec side by side with the calculated expected recharge in hp/sec:
http://pastie.org/763221
Here's a little graph to go with the data:
http://dl.eve-files.com/media/corp/Entity/shieldcurve.png
So, to summarize:
- shield recharge curve is the same as cap recharge curve.
- the formula mentioned in my earlier post is correct.
- the peak occurs at approximately 25%
Q.E.D.
*: Note on step 3 - There are various ways to do this with varying levels of accuracy. You can use screencapture/ocr software, or fraps it and process it all manually. You can even whip something up with AHK or ACTool that does this if you really want.
_
Got Item? | EVE API? | Cache? |
| |
|