These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Ships & Modules

 
  • Topic is locked indefinitely.
 

shield regen rate and stability as an indicator of in-mission DPS
Author
Previous Topic Next Topic
Wingmate
Perkone
Caldari State
#1 - 2012-03-05 17:13:47 UTC  |  Edited by: Wingmate
SO. disclaimer - while i am pretty good at logical thinking, i suck at the application of said thinking to mathy-looking things. so, halp plz =D

i know that it's possible to compute an exact shield regeneration rate, and to create a shield regen graph using the capacitor regen formula. i'd like a way to do this in excel or another mapping program. here's why.

if it's possible to compute the exact recharge rate at any point of a ship, in turn it's possible to determine exactly how much recharge is currently occuring when your ship is holding against enemies - and, in turn, exactly how much damage is incoming on your ship before and after resists.

why is this important? well, it'll allow me to map sleeper DPS data, which no one seems to have. it'll also allow me to determine just how much tank to fit for missions and plexes. this'll allow me to fit more dps on certain ships for certain plexes, or tank in something other than a drake.

so, in short, i'm looking for someone who can help me put a formula into excel that'll use a preset shield size and regeneration rate to create a regen curve that i can then reference for determining incoming DPS. anyone got some mathlike skills to assist?

edit: for your own research, this was answered at http://www.eve-search.com/thread/1365024#18, interestingly enough, by isan'na. i just don't know enough math to be able to break that down into something that i can put into excel - specifically, the difference between T and t, total recharge time and "time". if i get that, i can make it into something for a spreadsheet program, i think.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
Ira Infernus
Knights-of-Cydonia
#2 - 2012-03-05 17:35:06 UTC
You have alot of variables to consider.

Are you calculating the total DPS of an anom, deadspace pocket, etc?

If so you forget about aggro triggers and movement, DPS mitigation by transversal and real velocities, damage projection over specific distances, etc etc.

If you are just going to sit there and take it to get the values, then what is the point? - you should be able to quickly reduce incoming dps with your own weaponry anyway, so knowing an exact value is not important. You will also have to take your skills/implants into account, wormhole effects, etc etc.


It would be much easier to calculate DPS at a specific time... Do you have the recharge equation at hand?
Wingmate
Perkone
Caldari State
#3 - 2012-03-05 17:49:49 UTC  |  Edited by: Wingmate
if you read the thread that i linked - and specifically the post i linked - you'll see that it answers all of your questions.

i have the formula. i need help from a mathy person to solve it. so, most of your questions are moot =) calculating specific dps requires knowing the dps of the ships attacking. since no one has sleeper dps numbers that i've seen, it's impossible to do that.

by calculating based on shield percentage, i can determine dps at the enemy's optimal while both moving and not moving, since i just use a different static shield percentage wherever it stabilizes at for both situations.

as i said, by being able to determine exactly how much dps is incoming, i can not only plan for it myself, but make sure that my corpies that are either getting into WH space (or getting into anoms in serp space where i don't want to add up small dps numbers for a zillion enemies) are ready to tank enough damage while still being able to output damage. WH space is especially important, since it takes a few seconds to kill that first ship if you're soloing, and you need to make sure you can tank the initial spawns long enough to get into optimal, or pop the first ship. it also lets me compare ship tanks between races, since if i know a room has 775dps against a drake, i can convert that to an armor tank, or to an active shield tank, or a fit that does more dps as opposed to being all tank and no damage.

trust me, it's relevant, and important.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
mxzf
Shovel Bros
#4 - 2012-03-05 18:10:07 UTC
If you haven't taken calculus, then it makes sense you wouldn't understand it. If I recall my calculus correctly, and I'm a bit rusty, the t is the current time and (s)he's taking the differential with respect to time to determine the recharge rate.

I'm working on decrypting the math into something that'll be easy to plug into Excel, but I'm fairly rusty (and haven't used hyperbolic trig before). So anyone who's more familiar with it, feel free to chime in, lol.

And, as Ira Infernus said, IDK how much the formula will actually help you, since it's not really realistic information, but the math has interested me, lol.
stoicfaux
#5 - 2012-03-05 18:11:34 UTC
You could also try using the LogServer.exe and determining your remaining shield points to infer incoming DPS.

For example, http://community.eveonline.com/ingameboard.asp?a=topic&threadID=1277798&page=1#11 I think you will also only want to filter on "svc - godma" messages. It's been awhile, but their should be a shield recharge message (every second?) that tells you how much of your shield regenerated which also contains your current and max shield hp.

Pon Farr Memorial: once every 7 years, all the carebears in high-sec must PvP or they will be temp-banned.
mxzf
Shovel Bros
#6 - 2012-03-05 18:36:32 UTC
Ok, I managed to derive the value of t, I think, as ( T*ln(S/(s-Sc)+( (S/(S-Sc))^2 - 1 )^0.5) )/4.8.

After a bit of work, I believe I've got it. If the total capacity of your shields is in cell A2 and the current capacity of your shields is in B2 and your total time to recharge shields (after skills and all, the T value from before) is in C2, then the formula for instantaneous shields recharged per second, formatted for Excel, should be:

=1+(A2*4.8/C2)*((EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)-EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)))/(EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)+EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2))))*(2/(EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)+EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2))))

(if you want, I can break it down into a series of smaller equations for you to work with, if that would be easier)
Wingmate
Perkone
Caldari State
#7 - 2012-03-05 18:53:23 UTC
mxzf wrote:
Ok, I managed to derive the value of t, I think, as ( T*ln(S/(s-Sc)+( (S/(S-Sc))^2 - 1 )^0.5) )/4.8.

After a bit of work, I believe I've got it. If the total capacity of your shields is in cell A2 and the current capacity of your shields is in B2 and your total time to recharge shields (after skills and all, the T value from before) is in C2, then the formula for instantaneous shields recharged per second, formatted for Excel, should be:

=1+(A2*4.8/C2)*((EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)-EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)))/(EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)+EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2))))*(2/(EXP((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2)+EXP(-((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8*4.8/C2))))

(if you want, I can break it down into a series of smaller equations for you to work with, if that would be easier)


hey, this is exactly what i was looking for. and no, i haven't taken calculus since my junior year of high school. i have a masters now =)

i will confirm later today with a friend shooting me with missiles (so we get a constant dps number).

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
mxzf
Shovel Bros
#8 - 2012-03-05 19:06:20 UTC  |  Edited by: mxzf
Yeah, I'm not 100% positive about the formula, I just took that formula from the other thread, solved for t, turned the hyperbolic trig functions into basic math, and plugged it all in.

*facepalm* Excel supports hyperbolic trig though, I'd forgotten about that. Here's a much simpler version of that, much less mess:

=1+(A2*4.8/C2)*TANH(((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8)*4.8/C2)/COSH(((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8)*4.8/C2)

Edit: And, from some variables that should always cancel each other out, I managed to simplify it down a bit further still, into:

=1+(A2*4.8/C2)*TANH(LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/COSH(LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))
Wingmate
Perkone
Caldari State
#9 - 2012-03-05 19:08:53 UTC  |  Edited by: Wingmate
that looks a little nicer.

i mentioned this to my brother-in-law who is way smarter than i am (hi, matt!) and he was going to take a look too.

now i just have to incorporate damage profiles and resists, and i've got another page for my all-eve spreadsheet extravaganza =)

edit: your second one is great. your third one gives me errors.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
mxzf
Shovel Bros
#10 - 2012-03-05 19:12:42 UTC
Yeah, it looks nicer because, for instance,

tanh = (e^x)-(e^-x)/((e^x)+(e^-x)) where x was ((C2*LN((A2/(A2-B2))+SQRT(POWER((A2/(A2-B2)),2)-1)))/4.8)*4.8/C2, lol. So by simply using tanh() instead of plugging all that in, it cleaned it up a LOT, lol. Same kind of thing for sech too (except using the fact that sech(x) = cosh(x)^-1.
Wingmate
Perkone
Caldari State
#11 - 2012-03-05 19:20:45 UTC
so, when testing it, this gives me a max of 240% of average recharge at 30%, instead of 250% at 25%. is that the formula, or rounding?

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
Nalha Saldana
Aliastra
Gallente Federation
#12 - 2012-03-05 19:24:26 UTC
The max recharge is at about 1/3 of shield, not 25%
mxzf
Shovel Bros
#13 - 2012-03-05 19:37:01 UTC
Wingmate wrote:
so, when testing it, this gives me a max of 240% of average recharge at 30%, instead of 250% at 25%. is that the formula, or rounding?


What the people were saying in that other thread is that the peak is 2.5 at 25% but for safety 2.4 at 30% was the better number to use for testing. It sounds like the formula is working out right. I don't have a ton of specific data myself, I just reformatted that equation, I don't know how the code actually handles it.
Wingmate
Perkone
Caldari State
#14 - 2012-03-05 19:57:19 UTC
i think i got it, now, mxzf. thanks =) the safety factor also kind of factors in wrecking shots and randomness as well, so i guess it's not a bad idea.

Nalha Saldana wrote:
The max recharge is at about 1/3 of shield, not 25%


as has been proven on several occasions, the 33% myth is because of lag and refresh rates. the actual max is 25%, at 250%, not 30-33% at 240%.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
Fronkfurter McSheebleton
Horse Feathers
CAStabouts
#15 - 2012-03-05 20:20:34 UTC  |  Edited by: Fronkfurter McSheebleton
Data for every single sleeper, as well as about any other npc:

http://games.chruker.dk/eve_online/inventory.php?category_id=11


Sleepers are toward the middle. (easiest way to find them is ctrl+f)

thhief ghabmoef
Wingmate
Perkone
Caldari State
#16 - 2012-03-05 20:26:07 UTC
Fronkfurter McSheebleton wrote:
Data for every single sleeper, as well as about any other npc:

http://games.chruker.dk/eve_online/inventory.php?category_id=11


Sleepers are toward the middle. (easiest way to find them is ctrl+f)


i knew it was out there somewhere =)

this formula will still be good for figuring out dps without having to add up a zillion rat numbers, which is nice.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
Nalha Saldana
Aliastra
Gallente Federation
#17 - 2012-03-05 22:02:58 UTC
i reformatted it in another way and ended up with a peak recharge at 29.8% so i guess its correct as long as that is the same equation ingame too.
Wingmate
Perkone
Caldari State
#18 - 2012-03-05 22:09:10 UTC
as said above:

"What the people were saying in that other thread is that the peak is 2.5 at 25% but for safety 2.4 at 30% was the better number to use for testing."

that formula's got some error built into it to account for wrecking shots and the fact that the curve doesn't exactly fit, but is super close to fitting, the actual in-game chart.

i make spreadsheets for pretty cheap. contact me for more info.

https://forums.eveonline.com/default.aspx?g=posts&t=197433
Nalha Saldana
Aliastra
Gallente Federation
#19 - 2012-03-05 22:10:43 UTC
Oh ok misunderstood that before. This math way beyond me anyway :P