These forums have been archived and are now read-only.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

ECM Falloff Formula

Author
BonFire Circle
Sebiestor Tribe
Minmatar Republic
#1 - 2017-03-27 15:12:37 UTC
Could someone please provide a link to the formula for ecm effectiveness after optimal range? Eve-U just basically gives two values and doesn't specify if it's linear.
Soel Reit
The Scope
Gallente Federation
#2 - 2017-03-27 18:23:19 UTC  |  Edited by: Soel Reit
http://wiki.eveuniversity.org/Turret_damage#Combined_effects_of_hit_chance_and_damage_distribution

falloff of ECM module is applied linearly
diminuishing your sensor strengh and thus the chance to jam every time


single module
Chance to Jam = ( Your ECM Strength / Target's Sensor Strength ) * 100%
multiple modules Chance to Jam = (1 -(1 - Your ECM Strength / Target's Sensor Strength ) ^ The number of jammers of this strength)*100%

http://wiki.eveuniversity.org/ECM



TL;DR
apply the same falloff that guns have for dps and apply it on sensor strenght of the ECM module.
then calculate the chance you get to jam with the new sensor strenght


Shocked
pretty sure someone will come along and correct me Cool
Buggs LeRoach
DHCOx
#3 - 2017-03-27 22:50:36 UTC
Soel Reit wrote:
http://wiki.eveuniversity.org/Turret_damage#Combined_effects_of_hit_chance_and_damage_distribution

falloff of ECM module is applied linearly
diminuishing your sensor strengh and thus the chance to jam every time


single module
Chance to Jam = ( Your ECM Strength / Target's Sensor Strength ) * 100%
multiple modules Chance to Jam = (1 -(1 - Your ECM Strength / Target's Sensor Strength ) ^ The number of jammers of this strength)*100%

http://wiki.eveuniversity.org/ECM



TL;DR
apply the same falloff that guns have for dps and apply it on sensor strenght of the ECM module.
then calculate the chance you get to jam with the new sensor strenght


Shocked
pretty sure someone will come along and correct me Cool


ya , you're wrong . it's like you didn't even read the page you linked ..
Soel Reit
The Scope
Gallente Federation
#4 - 2017-03-27 22:58:20 UTC
i read it Roll

but i just wait someone like you that is able to read better than me so you can explain Cool
i love my ez life
BonFire Circle
Sebiestor Tribe
Minmatar Republic
#5 - 2017-03-27 23:29:27 UTC
So is this curve the answer?
Soel Reit
The Scope
Gallente Federation
#6 - 2017-03-27 23:30:47 UTC
maybe better wait for the guru Cool


Zhilia Mann
Mephiztopheleze
Laphroaig Inc.
#7 - 2017-03-28 03:49:26 UTC
ECM Guru Checking In....

due to the all-or-nothing RNG-Jesus nature of ECM, stay inside your optimal wherever possible.

damps and TD/GD do not have the same kinds of penalties for falloff as ECM.

Occasional Resident Newbie Correspondent for TMC: http://themittani.com/search/site/mephiztopheleze

This is my Forum Main. My Combat Alt is sambo Inkura

Chainsaw Plankton
FaDoyToy
#8 - 2017-03-28 06:50:50 UTC
I assume it is linear, if you really want to test it get something with 8 mids and start counting successes/failures not sure if the log provides ecm entries, if it does at least you can automate the counting part. Someone used the log server output ages ago to calculate if eccm worked vs gurista npcs not sure if that would still work.

I'm also guessing that you could get the rates by using a target painter at various ranges and seeing what the sig bloom is.

with guns there is a curve to the damage/range as hit quality is effected as well as hit chance. so optimal + falloff is a 50% chance to hit, but ends up being something like 38% dps.

@ChainsawPlankto on twitter

Zhilia Mann
Tide Way Out Productions
#9 - 2017-03-28 15:12:16 UTC  |  Edited by: Zhilia Mann
It's standard falloff mechanics with a binary hit/miss equal to the gunnery formula's hit chance reduction due to falloff.

In math terms, optimal always hits[1], and falloff has a chance of hitting[2] of:

0.5^({\frac{max(0,range-optimal)}{falloff})^2

Or if you want that to not look like **** plug this in:

0.5^{\left(\frac{\textup{max}(0,\textup{range}-\textup{optimal})}{\textup{falloff}} \right )^2}

Here:

https://www.codecogs.com/latex/eqneditor.php

You can see some sample curves generated here. If you really want I'll do one specific to ECM but obviously there isn't one yet. [Edit: now there is. Pick the proper tab. Should be obvious.] Note that you're basically looking at the right half of a normal curve. Wolfram alpha can give you a sense for the family of curves we're talking about here.

So now to [1] and [2], which is ECM-specific.

"Hit" in this case actually means "gets a chance to roll against sensor strength". I actually don't know if this is the right order or not, but basically there are two checks:

1. Does RNG think you rolled high enough to hit a falloff calculation?
2. Does RNG think you rolled high enough to land a jam against the target?

If both pass, congrats, you get to jam someone. If either fails, the cycle fails.
Zhilia Mann
Tide Way Out Productions
#10 - 2017-03-28 15:29:33 UTC
Chainsaw Plankton wrote:
I assume it is linear, if you really want to test it get something with 8 mids and start counting successes/failures not sure if the log provides ecm entries, if it does at least you can automate the counting part. Someone used the log server output ages ago to calculate if eccm worked vs gurista npcs not sure if that would still work.

I'm also guessing that you could get the rates by using a target painter at various ranges and seeing what the sig bloom is.

with guns there is a curve to the damage/range as hit quality is effected as well as hit chance. so optimal + falloff is a 50% chance to hit, but ends up being something like 38% dps.


Easiest way to test specifically for linearity would be to hit something (with ECM, painter, whatever) outside of optimal + 2*falloff. A linear model would suggest that this would never hit. Exponential decay says it does. (Hint: it does.)
Soel Reit
The Scope
Gallente Federation
#11 - 2017-03-28 18:17:04 UTC  |  Edited by: Soel Reit
guru has spoken!
all hails to the guru Cool
Chainsaw Plankton
FaDoyToy
#12 - 2017-03-28 20:57:36 UTC  |  Edited by: Chainsaw Plankton
doh, your right, and I should have known that from the RR changes https://forums.eveonline.com/default.aspx?g=posts&m=6190014

at least I think you are mostly right, I'm not sure they do a chance to hit roll anymore as it seems they just factor that into your strength instead. My testing of a navy griffin orbiting a minmatar rookie ship hasn't resulted in a missed jam yet. I'm a bit under 30% falloff which gives me more than enough strength to overwhelm the rookie ships weak 6.96 sensor strength. If chance to hit was still in play I'd have missed at least a few.

Edit: of course the log doesn't show missed jams... I still think I'm right, every time I looked I had jams with 5/5 ecm, but I was afk and not watching for a lot of them. And now I'm trying out at ~15km right now where I should have a ~30% chance to jam but hard to really say anything without collecting data and I'm too lazy to write it down right now.

@ChainsawPlankto on twitter

Zhilia Mann
Tide Way Out Productions
#13 - 2017-03-28 21:28:06 UTC
Chainsaw Plankton wrote:
doh, your right, and I should have known that from the RR changes https://forums.eveonline.com/default.aspx?g=posts&m=6190014

at least I think you are mostly right, I'm not sure they do a chance to hit roll anymore as it seems they just factor that into your strength instead. My testing of a navy griffin orbiting a minmatar rookie ship hasn't resulted in a missed jam yet. I'm a bit under 30% falloff which gives me more than enough strength to overwhelm the rookie ships weak 6.96 sensor strength. If chance to hit was still in play I'd have missed at least a few.

Edit: of course the log doesn't show missed jams... I still think I'm right, every time I looked I had jams with 5/5 ecm, but I was afk and not watching for a lot of them. And now I'm trying out at ~15km right now where I should have a ~30% chance to jam but hard to really say anything without collecting data and I'm too lazy to write it down right now.


Eh, ****. I guess I have to spend some time on Sisi. If they've gone and made it a strength decline instead of separate rolls I guess I'll have to modify the discussion. Still, good to know it's due for a test.
Soel Reit
The Scope
Gallente Federation
#14 - 2017-03-28 21:29:49 UTC
Zhilia Mann wrote:
Chainsaw Plankton wrote:
doh, your right, and I should have known that from the RR changes https://forums.eveonline.com/default.aspx?g=posts&m=6190014

at least I think you are mostly right, I'm not sure they do a chance to hit roll anymore as it seems they just factor that into your strength instead. My testing of a navy griffin orbiting a minmatar rookie ship hasn't resulted in a missed jam yet. I'm a bit under 30% falloff which gives me more than enough strength to overwhelm the rookie ships weak 6.96 sensor strength. If chance to hit was still in play I'd have missed at least a few.

Edit: of course the log doesn't show missed jams... I still think I'm right, every time I looked I had jams with 5/5 ecm, but I was afk and not watching for a lot of them. And now I'm trying out at ~15km right now where I should have a ~30% chance to jam but hard to really say anything without collecting data and I'm too lazy to write it down right now.


Eh, ****. I guess I have to spend some time on Sisi. If they've gone and made it a strength decline instead of separate rolls I guess I'll have to modify the discussion. Still, good to know it's due for a test.


this is what i substained without tests.
it's the most logical thing all considered!

but never trust CCP on logical things Roll
Buggs LeRoach
DHCOx
#15 - 2017-03-29 15:03:45 UTC
i always understood , in falloff , there was a 50/50 roll to see if the jam was attempted . if success , then you'd roll for a jam same as within optimal . here's a link and a table , that shows there is a separate roll , and it scales with distance into falloff .

https://eveinfo.net/wiki/inde~125.htm

Optimal Range and Falloff

Due to the recent ECM nerf, falloff now has a much greater impact than before. Below optimal the jamming probability works as shown above. At optimal+falloff that probability is reduced to half, while at optimal+falloff*2 the probability drops to approximately 0. Between these two values it scales in a non-linear fashion. For reference, consult the following table, where the first value is %distance into falloff and the second value is the chance to hit.

5% 0.998 105% 0.466
10% 0.993 110% 0.432
15% 0.985 115% 0.400
20% 0.973 120% 0.369
25% 0.958 125% 0.339
30% 0.940 130% 0.310
35% 0.919 135% 0.283
40% 0.895 140% 0.257
45% 0.869 145% 0.233
50% 0.841 150% 0.210
55% 0.811 155% 0.189
60% 0.779 160% 0.170
65% 0.746 165% 0.152
70% 0.712 170% 0.135
75% 0.677 175% 0.120
80% 0.642 180% 0.106
85% 0.606 185% 0.093
90% 0.570 190% 0.082
95% 0.535 195% 0.072
100% 0.500 200% 0.063



Reaver Glitterstim
The Scope
Gallente Federation
#16 - 2017-03-30 02:49:24 UTC  |  Edited by: Reaver Glitterstim
Simple answer: it uses the same falloff chance as turrets. If it's a hit, it checks jam strength vs. sensor strength like normal. If it's a miss, it's always a fail.

Meaning my blackbird can jam you at 200km more often than you might think it can.

FT Diomedes: "Reaver, sometimes I wonder what you are thinking when you sit down to post."

Frostys Virpio: "We have to give it to him that he does put more effort than the vast majority in his idea but damn does it sometime come out of nowhere."

Chainsaw Plankton
FaDoyToy
#17 - 2017-03-30 07:07:09 UTC
Reaver Glitterstim wrote:
Simple answer: it uses the same falloff chance as turrets. If it's a hit, it checks jam strength vs. sensor strength like normal. If it's a miss, it's always a fail.

Meaning my blackbird can jam you at 200km more often than you might think it can.

that is how it used to work, however I think with effectiveness falloff it just lowers the jam strength at longer ranges as previously discussed.

@ChainsawPlankto on twitter

Reaver Glitterstim
The Scope
Gallente Federation
#18 - 2017-03-30 07:53:33 UTC
Chainsaw Plankton wrote:
that is how it used to work, however I think with effectiveness falloff it just lowers the jam strength at longer ranges as previously discussed.

I seem to remember hearing some discussion about changing it to lower jam strength instead of using the old chance-based falloff, but it's definitely not linear. The former was something I had suggested a while back, but I don't know if it ever got implemented.

FT Diomedes: "Reaver, sometimes I wonder what you are thinking when you sit down to post."

Frostys Virpio: "We have to give it to him that he does put more effort than the vast majority in his idea but damn does it sometime come out of nowhere."