| Pages: [1] 2 :: one page |
| Author |
Thread Statistics | Show CCP posts - 0 post(s) |

Coreemo
Republic Exploration And Defense Self Sabatoge
2
|
Posted - 2013.05.02 03:49:00 -
[1] - Quote
If multiple ships are multispectrum jamming one target, will that target eventually be perma-jammed, given enough ships? |

1st-Garrentious WispBender
Federal Defense Union Gallente Federation
3
|
Posted - 2013.05.02 03:51:00 -
[2] - Quote
Coreemo wrote:If multiple ships are multispectrum jamming one target, will that target eventually be perma-jammed, given enough ships?
It is a game of chance. Similar to when I take a gallentean fleet of 40 against 1-2 wartargets. You see my friend, the Caldari are a very formative enemy and I risk losing the entire fleet.
As ECM is a similar chance based mechanic, there is always the chance of survival. |

chatgris
Quantum Cats Syndicate Samurai Pizza Cats
473
|
Posted - 2013.05.02 04:22:00 -
[3] - Quote
1st-Garrentious WispBender wrote:Coreemo wrote:If multiple ships are multispectrum jamming one target, will that target eventually be perma-jammed, given enough ships? It is a game of chance. Similar to when I take a gallentean fleet of 40 against 1-2 wartargets. You see my friend, the Caldari are a very formative enemy and I risk losing the entire fleet. As ECM is a similar chance based mechanic, there is always the chance of survival.
Awesome character, awesome name! Hairdo is pretty good too.
Wondering about the etymology of the name..
X Gallentius in the name and hair Deen Wispa in the name (and possibly the sunnies? been a while since I saw the portrait)
Where does the 1-st and Bender come from? |

IbanezLaney
The Church of Awesome Caldari State Capturing
350
|
Posted - 2013.05.02 04:36:00 -
[4] - Quote
chatgris wrote:1st-Garrentious WispBender wrote:Coreemo wrote:If multiple ships are multispectrum jamming one target, will that target eventually be perma-jammed, given enough ships? It is a game of chance. Similar to when I take a gallentean fleet of 40 against 1-2 wartargets. You see my friend, the Caldari are a very formative enemy and I risk losing the entire fleet. As ECM is a similar chance based mechanic, there is always the chance of survival. Awesome character, awesome name! Hairdo is pretty good too. Wondering about the etymology of the name.. X Gallentius in the name and hair Deen Wispa in the name (and possibly the sunnies? been a while since I saw the portrait) Where does the 1-st and Bender come from?
I was guessing the Bender bit was Garr Earthbender (I think thats his name - soz if wrong)
1st bit - not 100% sure but I have an idea.
EDIT: Maybe 1st Blood or 1st commander or something? Ones an alt and ones a booster the Gals use.
That's my guess. Fix this **** See Sea Pea. |

Jack Miton
Aperture Harmonics K162
1775
|
Posted - 2013.05.02 04:40:00 -
[5] - Quote
the chances are independent. basically in a hypothetical where 1 mod has 50% chance to jam, 2 would have 75%, 3 87.5% ect. due to the way probability works, it never hits 100% chance. Only way to actually perma 100% jam someone is if you have a 100% chance on a single mod (ie: your jam strength is higher than their sensor strength).
|

StoneCold
Somali Coast Guard Authority
153
|
Posted - 2013.05.02 07:02:00 -
[6] - Quote
Jack Miton wrote: [...] basically in a hypothetical where 1 mod has 50% chance to jam, 2 would have 75%, 3 87.5% ect. [...]
I have to disagree. If 1 mod got a chance of 50% then 2 mods stilll got the chance of 50% - but 2 times (Laplace).
Sidenote: Hypothetical you could view your 50% jam chance from your post in boolean 'Yes i jam' and 'no jam,sry guys'. If you-¦re now moving your ship in a perfect (and closed) system where all results are equaly distributed you-¦ll allready hit a 100% jam-chance with 2 mods (with 50% chance to jam).
Though i might be VERY wrong and just bored at work.
For Hire Psychotic Monk for CSM |

chatgris
Quantum Cats Syndicate Samurai Pizza Cats
473
|
Posted - 2013.05.02 07:13:00 -
[7] - Quote
Basic stats - http://en.wikipedia.org/wiki/Binomial_distribution
If you flip it around, try to find the probability of 0 successes, and subtract that from 1. |

Jack Miton
Aperture Harmonics K162
1776
|
Posted - 2013.05.02 07:44:00 -
[8] - Quote
StoneCold wrote:Jack Miton wrote: [...] basically in a hypothetical where 1 mod has 50% chance to jam, 2 would have 75%, 3 87.5% ect. [...]
I have to disagree. If 1 mod got a chance of 50% then 2 mods stilll got the chance of 50% - but 2 times (Laplace). Sidenote:Hypothetical you could view your 50% jam chance from your post in boolean 'Yes i jam' and 'no jam,sry guys'. If you-¦re now moving your ship in a perfect (and closed) system where all results are equaly distributed you-¦ll allready hit a 100% jam-chance with 2 mods (with 50% chance to jam). Though i might be VERY wrong and just bored at work. EDIT: No, it-¦s not additive. If you have a poor little bugger attacking your scorpion / falcon / rook fleet and you activate 1000 ecms on him (each 50% jam-chance) you still have a chance of 50% to jam him for each module. It-¦s not 99.9999999999%. you didnt do well at math in school did you? |

StoneCold
Somali Coast Guard Authority
153
|
Posted - 2013.05.02 07:50:00 -
[9] - Quote
aye :( a huge and noobish mistake denied me summa *** laude. So i m just a magna *** laude math noob like 80% of all humans :/. For Hire Psychotic Monk for CSM |

Bad Messenger
Nasranite Watch
459
|
Posted - 2013.05.02 13:31:00 -
[10] - Quote
i like to note that jamming cause cancer so better to not use it. |

Princess Nexxala
Quantum Cats Syndicate Samurai Pizza Cats
379
|
Posted - 2013.05.02 15:00:00 -
[11] - Quote
I endorse this message.
Bad Messenger wrote:i like to note that jamming cause cancer so better to not use it.
derp? |

Onomerous
Shockwave Innovations Shockwave Sovereign Industries
70
|
Posted - 2013.05.02 16:01:00 -
[12] - Quote
Quote:basically in a hypothetical where 1 mod has 50% chance to jam, 2 would have 75%, 3 87.5% ect.
^ these two ^
Quote:i like to note that jamming cause cancer so better to not use it. lol |

Haulie Berry
584
|
Posted - 2013.05.02 16:03:00 -
[13] - Quote
StoneCold wrote:Jack Miton wrote: [...] basically in a hypothetical where 1 mod has 50% chance to jam, 2 would have 75%, 3 87.5% ect. [...]
I have to disagree. If 1 mod got a chance of 50% then 2 mods stilll got the chance of 50% - but 2 times (Laplace).
You... you have to disagree? Because you think it's a 50% chance 2 times, you say?
Guess what aggregate probability that comes out to.
Go on.
Guess. |

Gizznitt Malikite
Agony Unleashed Agony Empire
2080
|
Posted - 2013.05.02 17:56:00 -
[14] - Quote
Actually, this is much more akin to a Geometric Distribution
TL; DR of the math below...
Number of jammers needed = log ( chance not jammed) / log ( 1 - p); where p = chance of single jammer working.
Example: If you want your opponent to only have a 1 in a million chance of not being jammed with a 33% chance of jamming per jammer, you need n = log (1/million) / log ( 1 - .33) = 34.4 = 35 jammers.
Math: P(i) = P(Jammed once given x jammers ) = p * q ^ ( x - 1 ), where p = probability of a single jammer working, and q = 1 - p.
P ( Not Jammed with n jammers) = 1 - Sum(from i=1 to n) [ P(i) ] = 1 - p - pq - pq^2 - ... - pq^(n-1) = 1 - p (1 + q + q^2 + .... + q^(n-1)) = 1 - p ( 1 - q^n ) / ( 1-q ) = 1 - p ( 1- q^n) / p = 1 - 1 + q^n = q^n = (1 - p) ^ n
So, is there a magical n such that ( 1 - p ) ^ n = 0? Only if p = 1, or n = inf. is this absolutely true, although you can pragmatically hit it much, much easier, because you really only need it to be close to zero, not exactly zero.
How many jammers does it take for them to have a one in a million chance of not being jammed? (1-p)^n = 1 / million, rewritten: n log (1-p) = log ( 1 / million) n = log ( 1 / million ) / log ( 1- p) n = -6 / log ( 1 - p) With a 33% chance of success, you need 35 jammers to jam out your opponent all but once every million cycles. With 18 jammers, they have a 1 in a thousand chance of not being jammed. With 12 jammer they have a 1 in a hundred chance of not being jammed |

Bienator II
madmen of the skies
1703
|
Posted - 2013.05.02 18:19:00 -
[15] - Quote
Coreemo wrote:If multiple ships are multispectrum jamming one target, will that target eventually be perma-jammed, given enough ships?
2 mods/drones are twice as good as 1 mod/drone at the target. No stacking penalty is applied. eve style bounties (done) dust boarding parties imagine there is war and everybody cloaks - join FW |

chatgris
Quantum Cats Syndicate Samurai Pizza Cats
473
|
Posted - 2013.05.02 18:28:00 -
[16] - Quote
Thing is, the Geometric Distribution is effectively a subset of Bernoulli's.
"The probability distribution of the number of X Bernoulli trials needed to get one success"
Pretty much the same thing, and for the given question the Geometric distribution is easier to calculate. However, I tend to use Bernoulli's because I can look at the distribution and see things like "20% of the time, I'll have 2 successes which means I'll be able to try and jam a different target with my free jammer". |

Deen Wispa
Justified Chaos
515
|
Posted - 2013.05.02 18:55:00 -
[17] - Quote
chatgris wrote:Thing is, the Geometric Distribution is effectively a subset of Bernoulli's. "The probability distribution of the number of X Bernoulli trials needed to get one success" Pretty much the same thing, and for the given question the Geometric distribution is easier to calculate. However, I tend to use Bernoulli's because I can look at the distribution and see things like "20% of the time, I'll have 2 successes which means I'll be able to try and jam a different target with my free jammer".
Be honest. You just wanted to throw in those words so you can sound smarter than the rest of us. High Five. Yeah! C'est La Eve . |

chatgris
Quantum Cats Syndicate Samurai Pizza Cats
476
|
Posted - 2013.05.02 19:16:00 -
[18] - Quote
Deen Wispa wrote:chatgris wrote:Thing is, the Geometric Distribution is effectively a subset of Bernoulli's. "The probability distribution of the number of X Bernoulli trials needed to get one success" Pretty much the same thing, and for the given question the Geometric distribution is easier to calculate. However, I tend to use Bernoulli's because I can look at the distribution and see things like "20% of the time, I'll have 2 successes which means I'll be able to try and jam a different target with my free jammer". Be honest. You just wanted to throw in those words so you can sound smarter than the rest of us.
I thought I kept my language pretty plain and to the point... :(
I actually do this pretty much every time I am sizing up a target with ECM drones or a griffin. |

JAF Anders
Quantum Cats Syndicate Samurai Pizza Cats
95
|
Posted - 2013.05.02 19:30:00 -
[19] - Quote
chatgris wrote:Deen Wispa wrote:
Be honest. You just wanted to throw in those words so you can sound smarter than the rest of us.
I thought I kept my language pretty plain and to the point... :( I actually do this pretty much every time I am sizing up a target with ECM drones or a griffin.
I'm really enjoying the technical dialogue here. |

chatgris
Quantum Cats Syndicate Samurai Pizza Cats
476
|
Posted - 2013.05.02 19:47:00 -
[20] - Quote
Also, I thought that a large percentage of eve players were Computer Science guys, and this is first year stats. |

X Gallentius
Justified Chaos
1359
|
Posted - 2013.05.02 21:34:00 -
[21] - Quote
I thought they said ecm is addictive. Yes it certainly is. Once a player learns of its true power, he never flies without an ecm alt again. |

Tsukino Stareine
EVE University Ivy League
117
|
Posted - 2013.05.02 21:41:00 -
[22] - Quote
chatgris wrote:Also, I thought that a large percentage of eve players were Computer Science guys, and this is first year stats.
I learnt this during first year of A-Levels.....
and ECM is boring, what's the fun in shootin something that can't shoot back? |

Deen Wispa
Justified Chaos
515
|
Posted - 2013.05.02 22:21:00 -
[23] - Quote
chatgris wrote:Also, I thought that a large percentage of eve players were Computer Science guys, and this is first year stats.
I flunked first year stats harder than a Squid drake fleets trying to learn how to nano. High Five. Yeah! C'est La Eve . |

IbanezLaney
The Church of Awesome Caldari State Capturing
352
|
Posted - 2013.05.02 22:28:00 -
[24] - Quote
X Gallentius wrote:I thought they said ecm is addictive. Yes it certainly is. Once a player learns of its true power, he never flies without an ecm alt again.
         Fix this **** See Sea Pea. |

Onomerous
Shockwave Innovations Shockwave Sovereign Industries
72
|
Posted - 2013.05.03 12:25:00 -
[25] - Quote
Tsukino Stareine wrote:[quote=chatgris]snip...
and ECM is boring, what's the fun in shootin something that can't shoot back?
The same could be said for most of the PVP in EVE: to win. Even one SD applied to the ECM boat can often works wonders. |

Andreus Ixiris
Mixed Metaphor
2508
|
Posted - 2013.05.03 12:47:00 -
[26] - Quote
ECM is not additive. As far as I understand it, it works like this:
Each point of jam strength on a single ECM jammer increases the chance of a successful jam and each point of sensor strength (native, skill, ECCM, etc.) on the target decreases it. Each seperate ECM jammer is, in effect, an entirely seperate "re-roll." The effects of one ECM jammer on one ship do not have any effect on the success of a jammer on another ship - in fact, they don't even have any effect on any other ECM jammers on the same ship. More jammers mean more re-rolls, thus increasing the number of chances you get to jam someone during a given 20 second period. Given that unlike sensor dampening, target painting and tracking disruption, the effect of ECM is a binary state rather than a sliding scale, the optimal method of ECMing multiple targets would be to carefully and methodically cycle through your jammers until you get a successful jam and then move on to the next target, since after the first successful jam cycle drops, all subsequent jam cycles on the same target are completely wasted until the first cycle ends.
Because of this system there is always a variable non-zero probability (usually small) that all jam cycles on a given ship will fail. Ironically, from a theoretical perspective at least, this makes ECM the least reliable form of e-war. Mane 614
|

Gizznitt Malikite
Agony Unleashed Agony Empire
2094
|
Posted - 2013.05.03 17:49:00 -
[27] - Quote
Andreus Ixiris wrote:ECM is not additive. As far as I understand it, it works like this:
Each point of jam strength on a single ECM jammer increases the chance of a successful jam and each point of sensor strength (native, skill, ECCM, etc.) on the target decreases it. Each seperate ECM jammer is, in effect, an entirely seperate "re-roll." The effects of one ECM jammer on one ship do not have any effect on the success of a jammer on another ship - in fact, they don't even have any effect on any other ECM jammers on the same ship. More jammers mean more re-rolls, thus increasing the number of chances you get to jam someone during a given 20 second period. Given that unlike sensor dampening, target painting and tracking disruption, the effect of ECM is a binary state rather than a sliding scale, the optimal method of ECMing multiple targets would be to carefully and methodically cycle through your jammers until you get a successful jam and then move on to the next target, since after the first successful jam cycle drops, all subsequent jam cycles on the same target are completely wasted until the first cycle ends.
Because of this system there is always a variable non-zero probability (usually small) that all jam cycles on a given ship will fail. Ironically, from a theoretical perspective at least, this makes ECM the least reliable form of e-war.
Yes, each jammer has a binary style of success (it either works, or fails). FYI: Tracks, Damps, and Paints, when used on a target beyond optimal range, also first experiences a binary style of success (it either works or fails).
No, there is NOT ALWAYS a "viable non-zero probability" of jams missing. P(Jam) = Jammer's Sensor Strength / Target Sensor Strength. If the sensor strength of the Jammer is a bigger number than the sensor strength of your target, you will ALWAYS have a successful jam (when jamming within optimal range).
In general, The jammer strength is 8-12 for BBs, Griffins, and Scorps, 10-13 for Kitsune's, and 12-16 for Rook & Falcon.
The base sensor strength of most combat/attack frigates is 8-10, Cruisers 14-16, BCs 18-22, BSs 21-25. Disrutpion Ships and Faction Ships generally have significantly higher sensor strengths.
FYI: Most people don't know this about ewar: When using EWAR on a target beyond optimal range, there is first a "does this EWAR work on the target check". The chance of success is 50% at Opt + Falloff, and 0% at Opt + 2x Falloff.
|

Secret Squirrell
Allied Press Intergalactic
8
|
Posted - 2013.05.03 17:58:00 -
[28] - Quote
Andreus Ixiris wrote:Ironically, from a theoretical perspective at least, this makes ECM the least reliable form of e-war.
Good thing we care about how effective it is, not how reliable =P |

Andreus Ixiris
Mixed Metaphor
2569
|
Posted - 2013.05.05 14:07:00 -
[29] - Quote
Secret Squirrell wrote:Good thing we care about how effective it is, not how reliable =P Yeah, pretty much. Mane 614
|

Muad 'dib
The Imperial Fedaykin Amarrian Commandos
932
|
Posted - 2013.05.08 11:15:00 -
[30] - Quote
I think ccp dont really care because its balanced on a large scale fight where its hard to spread out the jams as efficiently.
In a 5v5 with 2 ecm ships they just put one on everything or having preferred targets called in gang.
While thats not a great way to look at it, one thats even worse is ECM drones, who in a pack of 5 are FAR more effective than 5 any-other-ew because those are subject to stacking - there is no such stacking reduction in the modules effective ecm strength in regards to the points of ECM already applied. Cosmic signature detected. . . . http://img299.imageshack.us/img299/4375/mynewsig2.jpg CCP Hilmar CEO > "why am i sweating, why is the game doing this to me"
|
| |
|
| Pages: [1] 2 :: one page |
| First page | Previous page | Next page | Last page |