| Author |
Thread Statistics | Show CCP posts - 0 post(s) |
Dust Puppy
   |
Posted - 2005.02.14 22:32:00 -
[1]
Well the problem with having a chance based system in propulsion jamming is that warp scrambled ship will only have to warp once to escape. That means the longer the ship lives the better are it's chances of escaping which all makes sense in my books. The problem is really just tuning the numbers. Let's take a ship that can survive for 2 minutes. If the cycle is 12 seconds and the jammings strength is 95% it has a 0.95^10 = 0.5987 chance of being jammed the whole time of about 40% chance of escaping.
This of course raises the question of whether changing the bonus from duration to cap usage is really a good thing. If the duration was higher making the cycle 15 seconds the victim would only have a 1-0.95^8 = 0.3366 or 33.7% chance of escaping.
This also raises the question about the viability of ransoming someone. Obviously the longer you hold someone the more chance he has of escaping so people in that business would have to have one of those "100% chance of jamming" scramblers.
Of course as long as some scramblers have 100% chance of jamming I see no problems with this change.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.14 22:33:00 -
[2]
 Originally by: Dionysus Davinci
Originally by: Ithildin
1. Will AoE scrambling be a general scrambling - can I use it in empire wars without CONCORDOKKEN?
Don't count on it since you can't use a smartbomb. I doubt Concord will ever smile on collartral damage because you and your pal are having a little war.
I think ECM burst isn't classified as a hostile act but I think that's more of an oversight instead of a feature.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.14 22:55:00 -
[3]
On an almost unrelated note, is the pretty curves you plotted (not the hearts  ) based on the formula
=> 1/2(1-tanh(tau*x))
which I have plotted here. Come on I know you wanna tell me 
ps. Don't like the racial jamming things have to agree with Imhotep there.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.14 23:00:00 -
[4]
Originally by: Ithildin
Under the CURRENT system, CONCORD will slay you if you ECM, Dampen, and/or propulsion jam someone.
Well that makes sense and is the way it should be but still there are people that disagree on that you can see one thread here and I'm pretty sure that I've seen another one before that one. Bah last time I go offtopic I promise 
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 01:34:00 -
[5]
Originally by: Lallante
Edited by: Lallante on 15/02/2005 01:29:29
Edited by: Lallante on 15/02/2005 01:27:33
I crunched the numbers, and this is NOT going to make warp scrambling or jamming HARDER. Its going to make it much MUCH much MUCH easier in some circumstances, and a little harder in others (although itll be weighted towards the former).
Scrambling
Remember that damn sniper with 5 stabs? well a normal 20km disruptor will fair chance of holding him down. At the same time, there will be a small chance he can now escape from 10 strength scrambling.
TomB clearly says at the end of his post that the weight will be heavily in the scramblers favor. What this means is, there will be MANY more situations in which less scrambling does the job, and only a few when more scrambling fails to. You will get MORE TARGETS PINNED DOWN THIS WAY, NOT LESS! You will also have a great deal of extra range, which will be reduced in effectiveness but will work neverthe less: That frig that keeps buzzing past and going to 80km and cming back: You can keep your disruptor activated on him THE WHOLE TIME (even if most of it is outside falloff, ie 0% chance)
Jamming
My immediate thoughts were: multispecs are now useless. OMG NOT SO
HEres the graph (click thumbnail)
CURVED LINES:
These show the increasing chance of getting ONE Jam off (obviously the first plotted point on this line is the base chance at any given second). As you can see after 30seconds, you are more or less guaranteed at least one 5 second jamming session, even if you only have ONE MULTISPEC!
MODULATING LINES:
The essential premise here is that if the enemy takes 10 seconds to lock, then you have 2 "ticks" each 10 second cycle (cos multispecs take 5 seconds) to break that lock, hence the overall chance of jamming them increases for 2 ticks. After 10 seconds (2 ticks) they can relock anyway, so the chance resets, and so on. I promise you the logic is sound, even if it seems overly complicated.
-- uber cool figure --
As you can see, even with only 1 multispec on a BS, you will be able to keep him jammed on average 1 in 4 cycles, and so unlocked between 25% and 50% of the time (depending on lockspeed, with a lockspeed of 10seconds it would be about 50% of the time). To be honest, I think they will nerf this (lower strengths) because its TOO DAMN GOOD
Essentially by tweaking the numbers, The Devs can make it as easy or as hard as they like to warp scramble someone, If its found that people are getting away too often, they CAN JUST UP THE STRENGH OF SCRAMBLERS!
Good points there and I hope at least someone takes them into consideration. Of course you realize that arguing with some people on this board is like talking to a politician, they will never ever admit that they were wrong about something (I'm not one of them I always think I'm right  ).
I think the theory will work all that is left is just to tweak the numbers.
ps. still don't like racial warp jamming thingie.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 15:50:00 -
[6]
Originally by: Arthur Guinness
Matthew, could you please calculate for me how likely i am to get 2 commander spawns within an hour of npcing at a belt in stain? How likely i am to get 3 tech2 bpos?
You might be good at math, but reality does not always go along with propability math, it's called hard LUCK.
ROFL 
nice math job btw Matthew
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 16:18:00 -
[7]
Originally by: Helplessandlost
I'm confused on one thing...
With these new changes if I have stabs in low slots I can still be locked down with just one scrambler even if I have two WCS loaded? I'm just a little confused on this one
Yeah if you have a stabilizer in the low slot then you have strength of 2. If someone were to attempt to jam you with strength 1 he would get 100*1/2 = 50% chance of succeding. Don't be alarmed though you only need to warp once so he has only 25% chance of jamming you 2 times in a row, 12.5% to jam you 3 times and so on.
All those numbers will probably be tweaked later on so don't take those numbers as a fact. The main problem I see with this that by fitting two stabilizers you always have a chance of getting away as the strongest warp scramblers only has base 2. so even if you have 3 warp scramblers the ship still has the chance that all the scramblers will fail or 1/3^3 = 0.037 chance of warping away. If it survives long enough it's chances will improve.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 17:20:00 -
[8]
Originally by: ChernoChiel
Edited by: ChernoChiel on 15/02/2005 16:57:17
well, like everybody else, im not to happy about it...
but enuf with the whining, heres the usefull part of my post :)
a fallof would really scuk for the small and quick tacklers (crow, vigil, griffin, whatever).
They orbit at close range, and try to not get hit this way.
when the optimal is 30Km, and falloff 20KM, their outside the falloff and would always miss. (?)
so, i would suggest making the EW modules ALWAYS hit at optimal and below (so closer to the target)
always might be a bit too much, but maybe at the same rate as at optimal.
so, to put it shortly:
change the optimal for EW to matter only above optimal!
(will any Dev read this post at page 12?! hell, i tried :/ )
That is how it works. FYI it is also how it works on turrets only it is harder to track people that fly close to you, That's why I hate bees they are the interceptor of nature 
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 17:59:00 -
[9]
Originally by: Icarus100
Originally by: DigitalCommunist
Target jamming for 100% inside optimal range with a single module? How does that not make the scorpion uber?
Because if this gets through all the other ships can do it too?
Where do you get target jamming for 100%. I think I've read every single post here and not once have I seen that. I think you can get 100% jamming if a ship has a sensor strength of 6 or lower then you can with a single racial jammer. That is then you get 6/6 = 1 if it has more than 6 for example 8 then you only have 6/8 = 0.75 chance of jamming with a single module. If you use 2 racial then you get 1-0.25^2 = 0.9375 but no matter how many you will stack you will never ever get 100% chance of jamming. Although you do get so close that it hardly matters at all.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 18:05:00 -
[10]
How about the booster jammers similar to the shield boost amplifier . So instead of fitting 2 modules that work independantly you can fit a module to work with a certain jammer by increasing it's strength. So you could have a racial jammer that jams with the strength 6 and 2 of those amplifiers that give you 4 each you now have a single jammer that jams with the strength 14. Which means that inside it's optimal it would always jam a ship that has lower strength than 14. Just an idea.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.15 18:25:00 -
[11]
Originally by: j0sephine
This would make EW fall back to where it's now -- stacking modules for 100% jamming chance...
(and in the given example, using these 3 slots for regular jammers with strength of 6 each, against 14 point strong target still gives over 80% chance of jamming it on each cycle, even when they work independantly... hardly bad odds -.o
Well the way I thought it that if you couldn't get the critical jamming strength it would be better just to put more jammers on but if you had the slots for it you could stack amplifier to make one giant jammer that can jam most ship for 100% chance.
Thought about this originally only for scramblers though as the numbers are a bit low there. If you got a scrambler with scrambling strength 2 then you can scramble a ship with warp core strength 2 in 100% of the cases but if the ship has an extra warp stab you now have only 66.7% chance of scrambling and even if you have 2 of those you still have only 1-1/3^2 = 0.888 or 88.9% chance of scrambling which isn't very much when you factor in that the ship only has to warp out once to escape.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.16 13:56:00 -
[12]
Actually webifier could work like turrets. Meaning you have a bad webifying cycle and only slow the ship down by 20%.
__________
Capacitor research |
Dust Puppy
|
Posted - 2005.02.16 22:19:00 -
[13]
Originally by: Leam
Wait to see it until you say it will be useless uh? ppl like you said the same when ccp announced that there would be rats at gates... "oh no, we wont be able to travel through 0.0! pvp will die!",
I dont know what you call "useless". lets take tomb first post, where it says that a jammer would have 20% chance of jamming a apoc. Now let say that you're using 5 jammers. so you have 5 jammers with 20% of chance of jamming each. Im to lazy to make the calcs and give you the exact number, but i dotn think that the chances are too low. 20% the first one, if it fails 20% the next one, and so on
Actually that is not that much. For it to fail it's jamming all five of the jammers must fail so that is 0.8^5 = 0.32768 or 32.8% chances that the jamming will fail. Those numbers in TomB's post will of course be tuned in the next week so this doesn't really mean that much.
__________
Capacitor research |
| |
|