Jump to content

geordan

Members
 View Profile  See their activity

  • Content Count

    341

  • Joined

  • Last visited

Posts posted by geordan

  2. Posted

    Edit: the TIE Swarm player's MoV was just crazy.

     

    Round 1: bye

    Round 2: 100-27

    Round 3: 100-0

    Round 4: 100-0

    Round 5: 100-15

    Round 6: 100-51

    Top 8: 100-12

    Top 4: 0-100 (only loss)

     

    His only loss was to the dual VT-49 list without Engine Upgrade. Of all his matchups during the day you would have thought he would have had the easiest time with that one!

    Gus is pretty good at this game.

    Rydiak, Blail Blerg and MajorJuggler reacted to this

  3. Posted

    Soooo many turrets.

     

    No kidding.  I faced three DeciFels (of course, I was part of the problem, flying Chirafel) and did miserably.  Karma, as Jeff Wilder put it.  DarthDave faced four (!) with his TIE swarm.

     

    We like us some pancakes here in California...

  8. Posted

    im guessing he used Craken + 4x Tala Squadron all with concussion missiles? maybe a few assault missiles in the mix as well? am i close??

     

    It's this:

     

    https://geordanr.github.io/xwing/?f=Rebel%20Alliance&d=v3!s!67:10,-1:-1:-1:;64:13:-1:-1:;64:13:-1:-1:;64:13:-1:-1:;64:13:-1:-1:;64:53:-1:-1:

     

    Oddly, I completely tabled it in Swiss with Chirafel and took 3 total damage (+1 for voluntarily flying through a rock) on Chiraneau; my wife beat it in Swiss (on her way to her 5-0) but got matched up against it in elimination and lost.

     

    It's legitimately terrifying to face -- make a mistake and you're eating Concussions all day.

     

    It's so cool so see an ordnance swarm doing well!

    KommanderKeldoth, Sekac and The_Brown_Bomber reacted to this

  13. Posted · Edited by geordan

    It happened in a live streamed game. I believe in MN? Xizor went off the edge by a hair. TO let it slide.

    I saw that on stream, too. It sounded like the TO let it slide because the Xizor player was intending to push the limit afterward and boost, and ruled that since he couldn't boost, the entire action was undone.

    (I don't agree with that, by the way. You don't commit to pushing the limit before you perform your first action.)

     

    I didn't see the stream, could you elaborate? Did the Xizor player barrel roll off the board and the TO determined that since he couldn't boost back on, undo the entire action?

    Yeah, that's what I recall.

    Edit: Actually, maybe that wasn't it. (The vagaries of human memory!) Thinking about it more, I believe Xizor focused, then Pushed the Limit to boost, which put him off the board. The TO ruled that the boost failed, and therefore could be unwound, which is contrary to the standing ruling that you can in fact boost or barrel roll off the board. Either that, or it was ruled that it could be undone because it was part of a Push the Limit. Table audio was really quiet, so I couldn't really tell what happened.

    ViscerothSWG reacted to this

  14. Posted

    I bet there's a story here.

     

    It happened in a live streamed game. I believe in MN? Xizor went off the edge by a hair. TO let it slide.

    I saw that on stream, too. It sounded like the TO let it slide because the Xizor player was intending to push the limit afterward and boost, and ruled that since he couldn't boost, the entire action was undone.

    (I don't agree with that, by the way. You don't commit to pushing the limit before you perform your first action.)

  15. Posted

    I wouldn't necessarily call a bug "functionality."

     

    I intend to fix faction correctness on the basis of code integrity; the code was written with the assumption that validations were being run, and arbitrarily loosening those restrictions would introduce potential bugs.

  18. Posted

    Ah, I see.  I went the procedural route because I'm not so great at the probabilities.  (Also, implementing decision points like Gunner and conditional defensive token spending is easier, for me, when done procedurally.)

     

    ...I'll stop derailing this thread and we can go back to talking about how awesome double IG is.

  19. Posted

    Yikes that deviation is what's killing me. Thanks. Now kindly add in stealth device and Palpatine :P

    Sorry, Fel also has Stealth Device in that run.

    Palpy... yipes. We'll see :)

     

    Brute force calculation doesn't take that much compute time, and gives the exact answer. If I have a spare 20 minutes to an hour this weekend then I'll see about updating my code. I already have hooks for Stealth Device, Autothrusters, any number of focus and evade tokens, and getting attacked by multiple pre-defined attacks. I actually wrote the multi-attack code back after Worlds 2013 to see what the odds were of Dallas killing Paul's Biggs. The only thing that I don't have in my scripts already is the 2nd "gunner" attack for a particular attack if the first misses.

    Brute force is the best force! The sim runs 10,000 iterations in under a second; not being a stats person I don't know how many to run to get good signal. What do you recommend?

    Yeah, getting Gunner functionality in there was the wacky bit, but I believe it's working. I do want to see the output of your scripts, as a sanity check on my results.

    (The eventual goal is to put this up as a web service to receive sim parameters and spit out data for analysis.)

  20. Posted · Edited by geordan

     

    Does anyone know the math on two IG's with fcs and HLC's with/without focus shooting at a tokened-up Fel? Based on my extensive testing(in arcs twice ever) the odds of getting through are 0%.

     

    Not yet, been meaning to modify my scripts so I can run this. Depends if you are at range 3 or not and if Fel has Stealth. The chances are much greater than 0% with HLC FCS and 88B's ability.

     

     

    Sozin and I have been working on something to simulate this.  It's a Monte Carlo stand-and-shoot-each-other simulation.

     

    For this run, Soontir is fully tanked up with Autothrusters active, not spending his focus tokens on attack.  IGs start with a Target Lock, and will use their focus token on defense if necessary.  Soontir shoots at IG-88B first, then the bots return fire.

     

    This doesn't yet model the loss of Stealth Device on a hit.

     

    I hope the math/code is right.

     

    Forgive the output, it's still very much a work in progress.

    Soontir Fel
-----------
Hits Landed: 0.174 (stddev=0.461)
 0 ( 85.77%): ##########################################
 1 ( 11.30%): #####
 2 (  2.68%): #
 3 (  0.25%): 

Crits Landed: 0.122 (stddev=0.350)
 0 ( 88.64%): ############################################
 1 ( 10.56%): #####
 2 (  0.80%): 

Hull Remaining: 1.737 (stddev=0.986)
 0 ( 13.27%): ######
 1 ( 25.40%): ############
 2 ( 35.65%): #################
 3 ( 25.68%): ############

Shields Remaining: 0.000 (stddev=0.000)
 0 (100.00%): ##################################################

IG-88B
------
Hits Landed: 0.279 (stddev=0.550)
 0 ( 76.86%): ######################################
 1 ( 18.64%): #########
 2 (  4.22%): ##
 3 (  0.28%): 

Crits Landed: 0.060 (stddev=0.244)
 0 ( 94.15%): ###############################################
 1 (  5.69%): ##
 2 (  0.16%): 

Hull Remaining: 4.000 (stddev=0.000)
 4 (100.00%): ##################################################

Shields Remaining: 3.704 (stddev=0.593)
 1 (  0.62%): 
 2 (  5.32%): ##
 3 ( 17.07%): ########
 4 ( 76.99%): ######################################

IG-88C
------
Hits Landed: 0.846 (stddev=0.859)
 0 ( 41.30%): ####################
 1 ( 37.14%): ##################
 2 ( 17.30%): ########
 3 (  4.15%): ##
 4 (  0.11%): 

Crits Landed: 0.105 (stddev=0.317)
 0 ( 89.81%): ############################################
 1 (  9.88%): ####
 2 (  0.31%): 

Hull Remaining: 4.000 (stddev=0.000)
 4 (100.00%): ##################################################

Shields Remaining: 4.000 (stddev=0.000)
 4 (100.00%): ##################################################

    Edit: forgot to set IG attacks to HLC

×
×
  • Create New...