Suspicious Scores and Inappropriate Names

Started by knucracker, September 02, 2014, 10:14:05 AM

Previous topic - Next topic

Karsten75

#45
It's actually not a bad idea. It can be selectively applied to the controlled sectors of the game. I'll mention it to Virgil. :)

Edit: We looked, and while it's possible, the number of scores that would be wiped this way is actually very small. And it closes some loophole we use to do detect some things...

iamltr27

#46
Quote from: stoey on June 12, 2015, 07:59:21 AM



what about just...... what ever the time is in seconds for a Nullifier to be built and zap...... Then all scores deleted under this number

kindest regards
stoey  8)

The impossible scores can be useful for spotting patterns.  If player X has impossible scores on easy Prospector sites, then his incredible (but maybe possible?) scores on more challenging maps can be safely discarded -- if you filter out just the impossible scores then you lose the context that establishes X as a hack.

I prefer the approach Tex suggested back in December of scrubbing ALL entries for established hacks.  But even that has issues ... if it is as easy to spoof a name as it is to spoof a score then the hacks would likely respond by hiding behind the names of legit players.

stoey

#47
Quote from: Asbestos on June 12, 2015, 11:42:58 AM


There's still custom maps with shorter than normal nullifier build times, maps that don't even need a nullifier to be won, maps like Brain, which use CRPL, maps that use the old totem system... if you just deleted all the scores with a time under the bare limit of the time it takes to build and fire a nullifier, you would probably cut off a lot of real scores too.

ok asbe, i hear ya.    there are a some special maps.  Maybe it can be used as an indicator for further investigation not formula to delete scores... so....

RESULTS OF TIME TO BUILD NULLIFIER EXPERIMENT

27 Seconds to build nullifier
41 Seconds to build and fire nullifier

Only 4 seconds to build a nullifier on build speed 100. (if the map started on that full upgrade)



There i done my bit for the cause




(as with everything in science this experiment needs to be confirmed by other experiments to establish the consensus view on nullifier build times. chow)

stoey

Quote from: Karsten75 on June 12, 2015, 01:08:54 PM
We looked, and while it's possible, the number of scores that would be wiped this way is actually very small. And it closes some loophole we use to do detect some things...

i didnt see you wrote that, kars. Ok no probs. thanks for looking into my idea

iamltr27

Suspicious scores:

S200_36 : Bremoilo
Ed C**ter          1:10.5

S200_116 : Igonar
Ed C**ter          2:52.1

iamltr27

Suspicious scores:

S200_187 : Abertoft
von            6:00.2
tj               6:43.2
jaxx           7:00.5


S200_54 : Kinloneport
Dysoch              0:30.0
Dot                    0:52.5
Amazing Asian   1:01.0

iamltr27

Suspicious scores:

S200_144 : Langrusate
Torg          2:32.0

S200_147 : Kinjawfirth
Torg          0:43.2
Creepa'      0:46.6

S200_137 : New Treketh
Torg          0:40.4

Xalos

The tier-0 building speed is one packet per sixteen frames, or 1.875 E/s.  The charging speed is one packet per nine frames, or 3.333 E/s.  A Nullifier takes 50 energy to build, 50 energy to charge, and 2 seconds to fully discharge.  A Command Node takes 3 seconds to land.  Conclusion: the minimum possible time for Inceptus: Tempus is 3 + (50 / 1.875) + (50 / 3.333) + 2 seconds, or 46.667s total.

There are four times on Inceptus: Tempus which are below this minimum possible time.  I don't know if they were possible at some point in the past, but they're very definitely not possible now.

Grayzzur

CW3 Colonial Space Map 2402 (The Sector TIOCreeper-Level 1) --
A player posted a score with an inappropriate name. Currently #49, with a time of 52:19.6.
"Fate. It protects fools, little children, and ships named 'Enterprise.'" -William T. Riker

Sorrontis

ColonialSpace map 2482 "Creeper's Revenge 7"

Ady     time 10:30.5

due to CRPL, destroying all 9 ramping emitters (boost 2.5minutes each), minimum time to beat is 22.5 minutes + time to collect tech artifacts.
"If you want others to be happy, practice compassion. If you want to be happy, practice compassion."

applemachome

CS - #2560 "Abort"
radd - 58.5


There is no configuration to get less than 1:02.7 on this...I've tried 50 times.

warren


Asbestos

Quote from: applemachome on September 24, 2015, 10:52:23 PM
CS - #2560 "Abort"
radd - 58.5


There is no configuration to get less than 1:02.7 on this...I've tried 50 times.
I'm fairly sure that radd is a legitimate player. See the thread about skill vs. time to see how you can shave off a few split seconds.

halfbit

I have seen some; many - more or lees the same names with the same setting. -> an inhibitor ~0:46 where ever 'possible' or not w/o cheating

so how about a NSA approach:  select all maps and 1st, 2nd and 3d and there  names ... make a cluster and bingo

I am a "bad" player, I am interested in playing and getting my brain free for the next hours of work; I in principle dont mind cheaters, but it starts to get annoying  if there are always the same names solving it in ~46'' ...

I am not a fast player; so i don't mind if I am not one of the top ten (or 100) - just cheating is not fair ...


warren

#59
Note: 45.7 is a bit of a magical number. Unless a map uses CRPL or custom unit costs, (or a few other rare things) this is the best time possible. Scores of ~46 seconds are only suspicious if there are: More than one emitter greatly spaced and no inhibitor; no syphon and only one Command node may be built; the creeper is too deep and spreads too quickly.