Do you have: the Markus notice testing

... A place to ask for help on any topic whether it be starting out to player killing to IT issues.
Paigey
Posts: 150
Joined: Wed Dec 16, 2015 9:37 pm

Do you have: the Markus notice testing

Postby Paigey » Sat Dec 24, 2016 10:21 am

Hi There,

Does anyone have a copy of the notice testing that Markus performed back in the day.

Please share if you do. Thanks!

Maegon
Posts: 276
Joined: Wed Mar 25, 2015 3:15 am

Re: Do you have: the Markus notice testing

Postby Maegon » Sat Dec 24, 2016 11:33 am

The dropbox link no longer works, but this was posted by Markus.

https://sites.google.com/site/alaylades ... et-stabbed

Author: Markus [ November 11th, 2011, 9:17 am ]
Post subject: Re: Guide: How to not get stabbed
Quote:
- >85% notice

Where did you get that number? 85% is provably unreliable even with h.target.

http://dl.dropbox.com/u/20846822/Notice ... 100220.xls

Unless something changed since I did the testing, the threshold of reliable h.target is 89%. At least that's what a 400-size sample says. (I've heard anecdotal evidence that this can still fail, though I've never in my life seen that happen so I suspect user error such as notice turned off or wrong target).

However, because of the way practice costs work, the sane amount to get - if you don't want or can't afford 99% - is 91% for every setup except the ones ridiculously short on pracs. So that's what I tell people to get, 91%.

In general, there main ways to keep alive against a stabber are:
- being engaged on something
- moving: if you dont stay in any single room for 4 seconds, you can't be stabbed. Basically every time you enter a room, assume a stab timer just started on you. This is the true meaning of healthy paranoia. ;)
- spamming
- staying in no hide rooms (not 100% reliable, but safer than the pure math indicates because stabbies don't usually bother stabbing in them).

Elyse
Posts: 212
Joined: Wed Feb 18, 2015 1:47 pm

Re: Do you have: the Markus notice testing

Postby Elyse » Sat Dec 24, 2016 12:06 pm

Here's my copy of his spreadsheet.
From the file name, looks like the testing was done in February 2010.


"Notice
%" "Diagnose
target" "max fails
in a row"
85 2.00% lots
88 0.50% lots
89 2.00% lots
90 3.50% lots
91 5.50% lots
92 9.00% 56
93 12.50% 18
94 20.00% 28
95 26.00% 15
96 25.00% 14
97 32.00% 12
98 42.00% 9
99 47.50% 5

"Notice
%" "Diagnose
h.target" Sample Size
85 98% 100
88 99% 200
89 100% 400
90 100% 400
91 100% 400

Paigey
Posts: 150
Joined: Wed Dec 16, 2015 9:37 pm

Re: Do you have: the Markus notice testing

Postby Paigey » Sat Dec 24, 2016 12:59 pm

Thanks for sharing all. Very much obliged.

chavaon
Posts: 56
Joined: Sat Mar 21, 2015 9:08 pm

Re: Do you have: the Markus notice testing

Postby chavaon » Tue Dec 27, 2016 7:57 pm

Notice % - k vs 94 hide - Longest failure series
78 -
84 - 1
86 - 4
88 - 14
90 - 25 - 8
92 - 22 - 11
93 - 34 - 7
94 - 50 - 5
95 - 59 - 5
96 - 69 - 3
97 - 76 - 3
98 - 75 - 3
99 - 81 - 4


Notice % - k vs 99 hide - Longest failure series
99 - 44 - 5

Notice % - k hidden vs 94 hide
66
72 - 24
78 - 80
84 - 100

Notice % - k hidden vs 99 hide
72 - 1
77 - 40
82 - 83
84 - 97
86 - 100


Hmm I seem to have a slightly different one than Elyse from April 2009.


Astolfo
Posts: 325
Joined: Tue Apr 14, 2015 4:16 pm
Location: Lost in Space!

Re: Do you have: the Markus notice testing

Postby Astolfo » Tue Jan 03, 2017 9:07 pm

Anyone still in contact with him btw?

Taziar
Posts: 390
Joined: Sat Mar 21, 2015 10:28 pm
Location: Seattle

Re: Do you have: the Markus notice testing

Postby Taziar » Fri Jan 06, 2017 1:44 am

Speak of the devil...

Astolfo wrote:Anyone still in contact with him btw?


He just posted on wotmudarchive.org forums: http://www.wotmudarchives.org/forum/vie ... 2421#p2666

Taziar


Return to “Help and Newbie Area”

Who is online

Users browsing this forum: No registered users and 1 guest