Help - Search - Members - Calendar
Full Version: Open Tests
Dumpshock Forums > Discussion > Shadowrun
Pages: 1, 2, 3
Sphynx
After yet another annoying (2 games in a row) detection by an Int:2 Security Guard against the Stealth:12 adept, I've decided I -must- House Rule open tests.

Although I'm looking for suggestions, I also have one.

OpenTests: TN = 4. Most successes wins. If number of successes is a tie, highest roll wins.

Has any one else found the need to House Rule that rule?

Sphynx
ShadowGhost
It's called luck. I leave open rolls as they are. I've seen 8 dice rolled come up with a result no higher than four, and 2 dice roll a 25.

I've seen a Rating 8 Forged Credstick clobbered by a rating 4 credstick reader...8 Dice against Tn 4 losing to 4 Dice against TN 8.

I'd just look at it as the Stealth 12 adept hit the one spot on the floor that squeeks, a twig under leaves that snaps, or suddenly being illuminated by the headlights of a passing car while trying to hide in the shadows.

A skill of 12 means you're really, really friggen good, but it doesn't mean you're invisible or silent, or automatically succeed. Statistically, you'll clobber two dice with your 12. But on those rare occasions, you will lose.


Smiley
I feel your pain. I was stealthing my ass off last night with a 26 and STILL got spotted by some pissant Cascade Ork. Sphinx, it does seem stupid that someone with stealth of 12 could be so easily disocvered. Maybe adding your stealth skill number to your roll?
Jason Farlander
The problem with eliminating Open tests is that it makes a character using stealth almost impossible to detect if you continue using the visibility modifiers as they are written. A character with stealth 6 moving in partial lighting conditions will almost never be noticed buy an INT 6 guard with Stealth(Alertness) 4(6), insofar as the character's TN will be 4 while the guard's TN will be 6. You can forget the remote possibility of ever seeing someone in a ruthenium suit if their stealth is 4 or better.

Reworking visibility modifiers presents a challenge insofar as those modifiers carry over into combat situations, so you also have to make sure that you dont screw up the game in that regard.

I certainly wouldn't mind eliminating open tests from the game, but in this particular case I find it far easier to stick with canon, even if it does cause some minor annoyances.
Sphynx
QUOTE (ShadowGhost)
But on those rare occasions, you will lose.

The problem is that it's not a 'rare occasion'. As anyone who's played enough can tell you, the 12+ seems to come up more often when you roll under 4 dice than when you roll over 6 dice for some strange reason. Far too often (nearly every other game) stealth rolls flub. And yes, a stealth:12 should be nearly invisible. The chances of being detected per run should be closer to 1 in 1000, not 1 in 3.

Sphynx
Moonstone Spider
I recall houseruling them as a variant of opposed test, and that stealth worked like ED for characters. TN = 4, each success means your opponent needs one more success to spot you on a perception check (His TN being dependent on the environment and your actions as normal).

We had a disguise skill that worked the same way for makeup and acting your way. The more successes on your disguise test, the more successes an observer would need on a perception test to tell you weren't actually a repairman or CEO or whatever you were pretending to be.
Lilt
The rules make it hard to sneak reliably unless you have camofluage or similar perception TN mods. The perception 2 guard has a 12% chance of seeing the high stealth adept, but once you start adding modifiers (like a camo suit for +4) then it drops to 5%. Add the partially hidden perception modifier for another +2 and another +2 as the guard is distracted (dosen't care & thinking of dinner) the guard's chance drops to around 1.64%...

I personally rule that a single success means that the guard will look again, without the distracted modifier and maybe a bonus, but will think it's just a cat (or something) unless he scores a success again.
Smiley
QUOTE (Sphynx)

The problem is that it's not a 'rare occasion'. As anyone who's played enough can tell you, the 12+ seems to come up more often when you roll under 4 dice than when you roll over 6 dice for some strange reason.

YES! IT'S NOT JUST ME!
Herald of Verjigorm
If you get the high numbers by rolling 4 dice, just roll your stealth of 12 in three groups. If the low numbers are a disturbing fluke of probability and your rolling style, this should get around that.
Smiley
After last night's run, i think i just have dice gremlins.
Nikoli
I'm assuming the 12 dice tossing character is a Phys ad. If so the player should invest in traceless walk. severely limits what can give them away. You make no noise, nothing you are touching makes noise (guns still do since you don't touch the gass escaping nor would you want to)
No more pesky pressure sensors, no more pesky squeky floor boards, etc.
TinkerGnome
A stealth adept should grab Traceless walk and a camo suit right off the bat. Then upgrade to a stealth suit/cloak ASAP. Sure, 12 dice vs. 4 dice is fine... but add a 10 to the high of 12 and it becomes a lot harder.

Which has already been said, so I'll paraphrase as "me too!"
Jason Farlander
I found myself in a bruteforcing sort of mood (I also like rolling dice) so I just rolled a series of 100 opposed (not open) stealth/perception tests. I assumed a stealth of 6, an intelligence of 9 for the guard, and that the only modifier was a +2 perception TN mod (for whatever reason).

If ties go to the perciever, the guard succeeded 28% of the time. If ties go to the character using stealth, the guard only succeeded 14% of the time. Thats only with a TN difference of 2... imagine how bad it would get with perception mods of +4 or worse.
Lilt
Well as the result of the stealth test is the TN for the perception test, ties go to the perceiver.
Jason Farlander
QUOTE (Lilt @ Apr 13 2004, 01:48 PM)
Well as the result of the stealth test is the TN for the perception test, ties go to the perceiver.

Lilt: this was a test for a move away from an Open Test version of Stealth to an Opposed Test version. Numbers of successes assuming a base TN 4 from each test were compared - NOT the highest number in each test.
gknoy
I'd like to note that we are discussingthis very issue in another thread ("Speaking of calculating odds ...").

http://invision.dumpshock.com/index.php?sh...topic=3656&st=0

My graph ( http://www.anasazisystems.com/~gknoy/sotsw...enTestGraph.png ) shows the AVERAGE value in an open roll from any number of dice; interesting to note that 12 dice can expect to roll about 11 as a high number (on average). Note that I screwed up the axis labels - dice on X axis, roll results on Y.

Your perception 2 guard has an extermely low chance of beating an adept, ON AVERAGE. I'd say that if your people are regularly doingit, it's a factor of imperfectly rolled dice. smile.gif (in a batch of 12, maybe they don't have enough freedom of movement?)
Sphynx
50% of the time a person rolling 12 dice will NOT get a 12+.
10% of the time a person with a perception of 4 will.
Better than a 5% chance that 12 stealth is noticed by perception 4.

Though, experience tells me it's closer to 20% for some reason. nyahnyah.gif

Sphynx
Jason Farlander
QUOTE (Sphynx)
Better than a 5% chance that 12 stealth is noticed by perception 4.

And in broad daylight with no cover and an undistracted guard, I'd say that a mere 5% chance of being noticed by an average guard is pretty damn good.
Moon-Hawk
Okay, I've got a crazy idea. Bear with me, it's crazy.
Perhaps allow a karma reroll after they've been detected.
DM: Make your stealth test.
Player (rolling): My highest was an 8
DM (rolling): You're gonna fail.
Player: No way, who saw me, what'd they get?
DM: Eat me. You wanna reroll or not?
Player (spending karma and rerolling): I got an 11
DM: Nevermind, you're fine.

Usually, when rerolling with karma, you already know if you've succeeded or not and make a reroll using that information (admittedly, not always), so it's not painfully broken by that standard.
It's not perfect, but it's a really, really simple fix, and it ends all player whining, and isn't that really what's important? nyahnyah.gif (kidding)

edit: I should say that, as a house-rule, I allow karma to reroll all but the highest die on an open test. All in all it sounds generous, but it makes the whole open-test pass/fail issue seem to work out more like I think it should without any major changes.
John Campbell
So you occasionally get spotted despite all the min-maxing you can do to your Stealth skill.

Hear that? That's the world's smallest violin, playing just for you.
Smiley
I was wondering how long it would take for someone to declare min/maxing or munchkinism. Seems to be an inevitability.
Sphynx
Ok, John that quickly puts you on an ignore-all-posts-from-him.

First off, despite belief, the player is a mundane, not an Adept, started the game with a stealth of 4 and through karma only has it up to 12. He is the best of the best, better than even the min-maxxed adepts who can start at 12 (though their stealth isn't REALLY a 12, they only get to roll 12 dice). Although I can appreciate how someone would want to play a violin that a player's 100 karma (cost to go from 4 to 12 with a Quickness of 12) seems useless since every 3rd time he hides a lucky dice roll makes it useless, I tend to give my players not only more respect than that, I tend to want to keep the game fun for them.

Now that I've had to explain myself, I am going to come up with a House Rule because it's gotten crazy, and I don't enjoy seeing my players that annoyed that his rolls just don't matter. The question isn't if the current rule is good enough, it's coming up with a better system so that the extreme luck factor isn't a factor.

Sphynx
Smiley
QUOTE (TinkerGnome)
Then upgrade to a stealth suit/cloak ASAP.

A what, now? You mean ruthenium?
TinkerGnome
Hits the "Abandon Thread" alarm.

That's okay, I guess we already covered everything important on stealth rolls. Modifiers, modifiers, modifiers. There's a reason a good stealth adept has multiple vision modes and high athletics. Go where the guard does not look!
TinkerGnome
QUOTE (Smiley)
QUOTE (TinkerGnome @ Apr 13 2004, 01:32 PM)
Then upgrade to a stealth suit/cloak ASAP.

A what, now? You mean ruthenium?

Yes, the word that is hard to spell properly without looking it up wink.gif
Smiley
Gotcha.
gknoy
QUOTE (Sphynx)
50% of the time a person rolling 12 dice will NOT get a 12+.
10% of the time a person with a perception of 4 will.
Better than a 5% chance that 12 stealth is noticed by perception 4.

Though, experience tells me it's closer to 20% for some reason. nyahnyah.gif

You're quite right. Much better than 5% chance. I believe.

In a million rolls, 12 dice will roll an 11 or better at least half the time.

In a million rolls, 4 dice will roll 7 or better half the time -- which means half the time they roll WORSE than seven.

Of course, this is on average. I find it probable that you have not had a sample size of a million rolls. =)

The adept is likely to roll about 3-4 higher than the guard.

The guard is likely to beat the adept's roll about 25-30% of the time.
(IF i am interpreting my data correctly, and I am hazy on THIS particular part of it (and only this part wink.gif You all can look at my data and see if i'm interpreting that wrong; I mightbe comparingthe wrong things to make that judgement.)

Which seems to jive with your experience.
Jason Farlander
QUOTE (Sphynx)
Now that I've had to explain myself, I am going to come up with a House Rule because it's gotten crazy, and I don't enjoy seeing my players that annoyed that his rolls just don't matter. The question isn't if the current rule is good enough, it's coming up with a better system so that the extreme luck factor isn't a factor.

Sphynx

In any game with a dice mechanic, extreme luck will be a factor. That said, good luck with your houseruling - I hope you manage to make the game more fun for your players. Let me know if you manage to get around the fact that requiring a standard test skews the probabilities far too much in the stealth-using character's favor.
Moonstone Spider
How do you default on an open test anyway? Subtract 2-4 from the highest result?
Kanada Ten
Yep.
A Clockwork Lime
One of the few good things Blackjack ever came up with was the Joy of 4.
Moonstone Spider
I'll have to try that on a gullible GM some day.

"My best roll was a 6 on my Stealth test but I'm defaulting to quickness so I get +4."
blakkie
QUOTE (Moonstone Spider @ Apr 14 2004, 02:49 AM)
I'll have to try that on a gullible GM some day.

"My best roll was a 6 on my Stealth test but I'm defaulting to quickness so I get +4."

LOL. smile.gif

Anyway, yes the odds of open tests are somewhat counter-intuetive when your are used to standard Opposed tests. One thing i've found though with playing around with the open test odds spreadsheet Lilt and i put together is that after getting yourself a solid 6 openning dice TN modifiers tend to make more difference than adding dice. So go with modifiers like spirits, obsuring spells, cloaking gear, etc. Plus become familiar with the odds you are truely trying to beat. Once you understand the numbers you won't be relying on the "sure thing" that isn't, that will help a LOT.

EDIT: On the gear front simple camo beats the hell out of ruthenium on the cost/performance, as long as you have the oppropriate camo print for your surroundings.
Rev
I think i'll post this here too, just in case.

CODE

Percentage of wins by dice indicated on the left against dice indicated on the bottom in open test (dice on left got higher result):

12   90 82 75 70 66 61 57 54 50 49 46 44
11   89 80 73 68 63 59 55 52 48 46 44 42
10   88 80 71 65 61 56 52 50 47 44 42 40
_9   86 76 68 62 58 53 50 47 43 42 39 37
_8   85 74 66 61 56 50 47 44 41 39 36 34
_7   83 71 63 56 52 47 44 40 38 36 33 31
_6   81 68 60 52 48 44 40 37 34 32 29 27
_5   78 64 55 48 43 40 35 32 30 28 26 25
_4   75 60 49 42 37 33 31 27 25 23 22 21
_3   69 54 42 37 31 27 25 22 20 18 18 17
_2   60 42 33 27 24 19 18 16 15 13 12 12
_1   43 27 20 16 12 11 10  9  8  7  7  6

      1  2  3  4  5  6  7  8  9 10 11 12


CODE

Percentage of wins by dice indicated on the left against dice indicated on the bottom in most 4+'s test (dice of left got more 4+'s):

12   97 95 91 87 82 76 72 64 58 52 46 41
11   96 93 89 84 79 72 65 59 52 46 40 35
10   95 91 85 80 73 67 60 53 45 41 35 29
_9   93 87 81 75 68 61 52 47 40 34 28 24
_8   90 84 77 68 60 53 46 39 33 27 23 19
_7   86 79 71 62 54 45 38 32 27 22 18 14
_6   82 74 62 53 45 37 32 25 20 16 13 10
_5   75 64 54 44 36 30 24 18 14 11  8  6
_4   66 54 45 35 27 22 16 13  9  7  5  3
_3   55 43 33 24 19 13 10  7  6  3  3  2
_2   40 29 21 15 10  7  5  3  2  1  1  1
_1   21 15  9  6  4  2  2  1  0  0  0  0

      1  2  3  4  5  6  7  8  9 10 11 12


Take with a +-20% (of the reported percentage) grain of salt though as I did not rigerously qualify that I am using the random number generator within its limits, but I did make some lackluster effort in that direction and I messed with it a bit and things only changed by about 20%. Used 10k tests per combination.
Dax
Look at it this way Sphnyx. Every time I DM a game of D&D, the rate of nat 20's jumps from next to nothing to extremely common. I've seen a great deal of my villians meet their end at the hands of such lucky rolls.

Perhaps there is something to be said about GM karma/luck/whatever you want to call it.
Herald of Verjigorm
It could be luck, or...
Dax
Ahhh. Player vs Player. How I love that comic.

But on a more serious note, I don't belive its cheating. The group is made up of experianced gamers who have been playing since the early days of D&D, who have run through all the classic D&D modules, and who have 10 years more experiance than me.

We really don't cheat in our games. We just try to have fun and laugh at freaky turns of luck.
Moon-Hawk
QUOTE (Sphynx)
Ok, John that quickly puts you on an ignore-all-posts-from-him.

First off, despite belief, the player is a mundane, not an Adept, started the game with a stealth of 4 and through karma only has it up to 12. He is the best of the best, better than even the min-maxxed adepts who can start at 12 (though their stealth isn't REALLY a 12, they only get to roll 12 dice). Although I can appreciate how someone would want to play a violin that a player's 100 karma (cost to go from 4 to 12 with a Quickness of 12) seems useless since every 3rd time he hides a lucky dice roll makes it useless, I tend to give my players not only more respect than that, I tend to want to keep the game fun for them.

Now that I've had to explain myself, I am going to come up with a House Rule because it's gotten crazy, and I don't enjoy seeing my players that annoyed that his rolls just don't matter. The question isn't if the current rule is good enough, it's coming up with a better system so that the extreme luck factor isn't a factor.

Sphynx

I agree. Anyone who invests that much karma into a skill should really be able to perform that skill so freakin' well that they do not fail against average people. Part of the charm of SR is that anyone can beat anyone, theoretically, but someone who puts THAT MUCH karma into a skill shouldn't be spotted by Joe Average security guard, barring extreme luck. A house rule of some type is indeed necessary.
I agree that camo should be worn and all, but this guy with a 12 stealth skill should be James Bond sneaky. He should be able to sneak past a blockade wearing a tuxedo and come out clean enough to seduce the one beautiful female guard out of 200 that rolls fantastically and catches him.
blakkie
QUOTE (Moon-Hawk @ Apr 14 2004, 03:49 PM)
QUOTE (Sphynx @ Apr 13 2004, 08:19 PM)
Ok, John that quickly puts you on an ignore-all-posts-from-him. 

First off, despite belief, the player is a mundane, not an Adept, started the game with a stealth of 4 and through karma only has it up to 12.  He is the best of the best, better than even the min-maxxed adepts who can start at 12 (though their stealth isn't REALLY a 12, they only get to roll 12 dice).  Although I can appreciate how someone would want to play a violin that a player's 100 karma (cost to go from 4 to 12 with a Quickness of 12) seems useless since every 3rd time he hides a lucky dice roll makes it useless, I tend to give my players not only more respect than that, I tend to want to keep the game fun for them. 

Now that I've had to explain myself, I am going to come up with a House Rule because it's gotten crazy, and I don't enjoy seeing my players that annoyed that his rolls just don't matter.  The question isn't if the current rule is good enough, it's coming up with a better system so that the extreme luck factor isn't a factor. 

Sphynx

I agree. Anyone who invests that much karma into a skill should really be able to perform that skill so freakin' well that they do not fail against average people. Part of the charm of SR is that anyone can beat anyone, theoretically, but someone who puts THAT MUCH karma into a skill shouldn't be spotted by Joe Average security guard, barring extreme luck. A house rule of some type is indeed necessary.
I agree that camo should be worn and all, but this guy with a 12 stealth skill should be James Bond sneaky. He should be able to sneak past a blockade wearing a tuxedo and come out clean enough to seduce the one beautiful female guard out of 200 that rolls fantastically and catches him.

Perhaps part of the issue is proper TN modifiers for perception? Those odds assume alert observer watching a wide open area in perfect daylight and the stealth person is wearing...um a Tuxedo i guess. Being able to move totally undetected (no Success) 1 in 4 times (or something like that). You start putting in modifiers like dim light, terrain that offers lots of cover, and such and the odds rapidly drop.

Also i think Lilt is really onto something when he is talking about what the # of successes mean. 1 success means you become alerted, but not alarmed. You perk up your senses, you actively search for what put you on edge, etc. At 2 success you notice something tangible. A blur or flutter that doesn't allow you to identify it, but you are certain there is SOMETHING there.

Now if you look at the odds of those kind of successes it starts falling more in line with what kinda feels right.
Wireknight
The only thing I've really come to hate about open tests is how they're rerolled. Say you're Stealthd00d, the adept in question with an adjusted Steath roll of 12 dice. Let's say you really screw up, you don't roll higher than 5, and you want to reroll the whole thing.

It costs you 1 Karma Pool, right?

Wrong.

It costs you 78(1+2+3+4+5+6+7+8+9+10+11+12) Karma Pool, since it counts as one full reroll per open test die you roll, and reroll costs are cumulative rather than static.

It's rare that, for an open test, any serious sort of reroll can be done without emptying out the karma pool completely.
Moon-Hawk
Which is why I strongly suggest house-ruling such that a karma point allows you to reroll all but the highest die.
hobgoblin
i dont realy see the problem with either the open roll for stealth test as no matter how high a skill you have you are not invisible, you cant stand in front of a guard and jump up and down.

as for rerolling dice in a open test, 1 dice can be all that you need as your trying to get the highest single number. being able to reroll 12 dice to get higher then the 5 you got before by paying 1 karma is munchkin at best, totaly gamebreaking at worst. it would effectivly mean that you could get a nice skill, then keep spending karma until one dice hit the alltime highest roll in the entire game. someone do the odds to back me up here smile.gif

a open test just means your trying to do your best when the diffrent factors isnt under your control unlike when your trying to do something when you have more or less full control of the diffrent factors and more or less know how hard it will be to pull it of (or in other words a success test). just look the other places where a open test is used: the opening roll of a vehicle combar round (as the diffrent riggers try to get into as good a posison as they can in a continualy shifting enviroment) and in interrogation and intimidation tests, where you dont realy know how the other person will react but your trying to have him do what you want by either getting him to contradict himself or by scareing him silly.
Moon-Hawk
QUOTE (hobgoblin)
personaly i would rule a open test karma reroll like this:
1 karma and you get to roll 1 die, if its higher then the highest result you got allready then ok its the new highest, if you want to then you can pay 2 more for a new roll and so on. basicly your buying more dice by spending karma.

it makes more sense that way then the way wireknight looks at it.

That's the official rule for karma on open tests.
Wireknight understands this. The point Wireknight is making is that by that official rule, in order to reroll all your dice, it costs 78 karma to reroll 12 dice for an open test, but only 1 karma to reroll 12 dice for a regular test. A bit skewed, neh?
A Clockwork Lime
Again, you guys really should follow the link above to the Joy of 4. Fixes this annoying problem quite nicely, even if it was originally designed for something else. smile.gif

Basically, just turn all Opposed Tests into a standard test against a TN of 4 (modified by situational modifiers as normal to the side who they affect the most; in a Stealth Test for instance, lighting conditions would be applied to the Perception Test, whereas squeeky floorboards would be applied to the Stealth Test). The character who gains the most successes "wins."
hobgoblin
oops, sorry moon-hawk, didnt expect a answer so fast. i edited my post to try and more clearly get my point forward about why it si like it is.
TinkerGnome
If you're rolling 12 dice, about the only way to get hosed is with a botch. And a 1 die reroll should break that, no problem. EagerBeaver the security guard (rolling 4 perception dice) should have a minimum TN of 4 (roll of 2 + 2 for partially hidden or poor lighting... sneaking without at least some cover is silly). In that horrible case, the guard has a 93.75% chance to notice that something is there. He has no idea what it could be. It could be a devil rat, for all that he knows. He has a 68.75% chance to know something's there and think it's a person. It could not be. He has a 31.25% chance to notice that there's a person there and suspect it is an intruder. He has a 6.25% chance to know it's a person and an intruder.

In other words, even if you get beaten on that roll, you have to get beaten very, very badly to really get caught. Depending on the SOP for the place the guard is guarding, there might not even be a call go out for 1 success or even 2 successes. Even if there were a call, it'd be generic and low priority. If you kill/incapacitate the guard before he gets off another report, you'd likely have a fair bit of time before the dispatcher realizes he hadn't reported back in on that noise he was investigating.
hobgoblin
i fear that the real problem is that most if not all people forget to read 1 success as just the minimum needed (i thought i saw a pussycat, or something like that). spoting someone useing stealth is a perseption test, apply the appropriate rules smile.gif (yea i know its been stated, im bored...)
Egon
yep open test are all about luck. I had a sammy chummer have his predator in a stealth holster in the small of his back under a long coat taken away by the troll bouncer, INT of 2 or 3, right after our mage with an AK hanging like a third leg walked right in.

Oh the mage didn't want people to know he was a mage so during combat he would fire the AK in a more or less random way, and shield us and cast mana spells. Something about us gun down apposing mages first and with great prejudice made him nerves.
Rev
CODE

% where dice on left score two sucesses against the target number generated by taking the maximum result of the dice on the bottom.

12   79 66 54 47 40 35 31 28 25 23 20 19
11   77 62 53 43 37 32 28 25 23 20 18 16
10   76 60 48 41 34 30 25 23 20 18 16 14
_9   74 56 45 37 30 26 23 20 18 16 14 13
_8   71 54 41 33 28 23 20 18 15 13 12 11
_7   67 49 38 29 24 20 17 15 12 11 10  9
_6   63 44 32 25 19 16 13 11 10  9  7  7
_5   59 39 27 21 15 13 10  8  8  6  6  5
_4   52 31 21 15 10  8  7  6  5  5  3  3
_3   42 23 14  9  7  5  4  3  2  2  2  2
_2   25 12  7  4  3  2  1  1  0  0  0  0
_1    0  0  0  0  0  0  0  0  0  0  0  0

      1  2  3  4  5  6  7  8  9 10 11 12
TinkerGnome
So, combining the two sets of data, it looks like while Joe 4 dice can hear a super sneaky character about 1 in 5 times as a noise, he's only got a 3% chance of actually realzing that there's a person there? Sounds about right.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Dumpshock Forums © 2001-2012