lalato
Underboss
1546 Posts
    
Urbana, IL
 | | 06/06/2006 12:35 PM |
| I had a game last night where the majority of my +17 primary attacks missed an AC24. Why does dicebot hate me? Why? [)]
--sam | |
I'm in Urbana, IL | |
|
Avatar of Skirmishes
tullywi
Sergeant
982 Posts
| | 06/06/2006 8:00 PM |
| | Did you use the t10d20 feature? That makes it roll 10 times silently - only you see it. Then, roll a normal roll so your opponent can see. It seems to help me a little bit. Sort of like chucking a die through a wall or something. [:)] | |
| |
|
lalato
Underboss
1546 Posts
Urbana, IL
| | 06/06/2006 9:35 PM |
| nope... I didn't know there was such a feature. so you type "t10d20" in the dice roller box and that shows 10 silent rolls the other player can't see?
--sam | |
I'm in Urbana, IL | |
|
Avatar of Skirmishes
tullywi
Sergeant
982 Posts
| | 06/06/2006 10:15 PM |
| Yep! The official definition is tXXdYY where "t" is for "test", XX is the number of rolls you don't want you opponent to see, "d" is "die/dice", and YY is the type of roll. So, if you want to roll 5 six-sided dice, you would do "t5d6".
Whatever the results are - only you can see them. So, don't try to tell your opponent you got a 20 or anything like that! Only results sent to both players are valid.
cya around! | |
| |
|
robbdaman
Underboss
2380 Posts
| | 06/06/2006 10:21 PM |
| The dice bot hates everyone now and then. Trust me I've had some absolutely horrid games on there with tons of crit failures where every mini I had routed. It's sad stuff but you just have to realize that the dice bot is far from perfect.
R~ | |
Champion of the Titan
******************************************************************************************************************************************************
Successful trades with: Tickparasite, Iyceman, Faragdar The Wise's friend, avrivah, Drakkengi, brucemc, Krush, maniacal_mini_monger, hung4treason, Gandy, NarlethDrider, Kunimatyu, etc, etc, blah, blah, blah..... | |
|
Toxic_Rat
Sergeant
745 Posts
Lehi, Utah, USA
| | 06/14/2006 10:50 AM |
| | Has anyone checked to see that the dicebot is correct, at least statistically? Even distribution and all that would be nice to know, even though the dice gods can be fickle in who they support at any given moment. | |
Any time you try to make something idiot-proof, someone always goes out and invents a better idiot.
Champion of Siege Equipment | |
|
The Mighty
jai
Commander
3235 Posts
| | 06/16/2006 10:30 AM |
| | I ran some large scale tests of it when i rewrote it that tracked a couple thousand rolls. It is as good as it is going to get on our budget [)] | | | |
|
Venport
Sergeant
739 Posts
| | 06/16/2006 11:04 AM |
| | yeah $0 is such a huge budget to work with... and for the record the dice bot dose not hate you it hates everyone :-) | |
Sacramento DDM http://www.maxminis.com/forums/topic.asp?TOPIC_ID=16854
New world Project http://www.maxminis.com/forums/topic.asp?TOPIC_ID=20745
| |
|
Zippy
Underboss
1993 Posts
Whitewater, WI
| | 06/16/2006 7:30 PM |
| | It's really a love/hate relationship, just like real world dice. | |
There are 10 kinds of people in the world; those who understand binary, and those who don't.
Reference Thread, H/W List, Champion of the Catoblepas | |
|
HailSpork
Sergeant
545 Posts
| | 06/16/2006 7:57 PM |
| Is there any chance of getting perhaps an alternate test mode that gives you the amount of certain numbers rolled? For example, if I did a test on 5d20 and got an 11, two 8's, a 1 and a 3, it'd go
1: 1
3: 1
8: 2
11: 1
That way we could actually see what kind of distribution we're getting. | |
CHAMPION OF THE CURRY GOLEM! | |
|
The Mighty
jai
Commander
3235 Posts
| | 06/19/2006 8:07 PM |
| quote:
Originally posted by HailSpork
Is there any chance of getting perhaps an alternate test mode that gives you the amount of certain numbers rolled? For example, if I did a test on 5d20 and got an 11, two 8's, a 1 and a 3, it'd go
1: 1
3: 1
8: 2
11: 1
That way we could actually see what kind of distribution we're getting.
Not an entirely bad idea ... i will sit and "contemplate my navel" on this one | | | |
|
HailSpork
Sergeant
545 Posts
| | 06/20/2006 4:00 PM |
| | It may handy if it used a different input from the current test; perhaps a different letter or such, just in case some people prefer the original. | |
CHAMPION OF THE CURRY GOLEM! | |
|
Foxman
Sergeant
417 Posts
Calgary, AB, Canada
| | 06/20/2006 4:35 PM |
| I logged onto vassal for the first time today.
I thought I would try out the dice bot.
First Roll:
*** 1d20 = 1 ***
Sigh - This doesnt look so good... | |
It would be very unlikely for unlikely events not to occur.
- J. A. Paulos "Innumeracy" | |
|
Avatar of Skirmishes
tullywi
Sergeant
982 Posts
| | 06/20/2006 6:09 PM |
| quote:
Originally posted by Foxman
I logged onto vassal for the first time today.
I thought I would try out the dice bot.
First Roll:
*** 1d20 = 1 ***
Sigh - This doesnt look so good...
Actually that is GREAT as it had no outcome in a real game. It would be worse to have rolled and wasted a 20.
My first ball when bowling is almost always a strike. When I was younger, that first ball was always a practice ball and didn't count. Now that I'm older, I know not to throw any practice balls so that it counts. | |
| |
|
HailSpork
Sergeant
545 Posts
| | 06/20/2006 6:37 PM |
| Um... previous rolls have no impact on future rolls in a random environment. In a pseudo-random environment, the impact should be negligible, particularly on a PRNG as advanced as the mersienne twister (and on a limit of 1-20, thanks to the modulus allowing several redundancies on the actual randomly generated number)
Anyway, remember that 5% of everybody will get that result the first time they use it. | |
CHAMPION OF THE CURRY GOLEM! | |
|
Toxic_Rat
Sergeant
745 Posts
Lehi, Utah, USA
| | 06/24/2006 9:00 PM |
| quote:
Originally posted by HailSpork
Um... previous rolls have no impact on future rolls in a random environment.
No one really believes this do they? [)]
Yeah its just a die (or dicebot) but when that 20 comes up in a test roll it always does feel like a waste. | |
Any time you try to make something idiot-proof, someone always goes out and invents a better idiot.
Champion of Siege Equipment | |
|
madda
Sergeant
714 Posts
| | 06/24/2006 9:21 PM |
| Jai, there's no need to write new java code. I tested the dicebot using t10d20, t1000d20 and then t10000d20. Copy and paste into a text file, and using this "grep ^d20 dicebot.txt |awk '{ print $3 }'|sort -n|uniq -c" short unix command. I got:
<br> 554 1<br> 545 2<br> 554 3<br> 603 4<br> 504 5<br> 557 6<br> 531 7<br> 573 8<br> 504 9<br> 575 10<br> 584 11<br> 577 12<br> 551 13<br> 543 14<br> 551 15<br> 541 16<br> 527 17<br> 560 18<br> 573 19<br> 503 20<br>
As you can all see it's pretty darn good. (First number is the number of occurrences, second is the dice result). | |
Champion of Cockatrices.
I wish I never wished a wand of wishing. (Wishful thinking.)
Join the Eternal Campaign!
Chat about miniatures. | |
|
HailSpork
Sergeant
545 Posts
| | 06/24/2006 9:38 PM |
| | Call me supersticious, but I'd like to try running it on my own comp to see if my luck is the same. | |
CHAMPION OF THE CURRY GOLEM! | |
|
madda
Sergeant
714 Posts
| |
zephyrstorm
Sergeant
528 Posts
| | 06/24/2006 11:18 PM |
| quote:
Originally posted by madda
Then go ahead and do it. I know I did.
<br> 480 1<br> 502 2<br> 487 3<br> 492 4<br> 511 5<br> 528 6<br> 477 7<br> 499 8<br> 502 9<br> 459 10<br> 516 11<br> 541 12<br> 489 13<br> 515 14<br> 535 15<br> 489 16<br> 532 17<br> 496 18<br> 503 19<br> 447 20 My results came out similar. Slightly lower averages though, since I only ran the last die roll test (t10000d20). | |
Champion of the Hellwasp Swarm
New subscriber to rule #0: bshugg is always right. | |
|
Toxic_Rat
Sergeant
745 Posts
Lehi, Utah, USA
| | 06/25/2006 10:44 AM |
| Okay, while we're running numbers here :)
From Zephyrstorm:
<br># rolls Number % of total<br>480 1 4.80%<br>502 2 5.02%<br>487 3 4.87%<br>492 4 4.92%<br>511 5 5.11%<br>528 6 5.28%<br>477 7 4.77%<br>499 8 4.99%<br>502 9 5.02%<br>459 10 4.59%<br>516 11 5.16%<br>541 12 5.41%<br>489 13 4.89%<br>515 14 5.15%<br>535 15 5.35%<br>489 16 4.89%<br>532 17 5.32%<br>496 18 4.96%<br>503 19 5.03%<br>447 20 4.47%<br>10000 100.00%<br>
Average: 5.00%
Std. Deviation: 0.237%
Median: 5.01%
And from Madda:
554 1 5.03%
545 2 4.95%
554 3 5.03%
603 4 5.48%
504 5 4.58%
557 6 5.06%
531 7 4.82%
573 8 5.20%
504 9 4.58%
575 10 5.22%
584 11 5.30%
577 12 5.24%
551 13 5.00%
543 14 4.93%
551 15 5.00%
541 16 4.91%
527 17 4.79%
560 18 5.09%
573 19 5.20%
503 20 4.57%
11010 100.00%
Average: 5.00%
Std. Deviation: 0.241%
Median: 5.02%
For the two rolls that we all care about, it is interesting to note that in both tests, it is more likely to roll a 1 than a 20. Also, rolling a 20 is the least likely result in both tests.
The dicebot really does hate us. The upside to this is that it hates everyone equally. [}:)] | |
Any time you try to make something idiot-proof, someone always goes out and invents a better idiot.
Champion of Siege Equipment | |
|