Beware the GW legal team

[Ivesy_boy]

Well, a piece of nice software that gives the feel of a real BB game would be a great way to learn about the game. I discovered the game on the GW website one day and with the living rulebook in my possetion i was ready for a game, but without any minis. I found ski junkies tool and am now part of FUMBBL, and will buy the boxed set when it is released and start a league with some mates. The point is that by giving away tid bits of the game, makes people want the real thing. By getting the idea of the game without forking out for the box, I now know i like it, and will spend money on it. So maybe a free demo type game, or a small fee, designed to get you to play the game and buy minis.

[Zombie]

Pariah, Blood Bowl is a strategy game, not an RPG. In strategy games, power gaming is a good thing, necessary even.

Cheating, of course, is another story.

[Ancalagon]

Well took a few minutes to read all post...
First I have to say I've never played through internet but I agree the points shown here... sometimes I'd like to play a match against some of you, but I think nothing compares to the table game and having minis in front of you. This is the main reason why i'll go to blood bowl this march (should I make a shirt with "Ancalagon" on it? ) just to met people. Returning to first post and so... I don't think free blood bowl programs will harm GW, in fact they help on promoting this product.

[SBG]

Hi guys!

I went through all the posts, and I must say that I agree that the online tools are great for the game. I don't feel that the randomness of the dice rolls are that great (people who played agaist me heard me complain about it - again, sorry guys ! I'm just the most unlucky guy to play PBeM I guess...), but it allows people to play games they wouldn't be able to play elsewhere.

I played Dangerous Dave twice and even though I got brutaly beaten twice, I enjoyed his "company". I really wish I could cross the pond and play Spiky or Res and meet all of you, but just couldn't afford it or didn't have vacations left to go. So I dig those online games !

Fred

[RevolutionaryRat]

Speaking of BB computer games..does anyone recall the GW licensed one? Was it any good? Any idea if it would run on Windows XP?

Brian

[CyberHare]

I highly doubt that it would run on XP. Heck I have trouble getting it to run under win 95. It's an ok game if you really havn't got anyone else to play against but it's lack of flexibility made it a dinosaur before it was released. The rules don't completely follow any of the official rule sets either. As I recall, re-rolls were based on how many you could afford, not how many you had.

[GalakStarscraper]

I don't feel that the randomness of the dice rolls are that great

Just for the record.

The randomness of the dice rolls for the PBeM have gone through multiple challenges by several folks.

The last one was to record the first 25 dice rolls of 150 different game loads as it was believed that these were not properly random. Since we had already produced multiple data sets on the rolling 1000 times to see the numeric distribution which completely proved out the long term randomness of the PBeM tool.

We compared pair sequences in the PBEM for the first 25 rolls based on 6 roll sets. 6 roll set probablity is:
6 Matching Numbers: 1 / 7776 (.013%)
5 Matching & 1 Different: 5 / 1296 (.39%)
3 Matching & 3 Matching: 25 / 3888 (.64%)
4 Matching & 2 Matching: 25 / 2592 (.96%)
6 Different (1 of each): 5 / 324 (1.54%)
4 Matching & 2 Different: 25 / 648 (3.86%)
3 Different Sets of 2 Matching: 25 / 648 (3.86%)
3 Matching, 2 Matching, & 1 Different: 25 / 162 (15.4%)
3 Matching & 3 Different: 25 / 162 (15.4%)
2 Matching & 4 Different: 25 / 108 (23.1%)
2 Sets of 2 Matching & 2 Different: 25 / 72 (34.7%)
(thanks to PH for this math).

Here where the results from the 1st 25 rolls of 150 matches:

Code: Select all

-----------------------------------------------------------------------------------
             Analysis of 1st number rolled for a newly loaded game             
                         each should be ideally 16.67%                         

Number_Rolled    COUNT   <-- % 
1                   22  14.67%
2                   32  21.33%
3                   24  16.00%
4                   28  18.67%
5                   18  12.00%
6                   26  17.33%

                   150 100.00%
-----------------------------------------------------------------------------------
        Analysis of the first 6 numbers rolled for a newly loaded game         
                         each should be ideally 16.67%                         

Number_Rolled    COUNT   <-- % 
1                  141  15.67%
2                  152  16.89%
3                  147  16.33%
4                  162  18.00%
5                  152  16.89%
6                  146  16.22%

                   900 100.00%
----------------------------------------------------------------------------------

          Analysis of first 25 numbers rolled for a newly loaded game          
                         each should be ideally 16.67%                         

Number_Rolled    COUNT   <-- % 
1                  627  16.72%
2                  651  17.36%
3                  664  17.71%
4                  653  17.41%
5                  584  15.57%
6                  571  15.23%

                  3750 100.00%
------------------------------------------------------------------------------------
                         PBeM Testing Occurrence Table                         
                Samples generated from the 20 stepped groupings                
            of 6 from the first 25 D6 rolls of a newly loaded game             

SET                                          # OF        %        %
TYPE                                  OCCURRENCES   ACTUAL EXPECTED

2 MATCHING & 4 DIFFERENT                      685  22.833%  23.148%
2 SETS OF 2 MATCHING & 2 DIFFERENT          1,095  36.500%  34.722%
3 DIFFERENT SETS OF 2 MATCHING                111   3.700%   3.858%
3 MATCHING & 3 DIFFERENT                      468  15.600%  15.432%
3 MATCHING & 3 MATCHING                        15   0.500%   0.643%
3 MATCHING, 2 MATCHING, & 1 DIFFERENT         445  14.833%  15.432%
4 MATCHING & 2 DIFFERENT                      113   3.767%   3.858%
4 MATCHING & 2 MATCHING                        21   0.700%   0.965%
5 MATCHING & 1 DIFFERENT                       14   0.467%   0.386%
6 DIFFERENT                                    32   1.067%   1.543%
6 MATCHING NUMBERS                              1   0.033%   0.013%

                                            3,000 100.000% 100.000%
-------------------------------------------------------------------------------------
                         PBeM Testing Occurrence Table                         
        Samples generated from first 6 D6 rolls of a newly loaded game         

SET                                          # OF        %        %
TYPE                                  OCCURRENCES   ACTUAL EXPECTED

2 MATCHING & 4 DIFFERENT                       30  20.000%  23.148%
2 SETS OF 2 MATCHING & 2 DIFFERENT             50  33.333%  34.722%
3 DIFFERENT SETS OF 2 MATCHING                  7   4.667%   3.858%
3 MATCHING & 3 DIFFERENT                       30  20.000%  15.432%
3 MATCHING & 3 MATCHING                         2   1.333%   0.643%
3 MATCHING, 2 MATCHING, & 1 DIFFERENT          21  14.000%  15.432%
4 MATCHING & 2 DIFFERENT                        8   5.333%   3.858%
4 MATCHING & 2 MATCHING                         2   1.333%   0.965%
5 MATCHING & 1 DIFFERENT                        0   0.000%   0.386%
6 DIFFERENT                                     0   0.000%   1.543%
6 MATCHING NUMBERS                              0   0.000%   0.013%

                                              150  99.999% 100.000%
-------------------------------------------------------------------------------------
                         PBeM Testing Occurrence Table                         
                Samples generated from the 4 complete groupings                
            of 6 from the first 24 D6 rolls of a newly loaded game             

SET                                          # OF        %        %
TYPE                                  OCCURRENCES   ACTUAL EXPECTED

2 MATCHING & 4 DIFFERENT                      138  23.000%  23.148%
2 SETS OF 2 MATCHING & 2 DIFFERENT            223  37.167%  34.722%
3 DIFFERENT SETS OF 2 MATCHING                 19   3.167%   3.858%
3 MATCHING & 3 DIFFERENT                      101  16.833%  15.432%
3 MATCHING & 3 MATCHING                         5   0.833%   0.643%
3 MATCHING, 2 MATCHING, & 1 DIFFERENT          77  12.833%  15.432%
4 MATCHING & 2 DIFFERENT                       25   4.167%   3.858%
4 MATCHING & 2 MATCHING                         5   0.833%   0.965%
5 MATCHING & 1 DIFFERENT                        1   0.167%   0.386%
6 DIFFERENT                                     6   1.000%   1.543%
6 MATCHING NUMBERS                              0   0.000%   0.013%

                                              600 100.000% 100.000%
-----------------------------------------------------------------------------------

Just to be clear on the subject.

Galak

[christer]

I don't feel that the randomness of the dice rolls are that great

Also for the record.

Javabbowl uses the standard java.util.Random class for its dice rolls. Quoting from the documentation of said class:

The class uses a 48-bit seed, which is modified using a linear congruential formula. (See Donald Knuth, The Art of Computer Programming, Volume 2, Section 3.2.1.)

As some of you programmer type people might know, the book referenced is pretty much a volume of the "bible" of programming, and I'm pretty sure that the algorithm used is pretty much up to par with the best pseudo-random number generators known.

Ofcourse, pseudo-random means that it's not truly random. True randomness is impossible to do on computers. At least today.

Given the fact that the particular algorithm used is detailed in Knuth's book and the fact that Sun decided to make it standard has me convinced there's nothing wrong with it. I'm sure Knuth and the people at Sun have more knowledge in this area than I do.

In essence, I trust it to result in the grade of randomness required to play blood bowl

-- Christer

[Zy-Nox]

Ofcourse, pseudo-random means that it's not truly random. True randomness is impossible to do on computers. At least today.

Yes that is why i rolled two double skulls in one turn
And also 6 in one game

[GalakStarscraper]

Javabbowl uses the standard java.util.Random class for its dice rolls. Quoting from the documentation of said class:

The class uses a 48-bit seed, which is modified using a linear congruential formula. (See Donald Knuth, The Art of Computer Programming, Volume 2, Section 3.2.1.)

As some of you programmer type people might know, the book referenced is pretty much a volume of the "bible" of programming, and I'm pretty sure that the algorithm used is pretty much up to par with the best pseudo-random number generators known.

Likewise the random number generator on the PBeM tool was built by doctor level studies on creating pseudo-random numbers specifically using the Delphi language. For those that care the random number generator for a single D6 roll is:
1) Generate a random number between 10 and 109 (set to X)
2) Pick X random numbers between 0 and 127772 (sum them to Y)
3) D6 roll = (Y mod 6) + 1

for those that don't know ... mod is the remainder so: 29 mod 6 = 5 and 30 mod 6 = 0.

Galak

[SBG]

So, as I said, I'm the most unlucky guy playing this game !

Fred

[Martin]

Yes that is why i rolled two double skulls in one turn
And also 6 in one game

That would be computer imitating life. In my last game I started out with double skulls, rerolled it and took the guy down, next guy double skulls. That pretty much summed up how that game went. In fact ever since the start of overtime in the previous game my dice have been cursed. I needed a 2+ dodge to win the game in OT, got a 1. It's all due to my girlfriend touching my dice before OT. In some cases this might be a good thing but not here, she is a known cursor of dice. I have got to get ahold of a friends 13 year old and have her 'bless' my dice. She is the only one with the power to offset my GFs touch.

Martin

[Darkson]

Yes that is why i rolled two double skulls in one turn
And also 6 in one game

That would be computer imitating life. In my last game I started out with double skulls, rerolled it and took the guy down, next guy double skulls.

Beat that. Double skulls, re-roll, double skulls. Twice.. In two turns... And I only have 3 RR's Somedays, it just doesn't pay to play.

[Martin]

It's one of those days when you just want the game to get over quickly so too many don't get maimed or killed. I know that feeling too well. I was playing against a goblin team (we allow 4 trolls) his first 3 blocks injured a guy. He blitzed and knocked a guy over, who was then injured by a foul. So it's my turn finally and I only have 7 on the field. It kept up that way as the game went along. Should have known not to play someone who's girlfriend has just bought them new dice (unless it's me and my girlfriend).

Martin

[ravyn]

I know for a fact it will not run on xp. as soon as i finish my wife's win 98 Pentium 1 system will try it there if not will take one of the 486 i have laying around and set it up for blood bowl, mechwarrior 1, crescent hawk's inception and revenge.
Ravyn