Hi,

how do you like Dices in game ? I dont know why FFG makes dices that are imbalanced ?! They have different patterns on each side cut right into each side of the dice … Is that feature to make some sides "lighter" and some "harder" to roll ? I thought Descent will be played with balanced dice, each side with 1/6 probability to roll. Especially "X" side of blue die is super cut. X makes the die have great crater in a shape of X in it … i hate this dices …

Two items:

1. Are the differences in material removed from each die face enough to make a significantly noticable difference in die face probability?

2. If so, was this done intentionally to skew the die face probablities as part of the game design? If not, is each die internally counterweighed to compensate for the differences in material removed from each of its faces?

Personally, I'm not sure it makes enough difference to care.

I have made 500 rolls to test blue die just using my hands and dice tower. Results are listed below.

X - 77 (15%)

2 - 88 (18%)

3 - 82 (16%)

4 - 88 (18%)

5 - 79 (16%)

6 - 86 (17%)

ZXTR said:

You have done it on computer ? My test 400 attemts. Also all attempts was done same way … throw the die long distance on carpet so it will roll.

X - 76 (19%)

2 - 60 (15%)

3 - 50 (12,5%)

4 - 62 (15,5%)

5 - 68 (17%)

6 - 84 (21%)

Rasiel said:

Because they have had experience with printing directly onto the dice (the first edition of Doom) and the print wore away, leaving nought but coloured cubes of plastic - and rendering the game unplayable…

I'd MUCH rather have etched dice, thanks. And the bias they may or may not have would be negligable. People always remember the misses - remember, it's 1 in 6 rolls (statistically speaking). That's pretty dang frequent.

Using just Rasiel's numbers I get:

chi^2 = 10.993

chi^2/n = 1.83

chi^2/(n-1) = 2.20

Perhaps a small problem. (chi squared per degree of freedom should be about 1).

Combining the two results of ZXTR and Rasiel's, I get:

chi^2 = 5.968

chi^2/n = 0.994

chi^2/(n-1) = 1.194

This looks perfect to me. Rasiel if you are still concerned roll your dice 1800 times. (The large statistics matter). Ideally you would use others to help you roll since systematic errors matter.

For 1800 throws you should expect the dice to be about 300 each with anything more then 32(2 sigma) off being a problem, although to convince me it is not a fluctuation you would need to be probably 65 off (4 sigma) (i.e. getting one number more then 365 times or less then 235 times.)

Problem is not randomly throwing the dice … problem is what happens when you know how to roll the dice to get the result you want …

Some gamblers could tell its easier to roll some sides with this dice than with ordinary ones.

Rasiel said:

Get a dice tower or rolling cup then. Problem solved.

There also are Descent 2 dice apps for Apple and Android smartphones/pads.

You also can order blank dice and add stickers (I put a template for 19mm dice on BGG).

But, all in all, I would be astonished that the slight difference of weight between sides would have a serious effect upon play.

don't play with gamblers…

Rasiel said:

come on. nobody blames normal dice and they should be "unbalanced" as well right? one dot, 2 dots, 3 dots….