PDA

View Full Version : Ethical considerations of testing the weight-balance on (un)lucky dice?



rahimka
2015-06-13, 02:28 AM
During some video browsing, I stumbled upon a video about how to check whether your gaming dice are evenly weighted (basically floating them in a salt water solution to see if they consistently float with the same number on top).

At first I was eager to discover if any of my dice were giving me low numbers (I seem to get session-long runs of 2s and 3s on some of my d20s). But during the video, the first die which was demonstrated was weighted to roll 20s, and THAT got me thinking:

If I choose to test my "unlucky" dice, am I ethically obliged to check the "lucky" ones too?

What if some of my luckier dice are in fact loaded in my favor?

Am I obliged to give them up if I don't want to be "cheating"?

So can I just choose not test the ones I feel are "lucky"? Is that willful ignorance ethical if I'm testing the ones I suspect will be loaded against me?

What do you folks think are the gaming ethics in effect here?

The Evil DM
2015-06-13, 02:39 AM
Are you trying to demonstrate a conundrum faced by a lawful character. :)

Sarcasm aside, I have seen so many dice roll in my life time. I once calculated that I have seen well over a million die rolls.

The overall statistical invariance of dice - while possible - really doesn't amount to a whole lot. If you were playing in my game I wouldn't care either way. The dice are what they are and its not blatant cheating. A d20 with slight manufacturing flaws won't be all that far off due to its roundness.

Ceaon
2015-06-13, 03:37 AM
If you already feel you have dice that are "lucky" (regardless of whether they really are), isn't it already unethical to favor those dice over your "unlucky" ones?

hymer
2015-06-13, 05:19 AM
No need to make this difficult. Test every die you have (you know you want to), and toss any that aren't properly weighted. Having cheaty dice around is just temtping yourself.

JAL_1138
2015-06-13, 05:41 AM
Unless they're particularly flawed (massive air bubble, extremely flattened/egg-shaped, or have a malformed side the die can't land on) they're going to be broadly acceptable for gaming purposes. These dice become apparent after a few dozen rolls.

Most of my bad ones came from Wiz Dice (which I don't even count with my collection anymore, they were that badly air-bubbled) or Crystal Caste (I've got a normal d20 that will barely stay on...crud, I forget the actual number, I haven't used it in so long, but it's got a really badly convex face with a warped edge, next to a very concave face with the same warped edge, and doesn't ever roll the opposite faces; and I've got a "barrel" d20 that won't stay on 20 unless on a perfectly level (as determined with a spirit level), perfectly smooth surface, in a room with no air currents, when you've deliberately placed it on 20--in all other cases, even if it should land on 20, it will roll back to 2. Even if you set it on 20 and aren't very careful how you let go of it, it'll roll back to 2, always 2. Also have a barrel d10 and d12 with the convex-side problem; they won't ever land on the sides opposite the convex ones).

You only need to float-test the ones with no readily-apparent flaws to a visual inspection that still never seem to land on certain numbers.

I always recommend translucent but easily-readable (red, blue, purple, or green with white numbers) Gamescience dice with carefully trimmed-and-filed sprue marks, but they're a) hard to find due to production ceasing for a while and not really being back in swing, and b) labor-intensive. If you don't want to muck with that, Koplow makes a very good, though not very pretty, translucent die--good in the sense of consistent-and-even size, shape and balance (came out ahead of Chessex in measurements and stack tests).

(Always get translucent dice when possible, as they're easier to check for air bubbles--you don't need to float them.)

Do not use pipped d6es with rounded corners. They roll far more 1s due to the very design of the die, as has been demonstrated here (http://www.dakkadakka.com/wiki/en/That's_How_I_Roll_-_A_Scientific_Analysis_of_Dice).

EDIT: The float test can also give not-quite-false positives. An extremely small air bubble may affect float balance but not give a practical difference over a thousand rolls. Pretty much no RPG die is made to casino standards (and casino dice are made of plastic that chips relatively easily since they're rolled on padded surfaces and discarded after several hours (not days) of use, plus they're table-gougers), so some variance from perfect random is acceptable. It's impossible to make molded dice to casino spec because of heat-shrink, and even horribly-expensive, table-wrecking, machined metal dice don't have numbers filled with same-density material (affecting balance in a minute way) and rounded edges. That said, rounded-corner pipped d6es are so far off random that they really should be avoided.

hymer
2015-06-13, 07:43 AM
table-gougers

They make marks on the surface they get rolled on, or...?

JAL_1138
2015-06-13, 07:50 AM
They make marks on the surface they get rolled on, or...?

Yup. Casino dice? Scratch and/or ding heck out of a wooden table, rip the felt on a pool table. They're meant to be bounced off a rubber backstop and land on felt (that doesn't have a slate underneath it, unlike a pool table) and are large enough that to get them to roll without bouncing them off a backstop (instead of landing and skidding on a side), you need to hurl them with considerable force, exacerbating the scratch-and-dent or felt-rip problem. This is also part of why the corners of casino dice break off and the edges chip easily when they aren't rolled (thrown, really) in a craps table.

Metal dice are heavy enough to dent or scratch a table based on weight alone.

EDIT: They don't make *big* marks, but they make marks. (Little marks on the felt of a craps table or dings in the board under it don't matter; on a pool table, especially somebody else's pool table, they do).
Metal dice, due to weight, having pointier corners than a pool ball even if the billiard is heavier, and the throwing force used, can even make a little divot in the slate of a pool table. Can crack or chip a glass table, and will dent wooden tables.

LibraryOgre
2015-06-13, 08:08 AM
Dragonsfoot had a thread about this. (http://dragonsfoot.org/forums/viewtopic.php?f=11&t=70649)

Of note in the thread, here's the video... (https://www.youtube.com/watch?v=VI3N4Qg-JZM&feature=youtu.be)

And here's a link to an empirical test of Chessex v. Gamescience dice (http://www.awesomedice.com/blog/353/d20-dice-randomness-test-chessex-vs-gamescience/) of 10,000 throws each. The short form of it is that neither were really balanced, the Chessex dice were slightly less so, and the Gamescience dice had a bias against the number 14 due to how they were constructed.

(Just a note: If you head over to Dragonsfoot and start posting, keep in mind the forum's rules).

JAL_1138
2015-06-13, 08:19 AM
Dragonsfoot had a thread about this. (http://dragonsfoot.org/forums/viewtopic.php?f=11&t=70649)

Of note in the thread, here's the video... (https://www.youtube.com/watch?v=VI3N4Qg-JZM&feature=youtu.be)

And here's a link to an empirical test of Chessex v. Gamescience dice (http://www.awesomedice.com/blog/353/d20-dice-randomness-test-chessex-vs-gamescience/) of 10,000 throws each. The short form of it is that neither were really balanced, the Chessex dice were slightly less so, and the Gamescience dice had a bias against the number 14 due to how they were constructed.

(Just a note: If you head over to Dragonsfoot and start posting, keep in mind the forum's rules).

They didn't trim the sprue properly, and said so. You need to take an Xacto and/or an emory board to it.


Sprues got pretty bad and globby on them half the time in the last few years that Gamestation.net owned Gamescience, too, sometimes overlapping the edges.

EDIT: Relevant quote from the article on sprues: "If you carefully sand the flashing down on the GameScience dice you should get a result that is very close to being truly random."

ALSO EDIT: Rounded edges can have an effect, but Zocchi oversells it. Other dice issues are more generally due to outright manufacturing flaws, as I said, such as air bubbles or misshapen sides. GS are usually a little better about not having them other than needing the sprues trimmed, and use good die molds.

As I mentioned, if you don't want to fiddle around trimming sprues and inking numbers, Koplow'll do you fine and do use sanded/rounded edges. Chessex are slightly less so, but are fine because you're looking at what, ten or less rolls in a thousand, one percent or less of rolls? It's insignificant for RPG purposes. BUT: I have noticed more air bubbles and manufacturing flaws in Chessex than in Koplow, Koplow not being much worse than GS, and Crystal Caste having wonky sides far more often than Chessex. And Wiz Dice aren't worth buying unless they've really stepped up quality control since my batch; they were worse than my Crystal Castes. And those physically can't land on certain sides except under ideal circumstances.

ALSO ALSO EDIT: pipped d6's with rounded corners are not just those that have gone through the paint-sanding process. It's the ones you buy in Games Workshop or Chessex bricks that have the corners shaved off and the faces being circular on the trimmed cube. Those. Edges =/= corners; numbered Chessex d6es like the ones that come in 7-piece polyhedral sets still have their corners even though they have smooth edges from paint-sanding in a rock tumbler.

Maglubiyet
2015-06-13, 09:29 AM
If you buy a few Pounds o' Dice (http://www.chessex.com/Dice/poundofdice.htm) (or its equivalent) and use whichever ones are closest to you when you need to roll, odds are that any imperfections will balance out. You won't roll the same dice that frequently.

Or just use metal dice. Throw them at offending participants.

TheThan
2015-06-13, 01:59 PM
So basically dice imbalance isn't much of an issue unless the dice or horrendously misshapen or actually loaded...

seems like i pretty much wasted some cash on game science D6s for mini war gaming, when what i had was just fine.

JAL_1138
2015-06-13, 04:12 PM
So basically dice imbalance isn't much of an issue unless the dice or horrendously misshapen or actually loaded...

seems like i pretty much wasted some cash on game science D6s for mini war gaming, when what i had was just fine.

Bricks of rounded-corner d6es roll WAY too many ones. Doesn't have to do with the sanding/painting method, it's due to the drilled pips and truncated corners. Again, see this link. (http://www.dakkadakka.com/wiki/en/That's_How_I_Roll_-_A_Scientific_Analysis_of_Dice)

Truncated-corner, pipped d6es come up on 1 at 29% of the time. Should be 16.67%. They're WAY off. It's d20s and the other dice like square-cornered numeral'd d6es in standard 7-pc poly sets that you don't see much improvement with. Numbered d6es with square corners also beat drilled-pip d6es with square corners, although that wasn't tested in that particular link. With numerals there's less weight difference than with drilled pips.

So no, you didn't waste the money.

Segev
2015-06-13, 04:21 PM
It's weird to me that such d6s would come up 1s preferentially. I'd have expected the opposite, since the 1-side is going to have more mass due to only the single pip. Wacky.

ExLibrisMortis
2015-06-13, 06:04 PM
It's weird to me that such d6s would come up 1s preferentially. I'd have expected the opposite, since the 1-side is going to have more mass due to only the single pip. Wacky.
I read the webpage JAL provided, and it's quite interesting. As I understand, it (possibly, the writer of that page didn't test it completely) works like this:

The die rolls - turns to another face - a certain number of times, and after the final full roll, it has a certain amount of momentum left. This momentum isn't enough to make a full turn, but it will cause a wobble - the die lifts an edge off the table and falls back down. The amount of energy this wobble can absorb depends on the gravitational potential energy the die can achieve during the wobble, which again depends on the height of the centre of gravity of the die, which depends on whether the heavy part is up top or at the bottom.

Imagine a die with all its mass concentrated on one side face (infinite density, but let's not make it a black hole just yet). This will be happy to land heavy side down (and then roll on, with its tasty new momentum), but if it's rolling the other way, that'll take momentum and turn it into gravitational potential energy. If it does manage to get heavy side up, there won't be as much momentum left - some of it lost to friction, too - so there's not enough to complete another roll. The die will compress under the weight of the top, which maybe creates heat or something to get rid of the final little bit of remaining energy.

The discreteness of the die roll makes this possible. It won't work for an unbalanced sphere, because that doesn't really wobble. It doesn't fall back down onto one side either, of course, so it can still stop after a partial turn, slightly off-balance, on a surface with friction (which are most of them).

Ah well, at least any mistakes in this will cause real physics people to step in and explain :D.



As for ethical concerns, as long as you test your dice equally, I think you're good. Who knew that ethics could be easier than physics?

Rainbownaga
2015-06-13, 06:51 PM
I read the webpage JAL provided, and it's quite interesting. As I understand, it (possibly, the writer of that page didn't test it completely) works like this:

The die rolls - turns to another face - a certain number of times, and after the final full roll, it has a certain amount of momentum left. This momentum isn't enough to make a full turn, but it will cause a wobble - the die lifts an edge off the table and falls back down. The amount of energy this wobble can absorb depends on the gravitational potential energy the die can achieve during the wobble, which again depends on the height of the centre of gravity of the die, which depends on whether the heavy part is up top or at the bottom.

Imagine a die with all its mass concentrated on one side face (infinite density, but let's not make it a black hole just yet). This will be happy to land heavy side down (and then roll on, with its tasty new momentum), but if it's rolling the other way, that'll take momentum and turn it into gravitational potential energy. If it does manage to get heavy side up, there won't be as much momentum left - some of it lost to friction, too - so there's not enough to complete another roll. The die will compress under the weight of the top, which maybe creates heat or something to get rid of the final little bit of remaining energy.

The discreteness of the die roll makes this possible. It won't work for an unbalanced sphere, because that doesn't really wobble. It doesn't fall back down onto one side either, of course, so it can still stop after a partial turn, slightly off-balance, on a surface with friction (which are most of them).

Ah well, at least any mistakes in this will cause real physics people to step in and explain :D.



As for ethical concerns, as long as you test your dice equally, I think you're good. Who knew that ethics could be easier than physics?

Which also suggest the float test is flawed and OP's d20 may be unlucky after all

JAL_1138
2015-06-13, 07:06 PM
Which also suggest the float test is flawed and OP's d20 may be unlucky after all

The float test isn't to check for lucky/unlucky; it's to check for air bubbles in the plastic or serious imbalance. No matter which side comes up. But it can give false positives.

It's worth noting that the heavy side won't always come up (as in the case of a Crystal Caste "barrel" d20 I have that lands on 2 consistently--because the 2 is much, much lighter and there are practically no edges--even if it should stop on another number). It's specific to the die shape and the relatively small imbalance of the pips.

MrStabby
2015-06-15, 11:38 AM
I read the webpage JAL provided, and it's quite interesting. As I understand, it (possibly, the writer of that page didn't test it completely) works like this:

The die rolls - turns to another face - a certain number of times, and after the final full roll, it has a certain amount of momentum left. This momentum isn't enough to make a full turn, but it will cause a wobble - the die lifts an edge off the table and falls back down. The amount of energy this wobble can absorb depends on the gravitational potential energy the die can achieve during the wobble, which again depends on the height of the centre of gravity of the die, which depends on whether the heavy part is up top or at the bottom.

Imagine a die with all its mass concentrated on one side face (infinite density, but let's not make it a black hole just yet). This will be happy to land heavy side down (and then roll on, with its tasty new momentum), but if it's rolling the other way, that'll take momentum and turn it into gravitational potential energy. If it does manage to get heavy side up, there won't be as much momentum left - some of it lost to friction, too - so there's not enough to complete another roll. The die will compress under the weight of the top, which maybe creates heat or something to get rid of the final little bit of remaining energy.

The discreteness of the die roll makes this possible. It won't work for an unbalanced sphere, because that doesn't really wobble. It doesn't fall back down onto one side either, of course, so it can still stop after a partial turn, slightly off-balance, on a surface with friction (which are most of them).

Ah well, at least any mistakes in this will cause real physics people to step in and explain :D.



As for ethical concerns, as long as you test your dice equally, I think you're good. Who knew that ethics could be easier than physics?


Looking at this summary (can't follow the original link from here) this seems to make sense but possibly incomplete.

Again with lots of simplifications... Imagine the die rotating around an axis, and approximate this axis as running perpendicular to two of the faces. Now the biggest amount of excess momentum is going to be needed not when the heaviest side is on the bottom but when it is at the back of the die. Again as a huge simplification but I would expect to see 4 overrepresented numbers and two under-represented numbers - with the two under-represented ones being along the axis where the centre of mass of the die is offset the most.

TheThan
2015-06-15, 01:59 PM
I’m reminded of a guy I knew who had really badly formed dice. They would roll 6s 66% of the time.
Him and the owner of the hobby store I hung out with graphed it out. Rolling each dice like 100 times or something and that’s the result they came up with.

He didn’t use them very often though. pretty much because it’s really unfair.

Spojaz
2015-06-15, 03:09 PM
I tested all three of my D20's in the manner described, and it was suddenly apparent why I roll more 18's and 4's than I should. Two of them were really badly imbalanced, and floated repeatably on the edge between those numbers. I thought I just had weird luck. Now I feel like I owe my players an apology, and that I need new dice.

JAL_1138
2015-06-15, 07:09 PM
I tested all three of my D20's in the manner described, and it was suddenly apparent why I roll more 18's and 4's than I should. Two of them were really badly imbalanced, and floated repeatably on the edge between those numbers. I thought I just had weird luck. Now I feel like I owe my players an apology, and that I need new dice.

I'll repeat my recommendation of carefully sprue-trimmed (to avoid the reduction in 14s--this *must* be done or you'll get the results Awesomedice did, with a *massive* decrease in 14s rolled), greasepencil-inked (to further reduce weight imbalance) translucent (to check for air bubbles) Gamesciences...

Or if you don't want to muck about trimming sprues with an Xacto and emory board and filling in numbers with a greasepencil 'till your hands cramp up, just get Koplow translucents, which'll give you virtually the same degree of precision (higher than Chessex) as a very carefully-trimmed Gamescience without any of that fuss at all. Not as pretty (translucent GS are very clear and reflective, look almost like cut gems), but Koplow are virtually identical in terms of consistency.

I'm still loyal to GS, myself, but unless you just really like the look and feel of them there's no randomness reason not to just buy Koplows and save some money and effort.

EXCEPT. Except d6es. Even square-cornered D6es with indented pips, like so-- [::] --which is what Koplow sells in their big bricks still have a greater variation from random than d6'es with numerals, like so -- [6] --and especially GS, which have a very shallow numeral and thus the lowest weight imbalance. Still have to trim sprues, though.

If you need a few extra d6 but not a brick, Koplow sells 10-piece sets with three extra numeral-type D6 in them, for a total of four, and those are a good buy. Cost about the same as a Chessex 7-piece set.

Jay R
2015-06-16, 05:15 PM
To answer the ethical question:

Trying to get an unfairly good roll by misreading the dice is cheating.
Trying to get an unfairly good roll by knocking the die over is cheating.
Trying to get an unfairly good roll by telekinesis or magic is cheating.
Trying to get an unfairly good roll by buying loaded or other "cheat" dice is cheating.
Trying to get an unfairly good roll by testing your dice looking for a randomly produced weighted die is cheating.

Trying to get an unfairly good roll by X is cheating, for any value of X.

How bad is it if the die gives you a slightly better roll than average? I'm not a professional ethicist, and won't attempt to answer. I don't know if there is any generally accepted measure of how bad a given action is. And even if there were, we would not be allowed to discuss it on this list. But any of us can answer the simple question, "Is it cheating to use a die known to give unfairly good rolls?"

Talakeal
2015-06-16, 08:54 PM
Any advice on how to get this test to work? I am trying to test my dice and they sink like rocks no matter how much salt I add to the water.

LibraryOgre
2015-06-16, 11:51 PM
Any advice on how to get this test to work? I am trying to test my dice and they sink like rocks no matter how much salt I add to the water.

Others have suggested using corn syrup.

rahimka
2015-06-17, 06:10 AM
To answer the ethical question:
Trying to get an unfairly good roll by X is cheating, for any value of X.


I would generally agree. If one knows the dice are loaded in their favor then using them is "cheating" to some degree (even if, as people have indicated above, the actual affect of most imbalances is minimal).

But I don't know my "lucky"/"unlucky" dice are anything more than a superstitious gut feeling on my part.

Would you consider "using lucky dice" a valid X? (specifically in situations where the "luck" is just an untested gut instinct)

Would you consider "by not testing any dice that you feel are lucky" a valid X?

What I was trying to get at in my original question was the ethics of which dice I decide to even test: Assuming I intend to discard any dice I find to be imbalanced, IF I choose to test my "unlucky" dice (which I suspect might be hindering me), am I ethically obliged to test the "lucky" ones (which I suspect might be favoring me)? If I choose to some/any of my dice, am I ethically obliged to test them all?

My gut says "yes", but guts aren't rational and I'm curious what other (traditionally superstitious) gamer people think about it

Jay R
2015-06-17, 02:21 PM
I would generally agree. If one knows the dice are loaded in their favor then using them is "cheating" to some degree (even if, as people have indicated above, the actual affect of most imbalances is minimal).

But I don't know my "lucky"/"unlucky" dice are anything more than a superstitious gut feeling on my part.

Would you consider "using lucky dice" a valid X? (specifically in situations where the "luck" is just an untested gut instinct)

Would you consider "by not testing any dice that you feel are lucky" a valid X?

What I was trying to get at in my original question was the ethics of which dice I decide to even test: Assuming I intend to discard any dice I find to be imbalanced, IF I choose to test my "unlucky" dice (which I suspect might be hindering me), am I ethically obliged to test the "lucky" ones (which I suspect might be favoring me)? If I choose to some/any of my dice, am I ethically obliged to test them all?

My gut says "yes", but guts aren't rational and I'm curious what other (traditionally superstitious) gamer people think about it

I think you are emphasizing the wrong part of my statement.

"Trying to get an unfairly good roll by X is cheating, for any value of X."

You are not compelled to test your dice. You are compelled to stop using dice known to give unfairly high rolls.

Yes, using dice you have good reason to believe are "lucky" is trying to get an unfairly high roll. But until you have recorded hundreds of rolls, rather than counting on memory, all guesses about lucky dice are just ignorance. There is nothing ethically wrong with statistical ignorance.

Talakeal
2015-06-17, 02:49 PM
Others have suggested using corn syrup.

No luck.

Straight corn syrup is too thick and sticky for the dice to turn on their own, and if I water it down enough that this is no longer a problem the dice once again sink.

I am using Chessex opaque dice btw.

Anonymouswizard
2015-06-17, 03:13 PM
Do not use pipped d6es with rounded corners. They roll far more 1s due to the very design of the die, as has been demonstrated here (http://www.dakkadakka.com/wiki/en/That's_How_I_Roll_-_A_Scientific_Analysis_of_Dice).

The strange thing is, I was in a game using a 3d6 system where I was using my red pipped d6s from Orcs Nest (I don't know who made the dice) and another player was using a Games Workshop dice cube. The system was roll high, but avoid a 666 (666 was a crit fail and 111 was a crit success for fluff reasons).

My character was noted to be rather reliable most of the time until the screwed up, because my pipped rounded corner d6s seemed to roll a 666 at least once a session, and only once rolled a 111. This was due to a tendency to have at least one six in every roll, with the dice actually tending towards fours (246 was one of my more common rolls).

The other player's character was regarded as insanely lucky, due to consistently rolling 111 and never rolling 666. He had a slight tendency to roll low, but his character was so minmaxed he didn't really care.

So I think that throwing style has something to do with it as well, since I tend to throw with a flick, and people have abandoned these specific dice for not rolling enough successes in Shadowrun.

Spojaz
2015-06-17, 03:17 PM
No luck.

Straight corn syrup is too thick and sticky for the dice to turn on their own, and if I water it down enough that this is no longer a problem the dice once again sink.

I am using Chessex opaque dice btw.

I had to use nearly boiling water, A deadly amount of salt and about 2 minutes of stirring to get mine to float, but I noticed the worst offenders floated more easily than the others anyway, so yours might be fine. Just roll 'em a couple thousand times to be sure :smalltongue:

Beleriphon
2015-06-17, 04:06 PM
And this is why I use stainless steel dice. My d20 weights a good ounce, and I could probably bring down a very large Philistine with a sling if need be. :smallcool:

JAL_1138
2015-06-17, 09:28 PM
The strange thing is, I was in a game using a 3d6 system where I was using my red pipped d6s from Orcs Nest (I don't know who made the dice) and another player was using a Games Workshop dice cube. The system was roll high, but avoid a 666 (666 was a crit fail and 111 was a crit success for fluff reasons).

My character was noted to be rather reliable most of the time until the screwed up, because my pipped rounded corner d6s seemed to roll a 666 at least once a session, and only once rolled a 111. This was due to a tendency to have at least one six in every roll, with the dice actually tending towards fours (246 was one of my more common rolls).

The other player's character was regarded as insanely lucky, due to consistently rolling 111 and never rolling 666. He had a slight tendency to roll low, but his character was so minmaxed he didn't really care.

So I think that throwing style has something to do with it as well, since I tend to throw with a flick, and people have abandoned these specific dice for not rolling enough successes in Shadowrun.

Could be air bubbles, too. Those will throw the weight off to enough of a degree to account for it, and I've noticed open holes from air bubbles in the pips where the plastic gets injected into the molds in some Pound-o'-Dice d6es, all on the same particular pip. If there was an issue with air bubbles in that batch, they might be consistently on one side near the point where the plastic was put into the molds.



And this is why I use stainless steel dice. My d20 weights a good ounce, and I could probably bring down a very large Philistine with a sling if need be. :smallcool:

I hope you use a tray, dice cup with a lid, or tower. I won't let people roll those on my wooden table otherwise. They'll ding it to heck and gone. And it's not that the table is cheaply-made, it's just that hurling a 1-oz die with even vaguely-pointy corners is going to do that.

And I'd probably throw someone out of my house for using that die on the glass table. I dunno how much it'd cost to fix a chip in it and I'm not keen to find out.