May not be immediately obvious, after rolling, click and drag and it will roll again using your drag to generate the vector along which the dice are rolled.
This seems to be very laggy (on chrome), getting a very low framerate considering the simplicity of the polygons on my fairly beefy machine. I might have been spoilt after seeing the unreal 3 demo on asm.js ...
I rolled a few times, then decided to have a look at the code but unfortunately, between one request and the next, it's now responding with 503 Over Quota, so I guess it's the old HN hug of death...
Anyway - quick question: I noticed you mentioned using random.org, which is great and everything, but would anyone be able to comment on whether or not the physics simulation introduces biases that might reduce randomness?
Like, what if I always swiped from/to particular points on the screen? Does the size of the window matter? How about different framerates? I'm assuming that the random.org data is used to set up (some of the) initial conditions since I can't check the code right now.
I'm guessing that he uses random.org to randomize the initial values of the dice, how they're "being held". All the physics simulation does is roll them across the table.
So it should be impossible to precisely throw a value, since you can't control how you're holding the dice in your hand.
I have moved heavyweight libraries into separate storage. Quotas were exceeded by outgoing traffic, but statistics are counting every hour and now it works.
Do you think you could add one really silly little feature? An option to score by counting the number of dice which roll over some threshold, rather than summing the rolls. Defaulting to 5+, but adjustable. Perhaps by adding a ">5" to the end of the "NdM" string. This would make this actually really useful for playing Arkham Horror.
If you could add super eldritch skins to the dice models, that would make it really, really useful for playing Arkham Horror.
28 comments
[ 0.18 ms ] story [ 77.6 ms ] threadEdit: nvm figured it out this relies on WebGL.
I rolled a few times, then decided to have a look at the code but unfortunately, between one request and the next, it's now responding with 503 Over Quota, so I guess it's the old HN hug of death...
Anyway - quick question: I noticed you mentioned using random.org, which is great and everything, but would anyone be able to comment on whether or not the physics simulation introduces biases that might reduce randomness?
Like, what if I always swiped from/to particular points on the screen? Does the size of the window matter? How about different framerates? I'm assuming that the random.org data is used to set up (some of the) initial conditions since I can't check the code right now.
Thanks!
So it should be impossible to precisely throw a value, since you can't control how you're holding the dice in your hand.
Do you think you could add one really silly little feature? An option to score by counting the number of dice which roll over some threshold, rather than summing the rolls. Defaulting to 5+, but adjustable. Perhaps by adding a ">5" to the end of the "NdM" string. This would make this actually really useful for playing Arkham Horror.
If you could add super eldritch skins to the dice models, that would make it really, really useful for playing Arkham Horror.
I can't currently use Chrome on my computer because they don't support high DPIs :(
https://wiki.mozilla.org/Blocklisting/Blocked_Graphics_Drive...