Does your browser influence the game's RNG? Is it possible?

[Wyand]

The Simulated Violence competition brought some controversy and made me think: what if your browser also
affects the game's rng, not just bripod's javascript? I don't know how does these things work with Flash,
is it entirely server-side or not, so I'm awaiting the possible explanations by more experienced users.

[iancudorinmarian]

Paranoid people / people who can't accept a loss. I will explain a bit later how RNG actually works.

[dawn to dusk]

The EtG Sim differences have been known since back in the days of Computer Sim "PvP" event

See this post: http://elementscommunity.org/forum/pvp-events/pvp-event-computer-simulations-rules-(discussion)/msg1115557/#msg1115557

EDIT I have results for you all
NOTE THAT ETG SIM DOES NOT USE FLASH. This is very important to clear up any allegations that it's "better" to use any browser.

Firefox, specifically, screws up the RNG in the Sim. I went into single games to see what could have been messing up this RNG and this is what I found

Spoiler for Hidden:

Looking at the AW, the chances of this happening is 0.0001%. Now sure, maybe it was a fluke and somehow I actually did get this.

Spoiler for Hidden:

Here it is with Precog

Spoiler for Hidden:

Here's another one with Precog.

And you get the point. For some reason, Firefox screws this up, which is why the difference is shown so much. This never happens with Brimstone Eater, Gravity Tower, Steam Machine, etc. It only happens for the last 12 cards in the deck. Not sure on decks over 30 cards, someone can do that testing if they wish.

[iancudorinmarian]

The EtG Sim differences have been known since back in the days of Computer Sim "PvP" event

See this post: http://elementscommunity.org/forum/pvp-events/pvp-event-computer-simulations-rules-(discussion)/msg1115557/#msg1115557

That's very weird. Has no one got any explanation to it?

EDIT: Found this: https://lwn.net/Articles/666407/

Although even having said that, the range of values is still high enough to not matter. I doubt EtG or the simulator try to generate random numbers over 4.2 billion...

I think it's been patched? As far as the article suggests anyway.

[dawn to dusk]

Edited post, see the results there. I can only see the trend, and have no idea why it actually does this. It occasionally does get real hands drawn, but a lot of the times you will see clumps as shown above

EDIT: Tested Random.org, the numbers there are truly random. EtG Sim is just weird

[iancudorinmarian]

Ran like 20+ tests with mono life vs mono darkness and I've been getting normal results.

Spoiler for mono life:

Hover over cards for details, click for permalink

Deck import code : [Select]

5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bs 5bu 5bu 5bu 5bu 5bu 5bu 5c2 5c2 5c7 5c7 5c7 5c7 5c7 5c8 5c8 5c8 5c8 5c8 8pn

Spoiler for mono darkness:

Hover over cards for details, click for permalink

Deck import code : [Select]

5uk 5uk 5uk 5uk 5uk 5ul 5ul 5ul 5ul 5ul 5ul 5um 5um 5um 5um 5um 5um 5up 5up 5up 5up 5ur 5us 5us 5us 606 606 606 606 606 8pt

[iancudorinmarian]

I get normal results with a vast variety of decks on this version:



Sure, odd draws do happen, but not often enough to be considered rigged.

[Wyand]

Paranoid people / people who can't accept a loss. I will explain a bit later how RNG actually works.

I must add that originally it wasn't my idea, I just posted it, because it made me curious about RNG.
(Fun fact: the only truly random thing in the world is radioactive decay, i.e. which will be the next atom to decay
in a radioactive material.)

[kaempfer13]

quantum tunneling is also random to our knowledge. So rngstuff should be much easier to generate once quantum computers become the norm.

[Wyand]

Thanks kaempfer for mentioning that. I was thinking about that too, but wasn't entirely sure about it.
(My mind might had an uncertainty moment, ha-ha. )