[April Fool's] Wesnoth Random Number Generator Changes

[Ken_Oh]

My biggest concern is possibility of a Wesnoth player gaming the system by controlling how many web pages are on the net. Kind of like Google bombing, a malicious player might figure out how to Wesnoth bomb the whole internet.

[ivanovic]

My brother is a huge fan of Wesnoth, but he doesn't have an internet connection. Does this mean that he wouldn't have any noticeable changes with the RNG, because he would still be using the old one?

For the moment: Yes. We might find some way around this limitation, though this will required lots of additional work. For the moment you will definitely get (almost) "perfectly random" random numbers when playing on the 1.8.x multiplayer server.

My biggest concern is possibility of a Wesnoth player gaming the system by controlling how many web pages are on the net. Kind of like Google bombing, a malicious player might figure out how to Wesnoth bomb the whole internet.

Oh, it is not just the *number* of websites. The algorithm is really difficult to explain, but it is based on the data gathered by the google crawlers. So the *content* of the websites does also matter as well as the time of access and some other factors. It is not just a plain thing of relying on the number of websites or the amount of data.

[Gambit]

Finally.

[Sangel]

To be honest, I have my doubts about this project. While the Wesnoth RNG may not be "perfectly" random, I've seen no evidence that leads me to believe that this has a negative effect on gameplay. Furthermore, I see no reason to believe that improving the percentage accuracy of the RNG will do anything to stall the complaints of those who choose to believe that the RNG is somehow rigged against them.

Most people who take issue with the RNG (as opposed to the "outsize role of luck in determining battles", which is a game design complaint, rather than an RNG complaint) take issue with it on the grounds of single-player; that the AI is somehow "cheating" or unnaturally lucky. This issue does not apply to multiplayer.

While having superior random numbers in multiplayer games can only be a good thing, I don't think it will address the complaints that it is designed to counter.

[Schierke]

This is odd... Despite reading about the complains i never felt the RNG in Wesnoth worked so poorly, and as a mostly campaing player i can't say i feel too good about a (at least for now) multiplayer only solution.

But still, if the old RNG didn't seem so bad to me i guess it won't really be a problem for me, and if i decide to play some MP maybe i will get a better experience .

EDIT: Damn it, forgot the date lol.

[artisticdude]

To be honest, I have my doubts about this project. While the Wesnoth RNG may not be "perfectly" random, I've seen no evidence that leads me to believe that this has a negative effect on gameplay. Furthermore, I see no reason to believe that improving the percentage accuracy of the RNG will do anything to stall the complaints of those who choose to believe that the RNG is somehow rigged against them.

My thoughts exactly.

[CragHack]

I have the feeling, the RNG works extraordinarily bad today. Yesterday it looked still ok for me. Let's see how the random numbers are going to be tomorrow... Somehow, i'm confidential

[Radament]

Funny, I had a similar idea with Netscape Navigator, Google never occured to me though. Amazing.

[Captain_Wrathbow]

So, to use this new RNG, we will have to be constantly connected to the internet to get the random numbers?
What about someone with a frail internet connection that disconnects every 15 minutes? (actually not describing me, I'm just wondering)

[Gambit]

So, to use this new RNG, we will have to be constantly connected to the internet to get the random numbers?
What about someone with a frail internet connection that disconnects every 15 minutes? (actually not describing me, I'm just wondering)

Well they could always pirate/crack wesnoth, or play an older version.

[Unnheulu]

The reason it is not as good today is the date is 1/4. This represents the fact that all rng results will be a quarter of what is optimal for the player. I'm sure if you try again in three days it will be significantly better as it is 4/4, or a whole. So the rng results will be multiplied by 1. For the best results however, we recommend playing on the last day of each month, as then you get the highest multiplier.

[Gambit]

Have any numbers been thrown around for the price of the Premium RNG Service?

[jb]

Just to be clear, this new RNG doesn't effect the actual % of a unit's chance to hit. A mage with 70% chance to hit will still have 70%.

This new RNG will (in theory) make sure that 70% is a more accurately calculated.

Yes, the new system is a bit slow, but I'm sure it will be optimized eventually.

[Dave]

I didn't expect that because PROT_NONE sounds like a bad name to me - i would expect it to read as "no protection" .

Remember that C doesn't have namespaces, so the PROT_ part effectively acts as a namespace. It's a general C convention that if you have multiple values of the same type, they all start with the same prefix.

Have any numbers been thrown around for the price of the Premium RNG Service?

We are thinking of having several levels. For 99.99% accurate numbers you can pay $9.99 a month, for 99.9999% accuracy, you can pay $19.99 a month. We'd give a first month free trial to all users.

Though I'm not sure. It'd be great if someone could start a poll to see how much people would be willing to pay for good quality random numbers.

David

[doofus-01]

Finally! Premium service! Will this be extended to the i-Phone?