Chris
Level: Master Delver
Rank Points: 144
Registered: 08-12-2005
IP: Logged

Since the new snake pit is so much bigger than the last one, I've made a new smaller size copy that allows it to just about fit widthwise onto a 1024 pixel wide screen:

tiny, small, medium, and large

As before, hover the mouse over a snake's head to see its name and number.

techant
Level: Smitemaster

Rank Points: 854
Registered: 06-08-2004
IP: Logged

ErikH2000 wrote:

Pinnacle wrote:
Why are we moving so quickly?
EDIT: Six dead snakes (not on a plane)

I'm just testing. The snakepit will be reset at the contest start time, (Local Time:08-18-2006 at 01:00 AM) which as I write this is about an hour away.

-Erik

Heartattack
I had made the the snakepit my home page so when I got on my computer to check another webpage I freaked out when I saw I was dead and we were at turn 154 or something. I had to finish what I was doing all the time thinking I had messed up some how.. Glad it was only a test. (If you were trying to motivate us to pay attendtion it worked)

I forgive you Erik.
Please warn me next time.

____________________________
DROD RPG Released!! You must try it out it is Great!!

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

techant wrote:
I forgive you Erik.
Please warn me next time.

Sorry! I didn't think people would be paying attention in the hour or so I was messing with it.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

Okay, the second snakepit is up on the Snake Page. And I'm not testing anymore, so the commands you put in will be applied to the actual game.

Because of the snakepit's layout, no decisions will be possible until 24 hours have elapsed. This is so people in different timezones who are sleeping right now aren't put at a disadvantage. Niccus, Maurog, and Banjooie (resurrectees) have been placed in less than favorable positions. I encourage them to take a good look at their starting condition in the first 24 hours.

Except for the resurrected snakes, I believe the other positions are fairly equal in terms of relative advantage. If you don't like a particular subarena, you can probably choose a different one and plan a course there. It is still quite probable that some positions are better than others. This game isn't guaranteed to be fair; I've done the best I can. There would have to be a really terrible bug affecting many players for me to reset the game, but if you see something that looks wrong please report it, and I'll see what I can do.

There are no rules regarding what you can talk about, but I think most players would rather you not give out information that could affect future movements. Expect the atmosphere to get much more competitive!

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

Regarding what Coppro said, (and thanks for deleting it):

The disadvantage of the corner chambers is not what's inside of them. As you point out, a bump-only exit is possible. The corner chambers are actually red herrings to get the snakes to think they've passed the part that was intended to be hard.

It should be clear at some point to players that the middle area is a good place to be because of all the empty space there. The corner snakes have a shortcut to the middle area, but it will just send them back to the outside. So it is actually a time-wasting trap, made possibly fatal if multiple snakes decide to enter the central chamber simultaneously. The corner snakes will already have a disadvantage getting to the center because the other snakes will have had first route choice in the subarenas. So the corner snakes will have to work harder to avoid getting trapped in a congested subarena.

...or at least that is how I intended the setup to work. We will see what actually happens.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

---

[Last edited by ErikH2000 at 08-18-2006 12:49 AM]

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

Regarding Rabscuttle's position (again Mikko thanks for deleting):

Rabscuttle has actually been placed in front of a trap. It will be interesting to see if he goes into the middle area. If he dies in that chamber, he'll be mad at me I'm sure. I really like that trap though, and Rabscuttle's pretty smart so he's likely to figure it out.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

---

[Last edited by ErikH2000 at 08-18-2006 12:54 AM]

Chris
Level: Master Delver
Rank Points: 144
Registered: 08-12-2005
IP: Logged

I made some animations of how the snake pit plays out when all the snakes randomly decide whether to jump or bump at each step.

I don't think this is spoiling anything for anyone, but if you don't want to see them, please don't click the link.

Schik
Level: Legendary Smitemaster

Rank Points: 5423
Registered: 02-04-2003
IP: Logged

Chris wrote:
I made some animations of how the snake pit plays out when all the snakes randomly decide whether to jump or bump at each step.

I don't think this is spoiling anything for anyone, but if you don't want to see them, please don't click the link.

So.... who won?

____________________________
The greatness of a nation and its moral progress can be judged by the way it treats its animals.
--Mahatma Gandhi

Chris
Level: Master Delver
Rank Points: 144
Registered: 08-12-2005
IP: Logged

Schik wrote:
So.... who won?

I did of course! I expect.

It's kind of hard to tell, because I used YouTube to host the files, and they've gone all fuzzy.

But I expect it was me.

michthro
Level: Smitemaestro
Rank Points: 679
Registered: 05-01-2005
IP: Logged

I take it the snakes move in the order they're listed, and not according to their starting positions?

P.S. If so, and if no-one enters any moves, coppro wins.

jbluestein
Level: Smitemaster

Rank Points: 1670
Registered: 12-23-2005
IP: Logged

As far as the resurrected snakes go, could we call them something else? I like the term 'zombie snakes'. Or even better, 'brain-devouring zombie snakes'.

'Braiiinnnsssssssssss'



____________________________
"Rings and knots of joy and grief, all interlaced and locking." --William Buck

Mattcrampy
Level: Smitemaster

Rank Points: 2388
Registered: 05-29-2003
IP: Logged

I recall someone asked what the dead players thought:

The game either moves too fast or too slow. I can only see about half a day's moves ahead with other snakes in the pit at the current time - beyond that, it's pointless to put moves in because the other snake may do something you don't expect. That, and my strategy skills leave a lot to be desired.

I honestly had little idea how I was going to make it to the other side, finally worked out I needed to come at that square mentioned from the north, and then wasn't able to touch the internet for three days.

So, sadly I leave quite unsatisfied with my experience. I'd be happy to play with a turn cycle of about 10 seconds or so, but any longer and the game would get quite tiresome.

____________________________
What do you call an elephant at the North Pole?

jbluestein
Level: Smitemaster

Rank Points: 1670
Registered: 12-23-2005
IP: Logged

The mere fact that Rabscuttle's position is comment-worthy is interesting enough. I have some thoughts on it, but it looks like we're not discussing stuff too overtly yet.

It's a very interesting board, though.



____________________________
"Rings and knots of joy and grief, all interlaced and locking." --William Buck

Citrus
Level: Smitemaster

Rank Points: 1228
Registered: 01-08-2006
IP: Logged

All I can say is, Lucky him.

____________________________
Call me Citrus.

Niccus
Level: Smiter
Rank Points: 308
Registered: 07-02-2006
IP: Logged

Decisions decisions, get out as early as possible or use the time to coil away safely?

Gotta wait 12 more turns and see, I guess.

michthro
Level: Smitemaestro
Rank Points: 679
Registered: 05-01-2005
IP: Logged

hmmmm.. I know there was always going to be some unfairness, but my starting position looks really crappy. Maybe I'm missing something, but I'd much, much rather be in for instance Rabscuttle's position. I don't even have the option of playing BJJJ, since I can't count on Timo006 to do the same, and if he doesn't I'm dead. In my case that probably means bad news. (I'm not going to give away all sorts of ideas and tips by going into it). The only other player with a similar situation is jamesdenem, but he at least has three other snakes that can save him if he jumps. On top of that, I can't wait and see what the other players do at the first decision point (unless I get up somewhere between 3:00 and 5:00 AM, which isn't worth it). Oh well, guess I'll have to wait and see. Maybe it turns out to be a good position after all. Sorry for the bellyaching, but, you know, I'm a little disappointed with my starting position.

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

michthro wrote:
hmmmm.. I know there was always going to be some unfairness, but my starting position looks really crappy. Maybe I'm missing something, but I'd much, much rather be in for instance Rabscuttle's position.

Rabscuttle's starting position is possibly the worst one in the game. Anyhow, all I can say now is that people shouldn't worry too much about Rabscuttle, other than the cunning and sadism he brings to every game he plays.

I don't even have the option of playing BJJJ, since I can't count on Timo006 to do the same, and if he doesn't I'm dead.

It's not all about making those 3 jumps. You can use the perimeter tracks to move into a better area. Unless I misunderstand how the pit will unfold, everyone except the 3 corner snakes should be able to make choices on which subarena they enter.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

michthro
Level: Smitemaestro
Rank Points: 679
Registered: 05-01-2005
IP: Logged

ErikH2000 wrote:
Anyhow, all I can say now is that people shouldn't worry too much about Rabscuttle...

Well, let me put it this way: Being guaranteed of making it into the central area is a clear advantage. (Unless, obviously, there is a catch, which I don't see. In particular - and Rabscuttle doesn't need me to tell him this - I'm not suggesting that he should necessarily enter that area as soon as he possibly can.) I can't *prove* it, but there is a very strong intuition here. It's impossible to completely analyse the game and establish which snakes have the theoretical advantage (or afaik no-one has done that, anyway), so all we have to go on is intuition, heuristics, rules-of-thumb, call it what you like.

Besides, you probably could at least prove some statements to the effect that it is an advantage, such as:
At most n snakes will reach the central area, however the game goes. (Where n is relatively small.)
The winner will almost surely have reached the central area at some point (which is a statement that can be formalised in terms of a probability).

Oh, and it's not about Rabscuttle. I don't begrudge him anything, although he is the last player who needs any advantage. It's coincidence, perhaps unfortunate, that it happens to be him who has that nice juicy position. I'm only using that position as an extreme example of the contrast between some positions.

As for using the perimeter, well, it's true what you say, but that doesn't change the fact that I'd much rather be able to get straight into the center, or at least get into some open space, where I have more decisions to make, less chance of getting blocked, and less chance that by the time I finally get away from the side the outer areas and entrances to the central area are already pretty much filled up. (Some more heuristic ideas, which, as I said, is all I have this early in the game.)

---

[Last edited by michthro at 08-18-2006 11:12 PM]

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

Okay, I actually did screw up with Rabscuttle in my haste to get the 2nd snakepit up and running. !#&@%

I hate doing this, but I'm going to change one square on the arena so that Rabscuttle can't get to the middle. The square that will change will be the single box that has a subarena on its left and the middle area on the right.

Sorry, sorry, sorry.

But that's it. All the other unfairness present in the snakepit will just have to play out.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

Ugh, I want to mod myself down. Let's hope I didn't screw up too many other things. I am definitely done making witty predictions in unhidetime tags. But I'll leave those ones there so you can see how deluded I was.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

eytanz
Level: Smitemaster

Rank Points: 2708
Registered: 02-05-2003
IP: Logged

Moving away from discussions of particular players and into the abstract, I'm rather curious as to the inherent fairness of this game. I know Erik doesn't guaruntee anything and that's fine - but there's a meta-puzzle here which is interesting in its own right.

I think it's pretty likely that at least one or two of the serpents in the pit right now cannot win no matter what commands anyone types in. Note that on only very few moves does it matter what the command is - There has to be a jumpable obstacle ahead of a serpent and an empty square to its left or right, or a non-jumpable obstacle ahead of it, a jumpable obstacle to it's right, and an empty square to its left. In any other configuration, the moves don't make a difference. And of course, in the long run it may prove that many decisions have only local effect and do not affect the big picture.

I'm wondering if Chris or someone else who coded their own programs to run the arena could create a version that goes through every possible outcome. It may be interesting to see how biased this board actually is. Maybe I could try doing it myself but it's been 3 years since the last time I touched code and I feel pretty rusty.

____________________________
I got my avatar back! Yay!

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

eytanz wrote:
It may be interesting to see how biased this board actually is.

I was worried about the board being biased too. I made the outer perimeter rings of boxes as an alleviation for bad starting positions. It seems to me that they open up possibilities for individual snakes.

Some subarenas (the walled off areas just inside the perimeter) are definitely deadlier than others. A subarena with a hooked wall like the one in front of Michthro is an example. Or simply being in an overcrowded subarena with lots of snakes is bad business.

In the simulations I ran, I found that even when snakes were randomly moving, they had frequent opportunities to move to different subarenas. And it wasn't too difficult to get to the middle either. So generally, its not too hard to get out of a bad area.

If snakes had perfect ability to both know the best area and get there, then the value of being in each area would eventually equalize. (Except for areas which guarantee death, which in this snakepit are few.) So if you are in an area that is the best place to be for one snake, its value will decrease as more snakes enter it. When more snakes are present they gobble up empty squares that you want your snake to fill, and also mess up your planned paths when their behavior is incorrectly predicted. If any place is really good to be, more snakes will move into it and make it not so good. Even the worst subarena on the board is the best place to be if you are there alone with all that coiling space to yourself. Players don't have this perfect ability to know where to be and get there, but I suspect that the abilities people do have will be enough to keep the game reasonably balanced.

I could be very wrong about all of this, but that's my theory of how this snakepit design will work.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

eytanz
Level: Smitemaster

Rank Points: 2708
Registered: 02-05-2003
IP: Logged

Well, what I'm curious about is whether there are any snakes that cannot win, no matter what they do. You won't be able to tell impressionistically like you do - it could be that there's a serpent that can reach a lot of strategically good spots, but it so happens that he'll always die later anyway.

____________________________
I got my avatar back! Yay!

ErikH2000
Level: Legendary Smitemaster

Rank Points: 2794
Registered: 02-04-2003
IP: Logged

eytanz wrote:
Well, what I'm curious about is whether there are any snakes that cannot win, no matter what they do. You won't be able to tell impressionistically like you do - it could be that there's a serpent that can reach a lot of strategically good spots, but it so happens that he'll always die later anyway.

I strongly doubt it, but I can't really prove it. (Okay, it's probably possible to prove, but...) I think the possibilities open up too quickly at the beginning of the game, and every player will have at least a theoretical chance to win. Look at it this way: If you played any of the 43 snakes to win, and then purposefully maneuvered the other 42 to early deaths, then you could rig a win. I'm not saying that's a proof, but intuitively (to me anyhow) it seems it would not be hard for each snake to have at least one possible way to win the game.

-Erik

____________________________
The Godkiller - Chapter 1 available now on Steam. It's a DROD-like puzzle adventure game.
dev journals | twitch stream | youtube archive (NSFW)

Chaco
Level: Smitemaster
Rank Points: 3659
Registered: 10-06-2005
IP: Logged

Decisions will be made in less than 20 minutes!

____________________________
Quick links to my stuff (in case you forgot where it was):

eytanz
Level: Smitemaster

Rank Points: 2708
Registered: 02-05-2003
IP: Logged

File: beforemove.PNG (102.1 KB)
Downloaded 285 times.
License: Public Domain

Heh - I went to the snake page around the time I thought the move will happen to see what happened, and got this:



Good timing, I guess

____________________________
I got my avatar back! Yay!

---

[Last edited by eytanz at 08-19-2006 02:03 AM]

Rabscuttle
Level: Smitemaster

Rank Points: 2502
Registered: 09-10-2004
IP: Logged

File: dangit.JPG (47.1 KB)
Downloaded 275 times.
License: Public Domain

Aw.

Heh, I'm also just too late to loop around in the outer perimeter. Serves me right for sleeping in until 2pm.

---

[Last edited by Rabscuttle at 08-19-2006 06:10 AM]

techant
Level: Smitemaster

Rank Points: 854
Registered: 06-08-2004
IP: Logged

I don't know about the rest of you but I am having a really fun time... win lose or draw this contest is a blast Erik

____________________________
DROD RPG Released!! You must try it out it is Great!!

michthro
Level: Smitemaestro
Rank Points: 679
Registered: 05-01-2005
IP: Logged

Sorry, Rabscuttle. Somehow I think you'll still be ok, though.

Now that so many people chose to bump (which I think was a mistake in some cases), my position looks just as good as anyone else's, so I'm happy. From here on the game seems pretty fair to me.

eytanz wrote:
Moving away from discussions of particular players and into the abstract, I'm rather curious as to the inherent fairness of this game. I know Erik doesn't guaruntee anything and that's fine - but there's a meta-puzzle here which is interesting in its own right.

This game is fascinating from a game-theoretical point of view (and a blast from a "practical" point of view). On the face of it, it's not a full-information game, since two or more snakes may have a decision to make on the same turn, but I suspect that even so there is a snake with a winning strategy. Unfortunately, I doubt that it's possible to solve your meta-puzzle. It will definitely take a lot more than a brute-force search.

I think it's pretty likely that at least one or two of the serpents in the pit right now cannot win no matter what commands anyone types in.

This is a question that can be settled with a little effort, assuming that the anser is no. I like Erik's outlined proof, and suspect it will go through.

Note that on only very few moves does it matter what the command is...
.
.
I'm wondering if Chris or someone else who coded their own programs to run the arena could create a version that goes through every possible outcome. It may be interesting to see how biased this board actually is. Maybe I could try doing it myself but it's been 3 years since the last time I touched code and I feel pretty rusty.

It's true that on only relatively few moves does it matter what the command is, but even so, the number of possibilities is astronomical. Let's say 20 snakes get to make 10 decisions: That's (2**20)**10, which is in the order of 10**60, possibilities, which is hopelessly too many.

Maurog
Level: Smitemaster

Rank Points: 1501
Registered: 09-16-2004
IP: Logged

Some fun facts about the snake pit:

The snake pit area is 124 x 124 tiles = 15376 tiles.
The snake pit has 1709 walls and 2271 boxes for a total of 3980 obstacles.
This leaves 11396 free tiles for the 43 snakes to roam in.
Assuming that all snakes remain alive, they can exhaust this area in 265 turns.

With 19 moves already taken at this point, this leaves 246 moves.
246 moves at 1 move per 2 hours is 492 hours or 20.5 days.

Note that the snakes may all die without exhausting the area, finishing the game sooner. However, for each dead snake, there will be less area claimed each turn, theoretically expanding the total play time.

Happy snaking!

____________________________
Slay the living! Raise the dead!
Paint the sky in crimson red!