• Check out the relaunch of our general collection, with classic designs and new ones by our very own Pissog!

Thoughts on a solution to "infinite (time wise)" battles

This has never been an issue for me before

Keyword there. Just because something has not been an issue in the past doesn't mean that it can't become an issue.

It wasn't an issue for me in gen 4 but that doesn't mean anything useful. I had thought about it but it never happened to me before. It happens to me often now.
 
Scofield said:
This has never been an issue for me before

Keyword there. Just because something has not been an issue in the past doesn't mean that it can't become an issue.

It wasn't an issue for me in gen 4 but that doesn't mean anything useful. I had thought about it but it never happened to me before. It happens to me often now.
Click to expand...

So has it never been an issue for you, or has it been an issue for you?

Yes, technically any problem COULD happen, but if they're so rare they happen once every 100,000 battles there's no need to fuss over legislating rules for the most obscure of battle conditions.
 
The fact of the matter remains that the OP blatantly exaggerates the severity of the problem. If it happens about once a month, this battling condition is not common, and it doesn't make much sense for this to be very high on the list of priorities, if it is even on it at all.

Let's not waste Super's and whoever else's time. This occurrence is rare enough that these disputes can be handled manually, whether it be by simply weathering the storm in a ladder match, or having a rematch in a tournament match.
 
Dubulous said:
The fact of the matter remains that the OP blatantly exaggerates the severity of the problem. If it happens about once a month, this battling condition is not common, and it doesn't make much sense for this to be very high on the list of priorities, if it is even on it at all.

Let's not waste Super's and whoever else's time. This occurrence is rare enough that these disputes can be handled manually, whether it be by simply weathering the storm in a ladder match, or having a rematch in a tournament match.
Click to expand...

Once a month per person is actually fairly common; this would indicate hundreds of occurrences per month, and several per day. However, this problem is nearly impossible to the majority of players, since most teams do not have the requisite pokemon to even be in this situation (so the number of occurrences per month is probably in the single digits). I agree that we should not waste any programming effort on this problem, though I think that we should have a policy for moderators to follow if this is ever brought to their attention. I would say just coin flip the game, and if one player refuses to leave he gets kicked!

In a tournament it should be a rematch (though ultimately this is up to the discretion of the tournament staff).

Problem solved?
 
Is that once a month as in "once a month am I in a long PP war" or "once a month i'm in a 2v2 switch-to-stall PP war with zero residual damage on either side and full recovery"?

I haven't battled lately enough to know HOW often the exact "actually infinite" scenario happens (not just long battles like PP stalling), but considering the 400 ways to take damage in Pokemon I really doubt it. No one's produced a single log since I asked for one several posts ago.
 
How common or uncommon this situation is does not change the fact that our ruleset is flawed if it does not deal with infinites. The situation being more or less common simply makes it a lower or higher priority for PR to deal with, but, right now this is the only active thread. We're not so overwhelmed with other pressing matters that we can't sort this out. So lets stop bickering about how often it happens and decide on a solution.

I ask those of you who are reading just the OP's propositions to take a look through the rest of the thread and read the alternate proposed solutions. The second to last paragraph here, deciding the battle by who has the most patience (i.e. make no reference to infinites in the ruleset, people will quit when they get really bored anyway), and the chesslike clause are examples.
 
I agree with TAY's proposal. It's simple to understand, simple to implement, and mechanistically fair. We don't need a mountain of a solution to solve a problem that is a molehill.

Those of you who are confused about whether this problem happens should perhaps read the OP's posts a bit closer. It never happened for him in Gen 4 ("before"), but happened to him twice in one night while playing Gen 5.
 
better idea: give mods a "force tie" button

as long as they don't abuse it, i don't see how this would cause problems

that said, discussing this is largely pointless without the support of the simulator programmer, who would have to implement whatever decision we make
 
I said this on IRC, but I should probably post it here too. I don't know how difficult or not it would be to implement a "force a tie" command for moderators, but regardless, I'd rather not waste any of our extremely limited programming time on something like this. It rarely ever happens, and it's really not a big deal. Here's the solution:

Ladder play: Whoever has more patience or better internet connection wins. If you can convince your opponent to let a third party flip a coin to decide the winner, that's up to you guys. We will not force someone to bank on a coin flip if they want to "play it out". Worst case scenario: you lose one ladder match. Not the end of the world.

Tournament play: Consider the match a tie and play a new match. You may choose a different team or use the same one, up to you (unless the tournament has pre-team selection rules that prevent you from changing). You should probably save the log to verify that this event occurred in case there is a dispute.
 
Why should ladder play not follow the same rules as tournaments in this case? Them mattering less is no excuse for a ruleset which rewards something which is clearly not a part of the skill, or even luck management, of the game (who has to leave the computer first, or who gets bored. PP stall is entirely different since one player will still win.). Whether we have programming resources or not should affect PR's judgments on rules, as you yourself have set the precedent for (Cartridge Sleep clause being decided upon, despite it not being likely to be implemented). Additionally, it would almost certainly not be one of Smogon's programmers or scripters who would implement this (as I have previously stated, when I asked in the PO staff channel two scripters said they thought it was possible with existing functions. There are dozens of scripters across the servers, I am sure one of them is willing to whip up something for this and share it with us.). Dismissing a significant flaw in our ruleset, no matter how commonly or rarely it comes into play, and not attempting to come up with a ruling of how the rules should be for reasons without any competitive foundation (programming time) would be inexcusable in my opinion. At the very least we should discuss solutions, decide on the one which would be best, and if no one steps up to make it a reality put it on the wish list/deal with it as best we can for now.
 
If someone steps up to the plate and is willing to program something that allows moderators to force a tie for cases like this, then I would love that. I was speaking realistically in my post, judging by the lack of programming resources we've had in recent history. Of all the problems we have with our PO server, though, this one literally should be the least of our concerns.

However, should someone want to write up something for this, I'm fine with it being implemented. Speaking as a Tournament Director, I think it's fairly obvious that cases like these are ties (well, more like a standstill, but the results should be the same). My post about ladder results just incorporated pragmatism to the solution. If you're telling me that I'm wrong and that we have people who will do this, then lets get it.
 
Before we get to it, we should probably agree on the best solution. There have been half a dozen ideas brought up, and even though I personally agree that a tie is the only sensible ruling (and that the tie should only be called if the game is actually unwinnable for both players without the foe's co-operation), there have been few posts supporting it. Only Eo's, mine, SDS's (though command rather than button would be simpler), and your most recent show support. And out of those, only my first takes a first stab at wording the clause in a watertight way. I am not 100% certain that it's possible with current scripts, and can't promise that someone will write it, but from what I've heard and know of the scripters in the wider Pokemon community I suspect it can and will be done. Edit: At least a script which prevents the rating changes of a battle allowing either player to quit, even if it does not force an actual tie, is almost certainly possible with current scripts. And, whether or not it will be implemented in the short term, we should make a judgment and decide on the wording of the rule. At least to be sure there will be no arguments about unusual cases in some tournament, and be clear about our ruleset.
 
Yeah, a command would be simpler. In fact, how about something like the tournament-style "/cancelbattle <USER>" option except for something like "/forcetie <USER>".

That said, I'm not familiar with how something like this would be implemented, but again, implementation is the largest hurdle here.
 
CIM said:
I haven't battled lately enough to know HOW often the exact "actually infinite" scenario happens (not just long battles like PP stalling), but considering the 400 ways to take damage in Pokemon I really doubt it. No one's produced a single log since I asked for one several posts ago.
Click to expand...
Okay "fine"

Me vs trickroom. Look at date of upload
http://pastebin.com/LaTq7HnQ

Granted, I somehow managed to predict my way out to finish the match, but honestly, it was ridiculous enough and if it wasn't for that mistake he made at the end it would have been infinite.
 
NetBattle had/has an "Offer Unrated" option during battles, a similar "Offer Draw" option could be implemented, and no damage is done to either players rating or there is a slight drop. However there is the issue of when 2 friends match up on the ladder and just offer draw to make no noticeable negative impact to either players rating. This requires the use of our limited programming resources so it just isn't feasible at this time and so I agree with TAY's proposal and feel that is the best solution for the issue at hand.
 
Really, the solution that gets implemented is ultimately whatever the first volunteer to solve this problem via programming wants to do. So there's not that much more to discuss.
 
I'm definitely opposed to forcing a tie after a particular number of turns. I know that I've done battles that lasted at least 500 turns, when it was a stall team on both sides. The game was never looking like it would be a draw, however. We shouldn't implement a rule with a negative impact greater than the problem it's trying to solve.

We could do something similar to the three-move repetition rule in chess. If a position is identical in every way three times, the game is a draw. This would be a purely objective method that is capable of automation. However, we have to actually measure everything, not just HP. Additionally, it would be nice to have the ability to offer a draw.
 
The problem here is basicly with switching, since it doesn't waste pp. A "switch 10 times in a row"-type clause could also work.

And while we're on the subject of infinites, I'd also like to bring up infinite pp stall Tropius/exeggutor. It forces the same situation of one player switching infinitiely, and if they end up versus each other, it becomes an infinite stall with neither player even capable of ending it. To solve this type of problem, I would suggest banning leppa berry (that is the pp one right lol?).

EDIT: Sorry if I'm not supposed to have posting rights here, I got them with the last suspect test to vote. I actually just followed a link from PO and didn't even realize I was on smogon until I saw my badge lol.
 
Edit:

#issue 1: Well yes that's how infinite battles happen.

#issue 2: that's not the subject of this topic imo, but still i wouldn't like Leppa berry to be banned for something retarded like that. If it's only because the simulator doesn't allow forced ties (except by simultaneous timer runout or spending 1024 turns), then it's the simulator that's at fault and not the game.

I also agree that a big fuss shouldn't be made over it.
 
One of the reasons for banning Wob in Gen III is because it could trigger infinite, unwinnable battles (even if it was still found broken in Gen IV without it), and was pretty much the reason for banning Wynaut. And leppa berry is rarely, if ever, used outside of infinite pp stall harvest pokemon, a source of infinite length matchups which we are trying to prevent (not the only one, but still a source).

But since it is a rare occurance, and pp stall harvest isn't very common, it could be allowed, but we should still address it. Allowing for ties to be made by both competitors agreeing is probably the best option, since it lets us not ban anything and not worry about infinite battles since then they could just draw. As far as no one wanting to tie because they could end up stalling out their competitor, it won't happen (often) because of the opportunity cost of the time wasted stalling.
 
You must log in or register to reply here.
Back
Top