Curse

Gill Halendt · Jun 15, 2010, 11:40 AM // 11:40

Quote:

Originally Posted by lemming

Not really, the number of teams being formed every 30 second interval will still be the same.

Sure, but the wait could likely increase for some people not being assigned to any of the teams in the batch ready to start.

Martin Alvito · Jun 15, 2010, 12:36 PM // 12:36

Quote:

Originally Posted by Gill Halendt

Sure, but the wait could likely increase for some people not being assigned to any of the teams in the batch ready to start.

This will not affect the average time necessary to get a match by much.

It will increase the variance in your wait times. In RA, you would hardly notice the change. Players in FA and AB would notice that sometimes they match immediately and sometimes wait much longer than normal.

Killed u man · Jun 15, 2010, 12:36 PM // 12:36

Then add a counter:

When Timer Reset <= 3, "Priviledged".

Priviledged here meaning you WILL get placed on the next reset. Easy doable, can not be abused as you can not choose to get a reset, and makes sure you have no more than 3 resets.

Gill Halendt · Jun 15, 2010, 12:52 PM // 12:52

Quote:

Originally Posted by Killed u man

Then add a counter:

When Timer Reset <= 3, "Priviledged".

Priviledged here meaning you WILL get placed on the next reset. Easy doable, can not be abused as you can not choose to get a reset, and makes sure you have no more than 3 resets.

This could do, some form of priority in the queue.

Amy Awien · Jun 15, 2010, 01:59 PM // 13:59

Quote:

Originally Posted by Martin Alvito

This will not affect the average time necessary to get a match by much.

That was not the point, which was that true randomness will mean that sometimes people will have to wait much longer.

Quote:

Originally Posted by lemming

Not really, the number of teams being formed every 30 second interval will still be the same.

Random does not take into account that a player was not selected for the previous batch of teams, nor will it take into account that they weren't in the teams before that, or those before that. If you pick completely randomly, there will be people who are going to have to wait very long.

Quote:

Originally Posted by Killed u man

Then add a counter:
When Timer Reset <= 3, "Priviledged".

That might work, even though some form of non-randomness returns, I don't see how syncing would be workable.

Martin Alvito · Jun 15, 2010, 03:55 PM // 15:55

Quote:

Originally Posted by Amy Awien

That was not the point, which was that true randomness will mean that sometimes people will have to wait much longer.

Which I acknowledged. Look up what "variance" means.

But does it matter? Just because the claim is true does not mean that the effects are meaningful.

Quote:

Originally Posted by Killed u man

Then add a counter:

When Timer Reset <= 3, "Priviledged".

Priviledged here meaning you WILL get placed on the next reset. Easy doable, can not be abused as you can not choose to get a reset, and makes sure you have no more than 3 resets.

That would be a disaster in arenas like FA and AB where people already have long waits. Eventually, it would be a disaster in RA as well once the player base becomes small enough.

Lemming's solution is better without this modification than with.

Killed u man · Jun 15, 2010, 04:29 PM // 16:29

Quote:

Originally Posted by Martin Alvito

That would be a disaster in arenas like FA and AB where people already have long waits. Eventually, it would be a disaster in RA as well once the player base becomes small enough.

Why? I really don't see why it's a "disaster"? How will this make it slower for anyone? We're assuming the same amount of teams are forming per given time unit, so the same amount of people are going in.

I really don't know if you don't understand my suggestion, or you don't understand the idea behind it.

All this would do is that when people got "unlucky" for 3 timers (3 resets), is give them priority over other people. This can in no way get abused, even when the playerbase is low. It also wouldn't make anything slower is disastrous because if the player base is so low you can't form matches, you won't be able to form them no matter how you select people.

Again, I really don't understand what the issue with this is, elaborate?

Martin Alvito · Jun 15, 2010, 05:45 PM // 17:45

A lot of times, it already takes 10+ timers to get a match in FA/AB.

Uninformed players that don't understand the system will not be able to get a match if we switch things to the system you propose. Things would remain the same for you and me, because we would know to cancel after three timers and restart, but it would wreck the casual player.

If you meant >= 3, you should edit the above post.

Gill Halendt · Jun 15, 2010, 06:53 PM // 18:53

Quote:

Originally Posted by Martin Alvito

If you meant >= 3, you should edit the above post.

Guess that's it: he is suggesting some prioritization for players not being assigned to any random team for three times in a row or more, so that they can at least be sure to be in after the fourth timer has expired.

Killed u man · Jun 15, 2010, 08:29 PM // 20:29

Yeah, meant >=3, srry typed brain worked faster than my fingers there.

Ok, fully my bad, you had a valid point, and I eff'd up. I meant >=3

Martin Alvito · Jun 15, 2010, 08:42 PM // 20:42

If that's the case, then it's a good addition to Lemming's proposal. While it doesn't fix the problems the proposal introduces, it minimizes them - especially in the current RA.

I'd support that change.

Amy Awien · Jun 15, 2010, 09:47 PM // 21:47

Quote:

Originally Posted by Martin Alvito

Which I acknowledged. Look up what "variance" means.

Why don't you look up what "getting bored waiting and doing something else" means. I've read your post and concluded from it that you did not understand the effect it might have on people. Waiting too long is a negative personal experience - possible resulting in players moving to another game. The term 'variance' does not cover the actual meaning and effect of waiting too long.

Quote:

But does it matter? Just because the claim is true does not mean that the effects are meaningful.

Does it matter? Do people matter? Or how they percieve the game. How long will they wait before they go do something else.

*lemming* · Jun 15, 2010, 10:38 PM // 22:38

Quote:

Originally Posted by Amy Awien

Why don't you look up what "getting bored waiting and doing something else" means. I've read your post and concluded from it that you did not understand the effect it might have on people. Waiting too long is a negative personal experience - possible resulting in players moving to another game. The term 'variance' does not cover the actual meaning and effect of waiting too long.

Does it matter? Do people matter? Or how they percieve the game. How long will they wait before they go do something else.

I don't think you get how my suggestion works.

Martin Alvito · Jun 15, 2010, 11:30 PM // 23:30

I think he understands how the suggestion works just fine.

I just think he gets a little emotionally upset about the possibility that players might have to wait a bunch of timers to match every so often. Of course, that happens in plenty of formats already.

Never mind that the number of timers players wait would be reduced on a regular basis as well. Sure, the mean number of timers you wait would increase a bit. Just not enough to matter.

He's just being stereotypically American.

Tullzinski · Jun 16, 2010, 01:22 PM // 13:22

Quote:

Originally Posted by Martin Alvito

He's just being stereotypically American.

Hey not all americans want everything this instant! Some of us like to wait a few seconds for stuff...sometimes

Amy Awien · Jun 16, 2010, 02:47 PM // 14:47

Quote:

Originally Posted by Martin Alvito

He's just being stereotypically American.

Why is everyone of your posts decorated with that personal touch that is sometimes referred to as flamebait? Is there a reaction you are trying to get?

Quote:

Never mind that the number of timers players wait would be reduced on a regular basis as well.

It don't work that way, you're dealing with people, they're not going to keep notes on the waiting times and work out the average. They're just gonna be annoyed when they have to wait too long.

Quote:

Originally Posted by lemming

I don't think you get how my suggestion works.

I understand it just fine, you just seem to not quite understand what random means.

Please look at the suggested modification and try to figure out why just about everyone thinks that might be a huge improvement to your idea.

FoxBat · Jun 16, 2010, 02:59 PM // 14:59

The problem with Lemming's suggestion is that's how the game currently works. Modes are still syncable because there's that few people joining in the 30 second window anymore.

Possible exception for people joining in the T-2 seconds window. That's a loophole people have been using to increase sync chances in the merged district era and should be looked into. People joining in that time frame should just get a NOP and be grouped in the next 30-second go. Upping the wait to 1 minute might also help.

Martin Alvito · Jun 16, 2010, 03:19 PM // 15:19

@ FoxBat: Go get four or five players and sync an 8-man format such as JQ or FA. It will quickly become obvious that it's filling in order from the queue. When you land a full sync, you will be adjacent. When you are split, you will repeatedly be split into the same places in the party.

Quote:

Originally Posted by Amy Awien

You're just having a stereotypically oversized ego which can't admit having overlooked the obvious.

I've agreed with you on the facts from the start, which you don't seem to comprehend. We're disagreeing about the implications.

Suppose that 49 people click "Enter Battle" every RA timer. That means that players get left out in the cold for four cycles.

The odds of requiring four consecutive match timers to match would be 1/49 * 1/25 * 3/51 = 1/20825. The odds of needing three timers (given the assumptions) are 1/1225.

You'll hardly feel the change. Once in a very great while, you'll have to wait a while to get a match. If you're incredibly unlucky, it might be a very long time. But you're highly unlikely to have two long waits in the same week or month.

If you still think that's a big deal, then you really are being stereotypically American in your impatience. If you're a heavy RA player, you'll lose a few minutes of your life every month to match timers. But you'll almost certainly earn more than enough points to compensate unless you're currently syncing.

Amy Awien · Jun 16, 2010, 04:20 PM // 16:20

Quote:

Originally Posted by Martin Alvito

I've agreed with you on the facts from the start, which you don't seem to comprehend. We're disagreeing about the implications.

That's all fine, I have no problem with open discussion, but I am fed up with your childish, personal additions towards me. Please grow up.

Quote:

Suppose that 49 people click "Enter Battle" every RA timer. That means that players get left out in the cold for four cycles.

Where are your numbers coming from? Do you have access to ANet's data? How many players press 'Enter Battle' every 30 seconds, how many teams are formed? With those data you can calculate what the chances are of getting in, or not getting in a team.

Do you realize with the amount of players pressing 'enter battle' and the time-span we're dealing with, 1:1000 is a pretty big chance

Quote:

You'll hardly feel the change.

Without the data it's guessing. But consider this, why would ANet not have randomized it long before and in stead chosen a selection algorithm that apparantly makes syncing possible?

Quote:

If you still think that's a big deal, then you really are being stereotypically American in your impatience.

There you go again, "if you think Y then you must be X. When will you grow up and start discussing issues rather then calling people names when they disagree with you.

Btw, how old are you?

Martin Alvito · Jun 16, 2010, 05:50 PM // 17:50

Quote:

Originally Posted by Amy Awien

Where are your numbers coming from? Do you have access to ANet's data? How many players press 'Enter Battle' every 30 seconds, how many teams are formed? With those data you can calculate what the chances are of getting in, or not getting in a team.

But we don't have it, so in the absence of that, we do the best thing we can: speculate. The best approach, given the absence of data, is to make some reasonable assumptions about the math problem we are facing and infer the likely consequences of a change in the pairing mechanism from those assumptions.

The assumptions I made are useful because they force the possibility of long timer waits, and give us some idea of the possible odds of drawing a long wait given the join system we're facing.

Am I claiming accuracy? Of course not. But with this problem, the set of possible outcomes is pretty limited. What we can do is make reasonable, realistic assumptions about the problem and extrapolate probabilities about the risks of experiencing the issue you are so concerned about, the risk of drawing a lengthy series of timers in order to get a match.

Quote:

Originally Posted by Amy Awien

Btw, how old are you?

Older than you.

As I see it, we have two issues here:

1) You've been repeatedly, objectively wrong in your claims about what I've said. To be frank, I don't have to treat you with respect when you mischaracterize my statements. In intellectual discourse, that's disrespectful. Play nice, and I'll play nice back. Distort the truth, and the gloves come off.

2) You're focusing on extremely low probability events rather than on the net social result. We have a problem where players are behaving in a socially undesirable fashion because the system incentivizes the behavior. Lemming's proposal reduces those incentives. That proposal does not come without cost, but the social costs are objectively, demonstrably negligible if you accept the supposition that we should measure net social benefit as a fair distribution of glad points and Balth faction. Since that's always what people are on about when they raise the issue of syncing, I think that's a pretty fair supposition.

Permutating the assumptions I made admits the possibility of longer timer waits, but the chances of such waits are so negligible (much less ever experiencing multiple such waits) that they aren't really worth our consideration when trying to solve this problem of social choice.

The average player that doesn't sync is demonstrably better off under Lemming's regime given any set of assumptions you care to propose. Further, the odds of any given player being worse off given Lemming's regime are so remote (irrespective of the assumptions) that it's pretty clear that the proposal is socially improving, even given the risk of long match timers.

It's not that you haven't raised a valid point. It's just that it's irrelevant. In essence, you're complaining that injecting an insulin-dependent diabetic results in a wound that could become infected. While the statement is true, it's pretty evident that the diabetic is better off if given the insulin despite the risks.

Perhaps the issue is that you reject the definition of net social benefit that I put on the table. If your preferences are such that you'd rather get thrashed by g9+ syncs that don't suck, rather than ever experience a timer reset, I can't argue with that. But what I'm trying to point out to you is that those preferences probably aren't consistent with the player base in general. The existence of this thread tends to confirm the supposition I made; players' responses in the thread provide further evidence.