Group pvp v. one-on-one pvp, Can they co-exist? |
Group pvp v. one-on-one pvp, Can they co-exist? |
![]() |
14 Sep 2009, 06:22
| #41
Hi ChoKuRei,
Thank you for your response to my post. However, it was never my intention to criticize anyone, and I don't think I have done so. When you pm'd me the first time because you were offended by what I wrote, I edited my post...however, you did not edit your post as I asked in return. I apologized because you were offended, not because I had done anything wrong. At that time, I suggested to you that you start your own thread if you want in-depth evaluation of your system at this stage. I do not claim to be an authority on posting threads. What I'm asking is that for this specific thread that you follow the guidelines I presented when posting on this thread so we don't get sidetracked. We all want a better pvp system, so let's try to work as a team and do something completely different. The guidelines: 1. Summarize and present your pvp ideas on this thread. Include temporary "bandaids" or long-term solutions...general ideas or detailed structure. 2. Please remember that we are not debating specifics, arguing, or evaluating, but rather we are "collaborating" and "refining ideas". 3. Focus on developing your own ideas and communicating them as clearly as possible, rather than evaluating or criticizing other people's ideas. What I am observing from my perspective is someone coming up with an idea and someone else shooting it down right away and pushing forward their own system. This isn't a competition between ideas, it's a process of compilation of ideas that involves teamwork. It would be much more productive in the long run if you worked on your system instead of posting about other people's systems. Here is what I wrote in my previous post regarding Paul Graham's article...you must have missed it: "In this thread I am asking posters to use polite 'Refutation' only, if they wish to discuss someone else's post, so that we can focus on developing the ideas that are important to each person and not get sidetracked by unproductive contradiction." The reason why I posted the article by Paul Graham is because I believe that people can be polite when they disagree. I also believe that people can learn from each other when there is disagreement...when it's done in a polite way and when there is validity in what the other person is saying. When someone posts their ideas for the first time, they are often self-conscious. They need encouragement to keep developing their ideas, or they will stop working on their project if they have a negative experience. I have attempted to give you subtle hints throughout this thread without singling you at personally, because I am trying to be patient and polite. I'm just asking that you please follow my simple guidelines for this thread only, and I'm sure that it wouldn't be difficult for you to do so....it's just a matter of choice. ChoK, I perceive you as an intelligent adult who has some good ideas to share with us. However, you need to go into more detail so that we understand how your system can be implemented. Disadvantages: 1. Groups balanced by level isn't inherently balanced especially at 180+ eg a level 200 can generally take on 3-4 level 100s. 2. High levels won't be guaranteed a place when going to join existing prism fights. 3. Levelling wings will be harder from ranks 1-4. 4. It's not very realistic. For example, what is going to be the "balance" calculation for prism fights? Originally, you had it set so the levels are equal...e.g., lvl 200 v. two lvl 100's. However, you also recognized in your "Disadvantages" that the lvl 200 player would win every time. Your entire system rests on the formula for balance: "By means of a simple rule, both can be prevented: You can only join a prism fight if your level doesn't make the level difference between teams > 100 and your rank > 3." (ChoKuRei) Could you work on solving that issue, and then post your solution for us? I think that is the main reason why I have difficulty with your system the way it is right now. Once all those types of details are resolved, then we can start the evaluation process. Otherwise it wouldn't be fair to you for us to evaluate an unfinished proposal. I'd like to understand in more depth how your system works and how it could be implemented as a viable solution to our current problem. Thank you for understanding. anjulica This post has been edited by anjewel: 14 Sep 2009, 06:26 |
![]() ![]() ![]() |
15 Sep 2009, 00:41
| #42
QUOTE Disadvantages: 1. Groups balanced by level isn't inherently balanced especially at 180+ eg a level 200 can generally take on 3-4 level 100s. 2. High levels won't be guaranteed a place when going to join existing prism fights. 3. Levelling wings will be harder from ranks 1-4. 4. It's not very realistic. wink.gif I feel in your system that there would be much difficulty in resolving each of your disadvantages as stands. The focus is groups being balanced by total level, this is the root of your systems inefficiency. Leveling wings however can still be done in a 1v1 setting. In my proposal not only is each tier guaranteed access (all-be-it with an increased waiting period (as you would either have to join an ongoing Assault/Defense or wait for one to be full to join. Even then this could be fixed by choosing a 4-8 player group when you choose the prism to attack...thusly making things go faster. The more players the higher the reward, but the longer the wait. In any case you would have to wait 15-30 mins between each Prism(or be penalized with a lower position in the Queue). In any case, a tiered method is far superior to levels alone. |
Member
Legendary Crackler Crusher![]() ![]() ![]() |
15 Sep 2009, 04:24
| #43
Electricotter - I applaud you for contributing the first new idea in this thread in over 2 weeks.
I'm missing something from your description but I think we can combine our ideas into something even better. 1. Arbitrary numbers of people get 'preparatory' missions to do prisms. 2. The server then creates a balanced fight out of those who volunteer according to whatever algorithm Ankama chooses. This way, the system could be introduced now without the massive rebalancing that has happened for 1v1. Here's a simple balancing algorithm: Rating = (level ^ 1.5) + ((level ^ 1.5) * 0.3 * GvG win/loss record [caps at 20 and 80%]) . Assuming 50% win loss ratings across the board, such a system could pitch a single level 200 vs 2 level 100s & a level 83. Would that be a fair fight? I think so but you could increase it to level ^ 1.6. The 0.3 equates to a 15% adjustment in rating depending on GvG record of wins/losses. This is an essential factor to a) compensate for overpowered builds The perfectionist in me would be inclined to make special cases for the gap between level 99 & 100, dopples at 200, enis/osas/xelors but it isn't strictly necessary. The server then determines how much time to allow for extra people to take the mission. If the server is busy, allow only 20 seconds. If it's quiet, wait for 3 minutes. NPC can give some indication of this time with "Are you sure?" message if people don't want to risk the wait. So, here's the fight-arranging algorithm: If there's a balanced x vs 7 or x vs 8 (according to sum of Ratings), start the fight immediately. Else If there's a balanced fight amongst those waiting, start the clock. If a bigger balanced fight is possible in that time, restart the clock. Else start the first balanced fight. Players are given a 10 second warning before the fight starts. Assuming both sides have attackable prisms, the alignment which has momentum (ie won last prism fight) gets to attack. The highest level in the team for that alignment gets 10 seconds to choose which prism to attack. If they don't pick one, server picks at random. The main drawback to this (as you've identified) is that people can log off. As such I think major honour loss and being banned from GvG quests for 24 hours is in order ("You are a deserter and must regain your honour!") My inclination is just to teleport everyone to the fight immediately. If people make their own way there, it'll be aggro city (which in itself will unbalance the fights) plus slow everything down. Other drawbacks: - epic levels will still have to wait longer than low levels for a fight. - it will take a while for handicaps to be worked out. - the system favours balance over merit - if you suck, the other team could be handicapped at between 2 - 15% (big fight - 1v1) Advantages vs other 2 systems: 1. Relatively foolproof - you don't need to manage your team. You just sign up and wait until you're in a balanced fight. 2. Don't have to rush rebalancing of classes as: a) exponentially-increasing power by level is automatically compensated by algorithm. 3. No arbitrary level bands causing: a) players to alternate between good and bad at GvG as they level. 4. Unrealistic team-balancing is somewhat hidden from view. Multi-accounting would be prevented from entering 2 different fights simultaneously, or have 2 players on opposite sides. If the overall wait becomes too long for any one player (say 10x the time to allow extra players to join), they would be allowed to abandon the quest without honour loss. I would say wings up for this quest, just allow them to be put down at 50-75% reduced honour loss. Strokens and XP would be proportional to overall team level and one's individual level. To me, this seems so superior to the existing headhunter (& prism) system that I think there would be a strong call for 1v1 headhunter to be scrapped. What do you think? This post has been edited by ChoKuRei: 15 Sep 2009, 04:26 |
![]() |
15 Sep 2009, 08:29
| #44
I'm just adding this quick idea before I forget it, hehe
I was reading over the last couple of posts, and a simple formula popped into my mind regarding balancing, that could apply to Chok's system. Rather than go with SUM (of levels) v. SUM, we could try (SUM2) v. (SUM2) An example of SUM v. SUM would be a lvl 200 v. two lvl 100's, or a lvl 120 v. two lvl 60's. This is how (SUM2) v. (SUM2) would work: Level Values (obtained by "squaring" the lvl, or multiplying the lvl by itself) 402 = 1,600 (40 x 40 = 1,600) 502 = 2,500 602 = 3,600 702 = 4,900 802 = 6,400 902 = 8,100 1002 = 10,000 (100 x 100 = 10,000) 1102 = 12,100 1202 = 14,400 1302 = 16,900 1402 = 19,600 1502 = 22,500 1602 = 25,600 1702 = 28,900 1802 = 32,400 1902 = 36,100 2002 = 40,000 Example: lvl 200 attacker = 40,000 (lvl 2002) Defenders could be any of these groups: a) lvl 120 (14,400) + lvl 160 (25,600) = 40,000 b ) lvl 100 (10,000) + lvl 170 (28,900) = 38,900 c) three lvl 110s (12,100 x 3) + lvl 60 (3,600) = 39,900 d) four lvl 100s (10,000) = 40,000 Broken down, the formula is actually lvl2 + lvl2 + lvl2, etc. rather than (lvl + lvl)2. If we did it the latter way it would just work out the same way as SUM v. SUM. Another idea... I was trying to figure out a way to get around the confusion of joining with the level balancing idea, and came up with this. Unfortunately, it doesn't solve the problem completely, but could be a step in the right direction. As the levels change during the lineup, waiting for the fight, the lvl values "open" could be on the "defend" screen. e.g., If a lvl 200 attacks a prism, then 40,000 level points can be filled by the defending team. Unfortunately, this is still confusing...it would be easier to just announce how many lvls were available, however, it doesn't work with the formula. For example, if a lvl 120 and 160 attack (40,000 points), and a lvl 200 defends (40,000), even though the level points are the same, the actual levels are 280 v. 200. Anyway, here are a couple of ideas that will hopefully help trigger more ideas. anjulica Edit: Another little idea to simplify the level point values...is take two zeros off. For example, lvl 40 has a point value of 1600...take off the 2 zeros to put it at 16 This would make the "open level points" a bit more manageable...e.g., a lvl 200 attacks...400 level points are open. Each player's lvl points could be shown on the defense screen along with their own lvl. It simplifies the issue a bit...but doesn't solve it completely. Ooo...another idea I just thought of... Everyone joining goes into "reserves" when they click to defend. The different possible team combinations (according to level point values) are shown in the window. This post has been edited by anjewel: 15 Sep 2009, 19:36 |
![]() |
15 Sep 2009, 08:49
| #45
a simple way to fix the Whole PVP proble is to brign the old prism fight system back where anything outside a prism fight was useless and only ponits where given/taken from in the prism fight it'self. this was a very affective system. and the headhunter quest should be changed so you have to take out a prism in the area given to you by the headhunter person. this will stop poeple picking WHO they get to kill and/What class there target is as they chould get a high lvl group of a range of classes.
this also well reduced the amount of sram's/osa's/sac's/feca's/iop's pvping as there source of lvling as they are the only class i 99% of the time see PVPing. cable This post has been edited by cableseni: 15 Sep 2009, 08:58 |
![]() |
15 Sep 2009, 19:44
| #46
hehe, another little idea...based on the edit in my previous post:
After the time for defending a fight has ended, there is a one minute delay, at which time defenders can make decisions on who actually joins the defense. Idea #1: If different possible team combinations were shown on the defense screen...then players could click on the team they wanted to defend. If the majority of reservists pick one formation, then that is the team that is entered. Any players common to all group formations could be automatically moved up into the actual team list. If no one clicks, then the first team on the list is entered. Idea #2: Instead of a list, there could be a feature with the "mouse scroll over". Scrolling over each reservist highlights the other players who could combine with that player to make a team...or dims those players who can't combine. Clicking on that player means you've voted to send the player into the defense line-up. In cases of a tie, then the vote of the chars with the higher rank total is used. In cases where more than one team is possible with a player, all members from the various possible teams are highlighted (or non-eligible chars are dimmed). After clicking on the player to vote to send them into the defense line-up, the second teammate can be clicked on, which will narrow the team possibilities down even more. In situations where time runs out, the remaining spots are automatically filled in with one of the team combinations. The final members of a team under these circumstances would be either decided by rank, level, or who joined first as a reservist. I think that "dimming" the non-eligible players during the scrollover might be much more user-friendly than highlighting the ones that could be picked, as the list would look a lot smaller once it's been simplified with each vote. However, highlighting them with a red outline might work. ************** Something that came to mind when reading Electricotter's post: Instead of limiting instant transportation to prism fights, why not add it to the Headhunter Quest? For example, when players get a target from the militia, rather than get a scroll, they are transported onto a random map (or arena) for a 1 v. 1 fight with a player from the opposite alignment of similar level, who has also gone to the militia to get a target. In other words, they are "matching" targets (they both have each other as a target), triggered by requesting a scroll from the militia. For those who still like to hunt their targets, that option could still exist, but with low xp/rewards. This would make the automatic 1 v. 1 fight more appealing, as it would start quickly and have high rewards/xp. A few things this idea helps with: 1) no more scrolling through militia scrolls to find a class that can be easily beaten, 2) no more aggroing people who are afk, are in pod gear, or who are busy trying to do other things...the level of animosity overall would decline on the server, 3) "wings" would have truly been earned and mean a lot more...they would no longer be obtained by fighting a weaker class continually, but rather through fighting a random selection of opponents made by the server, and 4) players wouldn't be able to skip over friends, but they could play against their friends without animosity b/c the server chose them as opponents, and there was no choice involved (i.e., they haven't hunted their friends down with scrolls). This post has been edited by anjewel: 15 Sep 2009, 20:19 |
![]() ![]() ![]() |
15 Sep 2009, 22:48
| #47
QUOTE Instead of limiting instant transportation to prism fights, why not add it to the Headhunter Quest? For example, when players get a target from the militia, rather than get a scroll, they are transported onto a random map (or arena) for a 1 v. 1 fight with a player from the opposite alignment of similar level, who has also gone to the militia to get a target. In other words, they are "matching" targets (they both have each other as a target), triggered by requesting a scroll from the militia. I dislike this idea as faster fights mean quicker rewards from low level 1v1. The abuse is enough as is with having to hunt ones target. Only choosing those with active HHQ's is a good idea though. Also PvP could be more closely tied to the orders. Artisan =non PvP with conquest area access. Assassin= Head Hunter Quest (target and hunter). Soldier= GvG access. Unlearning potions could be made for the orders with an honor/align level penalty for consumption. They could be used once every 1-3 months or so. This would be a great way in allowing everyone to play the way they want to play. As for your new idea Chok, I like what was put forth but still feel that your algorithm would work fine in a tier framework, the numbers of like tier individuals wouldn't be an issue as they could jopin from any location via a clickable text, and the number of attackers could range from 2-8. All this aside however, The main complaint I have is the fact that one would have difficulty playing with ones friends in this format. Having several queues to choose from with a /a text stating the location and level of initiator would enable friends to join the same queues. I like Anj's level value system over the algorithm as there is more transparency via ease of comprehension. I'm not one for crunching numbers but it would be interesting to see how they stood up to each other. I think that one should start the queue at an NPC and have a perc defense-esque text appear. This would allow several friends to be in the same queue of a lesser size. And allow a list of possible queues with the initiators level listed(in place of a tier format). Lastly, the penalties and benefits I outlined such as a queue penalty to those who were in a Prism battle in the last 15-30 mins (from the battles end) is rather important to me. I'm not certain but it seems that a team of 8 lvl 200's v another team of 8 200's could constantly be in prism fights with one another. the same could be said for the 8 highest level braks and 8 highest level bontarians. It should be encouraged that everyone be able to take part in the system, tiers do a better job of this, but queue penalties and bonuses would be a useful tool as well. Alignment level and wings up or down (which shouldn't matter either way(wings down =no honor/less honor and reduced xp/strokens at worst) should be definite means of queue adjustment as should wing rank. These would only be meager adjustments but deserving of a place none the less. In final defense of an interactive /a text (/a must stay open with wings up) it would increase participation greatly. 1] When out on a hunt you could still defend Prisms (joining from within a dungy would not be allowed) 2] It would give those who wish to participate the ability to participate without extensive down time (which we do pay for...a 3 min queue is the longest one would have to wait) 3] You can coordinate queues to have a greater chance at being with friends. Also I think the log off system from my suggestion was pretty good too. It appears we are formulating some good ideas here, lets keep it up folks! |
| Lo-Fi Version | Time is now: 22 November 2009, 03:52 |