[Discussion] Dealing with p0 |
Lemodile « Censeur » 1368631380000
| 0 | ||
When the map editor was first introduced, transformice looked and felt very differently from now. Transformice still had old graphics, only a few servers, different forums (with little information) and only two map categories: the original maps made by Tigrounette and usermade maps. As people started to make maps, the second category of usermade maps became massive. And of course, some maps were better than the others. At this point P1 was introduced, to keep the better maps in the rotation for a longer period and also to give them more playtime. Soon came P2, which was the crème de la crème of the P1 maps. And soon that was split into P4-P9. The point of all of this was to separate the bad maps. The same purpose is served by the voting done on P0 maps. Slowly P0 has gained the title of "bad maps". To such an extent that all new players now automatically spawn in vanilla rooms, where they can't get "scared off" by the "horrible" p0's. P0 shares it's place in the map rotation with p1, which means these maps are pooled together and the game randomly selects one of these maps from this pool to be played. To make sure p1 gets a decent chance of being played in a room (and to reduce server load), this pool doesn't contain every single p0 map available. According to Sydoline, only the oldest 40.000 p0's are included. Which means that at any given time, the pool of P0’s the game gets to pick from is the oldest 40,000 p0’s at the last restart + the new p0’s exported since the restart. At the next restart, the oldest 40,000 are chosen again and the newly exported maps are no longer included in the pool. Also note that /info on any map not in the given 40,000 returns a Carte introuvable. The question that needs asking is whether or not this is the right way to deal with these maps. Some consequences of these measures may not actually have the desired effect of improving the overall map quality. Some examples: 1. Since P0 maps play less, it means that we mainly get to see "approved" maps. But this means that it takes much longer to filter P0 maps. Public votes less on lesser maps. I.E. the public essentially has less say in map quality. 2. Since players now spawn in vanilla rooms, the number of active numbered rooms has reduced. I.E. less players to vote on P0 maps. 3. It also means that, to get (your) new maps played, you need to get them permanent. You are dependent on the mapcrew which honestly only is able to handle the input from the submission threads, because many map makers don't bother with the forums. 4. By only seeing these old p0's, people get a wrong image of the average p0 map, leading to measures like the ones described above. Moreover, p0 is the only map category that actually has some sort of limit. Doesn't that sort of imply maps shouldn't stay p0 for forever? They should either be removed from rotation (p22 or p44) or made permanent (p1 or up). When we were part of the map crew, tig once told us that P0 maps should either be permed or deleted. They should not be left P0. This now makes more sense than ever. With a 40,000 limit, we can’t afford to let maps stay P0 and take up valuable space. Basically this thread is aimed at making the public aware of how the rotation is working and to request the map crew to change the way it handles P0’s and P1’s. Leaving a map as P0 should no longer be an option. It should either be P22 or P1. This means that the criteria for perming and deleting maps should be changed so the "gray" area -where no reason can be found to delete a map, even though you would never consider perming it- is minimized. To try and cut back on these "ancient" p0's, we have been going through the list of the 40,000 maps the server loads and splitting it into lists depending on how good they are. We report maps that even with the current guidelines should be deleted, we will submit maps that meet current p1 standards to the map crew and we will try to work to new criteria that should deal with all the maps we can't report or submit at this moment. With constant effort, we hope to bring the list of maps the game loads up to speed with the maps exported on a daily basis so that people actually get to see their maps in rotation. Replies and confirmation of action on this issue from the map crew would be appreciated. Any suggestions/discussions etc on the above are also welcome. We realize that even plowing through the current 40.000 maps is an immense task, but hope that both the public and staff members alike will see the value of this work. Hobothehill and Lemodile ------------ Things to discuss: Should we or should we not actively try to sort out p0 maps so that they either get permed because they're fine for rotation, or deleted because they're not? Should the criteria for p1 and/or reporting p0 maps be changed so that a map is either to become permanent or deleted? Should anything be changed about the current position of p0 maps in the map rotation? |
Fluffyshine « Citoyen » 1368632940000
| 0 | ||
They should P44 all the bad ones. *p44s every single p0* |
Typenick « Citoyen » 1368636240000
| 0 | ||
This reminds me of the old map police thread on vBulletin forums (anyone remember that? It came around just before map crew). Going through all the p0s that existed back then took us few weeks and doing the same with your 40k goal is definitely doable, although fatigue kicks in pretty quickly. While I support your idea I think that change in the system is needed much more than change in the maps (and it's easier as well). Keep in mind that those old maps that are played over and over accumulate loads of votes from players and after reaching a certain number getting it voted out is pretty much impossible, which makes us end up with rotation getting dominated with small number of old maps that never get deleted. I think that either the way the maps are chosen should be changed (40k oldest cache is a very bad idea) or we should get rid of p0s altogether and change them all to p22 (so they can then get permed, etc.). Since apparently we need the cache, make it at least shift a bit. Like, remove a handful of oldest maps every few minutes and load newer ones so it eventually cycles through all of them. |
Bakpaofgunz « Citoyen » 1368660060000
| 0 | ||
I really don't think splitting P0 maps into either deleted/tribe house maps and protected maps is a good idea. Not only does this completely wreck this category, but also does it ruin the entire concept of player-based voting and to a less important extent this would mean getting rid of a lot of nostalgic maps which have minor flaws to not qualify for protection. While sorting out the maps is a great idea to get rid of the unsuitable ones and protect the acceptable ones, most of the maps will be left in yet another big pool of mediocre maps which aren't suitable for either deletion or protection. Changing the guidelines concerning reporting P0/P1 maps won't do any good either, they already contain all the basic flaws and making them more complicated or developing them more would do more harm than good to my belief. The way we, the Map Crew, handle map deletion is not the origin of the issue here, the cause of the problem is the way the map rotation works. Also, the current position of P0 in the map rotation should not be changed in my opinion. The suggestion you brought up, or at least brought to attention, is to either P22 or P1 all P0 maps. This would essentially be the same as removing the P0 category altogether and putting every single exported map on a waiting list which someone, most likely a MC member, would have to check and then give approval or at least make a decision for every single map. Not only would this mean our workload would get increased by a ton, something which none of us is waiting for as this will lead to tremendous delays, but this would also mean the complete P0 category would be removed. On top of that passing through a big change like this would mean mass confusion for both the players and the staff members to change something that many, if not most, would not even notice. This still leaves the problem concerning the 40,000 maps unsolved. I was genuinely surprised to learn about the way this is being handled. It's something I personally have never noticed, and never would have unless it was brought up. The problem, in my eyes, is not that only 40,000 maps get picked, but that only the first 40,000 maps constantly get picked. We could of course go through all the maps and get rid of all the bad ones until we have 40,000 maps that are deemed suitable enough for this rotation, but this would mean that we would be stuck with those, as, from what I would estimate, the codes of these maps would range between @20k and @500k, maps which have been in the rotation for a long time and are virtually impossible to be voted out. Not to mention the incredible workload it would take to both set this up, but also to actually maintain it. In my eyes a potential solution to this could be randomizing the 40,000 maps that get picked after every server restart. The system as to how P0 and P1 are currently balanced doesn't need any changing or tweaking, it's ratios of showing up in the map rotation are fine in my eyes, but the maps shouldn't get monotone. With as much as 33k P1 maps I don't believe a single P1 map shows up that often (a long mathematical theory could be inserted here to demonstrate the actual chances to a P1 map showing up), and considering how P1 shares it's playtime with P0 the chances diminish for both categories yet again. However, I wouldn't say only the oldest 40,000 maps should be played. A randomizer would actually be great to add more variety, to what extent that's possible. But then again a server-sided change like that would require lots of time and discussion and approval from administrators, therefore for now I would really have to say that I prefer not to take any action, by which I mean filtering the 40,000 maps except for the obvious ones, until more light is shed on the subject. The Map Crew are aware of the ongoing subject and we have a discussion running in our private forums. I'm keeping an eye out for this thread and I would be happy to answer all questions that might occur or come up as I can imagine if anyone finds my story a tad vague and hard to comprehend. |
0 | ||
Well, p0 maps are to see if players like it, and if every map was p22 when exported, feedback from players would lessen? Players can already delete maps when they reach less than 50%, so everything is fine, but if the old p0s were permed, newer maps would get played more (like bad ones), so everything should be balanced. But I believe everyone should be able to play maps that were meant for the public as p0 and up. Many map creators' objectives are for mice to play their maps..right? And the p0 voting decides if they are worthy of perm or not because well, if all exported maps were p22, they just won't feel like real maps and perming and evaluating all those new p22s would be a hassle lol. Also some people don't want their maps permed more than p0. So in conclusion, keep maps as if. :D |
Bakpaofgunz « Citoyen » 1368660960000
| 0 | ||
Epalk a dit : Yeah, that's the issue. If only the oldest 40,000 maps get loaded, people don't play all the maps at all except for the time between when the map is exported and the next server restart. Epalk a dit : This is a dangerously irreliable statement. P0 voting is in no way a means to measure if a map would be worthy of a perm or not, you'll find this going through most of the oldest P0's. It mostly contains unsuitable maps like free cheese maps, afk gameplay maps, unaligned maps, ground spam maps et cetera. It would be more accurate to say P0 voting is a way to measure what the community likes, not what the community thinks is worthy of being protected. And lets be honest, everybody enjoys free cheese ^^' so those maps usually get up-voted while clever or good maps which are actually worthy of protection and which require some thought and skill tend to get down-voted as most mice recklessly rush through the map failing effortlessly, instinctively leading them to a no vote. |
0 | ||
Bakpaofgunz a dit : I understand now lol |
Hobothehill « Citoyen » 1368678240000
| 0 | ||
I think that in some ways, you have misunderstood what we are proposing Bak. The remove P0 entirely and set up a waiting list was already discussed before. I think mollo was the one who brought that idea up. Obviously that puts immense pressure on the mapcrew. Yes the simplest solution would be to change the system by which maps are selected. But you and I both know that it is notoriously difficult to get the system changed. Near impossible. Yes the sensible thing would be to pick 40,000 maps and then pick the next 40,000 maps next time and so on, until you reach the end at which point you go back to the start. Hopefully this will keep ahead of the map export rate. But in the meantime, the system has to be suited to meet the needs. What we are suggesting is to cut down the P0 pool such that new maps get selected into the 40,000 maps to be played instead of the old maps. Currently there are nearly 1 million P0 maps. Which makes this task nigh near impossible too. But you have to start somewhere. You realize that currently 960,000 P0 maps are not selected for playing in normal rooms? And the even bigger problem is that the old maps have been around so long, that its very hard to get rid of them by vote deletion. Say a map has 5000 yes votes and 5000 no votes. Thats 50%. To bring this below 40%, you need around 6000 no votes. You are talking another year or so in the rotation. And most old maps have over 20,000 votes. We don't have the liberty of a gray area when it comes to dealing with P0's. Maybe it can be reintroduced later but at this time, it just won't work. We end up playing the same maps again and again. As for the old maps we will lose because of this change. I was told that nostalgia has no place in P4-P9. But if it does, you could consider making a provision for old maps and put them into P1. P3 has a provision, no reason P1 couldn't. Or if that rankles you, edit the slight problem out of the map. Various P4-P9's have been fixed this way. No reason these old maps couldn't. Of course this puts a load on the mods. I have nothing against the system being changed. If some admin is willing to read through this and address the problem, good enough. But I don't think we can just do nothing. |
Lemodile « Censeur » 1368678600000
| 0 | ||
Just to quickly point out; the intention was never to have all p0's put into a category in one go. However, there's a huge amount of really old p0 maps that don't meet the criteria for deletion or perming, yet are profoundly bad when it comes to aesthetics or overall gameplay. Because these maps weren't so bad compared to other maps at the time and because it was easier to get a lot of votes for a map then it's nearly impossible to vote these out. edit; also see hobo's post. |
Papero « Citoyen » 1368681720000
| 0 | ||
Lemodile a dit : but its gonna take alot of time to delete them @_@ |
Johnlantern « Citoyen » 1368687600000
| 0 | ||
Hobothehill a dit : Funny thing, I was told by mapcrew the complete opposite when I mentioned that old art perms should be depermed if they no longer meet the higher quality maps we have now. I was literally told that some maps would remain permed in P5 for nostalgia reasons. As for the P0 debate, I don't think there should be a P0 category. I even dislike the idea of having the P1 category as it is now. I really think rotation should only consist of higher perms (P4-P9). I do, however, like the idea of changing P1 into a nostalgia category where maps that used to be 'up to par' would be placed (might even be a great place for those poor art maps I tend to hate #@330773 @317110 @263215). Of course this is a lot of work on the mapcrew. I know i'm not being reasonable at all, this is just my dream. |
Bakpaofgunz « Citoyen » 1368693960000
| 0 | ||
Hobothehill a dit : I understand that what you are proposing takes a lot of time and would need to be introduced gradually and in small steps, and while I do like the idea of filtering out the oldest 40,000 maps to allow more maps to be played, this would essentially be a loop constantly repeating itself if executed. 40,000 is only a small share out of the nearly 1 million maps in total which you brought up, and if we filter out these 40,000 it would not really change anything, as it would mean a given amount of different maps would take up the space of these 40,000 maps, maps which are slightly newer but still old and incredibly hard to vote out of rotation. I don't know the exact length of the list, but if I'm not mistaken I assume it reaches out somewhere from the beginning of the introduction of the map editor to codes up to @300-400k. Any code under @1.5m would be virtually impossible to vote out of rotation, which would mean this filtering which is being done by you right now would need to constantly be done again. While that isn't so much of a problem, I highly doubt there is anyone who has the time, dedication and motivation to consistently keep this up. Also a small note to Hobo: If I'm not mistaken maps get deleted when the vote count reaches below 50%, not 40%, right? Doesn't change that much but if true your calculations are flawed :P About the nostalgia matter: I have no idea what has been said about this before, but I doubt that nostalgia has no place in any of those categories. P3 up to P9 all have a fair amount of old, nostalgic maps (Just look at P6 for example, while we keep saying these are temporary fillers, which they are, it takes a long time to fill up this category and the maps which were recently added to bulk up the category are actually in the rotation for quite a long time, and some of them probably will remain there). I think there'll always be some place for old maps. Whether P0 is a place to also realise this will have to be discussed further. Johnlantern a dit : This is most unlikely, removing either P0 or both P0 and P1 would drastically discourage map makers from even attempting to make maps, considering how it's generally quite hard to achieve higher permanents and overall it's a small group of map makers who have developed their map making skills to a certain level so that it will most often be those people achieving higher permanents, destroying even more hope and motivation to the less skilled map makers, people who actually enjoy getting P1 maps. On top of that I feel as if introducing an idea like this would be very detrimental for the overall quality of the maps in P4-P9. |
Papero « Citoyen » 1368694620000
| 0 | ||
Bakpaofgunz a dit : no we dont filter out old maps @_@ we delete those which are p0 and are not worthy of staying *permanently* but i suggest you guys should return the cheeses too o-o |
Bakpaofgunz « Citoyen » 1368696300000
| 0 | ||
Paperonaruto a dit : You seem to have misunderstood, I was addressing the fact that Lemodile and Hobothehill are currently filtering out the list of 40,000 maps. Also, returning cheese is virtually impossible. It's already a great amount of work to get all the contest winners their prizes, to create and hand out cheese codes for cheese amounts ranging between 5 and 40 cheese is unnecessarily difficult and time-consuming. It's not our fault that the concerning player has made a faulty or unsuitable map, why should their cheese be returned? |
Papero « Citoyen » 1368696660000
| 0 | ||
Bakpaofgunz a dit : well technically you are removing their hard earned wealth? o-o and its not their fault their maps are faulty either. may i know are you only filtering out p0? and i said permanently delete it out of the system @_@. |
Johnlantern « Citoyen » 1368780240000
| 0 | ||
Bakpaofgunz a dit : The feelings of new map makers don't matter when selecting maps to be put into rotation, if one believes that the quality of rotation is what matters most when making maps permed. It doesn't matter if a player enjoyed getting their map permed, it's what the users have to play / would enjoy playing that matters. (Yes, I'm heartless) Perms should never be given out of pity lol Sure, you'll have amatuer map makers upset that their maps don't get permed (I'm sure this happens already), but the rotation, would essentially be more clean and enjoyable. Bakpaofgunz a dit : How so? If the requirements for making a map permanent in a specific category don't change, then bad maps should still not get through. The only 'drawback' to eliminating p0 and p1 would be more submissions in each perm category, which, in theory, would only make evaluating the map easier since you can clearly tell what the creator of the map was intending to do depending on the category they submitted it for (map submitted for P5, better evaluate it as an art map). This might even make perming requirements even more clear with less grey area. The only way it could possibly be detrimental is if mapcrew begins perming maps which don't meet the requirements for that category. And again: Johnlantern a dit : Johnlantern a dit : Johnlantern a dit : . |
Anythin « Consul » 1368828360000
| 0 | ||
Lemodile a dit : I won't get into this discussion too much because I don't have the knowledge about it, but all I'm really wondering is if some examples can be given of maps in this "grey" area. What are the considerations to not delete a map, but at the same time to not make it perm aswell? And if you do think a map shouldn't be deleted but at the same time doesn't fit any of the categories, wouldn't that basically mean there is a need for more categories? In my opinion it's not weird to delete a map if it has no chance of getting permed, as it results in the problems explained in above posts. |
Lemodile « Censeur » 1369847880000
| 0 | ||
Examples of what I consider to be "gray area" maps. Look them up on CFM. @1166619 - Obviously made to be a tribe house, but exported as p0. It has a cheese and hole that aren't hidden and the map doesn't break any other rules either. So it can't be reported. Does this mean it should be p1'd? You only use less than 1/4th of the map to get cheese back to the hole. The rest is "troll area". The decorations could easily distract a player from seeing the cheese, or think that it's hidden somewhere. In short; the map wasn't really designed for the map rotation to begin with. @1166609 - The gameplay's alright. However, the map is easily turned into something that could crash if the shaman adds a few boxes. There's a lot of random stuff going on offscreen as well. A player that accidentally hits the lava on the right gets transported to an area that can only be escaped by going offscreen again, causing problems for newer players. Yet none of these things are things you can report a map for. --- There are many more scenarios where a map doesn't break any of the rules, but where I still think it's questionable to want to keep them in rotation. I simply picked a random number now.. I think 1166620, and started counting down. It didn't take me long to find two examples. --- I don't really think there's a need to have more categories here. What would a category filled with maps that don't fit the rotation, yet aren't good for p1, have to offer for anyone? P22 is perfect for these kind of maps; they can still be played and edited, without other people being bothered by them. This is why I prefer having the rules changed so every p0 can, if needed, be categorized into either p1 or p22. |
Bhvana 1549747200000
| | ||
[Modéré par Batt_mellamy, raison : Necrobump / off-topic] |