Post by semolinapilchard on Jul 2, 2020 22:28:36 GMT
What area of New Blood are you suggesting this for?
Zombie server
What is your suggestion and how will it work?
Implement a system where the roundtime of each map is displayed beside each map during the voting process. Players will be able to judge a map based on how long it will play for, as well as its name.
What are the advantages of your suggestion?
Players will know he round times of maps while choosing them and will be able to take them into account while voting. Players will have a greater understanding of the maps they are voting for and it is less likely to lead to lots of players leaving when a longer map is queued. This is particularly important for.more casual or less frequent players who don't have time to get to know all the maps.
That could create a system where only the highly liked maps gets voted for though, which is the only issue I see with it. The roundtime idea isn't bad in my opinion, as it can show maps which can be quickly finished roundtime wise if someone is in a rush and can only play one or two games more.
Post by Subelectronite on Jul 21, 2020 17:11:17 GMT
I suspect that this may introduce favourability towards maps with a shorter time, which is not necessarily a bad thing but can influence the vote even if they don't remember what the map is like.
How about displaying the maps like/dislike ratio instead. For instance, [Map name] 100(likes) : 200(dislikes)?
The one issue with adding dislikes and likes to the vote would be that newer players would then potentially use that to judge a vote without making their own mind up or experiencing the map. I'd prefer the round time for the simple fact its informative, not opinionated, and thus just provides extra information for a player to make a map decision on. I just agree that the benefits to this suggestion would be minimal.
Post by semolinapilchard on Jul 21, 2020 22:03:53 GMT
Yes, I think roundtime is much preferable to likes to dislikes. Players deserve to know the roundtime of the maps they're voting on, whereas likes to dislikes are very arbitrary
I feel like maps with longer voting times could potentially get voted less times than maps with shorter playing times but I do like the idea because it would be nice to see how much time I would be spending on each map without typing a load of commands to see.
Instead of showing the round time for each level individually, the map voting pool should consist of 3 maps with the exact round time, based on player count. The round time could be displayed above or below all the maps instead for a label for every map. The "time based on player count" suggestion was certainly brought up in the past, and if it is indeed planned for post-rewrite, this suggestion could be implemented as well, but only in the form that I have described. Keep in mind that another suggestion is already being implemented which does a similar thing but displays Trial status instead, so if round time was shown as well it could potentially use up a lot of space in the chat.
Instead of showing the round time for each level individually, the map voting pool should consist of 3 maps with the exact round time, based on player count. The round time could be displayed above or below all the maps instead for a label for every map. The "time based on player count" suggestion was certainly brought up in the past, and if it is indeed planned for post-rewrite, this suggestion could be implemented as well, but only in the form that I have described. Keep in mind that another suggestion is already being implemented which does a similar thing but displays Trial status instead, so if round time was shown as well it could potentially use up a lot of space in the chat.
Altering the voting pool to be based on characteristics such as round time or player count sounds good in theory, but in practice will result in some maps appearing far more often than others.
If the voting pool were to be chosen based on player count, then given that the server is far more likely to be on fewer players than on more, we would primarily be offered small maps for the most of the day, and larger maps would rarely get played.
If the voting pool were to be chosen based on round time, then certain groups of maps will often appear together in pairs or in triples in the pool, and depending on how the code will work, maps with particular round times will appear more often.
The current system that chooses the pool entirely randomly is, I believe, the fairest system possible.
Instead of showing the round time for each level individually, the map voting pool should consist of 3 maps with the exact round time, based on player count. The round time could be displayed above or below all the maps instead for a label for every map. The "time based on player count" suggestion was certainly brought up in the past, and if it is indeed planned for post-rewrite, this suggestion could be implemented as well, but only in the form that I have described. Keep in mind that another suggestion is already being implemented which does a similar thing but displays Trial status instead, so if round time was shown as well it could potentially use up a lot of space in the chat.
Altering the voting pool to be based on characteristics such as round time or player count sounds good in theory, but in practice will result in some maps appearing far more often than others.
If the voting pool were to be chosen based on player count, then given that the server is far more likely to be on fewer players than on more, we would primarily be offered small maps for the most of the day, and larger maps would rarely get played.
If the voting pool were to be chosen based on round time, then certain groups of maps will often appear together in pairs or in triples in the pool, and depending on how the code will work, maps with particular round times will appear more often.
The current system that chooses the pool entirely randomly is, I believe, the fairest system possible.
I agree that keeping it random is the most fair, although I don't think fairness is enough to determine its effectiveness. Think of a hypothetical scenario like this: 3 players are online on the server. A round ends, and voting for the next map begins. Available maps to vote on are an 8 minute, 6 minute and 4 minute map. Player A is an experienced player that has already played through all the maps, and votes for the short, 4 minute round, due to the low player count. Player B is a new player that just recently began playing on the server and doesn't quite understand what round time means and if it matters, so they vote for the map that catches their attention the most: the 8 minute map that just so happens to be the one with the most interesting name. Player C is a decent player that completely dismisses the vote. Effectively, the votes are 50-50 between the very long and the very short map. If the very long map wins from RNG, the round becomes stale and boring. The beginning zombie would have very little choice of winning unless Player A is chosen (33%), and even then the sheer size of the map would make the whole round just not fun to play and players would eventually leave. Player A wouldn't be satisfied because a round has been wasted on a long and boring map, and leave the server. Player B wouldn't be satisfied as they would take the gamemode as boring, and leave the server. Even if Player C doesn't care and doesn't leave anyways, he would still be affected by the the others that had just left and would follow suit. This might be a bit too specific and too much of an exaggeration, but I've seen lots of situations like that play out in many similar ways. I believe that the map voting should be dynamically balanced for the player count to avoid such situations. I don't think maps would repeat too often as player counts would change gradually over time, and there could be some additional code written specifically to avoid matching up the same maps too often. Another way to avoid this would be to occasionally include a longer/shorter map (depending on current player count) as a third map as an alternative to two other maps with normal times, so that if people do wish to play a different map, they can vote on it to do so. Also, map queueing exists if people really need to play a "rare" map once in a while (rare in quotation marks mainly due to the chances of a specific map appearing in a single vote being quite low).
Altering the voting pool to be based on characteristics such as round time or player count sounds good in theory, but in practice will result in some maps appearing far more often than others.
If the voting pool were to be chosen based on player count, then given that the server is far more likely to be on fewer players than on more, we would primarily be offered small maps for the most of the day, and larger maps would rarely get played.
If the voting pool were to be chosen based on round time, then certain groups of maps will often appear together in pairs or in triples in the pool, and depending on how the code will work, maps with particular round times will appear more often.
The current system that chooses the pool entirely randomly is, I believe, the fairest system possible.
I agree that keeping it random is the most fair, although I don't think fairness is enough to determine its effectiveness. Think of a hypothetical scenario like this: 3 players are online on the server. A round ends, and voting for the next map begins. Available maps to vote on are an 8 minute, 6 minute and 4 minute map. Player A is an experienced player that has already played through all the maps, and votes for the short, 4 minute round, due to the low player count. Player B is a new player that just recently began playing on the server and doesn't quite understand what round time means and if it matters, so they vote for the map that catches their attention the most: the 8 minute map that just so happens to be the one with the most interesting name. Player C is a decent player that completely dismisses the vote. Effectively, the votes are 50-50 between the very long and the very short map. If the very long map wins from RNG, the round becomes stale and boring. The beginning zombie would have very little choice of winning unless Player A is chosen (33%), and even then the sheer size of the map would make the whole round just not fun to play and players would eventually leave. Player A wouldn't be satisfied because a round has been wasted on a long and boring map, and leave the server. Player B wouldn't be satisfied as they would take the gamemode as boring, and leave the server. Even if Player C doesn't care and doesn't leave anyways, he would still be affected by the the others that had just left and would follow suit. This might be a bit too specific and too much of an exaggeration, but I've seen lots of situations like that play out in many similar ways. I believe that the map voting should be dynamically balanced for the player count to avoid such situations. I don't think maps would repeat too often as player counts would change gradually over time, and there could be some additional code written specifically to avoid matching up the same maps too often. Another way to avoid this would be to occasionally include a longer/shorter map (depending on current player count) as a third map as an alternative to two other maps with normal times, so that if people do wish to play a different map, they can vote on it to do so. Also, map queueing exists if people really need to play a "rare" map once in a while (rare in quotation marks mainly due to the chances of a specific map appearing in a single vote being quite low).
A fine suggestion if this 'low-player regime' only comes into action when the server is at a lower player count (fewer than, say, 5), and if the community is happy for short-duration maps to appear more often than long-duration maps in such a regime.
However, such a system would require an additional map categorisation system, because round duration is not indicative of player count suitability. Linear course maps may have a long-duration, but many are very much suitable for three players. It is arguably easier to catch a human as a solo zombie on a linear course map, e.g. hell_escape/rainbow_course/encore than on a short pro-run map, e.g. nothing/farm/ctf_turbine. If the regime kicks in at a higher player count then we do have an issue of these easy but long-duration course maps appearing less often.