We will be monitoring the new Matchmaking System closely over the next few Wars, taking in your feedback and reviewing Alliance activity.
I would like to know if it is possible for Kabam to comment on why the new system is arbitrarily configured to initiate war attack phases at a time when almost no one was starting them before. In the old system, most alliances (at least most that were trying to fight three wars a week during seasons) were starting wars between 11am Pacific (the earliest you could start) and about 5pm Pacific (any later risked not finding a match or drifting towards the 7pm cutoff). The new system runs match making basically during this time but then sets the attack phase to start 20 hours later, which is four hours earlier the next day. That means attack phase starts between 7am and 11am Pacific in the new system. This is earlier than any alliance used to start attack phase (of the alliances fighting three wars a week), and by as much as four hours.
This seems to be the worst possible time to start attack, as it doesn't align with anyone's original preferred start time. Meanwhile, it seems to be completely arbitrary when match making is calculated, so if it was calculated between 3pm and 7pm Pacific, then everyone's wars would be starting between 11am and 3pm, which is on or very near when they used to start in the first place.
This seems like it is a completely arbitrary decision (you could have chosen any time to start matching) and yet Kabam seems to have chosen the worst possible option. Is there any good reason for start times to fall where they are falling now that no one seems to be considering? If not, why not just change it? It seems to be something that requires no structural change to the system, just literally what time the sequence starts.
Why wait to observe what happens in this case? Datamining can only show when people fail. It won't show if people succeed, but at higher personal cost. It won't show people setting alarms, or rushing to finish, or changing their life schedule to try to accommodate their alliance requirements. Datamining doesn't show pain. And in this case, it seems to be completely unnecessary pain.
@DNA3000 - Your assertion that the staggered start times were intended is likely correct.
According to @Kabam Lyra the game starts every war as soon as the match is found because this spreads the load out among the servers that run war, and the servers can't handle all of the wars starting essentially at the same time. So it must have been intentional for the wars to start over a period of time rather than all of them at the depicted start time.
But AQ starts for everyone at the same time and the servers can handle that.
I'm not arguing the technicalities over what might cause this, only that since Kabam gave an actual reason for doing it, and not an obvious ex post facto one, it must have been intentional that it was done that way.
Our war start time is now a full 8 hours earlier than we normally started them, it started 11am EST, we normally start them 7pm EST.
We start that late for a reason, we are all adults with jobs and lives outside of this game. Not all of us can take breaks throughout the day long enough to facilitate the first quarter of attack phase. This just isn’t tenable for us.
I realize it won’t be this early EVERY time, but even just once a week is way too much. We were willing to try working around a 3pm start time, but 11am is way way way too early.
Please make time slots or something. This is going to end up hurting our performance.
But AQ starts for everyone at the same time and the servers can handle that.
AQ + AW > AQ
That's one possibility, but I think the more likely possibility is that AW maps are more computationally intensive than AQ maps. AQ maps are all identical for all alliances (running the same map number, of course) while AW maps are all different due to defender placement. Also, AQ has to update one map per BG, AW has to update two (your attack map and your defense map).
Without knowing the exact implementation, sometimes there are hidden computationally explosive things hiding in the game when you run certain things that make them burn a lot more resources than you might think. AW maps might be one of those things.
Here’s some additional clarity to our previous announcement regarding the new timelines for Alliance Wars.
NEW MATCHMAKING SYSTEM - CLARIFICATION There are 3 Enlistment Periods each week. Alliances must opt-in to their next war during these times.
Enlistment will begin Sunday, Wednesday, and Friday from 3PM PST. (7PM UTC)
When your Alliance enlists, the Matchmaking window will start.
The Matchmaking window is from 11AM to 3PM PST (7PM UTC - 11PM UTC).
You may be matched at any point within this window, however the Matchmaking system prioritizes Alliances with higher prestige and war ratings.
Once matched the game will send the matched Alliances a push notification, please turn your notifications on.
As soon as 2 Alliances are matched, the Defence Placement Phase starts.
Once the Defence Placement Phase ends, the 24hr attack phase begins.
This should correct any conflicting information on the times from our original Post.
We will be monitoring the new Matchmaking System closely over the next few Wars, taking in your feedback and reviewing Alliance activity.
Over this period we will be refining and improving the system, we will post updates in this Discussion thread and append any new information to the top of this Post.
NEW MEMBERS RECRUITMENT - CLARIFICATION As long as you recruit a new member before the matchmaking window starts, they will be able to join your next war. The original behaviour is still correct. However there was a bug found, see below for a workaround.
NEW MEMBERS RECRUITMENT - BUG AND WORKAROUND There is currently a bug where a new member cannot join a war if you have already enlisted for your next war.
The workaround for this is to recruit new members before you enlist, or de-enlist and re-enlist after accepting a new member. We are working to get this fixed as soon as we can.
Question @Kabam Miike Maybeuse the workaround correctly for adding new members, we need to de-enlist, wait for attack /rewards to come, and in the small window between this and the begging of matchmaking, the change would need to happen right? This of course, still opens a possibility of not enlisting on time due to changes for alliances matched close to the end of the matchmaking phase. You guys need to look into this.
If this period of time was to be defined, let's call it, "Alliance Maintenance" window, after attack and before Matchmaking, then enlisting will still possible during this period but having the option to be ready before Matchmaking. Risk es kind of controlled.
I think there still needs to be a way to control attack time; but hopefully ideas flow fast and we can be ready for S8 before hand.
But AQ starts for everyone at the same time and the servers can handle that.
AQ + AW > AQ
That's one possibility, but I think the more likely possibility is that AW maps are more computationally intensive than AQ maps. AQ maps are all identical for all alliances (running the same map number, of course) while AW maps are all different due to defender placement. Also, AQ has to update one map per BG, AW has to update two (your attack map and your defense map).
Without knowing the exact implementation, sometimes there are hidden computationally explosive things hiding in the game when you run certain things that make them burn a lot more resources than you might think. AW maps might be one of those things.
I was just writing the same thing. There's a great many factors that increase the information needed to run AW over AQ. With Defense placement alone, each champ being semi-unique in the sense of where they are growth wise. Tier, rank, level, and sig level (if applicable). Then there's player Masteries that need to be taken into consideration and how they could potentially interact with the Node, per champ. There's literally 150 unique situations per side, making it a grand total of 300 unique situations per war vs the set Maps of AQ. Some of which will be similar, but very few would be identical. i.e. 1 player's champ gets placed on the same node as another player's champ in a different BG; that's also running the same mastery setup, at the same stage of growth.... yeah, that's likely not a common occurrence. So I agree @DNA3000 that there's clearly more needed to run AW over AQ.
I for one dislike this setup.. I usually start war matchmaking between 2:30-3pm Pacific (I'm on west coast) because I have many from east coast, Several from Europe and beyond... this time frame gets most out put. East coast are home from work, in Europe it's late evening.. I get off of work everyday at 2:30..
Our war started at 9am... this is utterly ridiculous. We don't lime it one bit.. this handicaps us because we lose hours of fight time
Can something be changed???
Thanks for looking after the rest of the world kabam 👍👍. From what I’m reading the earliest AW can start in my Timezone is 3:00am and the latest is 7:00am. That also means that’s what time it finishes. No ifs or buts, that’s the window. No work arounds, no anything. Killed AW for basically the whole east coast of Australia and anyone on or near that Timezone. Props.
Yeah.. I don't like this setup. Our war started at 9am Pacific time. I'm on west coast at work, off at 2:30, I usually start war approx. 2:30-3pm because I have many on east coast, Several from Europe and beyond.. that time is usually worked best for most.. East coast off from work, and Europe is late evening... this set up messes that up. We lose 4-5 hours of attack time. Not good.
Can we just please scrap this system and go back to the old one asap @Kabam Miike clearly no one likes the new system at all.
I like it. I know others that do. There are definitely things that need to be tweaked. In the end, players have shown they shouldn't be in charge of matching. I'd much rather go thru growing pains to get to a better system than going back to such easy manipulation of the system.
It's also pretty funny to watch people completely lose their **** over one meaningless war.
well it definitely affected us because we recruited members while the en listing was going on and none of them were able to join so we're in a war with five members versus a full team of 30 LOL
Easy fix would be that any current master team can't be matched with a team lower then plat 2, so master plat 1 /2 would get matched with each other, would allow teams to move and down more between those tiers , p3 & g1 should be matched and g2 & G3 matched, this would allow teams to fight against more equal teams, I know there will be some big teams that drop down to lower tiers like g2/g1 but if they want top rewards, with only 4 week season, no way they could move up all those tiers in 4 weeks. A master / p1 group, should never get matched with g1 or lower,
My 16 mil alliance enlisted for the war as normal during placement phase we noticed we weren’t on the tier 5 war map which is odd because we’ve been in tier 5 for the whole of last season. When it came to the start of attack day we looked at our opponent defence noticing they had placed r3 4*s all over the place, we instantly assumed they were tanking for the next season that was until we realised it was a 4 mil alliance that was in silver 3 with 750 less war rating than us
How can your amazing new war system match our two alliances together and how can it lower our map to there level when on the menu it clearly states tier 5
Think it's obvious by now that you guys need to go back and rework this whole approach (especially with AWA starting way earlier than advertised...).
So re-posting this suggestion:
one possible solution to the player movement (and start-time) challenges that these changes create:
Initiate matchmaking AFTER defense placement phase.
Make defense placement phase "open", meaning players joining an alliance anytime during defensive placement are able to place in that war (as long as everything is placed before matchmaking starts).
There's really no good reason you need a matched opponent when setting defenders, as you shouldn't have any more information after matchmaking than before that would modify strategy for placing defense.
This would carve out an almost 24 hour window after each war when people could move between alliances, allowing them to wait for rewards, finish own wars, etc. It would actually make the issue of coordinating alliance movements/recruits way easier than previously, where you need to create a gap by delaying AW start (or asking your new ally to do so in order to wait for you to join).
Theoretically this could allow alliances to start their 24 hr attack clock at a time that was variable but convenient for them, with the caveat that delaying their start would eat into their time to place defenders in the next one.
Matchmaking could still all be started at the same time for every alliance (and defenders would need to be locking in by that time), but Alliance XYZ could choose to start their 24hrs of attack with a 3 hour delay (eating 3 hours of their clock to set defense for the next one). Matchmaking has been done, but they're simply holding off until a time convenient for them (selected by officers ahead of time). You'd have alliances finishing attack at a different time (so real time scoreboard watching takes a hit), but all wars would be completed and results would be known before the next matchmaking began.
If I understand this suggestion correctly (to be honest I had to think about it for a while) the biggest problem is that it is possible for alliance A and alliance B, who are fighting each other in a war, to start their attack phases at different times. That seems problematic for two reasons. First, whoever starts later gets a significant advantage (they will know what the other side did and can react to that) and second, it might be technically infeasible to run a war in which both sides are not on the same schedule (as in, literally a limitation in the implementation of the game).
In the original system both sides "agreed" to the start time by starting matching at that time. In the current system everyone starts at different times but both sides in a single war are still guaranteed to start at the same time.
There's a theoretical way to fix this, but it layers a significant amount of implementation complexity. Allow everyone to set a preferred start time. When alliance A matches against alliance B, start their war at the average of the two preferred times. If alliance A wants to start at 9am but alliance B wants to start at 11am, the war starts at 10am. This gives some control to the alliances while still starting both sides at the same time.
You could also eliminate the ability to view the other side, but that then eliminates counterplay completely. In other words, you can't react to getting ahead or falling behind. That's honestly better than what we have now (giving the win to which ever side has the best hats would be better than what we have now) but it is still something I think we should try to avoid if possible.
It's also pretty funny to watch people completely lose their **** over one meaningless war.
If it was just one meaningless war, I'd just sit it out. And if it isn't important enough to express a strong opinion on, it should be doubly not important enough to express any metacommentary on.
I know u guys (@kabam) worknhard and do what u can to give us gamers a good experience... but, this new enlisting / matchmaking is not convenient for those in the Australasian regions, as start time becomes 1.00am for us, the beginning and the end of the war revolves around that time. Personally I don’t rhink anyone is going to get up every second day to play at that hour, we have jobs, families, committements, etc... we enlisted at the one we always start the wars, but it ended up at 1.00am.
Is there a fix for this? Did u guys know about that? (That it was going to affect half the globe)
Any word on this would be appreciated
I am against Kabaam being our Fuerer and telling us when we will be placing and fighting. As people around the world play this game. I am sorry you couldn’t figure out a better way to fix the matchmaking system while still offering freedom of choice. I mean I would rather elect to search for a match when my team decides, and if we have to wait to get an equally matched team. Then so be it. It is funny to me how a couple years ago the matchmaking was way worse. It is what it is in my opinion. As long as they are same tier it is a fair fight.
Comments
I would like to know if it is possible for Kabam to comment on why the new system is arbitrarily configured to initiate war attack phases at a time when almost no one was starting them before. In the old system, most alliances (at least most that were trying to fight three wars a week during seasons) were starting wars between 11am Pacific (the earliest you could start) and about 5pm Pacific (any later risked not finding a match or drifting towards the 7pm cutoff). The new system runs match making basically during this time but then sets the attack phase to start 20 hours later, which is four hours earlier the next day. That means attack phase starts between 7am and 11am Pacific in the new system. This is earlier than any alliance used to start attack phase (of the alliances fighting three wars a week), and by as much as four hours.
This seems to be the worst possible time to start attack, as it doesn't align with anyone's original preferred start time. Meanwhile, it seems to be completely arbitrary when match making is calculated, so if it was calculated between 3pm and 7pm Pacific, then everyone's wars would be starting between 11am and 3pm, which is on or very near when they used to start in the first place.
This seems like it is a completely arbitrary decision (you could have chosen any time to start matching) and yet Kabam seems to have chosen the worst possible option. Is there any good reason for start times to fall where they are falling now that no one seems to be considering? If not, why not just change it? It seems to be something that requires no structural change to the system, just literally what time the sequence starts.
Why wait to observe what happens in this case? Datamining can only show when people fail. It won't show if people succeed, but at higher personal cost. It won't show people setting alarms, or rushing to finish, or changing their life schedule to try to accommodate their alliance requirements. Datamining doesn't show pain. And in this case, it seems to be completely unnecessary pain.
I'm not arguing the technicalities over what might cause this, only that since Kabam gave an actual reason for doing it, and not an obvious ex post facto one, it must have been intentional that it was done that way.
We start that late for a reason, we are all adults with jobs and lives outside of this game. Not all of us can take breaks throughout the day long enough to facilitate the first quarter of attack phase. This just isn’t tenable for us.
I realize it won’t be this early EVERY time, but even just once a week is way too much. We were willing to try working around a 3pm start time, but 11am is way way way too early.
Please make time slots or something. This is going to end up hurting our performance.
Matchmaking window: 11a - 3p PST = 7p - 11p UTC...you got that part right.
"Enlistment begin" you have starting at 3p PST (7PM UTC), but 3p PST = 11p UTC, not 7p UTC, as you stated above, @Kabam Miike.
That's one possibility, but I think the more likely possibility is that AW maps are more computationally intensive than AQ maps. AQ maps are all identical for all alliances (running the same map number, of course) while AW maps are all different due to defender placement. Also, AQ has to update one map per BG, AW has to update two (your attack map and your defense map).
Without knowing the exact implementation, sometimes there are hidden computationally explosive things hiding in the game when you run certain things that make them burn a lot more resources than you might think. AW maps might be one of those things.
Question @Kabam Miike Maybeuse the workaround correctly for adding new members, we need to de-enlist, wait for attack /rewards to come, and in the small window between this and the begging of matchmaking, the change would need to happen right? This of course, still opens a possibility of not enlisting on time due to changes for alliances matched close to the end of the matchmaking phase. You guys need to look into this.
If this period of time was to be defined, let's call it, "Alliance Maintenance" window, after attack and before Matchmaking, then enlisting will still possible during this period but having the option to be ready before Matchmaking. Risk es kind of controlled.
I think there still needs to be a way to control attack time; but hopefully ideas flow fast and we can be ready for S8 before hand.
I was just writing the same thing. There's a great many factors that increase the information needed to run AW over AQ. With Defense placement alone, each champ being semi-unique in the sense of where they are growth wise. Tier, rank, level, and sig level (if applicable). Then there's player Masteries that need to be taken into consideration and how they could potentially interact with the Node, per champ. There's literally 150 unique situations per side, making it a grand total of 300 unique situations per war vs the set Maps of AQ. Some of which will be similar, but very few would be identical. i.e. 1 player's champ gets placed on the same node as another player's champ in a different BG; that's also running the same mastery setup, at the same stage of growth.... yeah, that's likely not a common occurrence. So I agree @DNA3000 that there's clearly more needed to run AW over AQ.
Our war started at 9am... this is utterly ridiculous. We don't lime it one bit.. this handicaps us because we lose hours of fight time
Can something be changed???
Can something be changed about this??????
I like it. I know others that do. There are definitely things that need to be tweaked. In the end, players have shown they shouldn't be in charge of matching. I'd much rather go thru growing pains to get to a better system than going back to such easy manipulation of the system.
It's also pretty funny to watch people completely lose their **** over one meaningless war.
How can your amazing new war system match our two alliances together and how can it lower our map to there level when on the menu it clearly states tier 5
If I understand this suggestion correctly (to be honest I had to think about it for a while) the biggest problem is that it is possible for alliance A and alliance B, who are fighting each other in a war, to start their attack phases at different times. That seems problematic for two reasons. First, whoever starts later gets a significant advantage (they will know what the other side did and can react to that) and second, it might be technically infeasible to run a war in which both sides are not on the same schedule (as in, literally a limitation in the implementation of the game).
In the original system both sides "agreed" to the start time by starting matching at that time. In the current system everyone starts at different times but both sides in a single war are still guaranteed to start at the same time.
There's a theoretical way to fix this, but it layers a significant amount of implementation complexity. Allow everyone to set a preferred start time. When alliance A matches against alliance B, start their war at the average of the two preferred times. If alliance A wants to start at 9am but alliance B wants to start at 11am, the war starts at 10am. This gives some control to the alliances while still starting both sides at the same time.
You could also eliminate the ability to view the other side, but that then eliminates counterplay completely. In other words, you can't react to getting ahead or falling behind. That's honestly better than what we have now (giving the win to which ever side has the best hats would be better than what we have now) but it is still something I think we should try to avoid if possible.
Very nice match making system . So much improved
The war schedule graphic suggests it.
With a grey maintenance filling out Tuesday/Wednesday
Seems it isn’t the case at all..
Just poor design
If it was just one meaningless war, I'd just sit it out. And if it isn't important enough to express a strong opinion on, it should be doubly not important enough to express any metacommentary on.
what is the reason?
It relly screwed us around.
And as for ‘fair’?
Man we are silver 3 and we are up against an alliance with 2 ‘Legends’
We never seen a legend before haha
This should be impossible in a top-down matching system. And I can't even imagine how you can screw up top-down match making.
Is there a fix for this? Did u guys know about that? (That it was going to affect half the globe)
Any word on this would be appreciated
... I got nothing here. If this can happen at all, even once, the entire system implementation has to be horribly broken.