Charybdis.9042 Posted July 19 Posted July 19 (edited) Just want to report that the EU matchup I'm on (2-4) has the NA point scoring system. We're fighting for 33/22/11 points, which can only be found on the NA table. Edit: and I just want to add that, with it being a flatter point scoring system, it is probably better than the EU one. Cheers, Snaxxi Edited July 19 by Charybdis.9042 2
Nauda.3678 Posted July 19 Posted July 19 Even better: While the match details actually show that we were supposed to get 33/22/11 points in the first skirmish the actualy points were 21/14/7. So these 2 don't even match. 2
Dawdler.8521 Posted July 19 Posted July 19 (edited) 6 minutes ago, Nauda.3678 said: Even better: While the match details actually show that we were supposed to get 33/22/11 points in the first skirmish the actualy points were 21/14/7. So these 2 don't even match. Assuming GW2mists is showing accurate points, all matchups have 21/14/7 points. Which doesnt make sense either unless the points take into account population for ALL of WvW and not individual matchups. We knew that both EU and NA would have had more points at "full" capacity and that none of the matchups had any variety in population during the first 2h seem very unlikely. Edited July 19 by Dawdler.8521 1
XenesisII.1540 Posted July 20 Posted July 20 (edited) Weird, I saw the wiki scoring the first skirmish for EU as 21/15/7, which added up to 42, but the first skirmish for NA is 6/4/2. https://wiki.guildwars2.com/wiki/World_Restructuring#Current_matchups Wasn't this suppose to be up to 33 points, since reset is usually the most populated time for NA of the week? Are they doing this by average of the overall population of all three sides over the two hours? Only explanation I can think of since reset tends to drop off after the first hour, but it dropped off that much that it's barely over the minimum of 9 points? Or did they set it so that you can only get the max points like between 8-10pm est? Seems like a broken system and another needless change. P.S Also looks like they're not doing this per server or match either, since every match is scoring the same points, even though we well know not every server carries the same populations every hour, even in prime times. So doesn't seem population weighted, but time zone weighted. Edited July 20 by XenesisII.1540 2
Chaba.5410 Posted July 20 Posted July 20 LOL! So what NA players should do for Saturday is to log in at 18:00, 20:00, and 22:00 CEST in order to maximize their VP. That starts at 9am PST, btw. 1
Charybdis.9042 Posted July 20 Author Posted July 20 (edited) 9 hours ago, Dawdler.8521 said: Assuming GW2mists is showing accurate points, all matchups have 21/14/7 points. Which doesnt make sense either unless the points take into account population for ALL of WvW and not individual matchups. We knew that both EU and NA would have had more points at "full" capacity and that none of the matchups had any variety in population during the first 2h seem very unlikely. I've just checked the API and it is reporting values that don't match the in-game UI, and as GW2Mists uses the API directly it can only report from there. The 21/14/7 score reported on GW2Mists is from the follow-on skirmish period, which suggests a fencepost error (off-by-one) or a timezone issue. Under the old scoring system these kinds of error wouldn't have been noticed as all skirmishes use the same scoring data. It isn't clear whether the UI or the API is correct 🤪 Just to clarify: the points scoring system has been described by Anet as a static scoring table (it is pre-calculated based on player population) and is not dynamically calculated by population changes. Source: second screenshot on this reddit post Edited July 20 by Charybdis.9042 1
Dawdler.8521 Posted July 20 Posted July 20 (edited) 1 hour ago, Charybdis.9042 said: Just to clarify: the points scoring system has been described by Anet as a static scoring table (it is pre-calculated based on player population) and is not dynamically calculated by population changes. That still means it's based on population ie what I said. Obviously it changes for each skirmish - otherwise it wouldnt change between skirmishes and we'd have the old regular system. So whether you call that change "dynamically calculated" or "pre-calculated" based on population... well that's really just semantics. EIther way yeah something is wrong. NA scores are even weirder now than EU with one team showing as 21/12/9, three at 21/14/7 and one at 19/12/11. None of it makes sense regardless of it counting one or multiple skirmishes lol. EU is behaving as expected even if the scores might be a little off though - scoring differences now the morning after reset are already as large as they used to be on mondays so it's tracking for ~3x higher than "normal" difference (matching the 66%->200% change). Edited July 20 by Dawdler.8521
Charybdis.9042 Posted July 20 Author Posted July 20 1 hour ago, Dawdler.8521 said: That still means it's based on population ie what I said. Obviously it changes for each skirmish - otherwise it wouldnt change between skirmishes and we'd have the old regular system. So whether you call that change "dynamically calculated" or "pre-calculated" based on population... well that's really just semantics. Not semantics at all. Pre-calculated is static and unchanging. Dynamic is responding to (near) real-time player numbers. Very different system behaviours. 1 1
Dawdler.8521 Posted July 20 Posted July 20 (edited) 40 minutes ago, Charybdis.9042 said: Not semantics at all. Pre-calculated is static and unchanging. Dynamic is responding to (near) real-time player numbers. Very different system behaviours. But it's not static and unchanging? It changes every 2h as per the slide? And to get the population, you'd of course have to get the player numbers some time during the skirmish. My guess would be that they check the population every 5m skirmish tick and then average that at the end of the 2h skirmish so they can weigh it against the VP score. Dynamic doesnt necessarily mean real time. WR in itself is dynamic compared to static worlds - but the change frequency is every 4 weeks... Edited July 20 by Dawdler.8521
Charybdis.9042 Posted July 20 Author Posted July 20 7 minutes ago, Dawdler.8521 said: But it's not static and unchanging? It changes every 2h as per the slide? And to get the population, you'd of course have to get the player numbers some time during the skirmish. My guess would be that they check the population every 5m skirmish tick and then average that at the end of the 2h skirmish so they can weigh it against the VP score. Static means that when you come back to the same skirmish slot it will have the same score (i.e. 1, 13, 25, 37 etc. will have the same scores, there being 12 skirmishes per day obviously). This is precisely what the Anet slide deck says by providing a 24hour window on the scoring. At no point in that presentation was a dynamic system mentioned (I watched all of it). Please don't guess 🙂 2
ManiacMika.9851 Posted July 20 Posted July 20 Dynamic would check the population every period (5 min or 2 hours) and change the score according to that. Static is when Skirmish 1 will always be the same as Skirmish 13, 25, 37, 49,... but can be different from Skirmish 2, 14, 26,... (This is what we have now) Flat is what the system was before (always +5 +4 +3 VP). Here are the tables with the swapped times: EU NA: Sk. 1st 2nd 3rd Start (PDT) End (PDT) Start (UTC) End (UTC) 1 6 4 2 7 PM 9 PM 02:00 04:00 2 15 10 5 9 PM 11 PM 04:00 06:00 3 15 10 5 11 PM 1AM 06:00 08:00 4 15 10 5 1AM 3AM 08:00 10:00 5 15 10 5 3AM 5AM 10:00 12:00 6 15 10 5 5AM 7AM 12:00 14:00 7 30 20 10 7AM 9AM 14:00 16:00 8 42 28 14 9AM 11AM 16:00 18:00 9 30 20 10 11AM 1 PM 18:00 20:00 10 15 10 5 1 PM 3 PM 20:00 22:00 11 6 4 2 3 PM 5 PM 22:00 00:00 12 6 4 2 5 PM 7 PM 00:00 02:00 NA EU: Sk. 1st 2nd 3rd Start (CEST) End (CEST) Start (UTC) End (UTC) 1 21 14 7 20:00 22:00 18:00 20:00 2 9 6 3 22:00 00:00 20:00 22:00 3 9 6 3 00:00 02:00 22:00 00:00 4 9 6 3 02:00 04:00 00:00 02:00 5 13 8 4 04:00 06:00 02:00 04:00 6 13 8 4 06:00 08:00 04:00 06:00 7 13 8 4 08:00 10:00 06:00 08:00 8 13 8 4 10:00 12:00 08:00 10:00 9 13 8 4 12:00 14:00 10:00 12:00 10 21 14 7 14:00 16:00 12:00 14:00 11 33 22 11 16:00 18:00 14:00 16:00 12 33 22 11 18:00 20:00 16:00 18:00 1
Charybdis.9042 Posted July 20 Author Posted July 20 Thanks Mika for the table - that matches what I'm seeing on EU scoring. If I had to take an educated guess then I'd say that the UI client is re-deriving the VP from the warscore data (I can't find VP data per skirmish in the API). This is where the "off-by-one" error is taking place, so the client is displaying different scores.
Dawdler.8521 Posted July 20 Posted July 20 (edited) 3 hours ago, Charybdis.9042 said: Static means that when you come back to the same skirmish slot it will have the same score (i.e. 1, 13, 25, 37 etc. will have the same scores, there being 12 skirmishes per day obviously). This is precisely what the Anet slide deck says by providing a 24hour window on the scoring. At no point in that presentation was a dynamic system mentioned (I watched all of it). Please don't guess 🙂 Ok then it’s dynamic. Obviously I’m talking about the continuous skirmish after skirmish VPs for this week, not looking what the history of skirmishes was 🙄 This score changes as it counts up every 2h. It’s also weighted by the population of those 2h, which changes and affect that count. The week isn’t done yet so we better hope it’s dynamic. Otherwise matches would have been done 2h into reset. Anet provided an example of a 24h scoring history. Edited July 20 by Dawdler.8521
One more for the road.8950 Posted July 20 Posted July 20 11 minutes ago, Dawdler.8521 said: Ok then it’s dynamic. It's static because it will always be the same VP at the same time of day. It doesn't change from skirmish to skirmish depending on how many is on, it has a scheduled difference based on when there's usually the most people on. But the points might change later, depending on how it goes, according to stream. 1
Dawdler.8521 Posted July 20 Posted July 20 (edited) 23 minutes ago, One more for the road.8950 said: It's static because it will always be the same VP at the same time of day. It doesn't change from skirmish to skirmish depending on how many is on, it has a scheduled difference based on when there's usually the most people on. But the points might change later, depending on how it goes, according to stream. What I am saying is, can you tell me what the VP score for all 3 teams in T3 EU will be at 23:00 on Tuesday evening? No you can’t because that score changes every 2h. But ignoring that, there is no “scheduled” pre-determined difference based on how many has been in WvW last month or whatever. In practice a skirmish at 03:00 would have the same modifier every day of the week but that’s not how the system works because it modify it by the population in that skirmish. If a skirmish tonight at 03:00 has no players on, it’ll be at low scaling. If you by some absolute strongman miracle 03:00 Monday morning manage to motivate all 3 teams to pack every border with queues then it’s gonna have high scaling. because it looks at the population. It’s not a schedule. Oddly enough, it’s dynamic. Edited July 20 by Dawdler.8521
One more for the road.8950 Posted July 20 Posted July 20 11 minutes ago, Dawdler.8521 said: What I am saying is, can you tell me what the VP score for all 3 teams in T3 EU will be at 23:00 on Tuesday evening? Well, I know that between 22:00 and 24:00 CEST it will always be 30/20/10 for 1st/2nd/3rd no matter tier. At least when they get the EU scoring to EU and NA scoring to NA instead of having them swapped.
Dawdler.8521 Posted July 20 Posted July 20 (edited) 24 minutes ago, One more for the road.8950 said: Well, I know that between 22:00 and 24:00 CEST it will always be 30/20/10 for 1st/2nd/3rd no matter tier. At least when they get the EU scoring to EU and NA scoring to NA instead of having them swapped. How do you know what the population will be at that time for every day in the future? If there was an average 500 players today 22:00-24:00 that skirmish would NOT have the same modifier as a skirmish tomorrow at 22:00-24:00 with 1500 players average (ignoring the fact we have no idea what the threshholds are, just example numbers). It's for the entire region and not per tier so that part would be true (this screws over unbalanced matchups where a tier have very low active pop and other tiers have very high active pop but that's another matter) . But it's still not a pre-determined schedule even if it's extremely likely it'll look similar day by day since player habits doesnt suddenly change. The modifier is determined after each skirmish based on the constantly changing population, not before. That's literally what the slide says. Population per skirmish. People seem to have mistaken Anets example slide with a template that's applied over every 24h now on out. Edited July 20 by Dawdler.8521
Odrson.1463 Posted July 20 Posted July 20 1 minute ago, Dawdler.8521 said: I understand your confusion because it was badly worded by the devs, BUT player count is (currently) NOT intended to be actively re-evaluated, as stated in the VOD, the table they published is the current used metric and HARD CODED into the scoring mechanism. If needed they will tweak these numbers but at no point did they make it sound like this was an automated system nor that these tweaks are run in between skirmishes. 2
Dawdler.8521 Posted July 20 Posted July 20 29 minutes ago, Odrson.1463 said: I understand your confusion because it was badly worded by the devs, BUT player count is (currently) NOT intended to be actively re-evaluated, as stated in the VOD, the table they published is the current used metric and HARD CODED into the scoring mechanism. If needed they will tweak these numbers but at no point did they make it sound like this was an automated system nor that these tweaks are run in between skirmishes. The guy in the VOD says that victory points are going to be weighted based on the player population per that skirmish and that they want award it based on the amount of players that are playing in them. If they have a fixed 24h template based on historical population (ie active at prime, dead at night), it doesnt really weight it by player population in the skirmish. It's just a guess what it's usually around that time. But I suppose the system is just far dumber than what I read into it.
Dayra.7405 Posted July 20 Posted July 20 3 minutes ago, Dawdler.8521 said: But I suppose the system is just far dumber than what I read into it. We as player would not understand a really dynamic scoring system, as we don't know the real player numbers and would only always wonder why the hell this was now this number and not any other.
XenesisII.1540 Posted July 20 Posted July 20 (edited) So 12pm est on a saturday had the most people online, while 10pm reset friday night had the least? Mind blown. /pats the loyal confused fan on the head --> Edited July 20 by XenesisII.1540
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now