This game isn't going to be revived unless they manage to actually bring in new players and keep them

[yood 345]

Just now, CookiePuss said:

Sorry, im like 90% empty thoughts.   3884 posts Joined Apr 2017

[CookiePuss 5378]

Just now, YooD said:

6 minutes ago, CookiePuss said:

Sorry, im like 90% empty thoughts.   3884 posts Joined Apr 2017

Dont look up my older accounts... It doesnt reflect well on me. 

[Rebelliousness 85]

33 minutes ago, Slickmund said:

I reckon there's a list of solution I can't even think of to help improve matchmaking, however I do strongly believe hard threat locking are a must at some point. Having mixed districts besides that to give players the choice to really play up, and learn from that and hopefully their opponents, would be lovely as well.

Hard Threat locking IS Threat Segregation, and at one of it's iterations tried to lock all the golds onto a Gold server and nearly killed the game until it was changed. threat segregation simply fractures an already small playerbase.  Players DO learn skills from seeing how better players do things.  Why then, when the original game system accomodated all threat levels and ranks would that system be argued against, when what was implemented in it's place was so unjust, broken and destructive?

For one thing, the original system is a terrible threat to the golds who need to play on bronze... because it would make them extinct, as a truly skill-based matching system would show what they are truly made of, rather than setting groups to op weaker, lower threat playerbase.  Otherwise I can't imagine who would argue against it.  At least in the original system, when you encountered a max rank Gold, you KNEW you were up against someone of prestigious skill.

[vsb 6171]

22 minutes ago, Rebelliousness said:

Hard Threat locking IS Threat Segregation, and at one of it's iterations tried to lock all the golds onto a Gold server and nearly killed the game until it was changed. threat segregation simply fractures an already small playerbase.  Players DO learn skills from seeing how better players do things.  Why then, when the original game system accomodated all threat levels and ranks would that system be argued against, when what was implemented in it's place was so unjust, broken and destructive?

For one thing, the original system is a terrible threat to the golds who need to play on bronze... because it would make them extinct, as a truly skill-based matching system would show what they are truly made of, rather than setting groups to op weaker, lower threat playerbase.  Otherwise I can't imagine who would argue against it.  At least in the original system, when you encountered a max rank Gold, you KNEW you were up against someone of prestigious skill.

the rtw threat system would allow dethreaters to go all out on low skill opp but then let them win the mission, and then the dethreaters would continue to stay low threat because they lost even tho they went 15 and 2 on some bronzes

a W:L based threat system is bad, idk how many times i can say this lol

[mynd 120]

Right now it would be nice if the servers would simply run reliably.

3 hours ago, Rebelliousness said:

and scoring changes

I hated most of the scoring changes; it ruined the value of Prestige/Notoriety multipliers and significantly devalued TK and AFK demerits.

[Uminee 15]

47 minutes ago, BXNNXD said:

a W:L based threat system is bad, idk how many times i can say this lol 

Elaborate maybe? Also, suggest alternatives?

If you're not basing threat on win/loss then you're either doing it by rank or by kill/death ratio, and those are both a lot more problematic than win/loss.

If you base threat on rank you're just encouraging rerolling, and punishing the max rank players for whom shooter games are not a primary skillset.

If you base threat on kill/death then you are misrepresenting the skills of players who focus on the objectives, and who play more supportive roles such as drivers.

Personally ... I've never liked threat at all. I understand why it's needed in a game with as much of a tactical learning curve as this one, but I don't feel like it should be a giant obtrusive symbol constantly affixed to a player to determine their worth. It just creates an environment of constant judgement and breeds toxicity. But by now it seems that this sort of treatment is what the current remaining population has come to expect/want. I wish it were not so, I feel like this game would be a lot more accessible to people who didn't have a predisposition to thrive on frustration if threat weren't even visible. Edited July 12, 2018 by Uminee

[vsb 6171]

1 minute ago, Uminee said:

Elaborate maybe? Also, suggest alternatives?

If you're not basing threat on win/loss then you're either doing it by rank or by kill/death ratio, and those are both a lot more problematic than win/loss.

If you base threat on rank you're just encouraging rerolling, and punishing the max rank players for whom shooter games are not a primary skillset.

If you base threat on kill/death then you are misrepresenting the skills of players who focus on the objectives, and who play more supportive roles such as drivers.

Personally ... I've never liked threat at all. I understand why it's needed in a game as much of a tactical learning curve as this one, but I don't feel like it should be a giant obtrusive symbol constantly affixed to a player to determine their worth. It just creates an environment of constant judgement and breeds toxicity. But by now it seems that this sort of treatment is what the current remaining population has come to expect/want. I wish it were not so, I feel like this game would be a lot more accessible to people who didn't have a predisposition to thrive on frustration if threat weren't even visible.

i think the system we have right now could be perfectly useful, but there needs to be a serious overhaul of both what actions accumulate score and how much score those actions give

but measuring threat solely based on one statistic is always going to skew threat for certain people and make it easily exploitable for others

[soLucky 7]

2 hours ago, Rebelliousness said:

Well Threat Segregation was exactly what it means... segregating the Districts.  It wasn't part of the original game. The threat and matchmaking systems have had a number of adjustments, but the most ruinous was the Threat Segregation patch, which completely introduced a whole new game.  the previous game wasn't being abused... it was the new system which created a system of abuses, like falling into a hole which the game never could get out of. The more they tried to fix, the more they broke the system.  This is why, at this point LO would be advised to evaluate the original game.

Both rtw and G1's first attempt at threat and matchmaking had their own issues which were abused pretty heavily.  I personally don't like the new system but I do think it is a bit easier on newer players than the older ones.

[Uminee 15]

25 minutes ago, BXNNXD said:

i think the system we have right now could be perfectly useful, but there needs to be a serious overhaul of both what actions accumulate score and how much score those actions give

but measuring threat solely based on one statistic is always going to skew threat for certain people and make it easily exploitable for others

Apb is not structured in such a way that it would be remotely possible to create an accurate metric for actions that contribute to a victory.

For instance, my main skillset is driving. When it comes to shooting, not so great. Under the current threat system, I've been gold for years.

Could you imagine the coding nightmare that would be required to determine a statistical value of, for instance ... timing the nitro on a vegas to use a random car as a ramp and leap up onto an overpass to intersect fleeing opposition, or reach a mission objective more quickly?

And yet, these are the kinds of insane variables that can take place in every mission, and as such would require an engine far more complex than the entirety of APB to accurately account for.

Or ... you could measure whether or not one's ability to pull off that stunt had any actual impact on the outcome of the mission ... and the most direct way to do that, is by seeing if they won the mission or not. After enough missions, one will be able to statistically extrapolate the general likeliness of someone winning or losing missions, and use that data to assign their threat level.

Is it perfect? Of course not, but any other option is either less perfect, or would require a quantum supercomputer and an enormous research grant. Edited July 12, 2018 by Uminee

[CookiePuss 5378]

4 minutes ago, Uminee said:

 for instance ... timing the nitro on a vegas to use a random car as a ramp and leap up onto an overpass to intersect fleeing opposition

 

come clean, this never happened

[vsb 6171]

3 minutes ago, Uminee said:

Apb is not structured in such a way that it would be remotely possible to create an accurate metric for actions that contribute to a victory.

For instance, my main skillset is driving. When it comes to shooting, not so great. Under the current threat system, I've been gold for years.

Could you imagine the coding nightmare that would be required to determine a statistical value of, for instance ... timing the nitro on a vegas to use a random car as a ramp and leap up onto an overpass to intersect fleeing opposition, or reach a mission objective more quickly?

And yet, these are the kinds of insane variables that can take place in every mission, and as such would require an engine far more complex than the entirety of APB to accurately account for.

Or ... you could measure whether or not one's ability to pull off that stunt had any actual impact on the outcome of the mission ... and the most direct way to do that, is by seeing if they won the mission or not. After enough missions, one will be able to statistically extrapolate the general likeliness of me winning or losing missions, and use that data to assign my threat level.

Is it perfect? Of course not, but any other option is either less perfect, or would require a quantum supercomputer and an enormous research grant.

i mean we already have a pretty good system of measuring actions that result mission winning, and it includes actually winning the mission

in your example you could do nothing but vegas jumps - 500m from the objective - and still win and still threat up

or the opposite, do nothing but get kills but lose the match despite going 30 and 3 and still threat down

[Uminee 15]

Just now, CookiePuss said:

6 minutes ago, Uminee said:

 for instance ... timing the nitro on a vegas to use a random car as a ramp and leap up onto an overpass to intersect fleeing opposition

 

come clean, this never happened

If it never happened,  I wouldn't be so proud of it that I find ways to work it into random topics on the forums even years later

[CookiePuss 5378]

2 minutes ago, Uminee said:

If it never happened,  I wouldn't be so proud of it that I find ways to work it into random topics on the forums even years later

Shit... well now I gotta try it some time.
... Im going to lose so many missions cuz of you.

[Uminee 15]

2 minutes ago, CookiePuss said:

Shit... well now I gotta try it some time.
... Im going to lose so many missions cuz of you.

Maybe, but I can guarantee the one time it works will be worth the reactions
 

7 minutes ago, BXNNXD said:

i mean we already have a pretty good system of measuring actions that result mission winning, and it includes actually winning the mission

Yes ... exactly ... but you had previously said that basing threat off of win/loss is bad. My entire post was explaining that 'pretty good system' that you were calling bad previously. I'm not certain if there's contradiction or miscommunication, sorry.

[vsb 6171]

Just now, Uminee said:

Maybe, but I can guarantee the one time it works will be worth the reactions
 
Yes ... exactly ... but you had previously said that basing threat off of win/loss is bad. My entire post was explaining that 'pretty good system' that you were calling bad previously. I'm not certain if there's contradiction or miscommunication, sorry.

W:L threat based system is what RTW had

not exactly sure how early g1 threat was calculated 

the current system is based upon score and while you do get a sizeable score bonus for winning, a majority of your score generally comes from accumulating score during the mission (for completing objectives, for getting kills/assists, etc), therefore -theoretically - your performance during a mission more accurately determines your threat

in practice i think theres not enough actions that give score and too much score being handed out freely for the actions that are measured, but its a decent framework id like to see tweaked rather than scrapped altogether

[Uminee 15]

Thank you for elaborating BXNNXD. That sounds a lot more reasonable than scrapping win/loss having any effect on threat which seemed to be implied.

Your responses and their incredible timeliness do make me think you skimmed through much of what I was saying though. Not your fault entirely, I can be overly wordy.

Of course having vegas stunts effect one's score would be ridiculous. However it's something that is possible to do, requires considerable skill, and can score a victory. I had that as an example of how it's impossible to calculate everything that a player does to factor into winning. People with certain skill sets get neglected by imperfect scoring systems, whereas direct win/loss doesn't care how you got it done, it just knows that you can. There are pros and cons to each I'm sure.

[vsb 6171]

9 minutes ago, Uminee said:

Thank you for elaborating BXNNXD. That sounds a lot more reasonable than scrapping win/loss having any effect on threat which seemed to be implied.

Your responses and their incredible timeliness do make me think you skimmed through much of what I was saying though. Not your fault entirely, I can be overly wordy.

Of course having vegas stunts effect one's score would be ridiculous. However it's something that is possible to do, requires considerable skill, and can score a victory. I had that as an example of how it's impossible to calculate everything that a player does to factor into winning. People with certain skill sets get neglected by imperfect scoring systems, whereas direct win/loss doesn't care how you got it done, it just knows that you can. There are pros and cons to each I'm sure.

nah i read through all your posts since i basically live on the forums, if it comes off like i skimmed anything its prolly my lack of brain cells

i agree that each system has pros and cons, i personally think that the more variables can be recorded and scored the better threat determination will be - kind of like fairfight using a ton of different stats to determine cheating, instead of just banning anyone with a high k/d ratio

[TheMessiah 430]

title

[UnoriginalPerson 11]

Forgot to put this in my original post. Not saying this thread is full of them, but for the elitsts who I know browse the forums:

https://en.wikipedia.org/wiki/Dunning–Kruger_effect

 

[CookiePuss 5378]

3 minutes ago, UnoriginalPerson said:

Forgot to put this in my original post. Not saying this thread is full of them, but for the elitsts who I know browse the forums:

https://en.wikipedia.org/wiki/Dunning–Kruger_effect

 

I like the synergy of your name and your post.
 

[mynd 120]

I'm personally tired of Win/Loss MMR style systems.  They're not reasonable in team-based games, especially one as unbalanced as this one.

I think matchmaking should just be thrown out the window entirely.  Focus on the backup system and getting district pops back to 50/50.  Backup should only be allowed if there's a reasonable difference in team scores, ie one team easily beating the other, and should probably favor attackers a bit more than defenders.

In practice, that would mean that teams struggling to complete objectives and dying a lot would be allowed to call in backup.  No idea is perfect, and my idea comes with issues such as being ineffective in scenarios where the time to complete a stage might be too short.

It does bother me when people call backup just because they can though, even when already in an advantageous scenario.

[Nite 261]

20 hours ago, BXNNXD said:

in practice i think theres not enough actions that give score and too much score being handed out freely for the actions that are measured, but its a decent framework id like to see tweaked rather than scrapped altogether

This reminds me of the time when I suggested back in the old forums that points could be earned for actions that specifically count as defending objectives to try and balance out point accumulation and instead we got some watered down solution where you get x points every so often for simply being the defending team. Wasn't totally happy with how that turned out, in retrospect (was still better than nothing though)

[vsb 6171]

2 minutes ago, Nite said:

This reminds me of the time when I suggested back in the old forums that points could be earned for actions that specifically count as defending objectives to try and balance out point accumulation and instead we got some watered down solution where you get x points every so often for simply being the defending team. Wasn't totally happy with how that turned out, in retrospect (was still better than nothing though)

please include a trigger warning i got ptsd flashbacks from opening that sayvara post

an interesting idea tho, i can see how it would help balance the score so that successful attackers arent automatically the top 50% 

[Faithh 2]

just maybe bring the old threat system back ? with gold 1 to 10..... maybe peeps have then a reason to play again since many players also left cause of that update 

[SelttikS 224]

mmr wise winning should count but so should overall performance during the mission. A well fought mission vs op that is of like mmr should be more valuable than a stomping on noobs.