Revoluzzer

I guess it's easier for anticheat-developers to test drive stuff like this on smaller titles like APB. Less risk to burn your reputation if it does go wrong. But a great case study to sell their product to big games if it goes well. Since I don't like the idea of running non-critical software with potentially system-breaking access, I'd much prefer a purely server-sided approach using big data and neural networks. Going with a shadow-ban approach is the better way to go, too, I believe. If it doesn't impact matchmaking too much, which it will at such low player numbers. So you'll need to get that sorted out somehow as well. Outright banning people has led to false bans. Not outright banning people has led to cheaters running rampant (and/or people feeling this way). To pit potential cheaters against other ones might even lead to more reliable data. I'm still curious if FairFight couldn't accommodate for the fact that APB gives players what is essentially wallhacks in some missions and thus couldn't make proper use of the data, since everyone at some point showed signs of being at least a wallhacker. Also third person view allowing players to track others "through walls" when peaking around a corner using the camera, while the line-of-sight between (LOS) barrel and target is obstructed. And subsequently "instantly locking on" to a target once the player moves out of cover, thus technically establishing LOS according to some data, while already having LOS according to other data.

The colours are only representations of underlying Glicko values. "Removing" green threat means expanding the other colours to occupy value-ranges previously occupied by a different one. A system which dynamically keeps all active players (e.g. everyone who participated in opposed missions in the past 30 days) evenly distributed across the entire threat-range would help, I believe. The colours aren't really relevant. Could just as well have each represent 25% of the full range. Though a more natural distribution would have the majority of players be in bronze or silver and only the very top end in gold and the very bottom end in green. Again, the colours don't really matter, mind you. Removing them altogether and leave the matchmaking to work in the background would probably be the healthiest choice. Something like soft reset is a mechanic already in place, called confidence. And what this does, is slowing down your threat-"progression". In other words, as confidence increases, threat-mobility decreases. A new player will have a confidence-value of 0, thus their threat swings up and down rapidly. A long-lived player will have a high confidence value and their threat moves like molasses. If the value reaches 1, you have to lower it before your threat starts moving again at all. This confidence value is also the reason why it is difficult to de-threat at first and then becomes very easy to gain threat again. (Copied from here.) Supposedly confidence will increase when you do not log in / participate in opposed missions for a prolonged time, as well as threat slowly decreasing. But I could not confirm this so far. It was only hinted at in past blog posts about the matchmaking system. A hard reset is nonsense. We had two of those and they barely qualified as stopgaps. The system is already self correcting. Once mechanics are in place which hinder players from gaming the system, it will heal by itself. Hard resets are great for people who want some easy matches. Once they're back in the matchmaking-bracket where they belong (and opponents are tough again), they will complain. If phasing ever becomes a reality the population limit on districts won't matter either, because matchmaking will be more global and the pool of available opponents potentially even greater than 100.

I believe it was Heavy Hitters first, then Furious Fighers, then Sharp Shooters. Heavy Hitters had several rounds of invites, I wasn't in the first batch. Perhaps the second, possibly the third. Furious Fighters had a few quick rounds, I think and Sharp Shooters got in all at once. But as you say, it's been quite a while since then.

The beta-UI was fairly similar to the one in that video. Looks like they updated some of the UI before they changed the security truck model, even. The radar is different here, for example.

The maps are designed for 4v4 at most. And I think most of the districts manage to find a good balance between authentic cityscape and tactical playground. Of course all of this falls apart in a 10v10 situation, because coordination usually goes out the window and respawns happen too frequently to keep the defending team away from objectives long enough to complete them. I agree with all but 2v2. That's peak competitive gameplay, as far as I'm concerned. Weapon choice matters most, because you can't cover all bases. Focusing on / teaming up against one opponent at a time matters most in this setup (which is very reminiscent of the very old days). Saving a mission while your teammate is respawning is still very much a possibility, which keeps it interesting at all times. Also I believe 2v2 is a sweet spot for balancing, because you get a lot of teams per district at that size. 4v4 is much more casual in comparison and 3v3 strikes a sweet balance between the two.

If Matt wants to woo us: Show, don't tell.

The actual simple explanation is: At the end of a match, the system will expect you to perform at a certain level. You meet that level, your threat doesn't change. You excel, your threat rises. You miss the mark, your threat degrades. But here's the thing: You don't know what the system expects of you. You can kinda sorta guess, based on the threat-colour of your opponents. But in OP's screenshot, everyone could be just around the tipping-point between Silver and Gold. Which, mathematically, could be nearly identical. Also the amount your threat moves depends on how confident the system is about your currently assigned threat-level. Confidence rises when you consistently meet expectations and lowers when you fail to do so.

Most favourite must be the VBR 'Temptress'. Not a huge fan of the looks, I'd prefer the M14 EBR body over the Ruger Mini. But I really enjoy that fast firing, semi-automatic gameplay it offers. And the versatility both in CQC and at mid-range. Least favourite probably the N-Tec, because it was historically way too prolific and hampered my enjoyment of other assault rifles in the game. Using those meant deliberately putting yourself at a disadvantage. Shame, too, because I'd love to enjoy the COBR-A if it weren't a polished turd.

Crosshairs still turn red, too, I believe.

I think the lore for APB is pretty fleshed out and intricate. Not everything is spelled or laid out and you might have to fill in the gaps here and there. The story or storytelling however is basically non-existent. @mtz's post is really great and acknowledges a layer of depth which I have never really grasped before. I have read the contact-e-mails whenever they arrived and quickly forgotten about them. And I largely ignored their voice-lines at the start of a mission. So APB did never tell me a story. It just provided a canvas for my own.

Hard disagree here. The arcade-y 3rd-person shooting mechanics in APB are some of the best I've played so far. Before they introduced curve mechanics, they struck the perfect spot between simplicity and complexity. I'd wager the driving mechanics struck a similar note, but vehicle balance was always terrible.

That is news to me, indeed. Beyond the introduction of the score system, the only "shifts" I remember G1 applying to threat was moving everyone down a couple notches to improve the situation, primarily because the community felt like too many people were moving towards Gold. That didn't work out because the threat system did self-correct. So they did it again a year or so later. And then, iirc, a third time just before LO took over.

If there truly is a mechanism/flaw which pushes players into Gold threat despite a lack of skill (which I have not seen any explanation for yet), then tweaking matchmaking won't solve it. The only explanation I could see for it is this: New players will jump in, inflate the threat of slightly more experienced players, then give up if they can't get a foot in the door. In my opinion the threat system should already be self-healing, though. If it isn't, then I assume it's because the inactive population is still taken into account when it shouldn't be. Only taking the active current population into account (say actively playing a mission within the last 30 days or so) should solve that particular issue. And with the elimination of threat-based districts, they wouldn't even need to adjust their threat-preference dynamically based on their population.

Besides throwing more people into the pool (which is a good thing when it comes to matchmaking), it will also increase the difficulty of gaming the system. Dethreating and hiding in Bronze districts for easy matches won't work anymore. Of course the next logical step would and should be to remove visible threat altogether. From the sound of it, they will consolidate all players into a single pool. The regions will be represented by physical servers, but those can be spread finer (e.g NA-East, NA-West districts, both in the NA region). Everyone can play in all regions (on all servers), no matter which region the character originates from. So I could take my EU-region character and play on a NA-East server. A player from the NA-region could have the same name as I, so I'd have the appendix @EU while I play in the NA-region. With cross-district-matchmaking, it wouldn't really matter which district you get put into. Your next mission might not even take place in the current one. So why not have everyone play on the same servers? Even if one group might be significantly smaller, they'd feel like they play a well populated game. Which performance-critical activities happen in social district? Aren't they already much less performant because the number of heavily customised avatars is much higher?

Daily rewards aren't for veterans or people who play on the regular anyway. They're incentives for people who otherwise wouldn't, so players like you have someone to play with/against.