Talla

Thanks alot for the response. Do you know how long would it take to run out of memory? I'm wondering if the length of time before the crash depends on the memory capacity of the card. Could just be a bad garbage collection reporting of the available/cleared memory, and a misrepresented integer overflow. A question for someone who read all the engine blog posts : How many existing redistributable libraries were able to be updated to 64-bit? vs. Which parts of the custom game code has to be entirely rewritten?

A while ago, I read the new RTX cards weren't able to effectively render in APB. With a new AMD GPU release on the horizon, could anyone provide feedback on the performance of AMD GPUs, RT 5000 or newer. Cheers!

Will parts of the codebase ever become source-available? Let us help you! We all want to achieve the APB endgame vision. Contributing a few pull requests would be better than refreshing the admin tracker on a daily basis, waiting for things to happen.

Fairfight was good when it came out. Nowadays, it would be nearly useless. EAC isn't bad but it isn't groundbreaking either. Pretty invasive & signature-based, which means indefinite cat/mouse detection cycle. If I were an anti cheat developer, I'd just give up at this point. Cheats are too far ahead nowadays. How the hell would you even detect computer vision auto-aim? Anti-cheats need to be designed INTO games during development, not integrated as a "one size fits all" add-on...

.45 with Chambered round mod is a blast. Easily a top 10 gun in terms of gameplay.

Hey at least the lights are still on.

Here I was expecting a highly experimental, potentially revolutionnary idea. Received highly experimental, potentially revolutionnary shitposting instead. I'm happy with that answer. Thanks for sharing

Welcome back, I'd like to understand your line of thought in detail. Please explain.

Yes, but actually no. https://unitgame.com/about_us.html Still, making a pull request on APB's public repo would definitely be somewhere on my life's bucket list.

A monument for all devs to remember, despite the tremendous talent of the team, there is no such thing as "Too big to fail".

Former employee analyzing how Realtime Worlds went bankrupt. https://web.archive.org/web/20200814025847/https://lukehalliwell.wordpress.com/category/realtime-worlds/

Nothing prevents another company to make a spiritual successor under a different IP (Besides the sheer amount of manpower required to make a game of this scale).

The maps, models for cars & guns, animations & textures could be exported to a new engine. There would be some heavy lifting to be done regarding all the custom code: Character creation, Symbols, Clothing a.k.a. skinning, LODs, physics... maybe even network... Starting from scratch would be best, but there is so much to do... If a "10-devs or less" team could update this game engine in 4 years by themselves, I'd be impressed.

It is just another side effect of a stagnant active population. Veterans & hardcore players optimize everything to win, and end up scaring new players / solos / casuals away. Nothing new here. Enforcing solo matchmaking would indeed make matchmaking teams fairer (More possible combinations of solo player rankings), but you would probably futher kill the game population, lower the matchmaking sample size and make things worse. You simply can't matchmake fairly an instance with 10/10 pop and ensure players find a match within 5 minutes. I'd agree with this. Create instances where grouping is impossible, and calling backup is the only way to construct a team. I think this was planned in the matchmaking rework with instance phasing. Grouping would only be possible in Social. The team gets attached an historical or sum of individual ratings. You then phase that group to the district with the appropriate team battle ratings. That would ensure full premades, duos or solos quickly find matches on a comparable scale.

Anyone has a recording of old APB Twitch stream, when Tiggs used the server-side spectate feature? It was a sight to behold. My jaw dropped as low as its FPS. It is simply not good enough to judge between a closet or a legit player (at that time). Monitoring GMs cost alot in the long run, even if you outsource this work to some 3rd world offshore company. You want to rely on automated anomaly detection tools. This game has too many security design oversights to prevent cheating.

Glad to see someone gets it. The engine update was needed for console deployment mostly. Having a unified codebase also allows devs to send updates more confidently and faster. In the backstage, it's still Unreal 3.5. Cheat makers already have no difficulty to hook onto it, and we'll be back to the good ol' signature-based detection that is actually worse than FairFight.

Also true

This game was very ambitious at the time. It's so ambitious that no one in the gamedev industry even tried to reproduce it, yet. I'm eager to see someone try with modern development tools.

Imagine if the devs went live with that idea :

50m is not terrible. Tap-fired at the proper rate, that thing was a monster.

RHD LS1 "Corsair" for the fun factor. Mobility + Pin-point precision + Cover/defensive gameplay. tied with .45 ACP with chambered round mod from the xmas gun game

Old videos of network going bonkers when we switched from Overkill to Softlayer servers, post NA merge. Another bug from bad networking https://www.youtube.com/watch?v=dLjMFugNIco A bug that used to occur when exiting cars gave you maximum falling velocity and ensuing Gomba stomp https://www.youtube.com/watch?v=lGGwUSSvLPE

Colby CSG + .45 AP Once upon a time, SHAW 556

So the description saying: "Doubles base health regeneration for 5 seconds" is actually not applied on medspray? It's only removing the delay? Receiving 2 obir burst is a easy way to test. If I recall, the sound effect triggers when regen delay is reached? Or when full HP is reached?

@xiphos @MonkaS I was part of those xmas 2016 false Punkbuster bans. FairFight did not advertise my ban. It was a silo PunkBuster error. I remember digging and seeing how the PB package for APB was never properly configured in the game files. I really doubt PunkBuster would do any kind of ESP tracking. It monitors the memory for interfering processes, not player inputs according to context.