NKato

Did you miss the part where the vehicle also blows up on its own five seconds later?

At this point, it feels unrealistic for heavily armored vehicles to simply combust in the form of a secondary explosion, when there are no indications that the vehicle has been carrying something particularly explosive. I would also prefer to have a higher chance of survival in these armored vehicles.

The light vehicles we have in the alpha are obviously designed to be resistant to explosives, even of the shaped charge variety. Here are some tests I ran with different explosives on the Hunter MATV: Anti-Tank Mine: When driven over, blows out tires, and disables vehicle for five seconds prior to secondary explosion (rendering said vehicle into junk). Explosive Charge (C4): When detonated beneath it as it is driven over, same result as Anti-Tank Mine. Except damage is more critical and the vehicle detonates with fatal force before I can get away. Explosive Satchel: Same result as the Anti-Tank Mine. Except that this time, when I hit Eject, I got run over by the Hunter. Then when I laid there injured, it blew up, killing me. APERS Mine: Dinky. Ineffectual against the vehicle. Zero damage. APERS Mine (Bounding): When driven past with the Hunter, it sprang up and did damage to the tires. Still Combat-Effective. M6 SLAM Mine: Negligible Damage to the Hunter. Claymore Mine: No Damage. (Well duh, it's ball bearings.) I think that the MATV's bomb resistances need to be adjusted. My recommendations: Anti-Tank Mine: Needs to do damage to the engine and hull, while not damaging the tires. AT Mines are directional shaped charge explosives. Explosive Charge (C4): It's a concussive blast, so it should do light damage to the engine and severely damage the tires (as they are inflated with air), but still have at least some tires remain operational so that the vehicle can be driven out of the blast zone. Explosive Satchel: Same as the Explosive Charge, but with definitive disabling damage (blown tires, critically damaged engine). My main gripe with the MATV and the higher end explosives is that despite their obvious explosive-resistant designs, the vehicles are disabled far too easily. And my biggest complaint is how they explode so soon after being completely disabled. This, to me, feels inappropriate for a vehicle that does not typically carry large amounts of explosives. We have seen MRAPs in Iraq and Afghanistan survive explosions that left massive craters in the highway, and they were immobilized, but still combat effective (they have CROWS turrets) that enabled them to defend themselves in a heavy combat engagement. Therefore, I offer a solution: It would make more sense if the vehicle simply caught fire when its health reaches zero, and burned itself out over a period of time (five minutes?). This would require the use of fire extinguishers to save an MATV from total destruction, and the availability of an Engineer to repair it back to combat effectiveness. What do you think? Is this a workable solution?

Oculus Rift support for ArmA 3, make it so! </Picard>

A battleship.

My advice to BIS is to consider completely rebuilding the AI code if they have to. In the long run (10+ years from now) it will be more beneficial to the company since they'll have a documentation policy in place at this time. Without that documentation, you may as well be staring into Pandora's Box. P.S. I don't see why they can't just, you know, rip out the old AI and replace it with xAItment like they did with VBS2. It can't be that expensive, can it?

I am already in the planning and design stages of a special mod for ArmA 3. I have a modeler lined up for production of the asset. Now, the question is...will I be able to get the programmer I need to help me with this project? ;) Here's a hint: In regards to the project, I plan on doing a separate forum thread for it, so that I can get feedback from the ArmA community on the design elements of the project. Please look forward to it! :)

Let's not forget that it would be good to have overwatch snipers, people who would watch the sides of a building (the windows, to be exact) with guns ready to take out any targets when necessary. This was a feature in SWAT 4, and the concept of having people watch the building externally while a squad goes in, would be very much doable.

I just thought of something else. You know what would make this city an even worse nightmare for infantry? Subways. An underground system of sewers and mass transit networks. That would be an interesting thing for BIS to implement, and it would actually give us a real function for bunker-buster weapon systems.

One quick mission scenario for such a city map would be to be a ground soldier leading a squad, fighting against 12 enemy helicopters patrolling the city. This is reminiscent of the finale of the GITS: SAC finale for the second season. That said, I think the vehicular AI will need to be tweaked a bit for urban combat. Indiscriminately shelling a building to try to kill the people inside it would be out of the question. It would need to be done in a manner as such that the AI will only open fire when it has a confirmed line of sight on the target. This way, collateral damage is kept to a minimum. Of course, it would be a good idea to have three different AI configurations for urban warfare: "Indiscriminate" where the Ai will just blow up anything to get at the enemy, "Discretionary" where the AI will sometimes blow up buildings, and fire directly on a position even though it has no visual contact, and "Cautionary" where the AI will avoid shooting unless it has a confirmed line of sight.

I think the main problem is that they are still stuck in an old school of programming when it comes to waypoint and pathfinding for AIs, especially when it comes to three-dimensional combat in buildings - as a result, the AI is unreliable by current standards. They will need to rip out the old AI code and replace it with something completely new for a proper urbanized FPS from BIS to become reality. There's a bevy of reasons why many urbanized FPS's are heavily scripted in singleplayer, and rarely involve co-operative play in multiplayer, and this problem is one of them.

So they want us to sit through three generations of pissy indoors AI? Scratch that, four generations. I forgot to count Operation Flashpoint/ArmA X.

The thing is, ArmA 3's graphics engine is based off of ArmA 2, and I am concerned that the graphics engine will not handle a high-poly environment well. See, when I play Chernarus in Combined Operations, I get the occasional hiccup, but Takistan has zero problems for me because of the near-complete lack of forests to increase the rendering overhead. That's why I feel a graphics engine that can "scale down" graphics with adaptive culling (we already have a version of that in ArmA 2) or a better LOD. Frankly, the very fact that we're going to DirectX 11, and that high-end DX11 cards run DX11 games quite well, tells me that if the game's graphical assets (textures, models, etc) are optimized in terms of LODs (Levels of Detail, less polys for the model in the distance, more polys up close), ArmA 3 will have fewer problems adapting to a heavily urbanized metropolitan environment. @SGTice: I personally think going for a smaller island, around 30 to 50 square kilometers, with two cities, one large (25 sq. km footprint) and one smaller (10 sq. km footprint) situated on opposing ends of the island, would be a good test case for Bohemia to use in developing the technology to support urban warfare. ---------- Post added at 01:26 AM ---------- Previous post was at 01:24 AM ---------- Exactly like that. They will need to integrate Xaitment into the ArmA 3 editor if they want to support true urban environments. However, there are still problems with AI collision detection - when the AI goes into that high rise building, you can see him clipping into the inner wall of the stairwell. That will be a problem.

My personal assessment of BIS' ability with their existing technology? Doable. There just needs to be a lot of work done on the AI programming so they have two subsets - one for in the field like we have traditionally had, and another set of AI programming specifically for indoor combat. In my honest opinion, it's possible to consider developing a separate line of games specifically designed for urban warfare while using a heavily modified version of the ArmA engine to ensure that the game retains the same amount of customizability as the ArmA series. This would probably be the easiest route to take - keep the scripting of ArmA, but develop a new AI, pathfinding, improve the graphics engine so that it works specifically for urban environments, and so on. It's a lot of work, to be sure, but that doesn't diminish the potential payoff.

It also explained what limitations there would be in the interiors - 15 floors in all would only be modeled, for a high-rise. If you've been inside an office skyscraper, you'd know just how bland the hallways are. With ArmA 3 supporting DirectX 11, it's possible to have all the hallways be a single flat poly for each surface (4 surfaces in all for the floor, walls, and ceiling), with the exception of doorways and intersections, and use tessellation to give depth to the wall textures. The end result: Interiors are going to be VERY low-poly compared to a BMP-3 vehicle. In fact, some skyscrapers could be low-poly in the first place, thanks to the use of tessellation.