Evil_Echo

Energy like blast, radiation use functions that varies by radius^(-3). Some other aspects like size of fireball vary by yield^(1/5). Just FYI.

Perhaps the OP thinks he speaks for many. Don't EVER assume you speak for me. In regards to the so-called improvements he claims the entire community desires.... 1. Not all modules need MP compatibilty. Certainly some do, BIS seems to be doing fine in their efforts to improve the situation. 2. Swimming with a dozen kilo or more of gear is unrealistic. Expecting to maintain bouyancy with such a load without external aids is complete nonsense. 3. You don't run or walk toward a live grenade. If anything you drop to the ground after a throw. 4. The 3d editor is clearly experimental. What is there is usable enough for a competent coder. Meanwhile I can wait for whatever BIS chooses to do with this concept. 5. Release the MLODs? For what reason should they give up valuable IP? If you want a model go make your own. 6. See no problem with player vs player CQB. The AI has it's issues in tight areas, but why spend time there when much more can be gained in other areas? If you want more of a challenge then play vs humans.

Always appreciate your efforts Mr-Murray. Do keep up the good fight on this.

Where did you get your facts from? I use multiple curves, based on the various weapon effects. Calibrated to real data and functions from legitimate unclassified sources. And that curve is not hyperbolic, its a inverse-power function. Clearly both of us cannot be correct, so is the US Dept of Energy wrong or are you making things up? I respect your attempt to develop a better nuke, that is commendable. But wish the result would more closely model the real thing. If that is not the goal, then just be upfront about it to not deceive those expecting something else.

You did imply it strongly by talking about the need to improve the physics, including scripts, plots of force curves, and mentioning "real looking forces". But let's not haggle semantics. While some things are simplified, BI does not use arbitrary functions in their code. Measuring damage at fixed intervals from a bomb or satchel blast will clearly show how it works. Basic physics shows that, ignoring absorption by intervening media, weapon blast damage should be proportional to 1/distance^3, a curve that drops dramatically. Your damage curve is practically linear - which would cause far too much damage at long range. You could still use stepped zones as quick approximation, just adjust your numbers to match the inverse cube relation I mentioned. However using nested nearObjects commands is quite expensive computationally. FYI.

I'm sorry, but this is not physics. The numbers looks arbitrary, stepped zones vs actual range calculations, etc. It may be an improvement, but please don't call it realistic.

Have done the trick of having a man on the ground lob smoke and the AI pilot searching for smoke. Works well.

Request for ability to get non-classed objects... You got it. While possible now, it takes two nearestObjects calls ( [] to get everything, ["All"] to get the classed objects ), and then a subtraction operation to find the difference. Very expensive in terms of CPU and memory allocatied. I'm just asking for a more efficient direct method to accomplish this task and lighten the burden on the players. Or assign items like fences and trees to a real class and get rid of non-classed items altogether.... Request to explictly kill a FSM. Again true, you can code them to catch a signal. But that adds a lot of overhead to your design if you wish to be able to trap that event everywhere and does nothing for FSMs you did not code up or have gone rogue on you. IMO you need a way to kill the FSM with certainty, something like your idea of killFSM handle.

Very nice indeed. Perhaps time to make a guide on the BI wiki?

the option to get non-classed objects via a nearestObjects call, perhaps by passing objNull or something in place of the array of classes. ability to detect objects in a direction/range/angle vs circle centered around a unit. biggest request - ability to explicitly kill a running FSM.

+1 to Gnat's suggestion.

Ease of modding does not necessarily mean loose security. The reverse argument is definitely true. You need good security built in, good security tools to manage, AND a admin that is takes responsibility for their site and does their best to make it secure. Anytime I see a sloppy site, it's laced with cheaters. It gets added to my do-not-visit list. If the admins are clueless or don't care I take my game elsewhere where the people do care about making ArmA2 a good experience. If more people did the same, then you'd see the bad sites die off and good riddance to them all.

Especially with youtube advertising the cracked versions. Sad they are not more pro-active in policing what is posted. In regard to cheaters - it's not just BIS, the server admins need to clamp down and secure their machines. If they don't use the tools at hand then they'll get pwned and deservedly so. If you don't want to have your house robbed you take measures to secure it - whining about losing your tv if you left the door unlocked does not create a lot of sympathy.

Interesting idea....

Probably seed to the random number generator. Which would effect what and how much things get leveled.

Ok then - the issue of compatibility has been discussed enough. Let's move on.

I'd kbTell something in the other topic.

There is no most efficient way. That is religion. I use tags based on name of speaker ( D1 == Delta 1 ) and a sequence number, but that gives you very little context to what is being said, only the order. My only advice is come up with a scheme that works for you and be consistant about it.

First off, I don't want to trigger any flame wars - going to tread very lightly in this minefield. 1. Both mods are hugely popular and rightly so. 2. Both encourage portability and ease of interface with other mods. 3. Both have good technical support, provide quick turn-around to reported issue, and have an active presence on forums. The issues.... 1. Trying to make any mod compatible with another is a lot of work. Unless the teams are in close synch, one side will always be playing catch-up to the other. 2. ACE and MM take very different technical approaches to how they write mods. That's not a bad thing, but it does make interconnecting the two more difficult. 3. In terms of documentation or interfaces to access inner components, both mods have sections that are opaque. Without enough information/tools, you can't do much to bring the two together. All that being said, there is much that could be done with ACE and MMA to improve how their mods work together. 100% compatible may not be achievable, but more compatible is certainly possible if the right people get together. I'm definitely not one of the right people. Mando decides here.

The FSM is just for when the character is pure AI, sqf for when a player runs that slot. So it may not be appropriate to add multiple choice info ( in the menu ) for an AI that will never see that menu. It would be more logical for the FSM to just pick an action in that situation ( perhaps via random number ).

Based on dialogs between Mando and I have over details of the nukes, I'd say he is a person that strives for realism. However striving is a goal, not an endpoint - at times you run into limits in the game engine and have to compromise. Note well, Mando does not replace models or sounds as a policy. That leads to maximum portability and compatibility. It also means the uber-alert RWR light on upper left part of the Deathmaster 5000 panel will not light up, since the original mod maker did not make it light up either. Since MMA is a toolkit, adding your own code to make the blinkenlighten work is quite possible. Likewise for adjusting any weapons parameters. In one of my missions I wanted a PAC-3 varient of the Patriot missile. So tweeked the ranges, speed, etc of a copy to fit my needs. Easy and fun to do.

Good to see you two have made so much progress. Nice work. Blank phrases are one way to do a monologue. A says something, B replies with a blank phrase ( kind of like nodding head or something so indicate "I hear ya" ), A replies to B's blank phrase with more chatter, etc. Also if you attached a sound file with x seconds of utter silence that should alter the pace, since the BIKB does wait for each sound to finish before moving on to next part.

Believe AZcoder is correct about this. HD - did you also check whether you are using SetIdentity in the init field of JohnRayner or via your init script?

You're welcome. Congrats on getting it working. I use setIdentity and description.ext for my stuff. Seems to work mostly.

The knee is eye-candy, just keeps the flow diagram neater. The grey/green ( blind coder here ) diamond on left is a true conditional, the yellow versions are conditional. The idea is that a true conditional is always true, you use that for situations where you want the code to flow on no matter what. In this FSM, it's used for managing defaults, when none of the other tests match up. Note that both types of conditionals have priority fields in them, high priority is checked first and will break ties. The true conditional was set to the default priority of 0, the other conditionals were set to 1. So it fulfills the roles of "default" in a sqf switch statement. The code in the start state is debugging code to log what is going on. I do similar things myself using diag_log calls. Great way to isolate problems in complicated code. _sentenceId in ["instructions_P_0"] The "in" test is checking to see if the current value of _sentenceID matches one of the elements in the list given in square braces. Great if you have desire to match more than one case. Otherwise you just as well use _sentenceId == "instructions_P_0" which is what I normally do, but the other way is a good trick to know about in case you ever need to test against multiple strings at once. DankTank: I understand the feeling of intimidation by all this, at first it appears overly complex. But like many complex things, once you break it down into smaller tasks the situation becomes more managable. The whole FSM thing can appear that way, but these files are quite useful beyond conversations, as are stringtables. Mastering them opens up a number of doors, especially in higher performance since FSM task scheduling does not suffer some of the restrictions that sqf scripts encounter.