Matrix Games Forums

boogabooga wrote: Mon Dec 26, 2022 8:04 am One trick to use when your enemy is using SARH missiles is to turn off your auto evasion. The enemy will have to break radar lock to try to dodge your missile, meanwhile, you can maintain your lock for as long as you need to. Of course, this tactic has obvious risks and you might need to jump in and turn the evasion back on if you are getting into trouble. This is a bit of a cheat though, since the AI can't be as clever as you as to when to dodge and when not to.

BDukes wrote: Mon Dec 26, 2022 3:14 pm I use 25% of max automatic firing for fighter vs. fighter combat in general and have had good results in terms of making sure missiles aren't outrun.

HalfLifeExpert wrote: Sat Dec 24, 2022 9:25 pm After trying a number of tactics (all with a Prowler in support), I got best results from using the BVR logic 'Follow Missile Straight In". All 4x Foxhounds splashed in the ensuing battle, but 3 of 4 Tomcats lost.

Is this correct for very long range AAMs like the Phoenix, Amos, and future AIM-260 JATM?

Overall, what's the best way to handle long range BVR engagements against enemy aircraft that have the same range or longer AAMs against you?

SeaQueen wrote: Tue Dec 27, 2022 12:57 pm

BDukes wrote: Mon Dec 26, 2022 3:14 pm I use 25% of max automatic firing for fighter vs. fighter combat in general and have had good results in terms of making sure missiles aren't outrun.

Yeah... but is that your only consideration?

thewood1 wrote: Tue Dec 27, 2022 12:48 pm I agree with that but at the same time, The AI should have the same tools as the player and visa versa.

There's an interesting discussion that happened on Tanknet.org a few weeks ago with a dev talking about AI complaints. The dev is a former tanker and stated that in his command he had people that made mistakes that game players would be screaming about if an AI did them. I have always said that show me even the worst AI in a game and I can find a human that can be worse. People make stupid decisions and moves all the time in games.

SeaQueen wrote: Tue Dec 27, 2022 12:57 pm

BDukes wrote: Mon Dec 26, 2022 3:14 pm I use 25% of max automatic firing for fighter vs. fighter combat in general and have had good results in terms of making sure missiles aren't outrun.

Yeah... but is that your only consideration? Sometimes I don't care if they outrun my missiles. The fact that they sometimes score a kill is sort of a happy byproduct.

If I'm using some low RCS aircraft then maybe I'd prefer to shoot before I'm detected and deny the badguys a shot? I'm willing to accept that I'll miss more often, but they won't hit me either.

If I'm setting doctrine for a striker, I'm happy to have the bad guys dodging my missiles because if they got past my escorts, then denying them a shot is perfectly good. I don't care if I kill them so much as they give me time to drop my bombs, and hopefully my escorts can roll in on them and finish them off.

I think sometimes the obsession with racking up kills is a little bit over exaggerated in the CMO community. I think some of it has to do with how scenarios are often scored (weighted loss exchange ratios) and some of it has to do with the overlap between the CMO community and the flight simulator community. At any rate, in a scenario (and in real life warfare), the mission should come first. The tactics are dictated heavily by the mission. I might go for the 25% max range, if I was willing to accept more losses in exchange for greater enemy attrition. Usually, though, the kinds of scenarios I find interesting demand greater balance between risk and mission success. That means that 25% max range isn't always the right answer. It also depends on the threat. If I'm flying F-22s versus MiG-21s with AA-2s then I am probably willing to get in a lot closer (and have more effective weapons) than I would be if I was flying against Su-35s with AA-12s? This is all stuff to think about. Remember, I want to kill the badguys, but I want to go home more.

My point is, there's more variables to consider than just maximizing the Phit for your missiles. A good scenario ought to reflect that. A good player's tactics adjust with the mission and the threat.

HalfLifeExpert wrote: Sat Dec 24, 2022 9:25 pm (Running Build 1300.1 FYI)

So I've returned to the classic scenario 'North Pacific Shootout'.

In the opening clashes, its Tomcats vs Foxhounds.

Even though the Tomcats outrange the Foxhounds by about 10nm (Phoenix Missile vs Amos Missile), In most of my attempts, the Tomcats got massacred, with maybe one Foxhound lost.

After trying a number of tactics (all with a Prowler in support), I got best results from using the BVR logic 'Follow Missile Straight In". All 4x Foxhounds splashed in the ensuing battle, but 3 of 4 Tomcats lost.

Is this correct for very long range AAMs like the Phoenix, Amos, and future AIM-260 JATM?

To effectively charge in to get close, only maneuvering at the last minute to evade the very long range AAMs?

I just wanted some feedback on this, especially with the arrival of boost-coast missile simulation with the War Planner/Tiny update.

Overall, what's the best way to handle long range BVR engagements against enemy aircraft that have the same range or longer AAMs against you?

BDukes wrote: Wed Dec 28, 2022 2:27 pm Ok. So in terms of making ideas into actionable things. How do you calculate the spacing/timing between escort and striker to allow that? It's variable depending on weapons parameters, so the goal is to develop some easy case rules of thumb. I don't expect you to have them on hand, but if you do, could you do us dummies a solid?

Yeah maybe. Kills are the easiest thing to quantify. The other stuff is much more complicated for hobbyists and its probably good enough. I think I probably set my mind's victory expectations beyond what the game tells me and guessing most do. So not really sure I care enough about this, never mind obsess.

My goal is to build something good enough that I don't think about it the night after.

SeaQueen wrote: Fri Dec 30, 2022 2:41 pm

BDukes wrote: Wed Dec 28, 2022 2:27 pm Ok. So in terms of making ideas into actionable things. How do you calculate the spacing/timing between escort and striker to allow that? It's variable depending on weapons parameters, so the goal is to develop some easy case rules of thumb. I don't expect you to have them on hand, but if you do, could you do us dummies a solid?

Consider the threat. If the thing your escorts are trying to protect is facing a threat of a 25NM missile (launch range), then you probably want to position your escorts so that they have a chance to destroy the launching aircraft AT LEAST 25NM out. The further forward from 25NM the more time and distance they have to kill it.

In terms of time. If you figure 480 KTS, the default speed of a fighter in CMO, is 8 NM/min. Then to have them at least 25 NM out translates into about 4 MIN spacing. If the threat has a longer missile, you need to go further out, if they have heaters/guns only, you can bring them in closer. The only problem with pushing escorts further out, is that eventually they get spread pretty thin, and then you either need more escorts or you run the risk of them sneaking in the sides. That's how I think about it. It's just time, distance and geometry.

N.B. In the CMO database an AA-12 ADDER-B is about a 60NM missile, while a PL-15 is 108NM. Meanwhile an AIM-7F is a 38NM missile and an AIM-120D is an 87NM missile. As you can see, given the variance, this demands considerable play in the placement of escorts relative to what they're defending.

Yeah maybe. Kills are the easiest thing to quantify. The other stuff is much more complicated for hobbyists and its probably good enough. I think I probably set my mind's victory expectations beyond what the game tells me and guessing most do. So not really sure I care enough about this, never mind obsess.

Not really. If you look at the victory conditions in table top, pen and paper wargames (e.g. Harpoon) versus what I see in many CMO scenarios, it looks like pretty weak sauce. It's much more complicated for people who don't know LUA.

My goal is to build something good enough that I don't think about it the night after.

I think that's a worthy goal. Typically, once I've done my initial thinking and planning, I save the game and then in subsequent attempts at a scenario I make adjustments within the saved game instead of from the loaded scenario file. That way I don't have to start from scratch each time. It accelerates the whole process.

SeaQueen wrote: Tue Dec 27, 2022 12:39 pm

boogabooga wrote: Mon Dec 26, 2022 8:04 am One trick to use when your enemy is using SARH missiles is to turn off your auto evasion. The enemy will have to break radar lock to try to dodge your missile, meanwhile, you can maintain your lock for as long as you need to. Of course, this tactic has obvious risks and you might need to jump in and turn the evasion back on if you are getting into trouble. This is a bit of a cheat though, since the AI can't be as clever as you as to when to dodge and when not to.

Not relying exclusively on the AI is NOT a cheat. It's being smart. It's being a good player. The AI in CMO is crude. It doesn't understand tactics as well as you do. IT NEVER WILL. It is intended to be the "good enough" solution for when your focus, as a player, is directed towards where it is most important for you to direct your attention. The game was not designed or intended to be run completely hands off. If it was, it'd would look nothing like what it looks like. EMBRACE IT! There's a really weird streek in the CMO community that is almost apologetic for the AI not being capable of making perfect decisions in all situations. Stop stressing! No pilot is perfect. No mission is executed perfectly. That's why there's a debrief. No AI is capable of perfectly understanding all possible situations and reacting in some mythical "optimal" way. Not all pilots of identical skill levels will react the same, anyhow. Tactics, doctrine, and training vary by nation and service branch. There isn't necessarily one solution to any given problem. It is not a bad thing to inject yourself where it's important, if you think you can do better than the AI (for varying definitions of "better"). That's what wargaming is all about. It's about injecting the human decision making into the simulation.

Relax!

boogabooga wrote: Fri Dec 30, 2022 11:06 pm I think that if I were wrong, there would not be such an impetus for real time multiplayer?

Tcao wrote: Thu Dec 29, 2022 10:38 pm
Can you please share more information on what caused tomcat got the massacred? Like the target outrun the missiles? Your fighters running out of missile first? Or the tomcats only got the chance to release one salvo of AIM-54 then after that it keeps at defensive maneuver (suppressed?).

it's funny that in the QB air to air 1985, Tomcat won most of the time because Foxhounds don't take crank (this could be a bug)

boogabooga wrote: Sat Dec 31, 2022 1:24 am Likewise, could you please share your QB settings and version? The QB version that gets shipped out in the patches is way old. A newer version is available on the relevant thread.

BDukes wrote: Fri Dec 30, 2022 3:45 pm Yeah. Hobbyists do stuff for free, though, so it is not a bar you can fairly set. Lua is a technological hurdle, and it's a big
ask to make a hobbyist go learn a programming language because SQ wants a deeper victory condition model.

My experience with Lua people is there are programmers (and their culture) who do and share to kind of move everybody forward, and then they're are people who just take or view it as currency to get a job, etc. etc. In the latter case, why would you do anything for them, if there is no return? The pro thing has added a bit of height to the hurdle. This is not a dig at pros or any person, just something to think about when thinking about why hobbyists do or don't do something. This is just my take. It could be wrong!

SeaQueen wrote: Sat Dec 31, 2022 6:25 pm

BDukes wrote: Fri Dec 30, 2022 3:45 pm Yeah. Hobbyists do stuff for free, though, so it is not a bar you can fairly set. Lua is a technological hurdle, and it's a big
ask to make a hobbyist go learn a programming language because SQ wants a deeper victory condition model.

The problem as I see it is that even the pros don't want to learn to program computers either.

My experience with Lua people is there are programmers (and their culture) who do and share to kind of move everybody forward, and then they're are people who just take or view it as currency to get a job, etc. etc. In the latter case, why would you do anything for them, if there is no return? The pro thing has added a bit of height to the hurdle. This is not a dig at pros or any person, just something to think about when thinking about why hobbyists do or don't do something. This is just my take. It could be wrong!

The problem with that thinking is that LUA isn't used extensively in the professional wargaming community. In fact, they're sometimes hostile to it (hence the, IMO, over exaggerated demand for multiplayer). The professional wargaming world is dominated by people who studied things like history, international affairs, international security studies, and the like. Those aren't the kinds of people who are typically comfortable picking up computer programming languages and doing interesting and creative things with them. They are not typically thrilled about being forced to understand the radar equation to play the game at an advanced level. They're more like <vocal fry>math is hard</vocal fry>. That's why they hire people like me; to do their homework for them. I'm the nerd who did their homework.

There's also a school of professional wargaming that is anti-data. They aggressively assert that the thing that really matters is not the technological aspects of warfare but the human decision making. At the strategic and to some extent the operational level they are often correct in the sense that technology doesn't not drive decision making to the extent that economics, politics, culture, the personalities of leaders, and other human factors drive decision making. These types prefer to outsource the adjudication of combat to a committee of experts in the white cell who somehow arrive at results through mysterious means. That thinking breaks down hard at the tactical level (where Command most comfortably resides) where decision making is often driven explicitly by technological considerations, physics, geometry and math. Those people don't like Command, and resist its use (and misuse, sometimes). Those sorts also go further in that they resist even collecting data on their wargames. They don't like people spending too long trying to analyze results in detail, and usually underbudget for quantitative analysis and any sort of deep dive there. They're cautious about the kinds of iterative question asking where people present results, people ask questions about it, and then you go back, do more analysis on the same data, and answer those questions. The notes taken by note takers are enough.

I think for those on the outside, professional Command may seem to be something it's not. I don't think that professional computer adjudicated wargaming has really reached its full potential yet because the generations are changing. The way I see it, is that CPE is part of the vanguard of a big cultural change within the professional wargaming community, which has been dominated for decades by old guys who want to talk about The Blitz all day and run pen and paper wargames on hexmaps. Younger people are less technophobic, more comfortable with video games, and the possibilities that computers allow for.

So don't worry about putting code out there. You don't really know what kind of code is actually needed, and the fact that you wrote something that happened to be useful shouldn't be thought of as a lost opportunity to profit. Truthfully, professional Command wargames are often cobbled together off of limited data sets, by people who would really prefer to be working on a different type of wargame entirely, and are suspicious of writing software at all. They'd really prefer a computer wasn't involved. That's changing, but only very slowly. If you really want to go the route of professional wargaming using computers, write TONS of LUA code. Get REALLY smart about it. Experiment with it freely and show people what your experiments tell you. That's building your resume, so that if you pursue a job as a professional wargamer, you have stuff that you can point to that demonstrates how you think about the problems it entails, and that helps people assess your potential professionally.

boogabooga wrote: Sat Dec 31, 2022 1:24 am

Tcao wrote: Thu Dec 29, 2022 10:38 pm
Can you please share more information on what caused tomcat got the massacred? Like the target outrun the missiles? Your fighters running out of missile first? Or the tomcats only got the chance to release one salvo of AIM-54 then after that it keeps at defensive maneuver (suppressed?).

it's funny that in the QB air to air 1985, Tomcat won most of the time because Foxhounds don't take crank (this could be a bug)

Likewise, could you please share your QB settings and version? The QB version that gets shipped out in the patches is way old. A newer version is available on the relevant thread.

I think a problem with these pre-Tiny scenarios might be that the Opfor is still set up to fire out to max range. That's going to play out 'weird' in a boost-coast engagement, especially if you modify your own WRA to some percent of max range. The asymmetry means that we would really need to see the details of how the engagement is playing out to say much more.