Then I don't know what you're trying to say, because you're far off the mark one way or the other. LOMAC was LOMAC. FC1 and FC2 were released later, each one of them added something that people liked. FC3 will undoubtedly add more features.



Sure, do, because there's a very obvious disconnect between what you're saying and what was done.



There are changelogs. If you really want, I can go looking for them. Did you read them when the releases were made?



Missiles aren't magical. They do lose energy that fast under the right circumstances. As for the number of patches, that's really not up to the customer. At some point the money that was spent on a given game runs out and you either release something new, or abandon the project. ED has always done their best to have a very workable product with patches.



There's plenty more to them than that. I suggest looking at what people have written about BS2 AI for example. Picking one behavior to complain about isn't anything other than picking on a pet peeve. I know for a fact that the AI in FC2 are more capable than those in FC1 or LOMAC. You might not see it, but it's there.



No, I mean the new sound engine developed by ED.



I forget what the point was also in this case.



Duly noted...



No, I don't care if you like it or not. I just found that your opinion wasn't based on facts. The amount of work that went into FC2 was stupendous, and frankly I just don't recall the testing we did for FC1 or BS1 any longer as it was a while ago, but lots of things were done there also.

Hopefully making new maps will be easier with EDGE.

Yes, the AI is better, but still has massive problems that should have been fixed a LOOOOONNNNG time ago (like before LOMAC 1.0 went Gold).

Example: after carefully programming a 200ft AGL TFR route through mountain valleys and passes towards an IP, within the elaborate and very detailed mission editor, the actual flight within the mission should NOT hit afterburners and climb LITERALLY 5,000ft above the next ridgeline, before nosing-over and diving toward the next valley and it's next ridgeline ... then to do it all over again ... like some sort of heavy-metal yo-yo.

Firstly, that's not flying TFR, or even low level flight when that happens. So what is the point of the radar shadow modelling then, if you can't even take tactical advantage of it in practice?

Secondly it wastes copious amounts of fuel, for no reason at all.

Thirdly, it is not even remotely realistic - which just blows my mind, considering every one keeps going on, and on, and ooonn about realistic avionics and weapons behaviours!

The AI developers simply aren't asking themselves this simple question;

"would a pilot actually fly anything like this, in such a situation?"

It blows my mind that after 6 or 7 years, this SIM is still doing these things. I can't believe no one in all that time INSISTED these absurdly obvious major and probably simple to fix flaws be reworked and resolved. There's so much that's right and very impressive about FC2.1, and so much that makes me shake my head in disbelief, that it still doesn't work properly and tactically, at a fundamental level, after all these years, new versions and patches.

I have serious doubts FC3 will fix these very obvious major flaws. I get the impression that people developing are now so habituated to what FC2x.x does so wrongly that they've become blind to these flaws, and simply accept them, or don't even notice how ridiculous they are any more.

All the while everyone is bowing down to the golden-calf of greater realism.

I don't know how ED expects to make a western multirole fighter expansion work when it won't/can't fly or fight in a realistically tactical manner right now, or survive?

How's about we forget all about new aircraft and avionics for a wee bit, and just develop a tactical flight and fight realism upgrade - instead?

If this occurred it would totally revolutionise what you could do with FC, and with mission planning. It would be totally addictive, highly enjoyable, making tactically realistic missions, that actually make sense, and work. You simply can't do this in the current FC iteration.

As far as I'm concerned there are four essential components to any attack mission plan:

1. Fly the planned mission, as planned
2. Strike the target
3. Suffer nearly zero losses when planned right (you know ... like in real life)
4. RTB

FC2.1 is OK at 2 of these four things - #2 and #4
FC2.1 is Dreadful at #1 and #3

The FC2.1 sim and the mission editor should allow you to PLAN, and also FLY such a mission. At present you can indeed plan such a mission ... you just can't FLY it as planned ... or rather, the AI can't execute the mission plan that you very carefully created for it.

This drives me bonkers! I so want to use the mission editor, but the AI simply can't fly the missions I plan for it.

The mission editor can plan anything I want, but the AI can't execute it.

It's easily the biggest fatal-flaw I see in FC2.1

It was also the biggest flaw in LOMAC 1.0.


But why is it like this STILL?

Please post a detailed account of your problems (maybe even a track or 2) in here and I'll see what I can do.

http://forums.eagle.ru/showthread.php?goto=newpost&t=81721

Nate

Snip the part about AI not being perfect. No AI is perfect and there's always room from improvement.



Incorrect. In real life, you plan to have losses because the enemy isn't asleep at the wheel and they're not using sticks and stones against your planes. If they are, bonus for you. This is why multiple strike aircraft are assigned to targets. The harder the target, the more assets.
If you're not suffering losses against a reasonably defended target (nearly zero .. heh) then the defense is poor. Plans do not survive contact with the enemy.



Because AI isn't easy (specifically, it's quite time-consuming). In any case, is TFR style flight your only issue?

Guess this means the AI won't be addressed, maybe in FC 3.5?

AI has been getting addressed with every iteration.

I'll do that Nate, a track or two may be best, with a description of what I'm trying to acheive with that flight.

I said nothing to assert the AI should be, or must be "perfect".

I said there a numerous VERY OBVIOUS FLAWS that should have been fixed years ago, in early development. I also said that what is planned should be what is flown, especially if you have not yet entered direct combat, or suffered any battle damage or failure modes.

If I set the AI pilot proficiency to the highest level within the mission planner, I reasonably and rightly expect that AI pilot, sans mechanical issues, to fly my planning, almost exactly as planned, or die trying to.

You're just minimising what I'm saying, or are a bit blind to it all. In real life you may have a small number of losses if things go badly. So you plan for incremental attrition as acceptable-losses for a given target value and priority.

But you never throw forces, and scarce irreplaceable resources, into a battle in such a haphazard and uncoordinated way that a complete wipe-out is not only likely, but actually occurs (be lucky if you only ended up in prison for that). Especially when loss of the control of the air means you lose everything else soon afterwards.

A complete wipe-out of a UK Squadron would be a sign of extremely poor tactical planning, or coordination, in preparation for the strike, or poor execution due to insufficient training. (and I made that point clear with examples, over a ED's place). If you are not protecting your forces, and planning their best chance for survival each flight, but instead are leaving them open to wipe-outs, you're just a "LOSER".

Especially against well-defended targets.

This why real-world TACTICS in plan generation, and well-trained EXECUTION of those tactics, makes all the difference to survival and accomplishment. This is why nations invest in photo, radar and LIDAR recon units and Sats to develop the required intel, to exploit it tactically, in detailed planning, to avoid any wipe-outs, so that incremental manageable losses, or no losses at all, are more often than not, what results.

That IS what every professional airforce does.

Don't pretend to me that any of these are prepared to accept the moronic 'expert' level flight behaviours and resulting losses we see in Lock On. When the AI can't fly as planned, then tactical planning for achieving a strike target's destruction, and the survival of almost all aircraft, becomes impossible.

For small airforces this is absolutely essential, and much effort is put into achieving tactical surprise and control, even if using functionally inferior equipment. If an enemy can be blindsided, and you're in and out fast, the target is degraded or destroyed, and you've most likely lost no one. This is why maximising use of terrain, jamming, and concurrently destroying EW and AEW radars matters. These things must all occur concurrently, planned to the second, at different geographical locations, via detailed coordinated planning, of several flights, to take all of these down together for sufficient time, and without transmitting sufficient warning, in order to get the target destroyed, and get out -- fast.

A few tens of seconds of exposure and warning and firing opportunities, is all you should allow the enemy to have, in order to kill a target. Give them any more, and they can and will wipe you out.



And my last sentence is why disciplined TFR flight, matters so much. It stops sensors seeing you and it stops weapons getting to you, and makes it much harder for fighters to find and nail you.

And on the contrary plans DO survive contact with the enemy, the plan simply degrades. The longer the contact the faster it degrades, so you keep the contact brief as possible, so your plan can work, without too much degradation.

You are tactically managing the plan's foreseeable degradation, to be out of harms-way before the other guy can get their act together to defend EFFECTIVELY. Which means you fly so you don't get seen until the last moment, then attack fast, from several directions at once, then EGRESS - immediately. Strike aircraft rarely come back for a second-pass. One or two targets, and one or two precision delivered weapons. If you fail to kill the target, you plan another strike, you do not go back and fly around the target area aimlessly, until you're shot down.

Do you have something against making the AI work properly in Lock On?

Because I have flown other sims and used other planning systems that really can do this.

In Lock On, and now in FC, these things do not happen properly, even if you specifically plan it in detail for things to happen on time, to a multi-package strike schedule, because the AI can not EXECUTE their plans, as planned. They will not maintain the headings, waypoint fly over times, alts and speeds alloted. They take alternate routes and altitude profiles and vary speed, and climb thousands of feet over a steep ridgeline, and then end up where they shouldn't be, and can't get back on course and schedule. So it is not possible to execute a tactical plans where one thing occurs, which makes another thing possible to occur, with almost or actually no losses.

That's how a real air battle plan would be developed.

The idea with strike tactics, as with boxing ring bouts, is to hit even an alerted opponent so fast, with such surprise and insufficient reaction times available, that you land a precise blow that knocks that target out. And you thus don't get hit by the other boxer in the process. Most of you tactical moves and flights that day are purely sparring, to set your opponent off-balance so you can make one or two big hits, then EGRESS (with your own jaw intact).

The mission planner in FC is terrific, it looks impressive, it works, except it doesn't work, because the AI won't fly a TFR flight plan in a valley, as planned. It they did a SEAD/DEAD flight could safely approach and kill the EW and SAM radars, then the fighters can move in, to provide high-cover, as the strikers move in on the target (TFR) in one pass, and then exit the area immediately and RTB, to the base planned.

If you do it right, you lose no aircraft, or almost no aircraft.

They should not still be anywhere near their missed target 2 minutes after their allotted time on target. They should not fly around and around a missed target aimlessly, until they're shot down.

I think it is right and valid to be amazed that these VERY OBVIOUS FLAWS have not been fixed yet.

And I'm also amazed that there are simmers out there who don't seem to understand how silly this is. Have you not played a tactically oriented strike simulator like Tornado, or even EF2000? In those, you had to evaluate the threats and intel, and use your brain to select appropriate and workable tactics and options that if planned right, you took out the target with almost zero losses. But if you didn't plan the tactics right, or didn't pay attention to the data, or improperly scheduled or coordinated each flight, and their roles and load-outs, you got wiped out.

I want this to work in Lock On - it would be a staggering combat sim if it did work. Lock On FC2.1 is easily fixable, it's just that no one ever bothered to do it. Everyone has gotten focussed on the fine details of avionics and weapon delivery modes, that are ultimately far less important than being able to execute a tactical mission plan, as planned.

I'm amazed there are people around who don't understand what's going wrong with the AI in Lock On.

It's as though all the AI pilots smoked a giant AI joint before they got into the air, and now can't concentrate or remember what they are supposed to be doing, or where they are, or where they have to go, by when. Somehow the tactical plan's waypoint orders are being shuffled to the bottom of these stoned AI pilots priority list!

They've got better things to do -- apparently.

I don't call that realistic. It's not even close. It's very far from realistic

Firstly, the aircraft mostly won't fly where you tell them to fly, at the TFR speed and level you require, and to the coordinated schedule you order them to in flight planning (some will get sort of near to what they should be doing, but the longer they fly the worse they get), and because they can not fly a disciplined TFR route, they get detected much too early and get wiped out by SAMs and fighters. Thus they are unable to achieve the second bit, which is to expose themselves as briefly as possible to make a coordinated multi-axis attack. So if they survive the ingress, they arrive at the target off schedule and out of coordination with the rest of the strike package, at the wrong place, time and alt, so they again get smashed. Then the third bit come. They have failed to follow to first two parts of the planning, so instead of making a successful rapid egress back into cover and withdrawal, they fly around like idiots trying to kill the target in a totally disorganised way ... until they're all shot down or else damaged and finally they limp away, as the long range SAMs and fighters try to nail them.

No airforce does this! Pilots fly to plan, they stay in cover if making a SEAD or interdiction strike, until they can't remain hidden any more. They hit fast with one pass, and they get out of dodge ASAP.

We are not talking about CAS in a warthog over a battlefield against charging armour here, we're talking about strike interdiction tactics against fixed targets to do real damage to airfields, command centres, Comms and EW sensors and logistics supply centers.

These are tactics that the Su-25 and F/A-18C and especially the TFR strike attack capability of the Tornado should be able to routinely achieve in Lock ON FC2.1, right now.

They should have been roughly doing this in LOMAC v1.0

Instead, Lockon, despite having a detailed radar-shadow model, still does not take advantage of it tactically, almost ever, because the AI's auto-pilot will NOT do what it is told to do. The AI keeps flying the aircraft, rather just than letting the AI fly its mission as planned. The AI simply keeps randomly intervening and deciding to do something other than what the flight plan says. The AI should do nothing unless that aircraft is under a direct attack.

What is the point of having all these fantastic SAM systems and Radar shadow modelling, if you can't the Ai fly a plan that can defeat these SAMs and EW and AEW and fighters?
These are forth-gen aircraft. Fourth generation fighter aircraft are not stealthy, they are the generation of aircraft that most requires the use of terrain masking and or high altitude and speed to defeat defensive systems of the day.

To not even be able to use the core tactical flight attack modes like TFR, means the AI pilots fly between 500 feet and 20k feet routinely, and this is why we get this absurd turkey-shoot and lack of any sort of viable tactical planning for strike packages which results in the current and past iterations of Lock On.

So I rightly ask, why is it STILL like this?

Cool thanks, note if the tracks are done in Warthog or BS2, they will most accurately reflect the most current state of the AI. Also try to keep the tracks as short as possible, so I can easily highlight the issues to the team.

Nate