[WAD] Subs not recharging at Periscope/Surface Depth

[Primarchx]

I thought earlier discussion was that folks wanted to stay on electric engines for a little while after they surfaced before switching to diesels? This lets a player pop up to periscope depth for a quiet looksie and then dip back down without turning on the diesel trip hammers or, if they stay there for a few minutes, the assumption would be they intend to go to diesels to charge.

[guanotwozero]

I reckon part of the problem is that we'd like to use our subs differently depending if we're close to enemies or not. I don't mean only detected enemies, but ones we suspect might be in the area and that we're trying to find. Batteries near to enemies, diesels/recharge elsewhere. If the AI could determine that and act accordingly, then we wouldn't need any manual override.

I've suggested a possible approach to this here.

[Shemar]

ORIGINAL: guanotwozero

I reckon part of the problem is that we'd like to use our subs differently depending if we're close to enemies or not. I don't mean only detected enemies, but ones we suspect might be in the area and that we're trying to find. Batteries near to enemies, diesels/recharge elsewhere. If the AI could determine that and act accordingly, then we wouldn't need any manual override.

I've suggested a possible approach to this here.

The problem with AI override is that it will never comply to every player's way of doing things because there is only one AI algorithm and about as many opinions about how the AI should behave as players [:'(]

The problem with this specific game is that while I suspect it would be a lot easier for the devs to give player control rather than program a good AI, working on a good player side AI also improves the computer opponent AI and that is why, I suspect, the devs prefer to work on and inprove their AI rather than just give us overrides. If we had override on everything they would hardly ever get any feedback or testing on how good the AI truely is.

[mikmykWS]

Part of the issue is our beta testers helped make a decision that impacted this early on. Now we have a second group that wants something else.

Definitely on our list to discuss this weekend. We know its important to you guys.

Mike

[guanotwozero]

ORIGINAL: Shemar
The problem with AI override is that it will never comply to every player's way of doing things because there is only one AI algorithm and about as many opinions about how the AI should behave as players [:'(]

The problem with this specific game is that while I suspect it would be a lot easier for the devs to give player control rather than program a good AI, working on a good player side AI also improves the computer opponent AI and that is why, I suspect, the devs prefer to work on and inprove their AI rather than just give us overrides. If we had override on everything they would hardly ever get any feedback or testing on how good the AI truely is.

Well, I don't think that the AI should necessarily be limited to one course of action.

It's perfectly fine to at least discourage micromanagement, so as to keep a level playing field between player and enemy AI. That's part of the spirit of the game. However, there could be a range of actions that can be configured by the player, and which enact automatically once configured. To keep that level playing field, such actions must also be available to the enemy AI, as well as an acceptable judgement/decision logic that can decide which to perform in what circumstances. I reckon that last part is by far the trickiest to implement.

For example, here, one action 'profile' would be to run on batteries until a charge threshold, then recharge on diesels. Another profile would be to run on diesels until close to enemies, then use batteries. Call those profiles A and B. These two are mutually exclusive, so a player configuration would be to choose A OR B. Once set, that behaviour is automatic, so no micromanagement required.

However, how would the enemy AI decide between A OR B? I think to figure that out, we'd have to get to the bottom of why would humans choose it - what factors would be taken into account.

Further to that, there could be additional configuration options, such as 'spy' or 'hunt'. The former would avoid combat and maximise observation, the latter engages where possible. Call them X & Y. These are mutually exclusive, but are both compatible with A & B.

Thus our config option now becomes ((A OR B) AND (X OR Y)), making it even harder for the AI. And that's just 2 groups of 2 options - we'd likely want more, though we could probably narrow it down to a most useful subset.

Maybe we need to hammer out just what behaviours we DO want, and be precise about WHEN we want them, so as to work out the decisions required if the AI is to mirror them. If any part comes down to pure player preference instead of decision-based logic, then that part could be a random factor for the AI. The human touch! [:'(]

[Shemar]

Well, I don't think that the AI should necessarily be limited to one course of action.

Of course it is. Any computer program is limited to a single responce under a specific set of circumstances*

It's perfectly fine to at least discourage micromanagement, so as to keep a level playing field between player and enemy AI. That's part of the spirit of the game.

Each person looks for different things. Personally I find the above statement in direct conflict with what I am looking for in this game. To me it reads "since the opponent AI is dumb, let's force the player to endure the same dumb AI on his own units". It is much better to achieve play balance by giving the computer opponent numerical or sitiational advantages (like time limits on the player), than to force him endure watching his units blunder into destruction.

To keep that level playing field, such actions must also be available to the enemy AI, as well as an acceptable judgement/decision logic that can decide which to perform in what circumstances. I reckon that last part is by far the trickiest to implement.

I would say impossible to implement. Computer AI only survives the scrutiny of human logic when hidden behind fog of war. Once fully exposed to the player's view no computer AI will ever be able to behave in a way that is not frustratig to the player, at least as long as we don't have a whole different computer technology. CMANO, like Harpoon before it, avoids the worst part of computer AI deficiency by breaking it down to relatively simple missions and putting the burden of giving some semblance of realistic reactions on the scenario designer.

For example, here, one action 'profile' would be to run on batteries until a charge threshold, then recharge on diesels. Another profile would be to run on diesels until close to enemies, then use batteries. Call those profiles A and B. These two are mutually exclusive, so a player configuration would be to choose A OR B. Once set, that behaviour is automatic, so no micromanagement required.

As an example of how quickly this breaks down is tha fact that the definition of "close to enemies" can be interpreted a hundrded different ways. Does the enemy have maritime patrol aircraft? Helos? Then immediately the definition of "close" as a matter of distance changes. Add to that the fact that while the computer has a strict model of uncertainly as to the position of enemy units, the player has a much wider and very subjective way of determining where he thinks enemy units are. So it is effectively impossible to program the AI to "run on diesels until close to enemies" because the AI has no way of determining what the player considers "close to the enemy" and it is therefore doomed to not act the way the player wishes.

However, how would the enemy AI decide between A OR B? I think to figure that out, we'd have to get to the bottom of why would humans choose it - what factors would be taken into account.

Further to that, there could be additional configuration options, such as 'spy' or 'hunt'. The former would avoid combat and maximise observation, the latter engages where possible. Call them X & Y. These are mutually exclusive, but are both compatible with A & B.

Thus our config option now becomes ((A OR B) AND (X OR Y)), making it even harder for the AI. And that's just 2 groups of 2 options - we'd likely want more, though we could probably narrow it down to a most useful subset.

All that is a huge amount of programmig work and even bigger amout of troubleshooting and beta testing. And it can all be simply avoided by a checkbox in the speed and altitude screen. Scenario designers can be given the same control over AI units so the can truely adapt the computer AI tot he scenario circumstances in a way generic AI never will.

Maybe we need to hammer out just what behaviours we DO want, and be precise about WHEN we want them, so as to work out the decisions required if the AI is to mirror them.

That is impossible to 'hammer out' as the different circumstances would take pages upon pages and include things like "how much time do I have left", "how close am I to winning the scenario", "what are the scenario objectives" as well as pure player preference in tactics.

If any part comes down to pure player preference instead of decision-based logic, then that part could be a random factor for the AI.

In what world would the problem of "The AI is acting dumb" be solved by "instead of the AI acting dumb in a consistent way let's make it act dumb in a random way".
_____
* Unless you make the AI randomply pick from a set of possible responses, but that is a horrible idea for player side AI (great for computer AI though).

[guanotwozero]

ORIGINAL: Shemar

Well, I don't think that the AI should necessarily be limited to one course of action.

Of course it is. Any computer program is limited to a single responce under a specific set of circumstances*

Yeah, but that means different courses of action, each of which is determined by circumstances. The circumstances can be reduced to a bit pattern representing game state, and any decision to pursue a course of action is a function of that bit pattern. Combinations/sequences of actions included.

It's perfectly fine to at least discourage micromanagement, so as to keep a level playing field between player and enemy AI. That's part of the spirit of the game.

Each person looks for different things. Personally I find the above statement in direct conflict with what I am looking for in this game. To me it reads "since the opponent AI is dumb, let's force the player to endure the same dumb AI on his own units". It is much better to achieve play balance by giving the computer opponent numerical or sitiational advantages (like time limits on the player), than to force him endure watching his units blunder into destruction.

As it's unrealistic to assume the devs will produce a HAL-like AI, there has to be an achievable approach. If you want to micromanage to the extent where you maintain an advantage over enemy AI, sure, that's your call. I'd prefer the playing field to be as level as possible, while still allowing as much scope as possible for player 'command level' decisions. I don't like the idea of AI 'cheating' as so many games do, thus the decision trees have to be comparable to what a human can do. My perception is that CMANO aspires to minimise micromanagement by automating such behaviour as much as possible. The challenge is to make that automation as realistic as possible.

Yep, these are our subjective opinions on what the game should be, but player feedback's good, right? [:)]

To keep that level playing field, such actions must also be available to the enemy AI, as well as an acceptable judgement/decision logic that can decide which to perform in what circumstances. I reckon that last part is by far the trickiest to implement.

I would say impossible to implement. Computer AI only survives the scrutiny of human logic when hidden behind fog of war. Once fully exposed to the player's view no computer AI will ever be able to behave in a way that is not frustratig to the player, at least as long as we don't have a whole different computer technology. CMANO, like Harpoon before it, avoids the worst part of computer AI deficiency by breaking it down to relatively simple missions and putting the burden of giving some semblance of realistic reactions on the scenario designer.

I agree, but still a useful level of decision processing could be achieved if enough factors are taken into account. I reckon that's part of the challenge of finessing the AI - adding more factors to the decision process so as to produce a more sophisticated result. It will never achieve human-level performance, but clever design means it shouldn't have to, to result in a challenging game. Good scenario design will be important.

For example, here, one action 'profile' would be to run on batteries until a charge threshold...

As an example of how quickly this breaks down is tha fact that the definition of "close to enemies" can be interpreted a hundrded different ways. Does the enemy have maritime patrol aircraft? Helos? Then immediately the definition of "close" as a matter of distance changes. Add to that the fact that while the computer has a strict model of uncertainly as to the position of enemy units, the player has a much wider and very subjective way of determining where he thinks enemy units are. So it is effectively impossible to program the AI to "run on diesels until close to enemies" because the AI has no way of determining what the player considers "close to the enemy" and it is therefore doomed to not act the way the player wishes.

Sure, I only gave those examples as they've already been discussed, without going into any detail of how they'd be implemented. Any 'profile' would consist of many behaviour components, each of which would be determined by game state, preference settings and decision logic. That's what I mean by hammering out what we want our units to do when acting automatically, so as to decide what would those components be, and what would determine their enactment.

EDIT ===

BTW I did suggest a means of judging being "close to the enemy" here.

However, how would the enemy AI decide between A OR B?...

All that is a huge amount of programmig work and even bigger amout of troubleshooting and beta testing. And it can all be simply avoided by a checkbox in the speed and altitude screen. Scenario designers can be given the same control over AI units so the can truely adapt the computer AI tot he scenario circumstances in a way generic AI never will.

I agree it could be a lot of work - that's why narrowing it down to a useful subset is important. Then the relevant player preferences could be implemented, including via checkboxes. If the AI is to compete realistically, then its decision trees should be able to choose the same preferences by some means, including by scenario design.

Maybe we need to hammer out just what behaviours we DO want, and be precise about WHEN we want them, so as to work out the decisions required if the AI is to mirror them.

That is impossible to 'hammer out' as the different circumstances would take pages upon pages and include things like "how much time do I have left", "how close am I to winning the scenario", "what are the scenario objectives" as well as pure player preference in tactics.

I still think we could reduce it to a useful subset, though that would mean rejecting factors we judge to be least important. And yes, that is subjective, but that's where the idea of player feedback would matter. Ultimately, though, it's the devs' call, and their guidance would be crucial.

Any improvement over the current 'single course of action regardless' would be a plus.

If any part comes down to pure player preference instead of decision-based logic, then that part could be a random factor for the AI.

In what world would the problem of "The AI is acting dumb" be solved by "instead of the AI acting dumb in a consistent way let's make it act dumb in a random way".
_____
* Unless you make the AI randomply pick from a set of possible responses, but that is a horrible idea for player side AI (great for computer AI though).

That's not what I mean. If certain player decisions cannot be reduced to determinism/rationalism, then they're effectively random, even if weighted. "I'll send this patrol over the sea, because I like the colour blue." If the computer does the same, then it's no different. Dumb? Maybe, but that's humans for ya. In any case, our world is regarded as non-deterministic, if the scientific consensus on quantum mechanics is to be accepted. Probabilistic decisions could be useful and realistic.


FWIW I think this particular sub issue would make a good test case for trying to achieve a better automated result. If that could be achieved here, then so too for any other area where the automation could be improved. But until that happens, manual override might be the best short-term fix.

[Shemar]

ORIGINAL: guanotwozero

Yeah, but that means different courses of action, each of which is determined by circumstances. The circumstances can be reduced to a bit pattern representing game state, and any decision to pursue a course of action is a function of that bit pattern. Combinations/sequences of actions included.

My point is that in any specific situation the AI can only have one response whereas each player will have their own response. It is therefore impossible for the AI to comply with all player wishes as those are often conflicting. There cannot be a consensus on how the AI should handle this (or any other issue) because that would imply we all play exactly the same way and use exactly the same tactics. In every possible situation where I would want my sub to snorkel and you would want it to stay on battery (for example), the AI will have 50% failure in doing what we want it to do, no matter how good it is.

Maybe I want to use my sub as bait so that a lurking fighter blasts out of the sky the enemy helo that goes after it or as a distraction so another sub approaching from the opposite direction can get through unnoticed. How can the AI possibly account for that, when every reasonable ROE would tell it to stay on battery?

[guanotwozero]

ORIGINAL: Shemar
My point is that in any specific situation the AI can only have one response whereas each player will have their own response. It is therefore impossible for the AI to comply with all player wishes as those are often conflicting. There cannot be a consensus on how the AI should handle this (or any other issue) because that would imply we all play exactly the same way and use exactly the same tactics. In every possible situation where I would want my sub to snorkel and you would want it to stay on battery (for example), the AI will have 50% failure in doing what we want it to do, no matter how good it is.

Maybe I want to use my sub as bait so that a lurking fighter blasts out of the sky the enemy helo that goes after it or as a distraction so another sub approaching from the opposite direction can get through unnoticed. How can the AI possibly account for that, when every reasonable ROE would tell it to stay on battery?

I'm not sure what you mean: do you mean the AI determining behaviour of player units?

If so, those would be set by preferences anyway i.e. I could set my subs to run on battery as part of a profile, you set yours to snorkel. Such behaviour would be automated, so doesn't really involve AI. Even getting our units to respond to gamestate is deterministic, such as the existing hierarchy of tasks. Strike can be temporarily overridden by Refuel, which can be overridden by Engage Defensive. If we want it to be different, we change the preferences. I could set my sub to dive deep if threatened, you to attack. There's no requirement for the AI to 'succeed' just for our preferences to be enacted.


Or do you mean the enemy AI? How it would choose to play its units, given the same set of circumstances? That's not the same as having to "comply with all player wishes", as it's doing its own thing. How it chooses its own thing is certainly important, and maybe needs more discussion. How it would choose to snorkel, run battery or whetever.

In any case, enemy AI choices, including responses to player behaviour, need not be deterministic. Choices could be weighted, depending on gamestate, and a response chosen by probability based on that weighting. Thus we can't say it's limited to one response. Potentially that could be extended to including AI 'generals', each of whom would have a different bias in how they weight things. Reckless warlords or cautious commanders - pseudo-personalities in terms of gameplay. Even without such generals, an unpredictable nature of enemy response would still be a plus.

The AI could (and maybe does) include a more extensive task hierarchy, such as 'metagoals' subdivided into smaller tasks. These could include tactics like baiting and ambushing. As the tasks succeed or fail, the goal hierarchy will evolve. Again, the decision mechanism to manage these goals could include unique game-instance bias and probability.

But sure, we can't expect the enemy AI to ever be as 'wise' as humans. Your trap will likely work. The best we can hope for is that when you down the first, the AI will alter its goals so as to recognise that threat and not be suckered in again.

[Shemar]

ORIGINAL: guanotwozero

I'm not sure what you mean: do you mean the AI determining behaviour of player units?

Player units

If so, those would be set by preferences anyway i.e. I could set my subs to run on battery as part of a profile, you set yours to snorkel. Such behaviour would be automated, so doesn't really involve AI. Even getting our units to respond to gamestate is deterministic, such as the existing hierarchy of tasks. Strike can be temporarily overridden by Refuel, which can be overridden by Engage Defensive. If we want it to be different, we change the preferences. I could set my sub to dive deep if threatened, you to attack. There's no requirement for the AI to 'succeed' just for our preferences to be enacted.

Maybe if you saw my play style you would understand how out of touch with how I play this solution is. I do not use any missions (except ASW patrols) I micromanage everything, take offs and landings, weapon targetting and releases, routes, speeds and altitudes, I conduct aerial combat by keeping tight control of every single aircraft and spend much more time paused giving orders than with the clock running. So while I don't mind Ai assistance as long as I can turn it off (all AI options are off by default when I play) any AI behavior that overrides my own commands or is out of my control is by default in conflict with how I like to play the game.

More to the point, the reason I play this way is that the number of factors that go into every single decision, including somethng as 'simple' as snorkel or battery, is not only orders of magnitude more than what any AI could take into account but often includes information, desires and motivation the AI cannot possibly have. It is therefore utterly impossible to make AI that I would not want to override on occasion and probably quite often, even if you were programming the AI specifically to cater to my own desires and play style.

As for enemy AI, I do not consider that to be my problem. I know for a fact enemy AI is usually a blundering fool, but fortunately limited intelligence and fog of war hide most of that. Computer game AI is closer to running around like a headless chicken than human controlled behavior, once fully exposed. For example in just the last scenario I played, 7 F-16s and Mirage 1s were killed without offering any resistance, just loitering over their airfield while a strike force of 9 Phantom IIs and Mirage F1s also died at the hands of 4 Harriers without firing a single missile in defense. So my opinion is clearly that priority should be given to make the AI at least effectively defend itself, rather than try to second guess whether I want my sub to snorkel or not.

[AndrewJ]

That's very similar to my play style too. Lots and lots of micro-management! In part because I enjoy the feeling of being in command of the sub, flying the aircraft, sitting at the ESM console, etc., and also because the human-assistant AI simply can't yet handle tactical decisions in an operational context. I don't mind handing off housekeeping details (and I'd like to hand off more), but for the rest I'm the tin-pot dictator, with more medals than a row of North Korean generals! If I give a command I expect my computerized minions to do it.

[guanotwozero]

ORIGINAL: Shemar
Maybe if you saw my play style you would understand how out of touch with how I play this solution is. I do not use any missions (except ASW patrols) I micromanage everything, take offs and landings, weapon targetting and releases, routes, speeds and altitudes, I conduct aerial combat by keeping tight control of every single aircraft and spend much more time paused giving orders than with the clock running. So while I don't mind Ai assistance as long as I can turn it off (all AI options are off by default when I play) any AI behavior that overrides my own commands or is out of my control is by default in conflict with how I like to play the game.

More to the point, the reason I play this way is that the number of factors that go into every single decision, including somethng as 'simple' as snorkel or battery, is not only orders of magnitude more than what any AI could take into account but often includes information, desires and motivation the AI cannot possibly have. It is therefore utterly impossible to make AI that I would not want to override on occasion and probably quite often, even if you were programming the AI specifically to cater to my own desires and play style.

As for enemy AI, I do not consider that to be my problem. I know for a fact enemy AI is usually a blundering fool, but fortunately limited intelligence and fog of war hide most of that. Computer game AI is closer to running around like a headless chicken than human controlled behavior, once fully exposed. For example in just the last scenario I played, 7 F-16s and Mirage 1s were killed without offering any resistance, just loitering over their airfield while a strike force of 9 Phantom IIs and Mirage F1s also died at the hands of 4 Harriers without firing a single missile in defense. So my opinion is clearly that priority should be given to make the AI at least effectively defend itself, rather than try to second guess whether I want my sub to snorkel or not.

Indeed, each player chooses their own playstyle, and you have yours. However, if you reduce the game to full micromanagement, you can't expect the enemy AI to ever be able to compete with you on an equal footing. As you do recognise that, it means CMANO will always be an assymetric game for you no matter how the AI is improved, unless and until micromanagement-level multiplayer is implemented. in that context, adding more direct control to subs would be more of the same, and I have no disagreement with that.

However CMANO does provide missions, postures and configurations to reduce micromanagement and shift emphasis from hot-seat to planning, and my own playstyle is to make use of that where feasible. Where I avoid it, it's because it's not yet up to the job, but I will do so when that changes. In that context my argument is to enhance that approach, and I see this sub issue as a good example where that could be done. FWIW I also agree to improving AI defence ability, as that will benefit friendly as well as enemy performance if a hands-off approach is used. Symmetry maintained.

As for priorities, well, that's really the team's call - all we can do is to report problems and make suggestions. I see no reason why bug fixes, finessing and major enhancements can't all be worked on in the same time frame, dev man-hours permitting. I don't doubt CMANO is here for the long haul, so all these things may come to pass.

I tend to see "Command" in the title as referring to "Ops Room" commander rather than "cockpit" commander, but that's just me. [8D]

[Shemar]

ORIGINAL: AndrewJ

That's very similar to my play style too. Lots and lots of micro-management! In part because I enjoy the feeling of being in command of the sub, flying the aircraft, sitting at the ESM console, etc., and also because the human-assistant AI simply can't yet handle tactical decisions in an operational context. I don't mind handing off housekeeping details (and I'd like to hand off more), but for the rest I'm the tin-pot dictator, with more medals than a row of North Korean generals! If I give a command I expect my computerized minions to do it.

If I could only get my minions to stop turning their radars AND SONARS on every time an itchy bitty missile is fired! Haha!

ORIGINAL: guanotwozero

However, if you reduce the game to full micromanagement, you can't expect the enemy AI to ever be able to compete with you on an equal footing.

The enemy AI will never be able to compete with me on an equal footing even if the CMANO developers hire 3 times their number and spend the next 10 years doing nothing but AI development. It just can't be done. So no I don't expect that, I expect the scenarios to put me in a distinct numerical or tactical disadvantage to offset the AI deficiencies or the scoring to be such that I need what amounts to overwhelmingly one sided kills to actually win.

However CMANO does provide missions, postures and configurations to reduce micromanagement and shift emphasis from hot-seat to planning, and my own playstyle is to make use of that where feasible. Where I avoid it, it's because it's not yet up to the job, but I will do so when that changes. In that context my argument is to enhance that approach, and I see this sub issue as a good example where that could be done.

I don't argue with your needs, just that your solution does not address mine [8D]

[guanotwozero]

ORIGINAL: Shemar
The enemy AI will never be able to compete with me on an equal footing even if the CMANO developers hire 3 times their number and spend the next 10 years doing nothing but AI development. It just can't be done. So no I don't expect that, I expect the scenarios to put me in a distinct numerical or tactical disadvantage to offset the AI deficiencies or the scoring to be such that I need what amounts to overwhelmingly one sided kills to actually win.

Maybe IBM will donate their latest chess grand-master beating machine and AI team for the cause? [:)]

I don't argue with your needs, just that your solution does not address mine [8D]

Fair enough! It does say a lot about the depth of this game where it attracts people with very different playstyles. [8D]

[mikmykWS]

ORIGINAL: Shemar

As for enemy AI, I do not consider that to be my problem. I know for a fact enemy AI is usually a blundering fool, but fortunately limited intelligence and fog of war hide most of that. Computer game AI is closer to running around like a headless chicken than human controlled behavior, once fully exposed. For example in just the last scenario I played, 7 F-16s and Mirage 1s were killed without offering any resistance, just loitering over their airfield while a strike force of 9 Phantom IIs and Mirage F1s also died at the hands of 4 Harriers without firing a single missile in defense. So my opinion is clearly that priority should be given to make the AI at least effectively defend itself, rather than try to second guess whether I want my sub to snorkel or not.

It does sound bad but I think having defensive missions set up might have helped. There is kind of two things going on. You likely have some definite weaknesses and exploits with the AI that the developers must fix, no denying that. You also have some definite weaknesses in scenario design primarily because the game is new and people are still working on what works and what doesn't. Both of these things will be remedied with time and player and developer improvements. Its what shakes out in the end that counts. It is just a matter if your willing to wait it out or help. I do not think banging on the developers or scenario writer will move that forward.

Thanks!

Mike

[mikmykWS]

ORIGINAL: guanotwozero

ORIGINAL: Shemar
The enemy AI will never be able to compete with me on an equal footing even if the CMANO developers hire 3 times their number and spend the next 10 years doing nothing but AI development. It just can't be done. So no I don't expect that, I expect the scenarios to put me in a distinct numerical or tactical disadvantage to offset the AI deficiencies or the scoring to be such that I need what amounts to overwhelmingly one sided kills to actually win.

Maybe IBM will donate their latest chess grand-master beating machine and AI team for the cause? [:)]

I don't disagree that our AI will never be as good as a human. That's not our goal.Our goal is to develop something reasonable that you guys will have fun with. We get there by learning about how people play the game and making improvements.

Thanks!

Mike

[Dimitris]

ORIGINAL: Shemar
If I could only get my minions to stop turning their radars AND SONARS on every time an itchy bitty missile is fired! Haha!

To better understand what is up with that, you have to place yourself in the following tactical problem:
Your surface group is electronically silent. You (somehow) detect an incoming weapon that appears to be heading towards one or more members of the group. Do you (a) hold EMCON silence and pray or (b) light up everything in order to get a more complete tactical picture (you detected one missile - who's to say there aren't 100 more out there?) and eliminate the _clear and present danger_ to your asset(s)?
(And before you say it - yes, there are more than these two options in RL. Let's focus on these two for now)

Harpoon usually did the former, and that was a source of endless "why is the stupid AI not lighting up its radars to blast that missile out of the sky?!!" bug reports. It also meant that e.g. once you knocked out the enemy's AEW cover the whole enemy group (if it maintained EMCON silence) was practically a duck shoot.

Command does the latter. Is it perfect? No. Does it cover each and every case? No. Is it _more rational if you put yourself in the shoes of the captain of the group flagship_? Yes, we think so. Do we intend to improve it so it covers more cases and what-ifs? Yes.

Thanks.

[Shemar]

ORIGINAL: mikmyk

It does sound bad but I think having defensive missions set up might have helped.

This is a quick reply before reading the rest of your post because I need to address it immediately. The above statement was not meant as a criticism to the game or the developer team. It was meant as a general comment on how computer game AI (any computer game not just this one) is bad at even things that to a human seem obvious and self evident and how much complicated work it is to make it behave even marginally 'intelligent'.

Edit having read the rest of the post: Like I said above that was not meant as a criticism. I am enjoying the game very much. Having said that, since you do mention me (or us) helping to improve the game, I am most certainly willing to do so.

[Shemar]

ORIGINAL: Sunburn

Command does the latter. Is it perfect? No. Does it cover each and every case? No. Is it _more rational if you put yourself in the shoes of the captain of the group flagship_? Yes, we think so. Do we intend to improve it so it covers more cases and what-ifs? Yes.

Guys please don't take my comments and rants against computer AI personally. They are not meant as attacks. As you said you will never get computer AI to act like a human, nor do I expect you to.

If you want to see some (hopefully constructive) criticism to all of the above, take this:
I hate having the AI override my decisions. I understand that giving us players complete control may be contrary to the design goeals you have for the game and even contrary to what many of the players wish, but from my point of view not having the AI override my orders based on some internal logic would take my game enjoyment from 70% to 95%, because honestly, in every scenario I have played that is the one big thing I have not liked.

Once again, let me stress that I understand that is probably not the direction you want to take the game, so please don't take my anti-AI commentary personally.

Addition:
To the actual point: I never had a problem with Harpoon staying silent. If I wanted the sensors on, I paused the game and turn them on. Just as I (would like to) do with CMANO. And in any case I would never turn every active sensor in the entire fleet on, including sonars, navigation radars and weapon directors. I would turn selective sensors on, starting with the search radars of the pickets and seeing what I got from there. Turning every sensor on not only does not add to the information I would get but also gives away the exact composition of my entire force. I don't know how much of that information is actually useful to the computer AI against me, but still, just for immersion purposes, it would be nice to have control over that.

Further edit since you mention the opponent AI: I am not saying don't have the AI turn everything on, all I am saying is give me a RoE option to turn that off on my own ships.

[Felcas]

Why don't use an option on doctrine to override the use of diesel even when contacts are nearby?