Naval air attacks-Why can we not set the target area?

Uncommon Valor: Campaign for the South Pacific covers the campaigns for New Guinea, New Britain, New Ireland and the Solomon chain.

Moderators: Joel Billings, Tankerace, siRkid

iceboy105
Posts: 97
Joined: Mon Aug 26, 2002 10:59 pm
Location: USA

Naval air attacks-Why can we not set the target area?

Post by iceboy105 »

We can set the target area for almost all attacks (air to ground, lcap, escort, ship to ship, etc.) but why not for air to ships? This doesn't make sense to me. Why leave this one out for the computer to choose? Seems to me fixing and letting the player assign a target if he wished would be a simple great add on and end everyone's frustration over planes attacking stupid targets which the player didn't intend. I recently had captured Port Moresby as the Japanese and set my planes to naval attack. Instead of attacking allied troop transports headed for Gili Gili right next door (which is what I wanted and needed as I did not need a strong allied force in the area) the computer sent my planes all the way to Guadalcanal and Tulagi to attack some cruisers and destroyers with strong AA. Too far, too wasteful, not needed, and not wanted. Sucks, sucks ,sucks!!!:mad:
mapr
Posts: 72
Joined: Wed Sep 04, 2002 5:38 pm
Location: Finland

Post by mapr »

Hi,

Something really should be done about this...

How about if you could atleast order what kind of ships your planes most likely would attack? Wery much high level decision...
I'd like see this done base by base... But I am one of tose control freaks who see allmost nothing good in computer commander decisions. Same ones who does run towards from enemy CV's but DOES NOT take your planes away from training missions... Castrated naval combat ships(not able to chase enemy TF's) And so on..

To topic:
In my PBM game my Japanese opponenp brought tanker TF to Gili Gili and couple task forces full of AP's towards PM. All I wanted to do was to keep PM unsuplied and keep Japanese infantry stuck there... Lots of it. My planes kept on attacking tankers at Gili Gili and AP's sailed quite safely to PM and back... I had lots of naval search and so on, no question about unsufficient recon. All I could have done would have been risking my CV's(or other ships) agains 3-4 for build up Japanes bases... And my Australian level bombers would have been enough to sink every ship sailing to PM if I had any means to tell them to do so...

-Antti

PS. Great Game ;)
Pps. But in may aspects wery much inferios to ancient War in the South pacific C64 version... As general UV is much better, but after playing War in the South Pacific(WITSP) UV was like coming back home after many years. Learned much more about UV from WITSP manual than from UV manual....
User avatar
mogami
Posts: 11053
Joined: Wed Aug 23, 2000 8:00 am
Location: You can't get here from there

Airgroup orders and turn phases

Post by mogami »

Greetings, It might be a product of what orders you are giving.
And/or misunderstanding the manner in which turns/missions resolve.
Each airgroup can have 2 missions assigned.
Primary and secondary. You will note that all land targets are available for selection as secondary targets but naval attack is only an option as primary.

Each turn, search planes launch, aircraft in other groups assigned naval attack sit waiting at airbase ready to launch if target is found. If no suitable target then they will arm/fly to targe from secondary mission. If "rest" is secondary target then the aircraft stand down untill afternoon search planes launch.

"How bombers fly combat missions"
First they aquire a target
group checks primary, if primary is a land target then the group will move to fuel/arm phase.

What can prevent launch
Weather, not enough escort fighters assigned. (will cancel entire mission)
Things that can reduce number of aircraft in mission
Too many aircarft at base (sizex50) reduce 25percent
Fail morale check reduce 25 percent
Fail leadership check-reduce 25 percent
Not in range of Air HQ (only Japan has Air HQ at start and they only have 1) reduce by 25 percent

All reductions are cumlative so a group with all these problems will ground it's self.

So If computer selects a target you would not have prefered it might be because the one you liked had too much CAP. I am aware sometimes groups attack targets and get slaughtered by CAP (AI did not know CAP was there or it was still less then other target).

The missions are launched as time passess. Groups might launch against 1 TF simply because it is spotted first (computer does not wait for a search to resolve before launching)

Picking the targets (piority) yourself would not improve results
(you might turn down a mission because search reports only a transport and fly on one reporting CV only to discover "transport" was battleship and "CV" was oiler.
If I had been an operational commander, I am sure I would have been found in the radio shack where reports come in.
(I might have slept there) However I still would have preffered the group leader to make the attacks (I would just like to know what was going on) (UV does this)
Image




I'm not retreating, I'm attacking in a different direction!
iceboy105
Posts: 97
Joined: Mon Aug 26, 2002 10:59 pm
Location: USA

Post by iceboy105 »

My planes were set on naval attack primary, rest secondary.
There were no allied carriers or base planes to provide cap over either fleet. In the case I presented none of the reasons you presented for poor target selection applied. The troop transport taskforce was spotted first and the weather was fine with no cap.

"Picking the targets (piority) yourself would not improve results"
I disagree. In this case I would have destroyed the troop transports headed for Gili Gili and the allies would not have landed and posed a threat to me. Even if I did attack a fleet that was spotted wrongly that is part of the fun of the game aka Midway, Coral Sea battles this happened frequently.

I think if you are going to provide this argument then why should we get to set the target priority for any attacks at all?
Just let the AI duke it out against each other. We get to decide every single attack target except for air to naval. Why cant we decide this too. Heck if a few people are set on letting the ai be in control then make this an option at the beginning of the game.

AI Control over targets
Manual Control over targets

This would end this argument for good.





""Greetings, It might be a product of what orders you are giving.
And/or misunderstanding the manner in which turns/missions resolve.
Each airgroup can have 2 missions assigned.
Primary and secondary. You will note that all land targets are available for selection as secondary targets but naval attack is only an option as primary.

Each turn, search planes launch, aircraft in other groups assigned naval attack sit waiting at airbase ready to launch if target is found. If no suitable target then they will arm/fly to targe from secondary mission. If "rest" is secondary target then the aircraft stand down untill afternoon search planes launch.

"How bombers fly combat missions"
First they aquire a target
group checks primary, if primary is a land target then the group will move to fuel/arm phase.

What can prevent launch
Weather, not enough escort fighters assigned. (will cancel entire mission)
Things that can reduce number of aircraft in mission
Too many aircarft at base (sizex50) reduce 25percent
Fail morale check reduce 25 percent
Fail leadership check-reduce 25 percent
Not in range of Air HQ (only Japan has Air HQ at start and they only have 1) reduce by 25 percent

All reductions are cumlative so a group with all these problems will ground it's self.

So If computer selects a target you would not have prefered it might be because the one you liked had too much CAP. I am aware sometimes groups attack targets and get slaughtered by CAP (AI did not know CAP was there or it was still less then other target).

The missions are launched as time passess. Groups might launch against 1 TF simply because it is spotted first (computer does not wait for a search to resolve before launching)

Picking the targets (piority) yourself would not improve results
(you might turn down a mission because search reports only a transport and fly on one reporting CV only to discover "transport" was battleship and "CV" was oiler.
If I had been an operational commander, I am sure I would have been found in the radio shack where reports come in.
(I might have slept there) However I still would have preffered the group leader to make the attacks (I would just like to know what was going on) (UV does this)""
User avatar
mogami
Posts: 11053
Joined: Wed Aug 23, 2000 8:00 am
Location: You can't get here from there

Target order

Post by mogami »

Hi, I guess I was not specific enough in target allotment discription.

The more distant target might have been the earlist found (it might have been spotted the night before so your strike launched before the closer transport TF was discovered

or

the weather in the transport TF hex might have been worse then the weather in the CA hes

or

(heres where most players have the problem agreeing with game program)
The target they attacked had a higher point value then the transport TF. It is in this regard players would like to set values themselves. However the airgroup might still have attacked the CA group if the above (weather/time of discovery) reasons come into play.

What it comes down to is during turn the airgroup commander says I have a target within orders (naval attack) at hex xx yy
I am going to launch. 15 minutes after he launches a new target more in line with what you (the player) would like to attack appears but the group has already departed. In tactical level games there is often the option to break radio silence to redirect a group. Prehap a routine that could do something along this line would be favourable? (This would generate as much pain as joy)
Image




I'm not retreating, I'm attacking in a different direction!
User avatar
Fred98
Posts: 4019
Joined: Fri Jan 05, 2001 10:00 am
Location: Wollondilly, Sydney

Post by Fred98 »

Mogami, my aircraft refuse to attack enemy naval units. I have been playing this game since June 2002. My aircraft have NEVER made an air attack on enemy naval units. EVER.

My aircraft locate them often but never attack.

In the “reasons for not flying” you made 2 important points:

The need for an air force HQ unit. This is not mentioned anywhere in the manual. In fact it has been stated many times that a HQ unit has “support” and "aviation support". Whether this is army, navy or air force HQ is irrelevant. Then you said that only the Japanese have an air force HQ at the start. If that is so it explains why allied aircraft cannot attack Japanese shipping. Please clarify this point.

Your second point, said that one squadron should be set to Naval Search. They will search. And another should be set to naval Attack. If the search is successful, then second squadron will then attack. I had no idea.

Unfortunately I have assumed that if I set a squadron to “Naval Attack” it will search and then if it finds something it will attack. They locate but NEVER attack enemy shipping. Please clarify this point.

Whilst the original poster asked about asked about assigning specific naval targets, I am asking a different question: how can I get an aircraft to attack a ship????? Isn't this central to the game??
User avatar
Piiska
Posts: 130
Joined: Wed Aug 28, 2002 2:44 pm
Location: Helsinki, Finland

Post by Piiska »

I suppose the main complaint the players have is caused by frustration of watching enemy TF that has been spotted for several turns doing its worse without anybody attacking it. Good examples are enemy carrier sitting for many days in one hex without a single attack attempt made on it, while those air groups that are in range, attack some other targets. Or an enemy invasion fleet sails quite unhindered over vast distances to your base, while the defending air groups waste their planes against impenetrable CAP of the CV taskforce.

I understand the fog of war, confusion during the battles and pure errors made by individual commanders and modeling of this is great in UV.

I love the fact that my subordinates make bad decisions occasionally and attack wrong targets. I served nine months as medic in the Finnish Defense Force and bad decisions by commanders and soldiers were a way of life in the military. In fact, every military makes BAD calls occasionally or maybe even regularly.

However, what I don’t agree with is that the theatre commander does not have a say on what TF a certain base should attack after a specific threat has been identified.

If the theatre commander (you) has decided to lay siege on PM, the base commanders in the area would be aware of this. They would have been briefed that any transport ships heading to PM are more important than cruisers somewhere else.

Now, if the cruiser TF is spotted before the transports I fully understand that the base commander decides to launch an air strike against it, but the next day when both the cruiser TF and the transport TF have already been spotted, the base commander should not attack the cruisers again.

If the contact report about the transports has reached the theatre commander I’m sure that the base commanders within a reasonable strike range would very quickly receive orders: “Drop everything else and attack the transports heading to PM regardless of your ready status or losses”. In this case the base commander would have to have a very good excuse indeed if the transports are not attacked, especially if he goes on and attacks another target.

In game terms the system could go like this. Any taskforce that has been spotted in the previous turn can be targeted, but this doesn’t mean an automatic attack. Targeting a taskforce with an air group would translate to orders: “To air commander of Lae. Enemy transport taskforce spotted yesterday at coordinates. x.x heading towards PM. Find and engage this taskforce”. After issuing such an order to an attack air group, it would wait for the results coming from search airplanes. If the taskforce is spotted again an air strike would be launched. If the TF is not found, the attack air group would miss the day’s action.

This way it would be a bit of a risk to order an air group to attack a specific target as it may well turn out to be a false contact report and while the search planes are at wild goose hunt, the attack planes sit on the airport doing nothing. Even if there would be other lucrative targets around.

This would also prevent frustration that people feel when an enemy TF sits for days and days right under your nose and nobody does anything.


To Joe:

Air HQs are not obligatory, but they are helpfull. This is mentioned in the manual. If you have set your attack Air groups for Naval Attack, but have put Naval Search to 100% there is no planes available to launch a strike. You need to have balance between search planes and attack planes. If you don't have specific recon planes such as Catalinas, try putting your Dauntless groups on Naval Attack and 30% search. That should do the trick.
mapr
Posts: 72
Joined: Wed Sep 04, 2002 5:38 pm
Location: Finland

Post by mapr »

Bad decions:
Yes, thats good... I'd give even more 'rights' to computer. To do more mistakes and to recover from player made mistakes. Like to see possiblity to parametrize what kind of freedom computer commanders do have. But also player should be given more options to work with...

Targeting TF:
Something like this could be good... Possiblity to blokade area would be wery nice. But what annoyed me in my game was that I did not agree with computer that tankers were more important target than AP's. As high level commander I'd have given order to sink every AP in range if there is no CV's to attack.
This should be done base by base, because in Solomons i'd possibly like to hit other kind of ships than in New guinea. This would be enough for me. Some kind of "Attact ranking" by ship types.

Of topic:
Same would go with Task forces, sometimes i'd like send them to area to hit transports with every bomb,shell and torpedo they have and leave other ships than CV's in unharmed if there is one more transport to sink... And possiblity to hit and run missions. Now your only shot is to quess where enemy TF could be and run there and back... Not good... During daylight my TF should be able to use it's recon planes to locate enemy and start hunting for it.

-Antti
Originally posted by Piiska
...
I love the fact that my subordinates make bad decisions occasionally and attack wrong targets. I served nine months as medic in the Finnish Defense Force and bad decisions by commanders and soldiers were a way of life in the military. In fact, every military makes BAD calls occasionally or maybe even regularly.

However, what I don’t agree with is that the theatre commander does not have a say on what TF a certain base should attack after a specific threat has been identified.

....

In game terms the system could go like this. Any taskforce that has been spotted in the previous turn can be targeted, but this doesn’t mean an automatic attack. Targeting a taskforce with an air group would translate to orders: “To air commander of Lae. Enemy transport taskforce spotted yesterday at coordinates. x.x heading towards PM. Find and engage this taskforce”. After issuing such an order to an attack air group, it would wait for the results coming from search airplanes. If the taskforce is spotted again an air strike would be launched. If the TF is not found, the attack air group would miss the day’s action.

...
iceboy105
Posts: 97
Joined: Mon Aug 26, 2002 10:59 pm
Location: USA

Post by iceboy105 »

More rights to the computer should be an OPTION at the beginning of the game. I dont want more mistakes!!!! But I agree with everything else you said :)
Originally posted by mapr
Bad decions:
Yes, thats good... I'd give even more 'rights' to computer. To do more mistakes and to recover from player made mistakes. Like to see possiblity to parametrize what kind of freedom computer commanders do have. But also player should be given more options to work with...

Targeting TF:
Something like this could be good... Possiblity to blokade area would be wery nice. But what annoyed me in my game was that I did not agree with computer that tankers were more important target than AP's. As high level commander I'd have given order to sink every AP in range if there is no CV's to attack.
This should be done base by base, because in Solomons i'd possibly like to hit other kind of ships than in New guinea. This would be enough for me. Some kind of "Attact ranking" by ship types.

Of topic:
Same would go with Task forces, sometimes i'd like send them to area to hit transports with every bomb,shell and torpedo they have and leave other ships than CV's in unharmed if there is one more transport to sink... And possiblity to hit and run missions. Now your only shot is to quess where enemy TF could be and run there and back... Not good... During daylight my TF should be able to use it's recon planes to locate enemy and start hunting for it.

-Antti

Diealtekoenig
Posts: 56
Joined: Sat May 18, 2002 8:42 am
Location: Port Moresby, New Guinea

Post by Diealtekoenig »

I think more control over ships on Naval Attack would make not only a better (more enjoyable/less frustrating) game but a more realistic game, but i would suggest doing it in the opposite way to what I think others are suggestiong, and I am not sure how to implement it.

I would suggest not "urging" AC to attack certain targets, but _forbidding_ them from attacking certain targets.

There are places I can see it would be suicide to send an airstrike, say against ships that are in a port with a huge airbase. It would seem as theatre commander (what I presume I am in this game) I could say "I forbid air attacks against ships in the Rabaul Hex" (or Port Moresby Hex) and allow A/C to attack targets they find elsewhere. As it is now, if I say "naval attack" the AC sometimes (not always) fly into enormous Land Based Cap (CV Based CAP they might blunder into, but they should be aware of the LB CAP).

I am not sure how practically this could be implemented. Designating specific hexes as forbidden seems cumbersome.

Perhaps one more level of orders: "Fear Land Based Air/Ignore Land Based Air" would work? That might effect not only where bases and CVs send airstrikes, but also effect whether ships on React will React only to hexes more than XX miles from _known_ (scanned/scouted recently) land based air over 50 AC or similar.
wobbly
Posts: 1095
Joined: Wed Oct 16, 2002 12:27 am
Location: Christchurch, New Zealand

Post by wobbly »

It seems to me that a previous poster (I know it isn't all my idea is what I am saying here - just can't find who said at this minute) had the right idea.

You should be able to target a specific type of ship or more specifically a specific type of task force mission; maybe with a secondary. This way if you know an invasion force is on it's way, you've just lost it's location, you can at least have a degree of control over what ships your air groups are giving priority to.

Scouting and chance, which is what the current system is simulating by giving AI control of targeting at the moment, will still be in effect as you will need to find teh task forces to target them; you'll just be allowed a little greater degree of control.

A typical air group mission status might be:Primary mission: bomb Naval - transport TF first, Any TF second; Secondary mission: bomb airfield - target Gili Gili.
[center]
Image
[/center]
SViking
Posts: 13
Joined: Sat Dec 14, 2002 12:37 am

Re: Naval air attacks-Why can we not set the target area?

Post by SViking »

I have to agree with the original poster: we should be able to set priorities for air-naval atacks...or the AI needs to be improved. Here's an example of the senseless tactics that are driving me nuts a times...

Playing as Allies in May '42, I positioned my carrier fleet SE of Gilli Gilli and a surface fleet a little closer in to intercept enemy transport fleets. Never mind that my surface fleet failed to move to intercept allowing the IJN to land troops. What really annoyed me was watching my recon spot two transport fleets (indentified at least one AP in each - one obviously already dropped troops) and a carrier fleet (two CVs spotted plus ship types), then having my carriers launch the following attacks:

- A handful of SBDs and all of my TBDs *UNESCORTED* attack the enemy carrier fleet - needless to say the enemy CAP wiped them

- All of my escorting fighters (~30) and the majority of my SBDs attack an enemy transport fleet - the one that had already dropped troops

:mad:

There is no excuse for this nonsense. The commander should be hung for treason. The errors seem obvious:

a) complete misuse of fighter escorts
b) attacking a transport fleet that had already dropped troops when an enemy carrier fleet and another transport fleet were nearby

This is why we need the ability to designate target priorties. The AI often seems incapable. Not always incapable*, but definitely often.


*moments later the enemy carrier fleet launched ALL of it's A/C against my carriers. Scratch 2 flattops. **sigh**
NorthStar
Posts: 217
Joined: Fri May 17, 2002 3:53 am
Location: New York, US

SViking,

Post by NorthStar »

Actually, I'd say the AI Commander (at the TF Level) made the correct decision. Ok, possibly attacking the APs at all wasn't a terrificbad, but at least he also attacked the CVs.

What caused you major problem was poor strike coordination. The escorts failed to link up with the bombers. This was not an uncommon occurance in real live, especially early in the war. If the game didn't simulate occurances like this, it wouldn't be a good reflection of the war.

Also, while it may be obvious to YOU which APs just unloaded, it wouldn't be all that obvious to SBDs at 10,000 feet. And strike pilots are not know for circling over ships for long periods of time identifying the perfect target. More often than not, they hit the first likely looking target and high tail it for home. Misidentification and attacks on less than optimum targets were rampent for both sides during the war.

The trick, as Mogami said, is to learn how to manipulate your settings to have the best chance of what you want happening. Stuff like this is part of the game, and you have to learn to deal with it.
SViking
Posts: 13
Joined: Sat Dec 14, 2002 12:37 am

Re: SViking,

Post by SViking »

Originally posted by NorthStar
Actually, I'd say the AI Commander (at the TF Level) made the correct decision. Ok, possibly attacking the APs at all wasn't a terrificbad, but at least he also attacked the CVs.
Huh? You really don't believe that do you? The smart decision would be to attack the CVs first. A marginal possibility would be to catch the full transports as well.

What caused you major problem was poor strike coordination. The escorts failed to link up with the bombers.
Actually, that's not true. The escorts did link up - just with a strike that didn't need them. Weather, btw, was clear.

Also, while it may be obvious to YOU which APs just unloaded, it wouldn't be all that obvious to SBDs at 10,000 feet.


Had the strike taken about a 90' heading to the right of their actual, they would have found the full transports. I didn't explain previously, but the two transport TFs were a few hundred miles part. So, the problem was, the strike went the bloody wrong direction to start.

The trick, as Mogami said, is to learn how to manipulate your settings to have the best chance of what you want happening. Stuff like this is part of the game, and you have to learn to deal with it.
...and I still say we need to the ability to set priorities. Call it another setting to allow us to get a better chance for the desired results. It doesn't seem like a big deal and not unrealistic at all. In fact, I would even take this a step farther - as one person suggested, I think we should have the ability to restrict attacks on targets. E.g., I may see another surface fleet, but choose to wait until I spot CVs.

This isn't just a matter of occasional bad luck, this was a well planned screw up by the AI. I could understand missing the CV fleet had there been bad intel, but that wasn't the case. I've seen this before. This was just a particularly good example of how bad things can get when the AI makes bad decisions.
NorthStar
Posts: 217
Joined: Fri May 17, 2002 3:53 am
Location: New York, US

Re: Re: SViking,

Post by NorthStar »

Originally posted by SViking
Huh? You really don't believe that do you? The smart decision would be to attack the CVs first. A marginal possibility would be to catch the full transports as well.
Granted. As I stated, it wasn't brilliant, but at least the CVs weren't ignored. Also, depending on how close the APs were to the CVs, it is entirely possible that the second strike was supposed to atack the CVs, but found and hit the APs by accident.

Actually, that's not true. The escorts did link up - just with a strike that didn't need them. Weather, btw, was clear.
Correct. But again, my point is that this is most likely bad luck -- the escorts joined up with the wrong group -- not a poor decision by the AI. That is, I don't think the AI decided to send the first strike in unescorted, I think it just happened

Had the strike taken about a 90' heading to the right of their actual, they would have found the full transports. I didn't explain previously, but the two transport TFs were a few hundred miles part. So, the problem was, the strike went the bloody wrong direction to start.
There are a number of reasons why this might have happened. Most notebly, the timing of detection of the two groups. If the full transports were detected later, the strike may already have been airborne.

...and I still say we need to the ability to set priorities. Call it another setting to allow us to get a better chance for the desired results. It doesn't seem like a big deal and not unrealistic at all. In fact, I would even take this a step farther - as one person suggested, I think we should have the ability to restrict attacks on targets. E.g., I may see another surface fleet, but choose to wait until I spot CVs.

This isn't just a matter of occasional bad luck, this was a well planned screw up by the AI. I could understand missing the CV fleet had there been bad intel, but that wasn't the case. I've seen this before. This was just a particularly good example of how bad things can get when the AI makes bad decisions.
Actually, I basically agree with you here. I think it would be great if we had manual control over attack priorites or could set exclusions, I jsut don't think it will make as big a difference as some people seem to think it will. The statement I take exception to is " ... this is a welll planned screw up by the AI".

The biggest problem with the game, IMO, is that the player is presented with a great deal of information out of context and with no information about what is going on under the hood. For instance, we see all of the search reports similtaneously at the begining of the air phase. However, when strikes are planned, the computer accounts for the ORDER and TIME which they came in. Since the player can not see this information, we are left cursing the AI for a dumb decision. So your statement about bad intelligence is not necessarily true. The player isn't given enough information to see how good or bad the intelligence is!

In point of fact, it is likely (agin, IMO, I'm willing to give the designers the benifit of the doubt) that the AI made a perfectly acceptable decision (AIs are never brilliant) based on information it had at the time. But since we don't have all the inputs it used, we can't judge.

Again, I basically agree with your premise. It's just that I am able in almost every instance to create plausible scenarios within the game rules as to why seemingly dumb decisions were made that would make them not so dumb. As long as this is the case, I'm not willing to say the system is broken in some way. And I hate to see others say (or imply) it either.
SViking
Posts: 13
Joined: Sat Dec 14, 2002 12:37 am

Re: Re: Re: SViking,

Post by SViking »

Originally posted by NorthStar
Granted. As I stated, it wasn't brilliant, but at least the CVs weren't ignored. Also, depending on how close the APs were to the CVs, it is entirely possible that the second strike was supposed to atack the CVs, but found and hit the APs by accident.

...

Again, I basically agree with your premise. It's just that I am able in almost every instance to create plausible scenarios within the game rules as to why seemingly dumb decisions were made that would make them not so dumb. As long as this is the case, I'm not willing to say the system is broken in some way. And I hate to see others say (or imply) it either.
Fair enough - there may be a plausible explanation for everything that happened that may exonerate the AI. So I will adjust my stand from certainty to one of deep suspicion...suspicion because this was not an isolated instance of apparently strange decisions, but only a particularly good example. Also, I've haven't noticed any attacks on secondary targets despite having identified a primary target during the turn. So the explanation of timing of indentification is questionable. Perhaps some further testing/observations will clarify the situation....or perhaps Matrix can clarify.


Actually, I basically agree with you here. I think it would be great if we had manual control over attack priorites or could set exclusions...
A Matrix rep responded in another thread that the item is on their list, though no work has started on it yet. If the AI is indeed acting rationally based upon available information, then the improvement should probably be one of exclusion, probably implied - i.e., by specifying the fleet types you want to attack and excluding all others - since I suspect that's the more likely situation. If you think about it, if the AI is working properly, then prioritization of targets could lead to the same ambiguities which are responsible for the current undesireable results. In any event, I hope (and trust) Matrix carefully considers the issues and defines the requirements before throwing something into the game.
NorthStar
Posts: 217
Joined: Fri May 17, 2002 3:53 am
Location: New York, US

Re: Re: Re: Re: SViking,

Post by NorthStar »

Originally posted by SViking
. . . If you think about it, if the AI is working properly, then prioritization of targets could lead to the same ambiguities which are responsible for the current undesireable results. In any event, I hope (and trust) Matrix carefully considers the issues and defines the requirements before throwing something into the game.


My point exactly (the more we talk, the more we agree;) ). The system already has a priority list "hard coded" into it. It was posted on another thread somewhere, but IIRC it goes something like CV, CVL/CVE, AP, BB, etc. . . .

Assuming everything is working correctly (an assumption I am willing to make) then shifting the top couple of priorities would not be likely to have a marked effect on results. The same uncertainties and randomness will continue to cause sub-optimal results.

Excluding fleet types is probably more reasonable (i.e. Surface TF or Transport TF), as long as we realize that misidentifications will still somethimes result in excluded fleet types being attacked.

Also, on a tangent, I would sugest that Carrier TFs should always be forced to have Carriers as their first and highest priorities. I don't really see a Carrier commander ignoring an enemy Carrier TF that can make his day very bad, very quickly in favor of a relatively harmless Transport group, no matter what orders come down from higher command.
SViking
Posts: 13
Joined: Sat Dec 14, 2002 12:37 am

Re: Re: Re: Re: Re: SViking,

Post by SViking »

Originally posted by NorthStar
Excluding fleet types is probably more reasonable (i.e. Surface TF or Transport TF), as long as we realize that misidentifications will still somethimes result in excluded fleet types being attacked.
Yep. You could also end up attacking nothing, though you know very well that one of the fleets contains CVs. I guess that's the point where computer AI breaks down. You would need one extremely sophisticated program to emulate an educated guess based on the current situation, past intel, opponent tendencies, etc. However, I'd rather face that problem than the current situation...and, assuming it's an option, at least it's at the player's discretion. Ultimately, giving the player the option for more discretion over air missions may be the best solution...maybe.
caine
Posts: 112
Joined: Thu Jun 13, 2002 12:27 pm
Location: Barcelona (Spain)

Post by caine »

I also think that more control oer target selection should be done.Being the game like it is now, it is incredible to see how 2 aircraft can attack a carrier group but a lot a group of battleships, for example.What could be the effect of launching 3 or 5 planes against a carrier group?
I strongly support the idea of manually selecting targets according to the available information.Without that, very odd situations occur.Giving only the option for choosing targets and relative strenght of the attack against an objective should be implemented.In my opinion it is almosty impossible to implement an AI for choosing the best target and without options available for each air group, like priorities.
mapr
Posts: 72
Joined: Wed Sep 04, 2002 5:38 pm
Location: Finland

Post by mapr »

I'd think that other issue concerning target selection could be target area selection. More i think of such option, more I'd like to have it. Even if practical result wouldn't be wery drastic. Now players have no control of area where planes attack. AI supposedly have some hadcoded ways to decide witch ones of possible targets to attack to optimise results...

Problem ain't too severe and concerns mostly bombers having range of 20+, but I find it somewhat annoying that bombers may do attacks anywhere on range and 360 degrees from their current location. Long range bombers placed at Lunga now may attack almost anywhere on the map. To me these bombers are quite random as a weapon... Not attacking where player would like them to attack. Or targets player would like them to attack. Wearing up themselves doing long range raids when it might be much more useful(atleast in players opinion) to keep planes downed and wait until targets are found nearby... And possibly on a certain area.

It would be nice if player could set target(and search) zone to each group, of cource AI could make it's own decisions, but would prioritise target zone that have been set much higher and concentrate more recon to the area and would most likely to attack targets in target zone. Setting search zones would be nice too... Would allow concentrated searching to some areas in the expence of others... Resulting that somewhere even a lifebout would not go unnoticed in thunderstorm and somewhere 10 CV TF could be sailing under clear skies for days and stay unnoticed.
Post Reply

Return to “Uncommon Valor - Campaign for the South Pacific”