Impact of strike logic (ANW 393)

[hermanhum]

ORIGINAL: rsharp@advancedgamin

The issue Herman is referring to is most likely a separate but related issue. Airgroups currently have a maximum 24 count and this is an arbitrary number.

I think that we may all be wrong on this. Here is a test file showing 30+ aircraft launching. All aircraft are the same with the same loadout. You can see a stream of nearly 40 planes launching when you run it in ANW SE from the Blue side with Show All activated.

[rsharp@advancedgamin]

Herman,

24 is only a limit of the number of planes in one airgroup. Do you see more than this in an airgroup launched by a mission? If that's the case then there is something different going on.

Freek,

I looked into why airgroups assigned to a mission are launched in sequence when they share a loadout and in parallel when they do not. Missions group them together and handle them as one airgroup in its representation but then use the original (split) airgroups when considering how to launch. Later it catches the unlaunched planes and sends them off.

In the case of assigning 3 airgroups to one strike mission:
So what we should have seen was one group of 8 planes with one loadout (2 of the 4 count airgroups shared a loadout type) and second group of 4 planes. This is an actual bug that I'm still looking for the proper fix.

In the case of assigning 1 airgroup to each of 3 missions with identical target lists:
This was your original solution. As I said, strike missions will not launch against targets already being pursued by other missions. I do see how this could be very useful when planning multi-axis attacks that require missions with differing delay times to achieve the one ToT. I'm considering a user option (ini setting) to enable the desired redundancy in the strikes. By default, ANW will not allow redundant strikes. This is my comprimise to allow desirable scenario editing and prevent any bad behavior being introduced. In 3.10, this would be configurable on a per mission basis.

Thanks and let me know what you think,

[FreekS]

Russell,

1. This is what I would need to do to ensure the full strikes launched in ANW: I will need to eliminate all redundent strike missions from all my scens. However this will not lead to the intended Time-on-Target in either 3.6 or ANW, in 3.6 it will generate a series of 4-ship planegroups and in ANW it will depend on having equal numbers of different weapon types assigned (i.e. need to edit the loadouts of all scens and therefore also rebuild all formation air patrols, as they are reset when you edit the loadouts).

If I do this also then I can get intended Time-on-Target in ANW but not in 3.6. So my conclusion is I need to do a lot of work to get partially the intended behaviour that I already had in 3.6. And is there actually a benefit of this new strike logic? (Why does ANW treat multiple identical missions different from 3.6? What's the benefit?).

2. I don't understand what the actual bug is you describe above; my observation was that with 24 planes (4 Harm, 8 Walleye, 12 Harpoon) assigned to the strike on the Battlecruiser, three airgroups started launching at the same time, and started flying to the target as soon as they were formed up (Harm first, x min later Walleye, x min later Harpoon, reflecting it takes x min to launch an additional 4 planes). Personally I don't see how you are going to be able to change this (and I worry about what consequences such a change would be if you did).

So once 3.9.4 has been released, then I would like to understand what I need to change in all my scens. At the moment, with the possibility of further changes in strike logic coming (in 394) I won't modify my scens.

3. Now, on you last point, I sort of understand some benefits of your idea for 3.10 to make missions configurable through an ini file. It sounds interesting (and very difficult to oversee the consequenses of). I think it needs a separate thread for people to discuss it. It would be nice to understand the list of configurable parameters and the possible impact of all. Also if it is done through the ini file, then would players not need to have the same ini-file settings as the designer to get the intended mission behaviour? I cannot think this stuff through with the little info I have. Now of course feel free to limit that discussion to the CCC group, but have they not been the group to support you introducing these new features that we have been repairing?

Freek

[hermanhum]

ORIGINAL: rsharp@advancedgamin

24 is only a limit of the number of planes in one airgroup. Do you see more than this in an airgroup launched by a mission? If that's the case then there is something different going on.

That may be the case for strikes with the "Forced Strike" option enabled, but that is not the case in either of the aforementioned examples. In each of these examples, a 'stream' of groups attacks the target. Each group consists of 4 planes.

The difference is that the first example
( http://forums.gamesquad.com/showpost.ph ... stcount=68 )
shows 24 planes out of 32 are launched while the second example

( tm.asp?m=1974249&mpage=1&key=&# )
shows nearly 40 in the air (10x groups of 4 planes).

I think that it clearly shows that the number of planes launched against on a strike is not limited to 24 planes.

ORIGINAL: rsharp@advancedgamin

In the case of assigning 3 airgroups to one strike mission:
So what we should have seen was one group of 8 planes with one loadout (2 of the 4 count airgroups shared a loadout type) and second group of 4 planes. This is an actual bug that I'm still looking for the proper fix.

I must disagree. The current behaviour in this example (fb.asp?m=1978929) with each type of loadout forming groups of 4 planes is how it has always worked. Designers who wanted to do something differently already have the tools available at hand if they did not like this behaviour. (i.e. Multiple strikes at the same target)

ORIGINAL: rsharp@advancedgamin

In the case of assigning 1 airgroup to each of 3 missions with identical target lists:
This was your original solution. As I said, strike missions will not launch against targets already being pursued by other missions. I do see how this could be very useful when planning multi-axis attacks that require missions with differing delay times to achieve the one ToT. I'm considering a user option (ini setting) to enable the desired redundancy in the strikes. By default, ANW will not allow redundant strikes. This is my comprimise to allow desirable scenario editing and prevent any bad behavior being introduced. In 3.10, this would be configurable on a per mission basis.

Emphasis added by HH

As I see, the situation is thus for the "strike missions will not launch against targets already being pursued by other missions" situation:

1) There can only be one default
2) There are two choices -- Allow or disallow redundant strikes.

The H3 default is Allow. So, why use the direct opposite as the default for ANW? If you can only choose one, is it not logical to choose the current default rather than turning things around 180 degrees? It is not as if there is a third option.

Anyone using the ANW SE can always toggle select the option to disallow redundant strikes. This would allow for easier transition of H3 scenarios to ANW users. Also, IIRC, a number of ODb scenarios also share the need for redundant strikes.

Caveat: I am assuming that only the Ship/Land strike behaviour is being discussed at this time. I am not certain how any of this discussion may or may not affect aerial interceptions.

[FreekS]

Jeez, I just ran the test scen with 'Focused strike' off instead of on, and guess what; still 24 planes launch but now in 5 groups of 4, first three groups (each with different loaout) then 2, then one. Just like in 3.6.

So the 'Focused strike' switch does things I don't understand, which have nothing to do with focus on one target. I think I'll give up trying to understand the logic of strike behaviour and wait for the Scenario Editor manual.

Freek

[rsharp@advancedgamin]

Freek,

As I see it, the change you will need to implement to get the same effect of your coordinated strike missions in 3.9.4 as 3.6 will be to add 1 line in the harpoon3.ini file. Additionally, you would have to recommend the same harpoon3.ini setting to the users. Not the best solution but a comprimise. In 3.10, the settings will be contained in the missions themselves.

Yes, there is not enough information for you to fully understand what parameterized missions will do. That will come as the feature is polished.

Herman,

There is no simple switch to just turn back on the 3.6 behavior. Don't expect me to change the default mission behavior without further study. Surely you can understand this as you've been preaching about it for some time.

Thanks,

[hermanhum]

ORIGINAL: rsharp@advancedgamin

There is no simple switch to just turn back on the 3.6 behavior. Don't expect me to change the default mission behavior without further study. Surely you can understand this as you've been preaching about it for some time.

The desire to examine the long-term consequences is certainly understandable. However, this mess was born from an arbitrary act and it seems pretty clear (at least to me) that only an arbitrary act is going to be able to rectify the situation.

Continuing to follow this path in the forlorn hope of salvaging an inelegant solution is just as likely to cause more future problems, IMO. Even though the solution may not be a simple on/off switch, biting the bullet now is far less painful than prolonging the agony (like the DD-X project).

[th3flyboy]

Pardon me if I'm just rewording what has already been said, why not just make the allow multiple strikes on same target to be an option similar to the allow nuclear weapons option in the launcher, and add that in the next patch? I know the 3.9.x series is feature frozen, but honestly, this sounds more like a FIX to me. It makes perfect sense to me to launch multiple missions on one target to arrive at the same time, the only other option that would really rectify that situation would be to add a timer on the mission creation screen to allow for you to specify the TOT, and have the TOT not go under the minimum amount of time needed for the whole strike package to get there. Then the package would be on target and attack at the same time. Furthermore, I believe that the arbitrary number should be changed to unlimited right off the bat, as that just ruins the kinds of massive strike packages I may want to make to simulate something like a cold war scenario where Soviet bombers are swarming the CBG with an attack. I may not have gotten a copy of 3.6 but I certainly understand what Freek and HH are saying, this appears to me like something that reduces the scenario maker's freedom from doing something that should be able to be done in the first place.

[rsharp@advancedgamin]

Hi,

Actually, your recommendation is exactly what I ending up doing for 3.9.4. I've set up an .opt file (just like UseNukes.opt) to allow, what I call, redundant strikes. I can see the reasoning for both the default and this optional behavior so I left it up to the user.

The other option of a ToT manager would definitely be a new feature and one I've even designed out a bit. I'm saving it for a rainy day.

As to the number of planes per air group, the real issue is the fuel in the first plane of the air group versus the fuel in the last plane to launch. Not a huge problem but then it will also just launch multiple air groups simultaneously to put just as many planes in the sky.

I have a few tests to run and then I'll put up a 3.9.4 release candidate if you would like to try.

Thanks,

[th3flyboy]

ORIGINAL: rsharp@advancedgamin

Hi,

Actually, your recommendation is exactly what I ending up doing for 3.9.4. I've set up an .opt file (just like UseNukes.opt) to allow, what I call, redundant strikes. I can see the reasoning for both the default and this optional behavior so I left it up to the user.

The other option of a ToT manager would definitely be a new feature and one I've even designed out a bit. I'm saving it for a rainy day.

As to the number of planes per air group, the real issue is the fuel in the first plane of the air group versus the fuel in the last plane to launch. Not a huge problem but then it will also just launch multiple air groups simultaneously to put just as many planes in the sky.

I have a few tests to run and then I'll put up a 3.9.4 release candidate if you would like to try.

Thanks,

That's good to hear about having that as an option, and I can understand the putting the idea for a TOT manager on hold, as I do understand that is a new feature, but it's nice to hear that that idea is being taken seriously, as in various other communities usually new feature ideas like that would be given the brush off. I guess I can see the planes per group reasoning, it's just to me there is one little flaw in that plan, setting up a mission with aircraft from multiple bases. That would seem to me like something which might need to be experimented with. Also I'd love to try that 3.9.4 RC, and I'll see what I can do about posting any bug reports I find.

[rsharp@advancedgamin]

Not only are we not brushing off the ToT manager, we've been dreaming on it for years. Still looking for the proper opportunity to implement it.

What brought me around to allowing redundant strikes was the notion that you could have a multi-base strike with several missions. You can manually calculate the ToT from each and delay the different missions so you achieve the simultaneous strike. Not incredibly hard to set up but the difficulty would be in getting it just right.

Thanks, I look forward to your feedback,

[hermanhum]

ORIGINAL: th3flyboy

It makes perfect sense to me to launch multiple missions on one target to arrive at the same time, the only other option that would really rectify that situation would be to add a timer on the mission creation screen to allow for you to specify the TOT, and have the TOT not go under the minimum amount of time needed for the whole strike package to get there. Then the package would be on target and attack at the same time.

The request for a new TOT feature would likely only apply to land strike mission because a TOT attack against a ship target is virtually impossible by the AI. You would need to know exactly when and where a ship is going to be detected in order to coordinate the various strike elements. Of course, there are some scenario designers who write them this way.

The TOT feature would thus only really apply to the Land Strike (where you know the exact location of all the targets). As pointed out, this is already fairly easily calculated. Although such a new feature might be handy and simplify the process slightly, the risk is always the unforeseen and unintended consequences. Since users already have the necessary tools to accomplish this task, I suggest that it is unneeded with more risk than anticipated low return on utility.

[th3flyboy]

ORIGINAL: hermanhum

ORIGINAL: th3flyboy

It makes perfect sense to me to launch multiple missions on one target to arrive at the same time, the only other option that would really rectify that situation would be to add a timer on the mission creation screen to allow for you to specify the TOT, and have the TOT not go under the minimum amount of time needed for the whole strike package to get there. Then the package would be on target and attack at the same time.

The request for a new TOT feature would likely only apply to land strike mission because a TOT attack against a ship target is virtually impossible by the AI. You would need to know exactly when and where a ship is going to be detected in order to coordinate the various strike elements. Of course, there are some scenario designers who write them this way.

The TOT feature would thus only really apply to the Land Strike (where you know the exact location of all the targets). As pointed out, this is already fairly easily calculated. Although such a new feature might be handy and simplify the process slightly, the risk is always the unforeseen and unintended consequences. Since users already have the necessary tools to accomplish this task, I suggest that it is unneeded with more risk than anticipated low return on utility.

Honestly HH, for now, it is not a good idea to add it, as it will be a risk during the process of fixing bugs to add something that may have a lot more bugs, but it would be a good idea down the line for beginners to help them get used to commanding. Furthermore, yes it would most likely be only for land targets, but you have to remember that this could also be used for CAPs and waypoints as well, as IIRC the USAF uses in mission planning a TOT on their waypoints, CAP plans, and their strikes. TOT at least in Falcon 4.0 means the time the aircraft is on that waypoint. It would be useful to be able to set that up for those kinds of missions as well, as it could come in useful for coordinating CAP operations so that you don't have gaps in your patrols.

[FreekS]

A few thoughts on the TOT calculator.
 
First, it could be for scen designer AND for player.
 
a. For a player I could imagine something like:
- select enemy ship just detected
- hit airops button
- up comes not only a list of all airbornplanes, but also of groups on the ground. Only the planes/grounded planes that have the loadout and fuel capable to reach the target are shown. Shown are also the TOT for each group. Selecting multiple groups would then make the planes launch at suitable times to coordinate with the TOT of the slowest selected group.
 
I feel that would add value.
 
b. This feature would also add value for the designer, as the ANW TOT is much harder to calculate than 3.6. In 3.6 it was easy, as outlined above, measure distance (2 clicks) devide by cruise speed (for many planes 500knots) and off you go. However the AI in ANW does not always use cruise speed. If it has enough fuel it may use Full or Burner (I think). And a mix of Cruise, Full and Burner. This was done (presumably) because the higher speed is safer for the strike planes.
So carefully coordinated strikes made in 3.6 MAY actually be less coordinated in ANW; and recalculating them to TOT cannot be done.
 
So haveing a TOT-calculator that takes into account the time spent at Cruise, Full, Burner speed of AI or player-controlled planes would be usefull. An alternative would be to carry out Ground strike and ShipStrike missions at Cruise speed only.
 
Freek
 

[hermanhum]

ORIGINAL: FreekS

A few thoughts on the TOT calculator.

First, it could be for scen designer AND for player.

a. For a player I could imagine something like:
- select enemy ship just detected
- hit airops button
- up comes not only a list of all airbornplanes, but also of groups on the ground. Only the planes/grounded planes that have the loadout and fuel capable to reach the target are shown. Shown are also the TOT for each group. Selecting multiple groups would then make the planes launch at suitable times to coordinate with the TOT of the slowest selected group.

I think that the idea to have it enabled for the AirOps button is nice, but might be a bit extravagant. If someone wanted to take advantage of such an option, expecting the user to create a mission from the Create Mission window and then enabling it as a switch (like the 1/3 rule) may be reasonable.

A potential downside of a ToT function like this might be the additional strain on the CPU. Already, the game seems to run quite slowly for anything but the smaller scenarios. The addition of extra calculation cycles to track the speed of the target, the distance to all units capable of launching an attack, the speed of the planes launching on their ingress (Cruise/Full/Afterburner), etc. seems a pretty heavy load because it would be in constant fluctuation as the target moved.

ORIGINAL: FreekS

An alternative would be to carry out Ground strike and ShipStrike missions at Cruise speed only.

I concur. It would be very useful to have an option emulating the previous strike behaviour at Cruise speed.

[rsharp@advancedgamin]

That's a good one. I would also like the feature where I could set a time and get a list of strike craft capable of reaching the target by that time. A filter.

I wouldn't worry about CPU cycles because, as noted, the ToT calculator, in any role with missions, would work with a fixed point. It could be a facility or it could be a reference point. After the initial calculation, it would only need to updated if the destination is updated.

If we did get to a point where we could use the ToT against slow moving targets then it would probably update periodically. Say every 5 minutes of game time. Just for dreaming sake, this would be user configurable so the game could adapt to the hardware available. In this case, many missions calculating at the same time would slow things down considerably on that update heart beat.

Dreaming aside, any implementation will go through several iterations which will build the parts necessary. The simplest would be a hotkey that allows the user to select a unit and a point. They would then give the ETA for the unit at different throttle settings at the current altitude. The most complex would be a complete strike planner with roles fulfilled by different mission craft. Lot's of space in between for useful tools for both the editors and players.

Controlling mission throttle settings is a feature in 3.10 but one thing at a time.

Great discussion.

[FreekS]

Russell,
 
But please don't forget the use of TOT-calculator by the mission designer.
 
If I have a set of strike planes devided over say two landbases, and I need to coordinate TOT, then I don;t know as a designer what altitude/speed the strikers will use. Some may use Full speed, others with less fuel may use cruise.
 
So the calculator should be implemented only if I can also control mission speed/altitude (i.e. 3.10)
 
If my planes are on a carrier than the designer also has to calculate where the carrier will be when the mission triggers. I usually make sure the carrier in that case is on delayed mission so I can set a plotted path and can calculate where the carrier will be when the planes launch. Its not exact though, and in the design phase I almost always have to replace the strike missions (strike, SEAD, CAP) several times with tweaked delay times.
 
A good calculator for both player and designer would thus be nice, combined with better control over mission speed/altitude.
 
Who knows, I might like 3.10!
 
Freek

[rsharp@advancedgamin]

3.10 will give you a lot more control as an expert user and editor. So you'll enjoy that. Get your default mission settings correct and leave them. A few missions will require exceptions and you can change their profile. That's pretty much what I see happening.

It will not include a ToT calculator of any sort. 3.10 is done as far as feature selection. Take that with a grain of salt, of course.

The carrier certainly would complicate things. I could see a new mode for the Launch Aircraft dialog that would work like you describe. It would have a new column for ETA which would consider

Delay time
Current course of the airbase (CV) and where it will be after the delay time
Distance to target
Standard mission speed and altitude of the current assigned mission (as stated, we may have to consider a consistent speed or use an average)

Further complicating things are mission delay variance and weapon release range. Such a feature would be more of a guide for the user than something the missions use to perform. Still a lot of value in such a feature for quicker and more accurate planning.

[FreekS]

ORIGINAL: rsharp@advancedgamin

The carrier certainly would complicate things. I could see a new mode for the Launch Aircraft dialog that would work like you describe. It would have a new column for ETA which would consider

Delay time
Current course of the airbase (CV) and where it will be after the delay time
Distance to target
Standard mission speed and altitude of the current assigned mission (as stated, we may have to consider a consistent speed or use an average)

Further complicating things are mission delay variance and weapon release range. Such a feature would be more of a guide for the user than something the missions use to perform. Still a lot of value in such a feature for quicker and more accurate planning.

Russell,

In the scen I'm making at the moment I've had to take all factors (weapon range, plane speed, plane navigating around nav zones, distance,...) you mention into account manually. End result is a SEAD mission from 4000nm away firing minutes before a ground strike mission from a carrier 300nm away. Both 7 hours or so into the game so the carrier has moved 100+ nm.

If we make the TOT calculator it HAS to be EXACT IMO. No value at all in an approximation. Better do it well in Harpoon 5.0 than poorly in 3.10!

Freek