pamak1970

August 8, 2005

I write this post to express my thoughts asking the members (and designers of course) to consider

if the feature i propose is considered important, If the idea, way to simulate it "makes sense"

and if it can be programmed with a reasonable effort.

I have read some threads in the past about various ways to insert the concept of areas of operations in the game (AoO).

Some use the name "command zones".

Although there was not a direct reply by the designers, i feel that they are not so eager to deal with this.

I noticed for example that their philosophy in general is against any artificial"- gamey limitations imposed on players in order to simulate a concept and i understand their point of view.

Using their philosophy as a point of start, i was thinking of ways to introduce this concept in the game.

My personal opinion is that AoOs are important enough to justify a reasonable amount of effort to simulate them, but since i am not a programmer i can not give the definition of " reasonable".

Therefore i will present my intial thoughts and let others decide.

We know that orders by most command levels define a certain area or zone for their subordinates to operate.

Among other advantages the AoO as a method of control of friendly units,helps preventing friendly fire incidents.

At the same time looking this from the opposite side of view, it helps engage enemy targets.

Units inside the same AoO are units of the same parent formation and have a better understanding of their position in relation with the position of other friendly units operating in the same AoO.

They have participated in common briefings before the start of operations when orders were issued, they have a better understanding of the mission of other friendly units in their area, they operate very often in the same command net ( radio frequencies) and get a quicker update of the movements of other friendly units.

All the above help them in identifying more easy the enemy positions and units and therefore engage them more quickly compared to situations when they have to engage "targets" inside AoOs of neighbour units.

The above is the main basis of my idea to simulate AoOs.

Since the new engine can deal with LOS of individual units, i think this can be exploited to simulate some of the benefits of AoOs.

In other words , if a unit spots a target

outside its AoO,it will have a much slower response (AI issue) to engage it.

If this happens then it will be obvious that the proper designation of AoOs (new type of command for the game) will be a crucial decision for the player and although there will be no rule to restrict him from quickly changing orientation , crossing boundaries and start engaging units on the other side of the map, still the disadvantage of slower response times to engage targets will discourage him.

The player could still be able to form new boundaries in order to redirect units during the battles and assign new AoOs but this type of command will neeed a significant time for execution (multiple turns)and therefore it will require anticipation and preparation.

Now this idea certainly needs some type of calibration and adjustments.

For the moment i am thinking of using it only in a battalion level (for multi battalions scenarios) so that things do not become complicated.

There will be some specific conditions that should be treated in a different way.

Two examples come to my mind for right now

First if a unit spots a target that fires on that unit, then there is no reason to suffer the delay penalty to fire back at that target even if it is located outside the AoO.

Second , if the target is deep and relative far away from the "expected" position of all other friendly units , there should not be any delay-hesitation for the friendly firing unit regarding the status of the target and it should engage the target without any delay even if it is outiside the AoO.

Now how can someone define this expression "deep and relative far way" inside the game?

One thought ( most compromises) i guess is to link the area of uncertainty with the time ellapsed.

For example,assuming that the set up zone for the friendly force at the beginning of the battle is south.

Then for the first turns any targets located anywhere in the northern 3/4 of the map is easily identified as hostile and there is no reason for any delay to engage it regardless if it is inside the friendly AoO or not .

After some time the same will apply inside only the northern half of the map.

After some time the same will apply inside only the northern 1/4 of the map and so on up to the time when any target anywhere inside the map will cause a delay to any friendly unit firing outside its AoO.

I can present other thoughts on the above but for the moment it is not nessesary.

First i would like to see reactions to see if people are really interested on what i presented until now.

After that we can talk about details.

[ August 08, 2005, 01:39 PM: Message edited by: pamak1970 ]

July 11, 2005

For anyone who has a relative new machine and is willing to see areas in 3D and zoom in ,here is the folowing link

http://earth.google.com/

July 11, 2005

Thank you for the site.

Has anybody seen the legend of any of these maps?

July 9, 2005

Nice to hear about the abilities of the new engine.

Now i do not know if anybody is interested with the conversation of HE effectiveness in general.

Anyway i will post the following link for anyone interested.

Although it is about modern artillery , most things apply for older ammunition also

As a side note it is interesting that according to

FM 7-90 appendix B , troops without overhead cover inside forests ,are still safer regardless the wood splinder effect.

This contradicts information from other sources.

Farthermore, this result is not due the effect of limited LOS cause of the presence of woods.

Notice also the use of PD (point detonation) fuse which will actually produce airbusts cause of the foliage-tree branches.

There are also other details like the effect of angle of fall which was mentioned in previous posts .

From

http://www.globalsecurity.org/military/library/policy/army/fm/7-90/Appb.htm#top

[ July 09, 2005, 01:37 PM: Message edited by: pamak1970 ]

July 8, 2005

Do feel free to explain how a guns inherent accuracy is dependent on velocity (with higher velocity being more accurate). I'd like to see how you reason that.

The higher the velocity the less time the projectile will be on air ,the less it will be affected by weather conditions , wind and so on.

This is the most critical issue regarding accuracy.

For the same reason (commenting on other posts ,if there is an option to use low or high trajectory to hit a target , you will most often go with the low trajectory,since the time of flight is shorter. .Additionally,high trajectory (in addition of the remark regarding easier spotting from counterbattery radars) has the disadvantage that the projectile travels along different heights with big enough variations in barometric pressures , wind speed and so on.

Also, i disagree with a post mentioning that you can have both low and high trajectory for any range .

If you see firing tables of various pieces ,you will see that this is not true.

July 2, 2005

I read this thread and it reminded me of one of the games i love.

I am not going to post its name but since we talk about various approaches in scenario design and tools ,i would like to give some information on this issue.

I will just point that the game is of a battalion level ,real time.

I will also have to clarify that all my comments are about scenario design and tools that help a player build them.

Similar tools and game interface are used to help player "control" his forces during regular play.

Nothing of the above is relative of programming AI.

This is much more complicated and a different subject

So, considering what i just said, i found that boolean logic (If-then-and -or else....) gave me the best flexibility as a scenario designer or as a gamer controlling my forces during play.

This is of course my personal opinion ,but i think it is worth something since I have some experience as an old wargamer starting with Avalon Hill board games and i have monitored very close the developments in all fields of wargaming entertainment including computer wargames.

Since i talk about boolean logic I will give some examples to present more clearly the possibilities a player or game designer have

See the following picture

Figure%202.jpg

That is one example regarding the control of movement in the game i am talking about.

You set the conditions that must be fullfilled in order to have a unit start movement from a waypoint towards another.

There are similar ways to control artillery (call artillery if"

or to set SOPS,like "fire if" or "retreat if", or to control formations.

Additionally the above conditions may apply to waypoints locations, along routes that connect waypoints or inside certain locations that a player draws on the map.

Let see some parameters of the above figure.

At the top we see the sentence "embark on route if" folowed by a combination of squares linked with and-or .

Now each square can be a certain condition ,so we can have a sentence like "embark on route if condition A is true or we can make it more complex like "embark on route if condition A is true and condition be is false, or (A true and B true) or ((a true and B true) or (c true and D force)) and so on.

Exprerienced designers can link even more complicated sentences by naming certain arrangements of conditions as "events" and then use sentences linking events -IF event A is true and or ....- see for example the second option in the figure

Now , someone might wonder what is this "condition A or B" stand for?

The asnwer comes from the rest of the image.

It can be any of of sentences you see there.

For example,suppose that someone chooses the 4th option

"Unit -this-has- reached checkpoint -blank".

The moment you choose this option you can modify it to match your preference.

For example "has" might become "has not" or "this" might become " 3rd platoon of second company"

and the blank square can be filled with any checkpoint (or waypoint) you have set on map during the game design or during play.

So your instructions you give to your computer might be

Embark on route (to the subsequent checkpoint two) if you are the second platoon of the first company and you have reached checkpoint one.

If you want for example to coordinate a movement by bounds with one unit covered by another from a certain location,your indstructions might be

"embark on route if unit X has reached checkpoint c".

If you see the figure, you will notice that you have a lot of tools at your disposal to control things.

You might want for example to preplan a fragmentary plan-about a possible counterattack in one sector.

Of course you want to be sure that you attack with superior forces against the enemy.

So you might want to use the option

" friendly operational forces in area A (the area you define on map) are > enemy operational forces in area A

You might want to be even more secure by adding to the above sentence

AND enemy platoons - or tanks in area A <4-assumming that you plan to attack with a force of 10 for example

I could give dozens of ways i used this tool to design scenarios or play the game.

For example, especially since the game is real time, i plan some fragmentary complicated plans during the Pregame phase.

I can link each of this plan to a "trigger"-the first option in the figure.

The trigger is like the "go code" you transmit to your troops during the game.

Whenever i feel that i have the right opportunity, i click one button and that simulates the "go word" in order for the troops to start executing a complicated maneuver.

This is important for a real time game,since you do not have the time to issue complicated orders during play.

I have also used the figure in order to simulate effects of boundaries and risk of fraticide.

The tools you have already see can be used in connection with "penalty zones" a designer has drawn on the map.

So i can simulate a situation where whenever a certain unit violates a boundary, it will have a certain chance to be destroyed and link this to a pop up message informing the player that he received friendly fire and that unit was destroyed.

Or the unit might jump from human control to AI control for a certain time, or even having a radio loss or whatever else.

The new version of the game will give even more flexibility in planning "penalty zones".

Another issue which was mentioned here regarding cheating is present in the above diagram.

In the option described as operational enemy units anywhere < 0, the game designer can specify if the computer will count the "true" number of enemy units, or if it will count only the units that were observed and spotted by the computer units.

Using all the above tools and more i can not describe here, i was able to let AI resolve battles from the beginning to the end based on my plans for both sides or test my scenarios against human players with very good results and a beleivable behavor from ccomputer cotrolled units.

Of course the game has already a good AI and this is certainly a prerequirement in order to see all these "plans" executed in a reasonable manner.

You can not use the above tools if AI of the game in general is weak.

I hope i gave a good picture of the potentials of this method of control.

My intention was not to advertise another game.

After all i could not compare CM with this game since they are different .

CM is a wonderful game and i think one of the best for wwii tactical scenerios ,probably the best.

However since we talk about various ideas for the new engine, i think it is worth examining the example i gave, taken from a totally different game.

[ July 02, 2005, 02:49 PM: Message edited by: pamak1970 ]

March 21, 2005

Too early would get my money. Also, it may be a problem they have not formulated a solution for. Some of what Steve has said about the difficulties of calculating LOS may apply here as well, since grazing fire inevitably has to take into account irregularities in the ground.

On the other hand ,since the whole issue is related with LOF ,LOS calculations, it is linked with the main core of the engine.

This means that they have to make a decision from the beginning ,regarding if and how they will deal with this issue.

March 21, 2005

It is interesting that there is not still any official comment regarding this aspect of game engine, which seems that either it is too early ,either the designers are not sure if they shoud or should not deal with this aspect,either they have not read the topic yet.

March 21, 2005

I provide a link with information regarding the subject we are discussing.

web page

You can search it farther.It is really very interesting.

March 21, 2005

Which might be one big reason why the Vickers squads liked to use plunging fire. I am inclined to think that grazing fire was more of a defensive tactic, designed as I said earlier, to interdict certain lanes of movement.

I think i will agree on that.

I guess that during an offensive the fire support base and the MGs are in quite a distance from their targets.

At such big distances you can not have grazing effect since you need to have a firing angle in order to send the bullets there.

Therefore the trajectory from the MG is not low and parrallel to the ground level and does not affect all the troops along the LOF.

Somewhere i read that grazing range was in general up to 400 or up to 700 m depending on the type of MG.

Another thing as an attacker is that you are not able to choose the battlefield and therefore it is more difficult to find the opponent vulnerable to grazing effect.

So i think the combinatiuon of the two above issues makes difficult to acheive grazing fire during an attack

March 18, 2005

Following the advise , i provide the link with the original discussion.

web page.

I agree of course about the problem of friendly fire and i also agree that ideally this proposition should apply to all small arms in general.

I focussed only on MGs cause i was thinking that there might be issues with processing PC power to apply the concept in general for all units.

This is just a personal impression ,not backed up by any knowledge in computer programming .

Anyway If someone reads the link i provided, he will get an idea of the nature and some possible complications of the idea we are talking about.

that is why i asked about any "official" news regarding the way that the new engine will resolve MG and small arms fire.

From the responses ,i understand that it is still too early for the designers to deal with this issues.

On the other hand and again without saying that i have any knowledge of computer programming, i was thinking that calculating LOF or LOS issues, may be part of the core of the game engine.

March 17, 2005

Sometime ago i had a discussion about this issue in another post.

Is there any official response regarding this aspect?

For people that may not understand, i talk about the issue regarding have a single line of fire affecting-attacking ALL the units along it, provided that they are at a relative short range and on about the same level with the shooter.

I feel as long as this issue is not simulated,the location of MGs in the game can not simulate realistic tactics.

For example, in reality MGs are generally located in such a way as to provide interlocking flanking and grazing fire,especially for FPFs (final protective fires) near the front of the main defense position.

In the game ,The effect of fire is not related by the type of it (like, flanking grazing).

It is only related to range and it only affects the area it lands (either as point fire or area fire).

The LOF is harmless along the rest of the path it "travels"

Therefore Mgs fire becomes always less effective as the distance to target increases ,even if someone acheives flanking or grazing fire.

So at the end it is more beneficial for players to have MGs aiming directly in front , than using an obligue angle in relation with the enemy avenue of attack.

In real life i think things are more complicated.

Take for example an enemy formation deployed and attacking towards friednly positions.

We can imagine the geometry of this formation as an orthogon parallelogram ,approaching our position.

A MG firing from directly infront is not so effective.

The LOF of bullets travels inside a relative small section of the area occupied by the formation and therefore there are less chances to hit a target located inside the formation's area.

The more the MG is in an oblique position, the longer the LOF "passing through" the formation's area and therefore the better chances to hit or pin members of that formation.

So if for example the LOF follows along the diagonal of this formation, it would be more dangerous compared to if it follows along its shorter side.

So , does anybody agree with the importance of simulating this effect,and if so, is it possible to accomplish this type of effect with the new game engine?

February 4, 2005

Originally posted by Battlefront.com:
Don't get me wrong... command and control delays, provided they are realistically modled, absolutely should be modeled. The point I was making is that these sorts of delays can't be blanket, abstract stuff or the game will lack realistic capabilities. For example, if a Platoon HQ sees an enemy tank coming up... and they have a radio, how long do you think it takes to get that intel to the Company HQ (prorviding, of course that the radio link works)? 1 minutes, 2 minutes, 10.34 seconds? Hard to say, but we need to keep this in mind when compared to a Squad seeing something and the Platoon HQ not having a good radio link to the Company HQ. In this case it could take 20 minutes for the information to arrive up the chain. Therefore, the system needs to model the reality of communications, not simply slapping arbitrary delays on stuff without the detailed command network and calling it good.

Steve

I would like to post my comments regarding command control delays .

These are not final thoughts ,it is just food for thought while we explore improvements for the new game.

First i would like to comment on delays regarding transmission of information.

In my opinion and since we have accepted that CMx2 will not be a command game ,i can not imagine how we can use these type of delays.

Since we agree that a player should have the ability to assume command of every element,this means that he needs to have real time information.

Otherwise he will not be able to have an effective command of squads or platoons.

Since the player will have this type of information when he assumes the role of a squad or platoon leader, he will posses the same type of information as a battallion commander also.

For me the issue of command delays should be more broad.

If we see the whole procedure of the process which includes gathering information-making a decision-execution, then i think it is more appropiate to focus delay on the last section of the procedure .

That means that the delay should be more focused on execution rather than gathering information.

After all the time needed to commit yourself to action reflects the sum of the time you need to perform all elements of the decision making procedure described above, including the time delay to gather information .

before i continue ,i would like to add something else about transfering information to higher echelons.

I do not beleive that simply because you have a radio link with higher HQ , you will transmit all information you gather from the battlefield in a matter of minutes or seconds.

The way i see it ,it is more complicated.

Neither a leader of a subordinate unit, nor the leader of a supervising unit have the time to transfer -monitor all pieces of information gathered during combat.

Although there are procedures where staff members for example of a battalion monitoring frequencies of subordinate units, it is simply unrealistic to expect that everything that subordinate units encounter on the battlefield,is transmitted to higher echelons.

This would be just an overflow of information which noone could handle ,especially during wwii but even today also.

I think that subordinate units transfer only important information which might affect the command decision of the higher echelon.

If for example a company with its organic weapons encounters relative weak resistense by some enemy squads during an attack , it is more probable that higher echelon will never get an exact piece of information regarding the position of those squads.

There might be a short report that the company is encountering weak resistance and keeps advancing towards the objective, with no specific details.

As long as the commander of the company feels confident about his mission and the resources he has, he will not spend time transmitting unnessesary information to the battallion commander.

So it seems that subordinate units will transmit information only under certain conditions.

One such condition is if the information is judged by the local subordinate commander that it is crucial for the battallion battle plan

A second condition ,is if the company commander feels that he can not accomplish his mission due to unexpected heavy resistance.

The second condition,will trigger a need to transmit a fairly detailed report to higher command with the intention,either to request additional fire support, either to request a change of mission for the company which will be close related to the maneuver of the company .

Now in combat mission we already have some cases with delay in fire support.

I will not deal with that although someone can propose certain improvements.

What we do lack is is a scale of various delays in changing of a mission -which again i link to movement.

For example, if a player as a company commander decides to change a certain path of movement there will be a certain delay.

The problem is that if a player as a battalion commander orders the same unit to change a path of momvement,we have the same delay as well.

In other words the engine can not understand when the player acts as a battallion commander and when the player acts as a company commander.

Notice that i prefer to focus more between battallion and company levels cause i think that the command delay is more significant between those levels than between company and platoon for example.

The more we go down in scale the more the command delay time compress .

So, now we have the issue ,how to make the engine understand which change of mission is coming from the battalion commander and which change of mission is comming from the company commander itself.

In other words, when the player assummes the position of the battalion commander and when the player is assumming the position of the company commander.

So, let see now what happens in real life.

Every mission assigned by a battallion to a subordinate company includes objectives and a certain area of operation.

A company advances or attacks against a certain objective which is the limit of its attack or advance ,with a frontage within certain boundaries left and right which seperate the company from adjustent units.

You defend an objective against enemy units approaching from a certain avenue of approach or direction of attack .Again the mission designates a certain area -defensive sector- with boundaries left and right. The lower bottom limit of the area is the key defensive position you were assigned to defend while the upper limit is related with the range of the weapons you have towards the enemy.

Similar things we have with the delay.

Left-right boundaries and initial position final position.

So, in all types of missions the battallion assigns a certain area of responsibility for the subordninate companies.

This also reflects the intention of the battalion plan.

For example ,there are narrow areas of resposibility along a main axis of attack and wider along a secondary one.

How the company will maneuver its sub-elements ,platoons inside this designated area, is mainly responsibility of the company commander.

However ,the company commander can not easily violate the boundaries that shape his area of responsibity,without risking putting the maneuver of the whole battallion in danger.

Farthermore, violations of this kind without approval from battallion command can lead to various bad incidents.

For example, friendly fire, since the company comander is not aware-or can not keep track of the battallion fire plan in adjastent sectors, or even the fire plan of the adjustent friendly companies.

So,he might find himself in the wrong place ,the wrong time.

Farthermore, adjastent companies and their sub-units which are not informed of this violation ,can easily assume that they have contact with enemy units and lead to friendly direct fire.

Not to mention that overconcentration of troops in a relative narrow zone can lead to unacceptable losses from enemy fire,or making slower the tempo of operations since there is doubt about the nature of the targets inside the zone.

On the other hand a company can certainly be more flexible in firing outside of its assigned sector.

Although targets outside of the assigned sector are considered secondary,there are noumerous instances where these targets can become primary.

For example two companies attack side by side .

It is possible that one company can acheive fire superiority in its sector,while the company beside it can not acheive it.

In this case, it is common to see enemy assets from the sector of the adjastent pinned friendly company shift their attention and fire towards the first company which is still advancing.

In this case, the company commander can certainly decide to engage targets outside of his original sector without any approval from battallion.

In fact there are cases where even before an attack there are certain units inside a company having the mission to provide coordination with the adjastent formations.

Many times this is coordination by fire, which means that units are focused on enemy targets outside of the boundaries of the company and in the near area of the adjastent formation.

However, as i said before , movement of a company or subunit outside of boundaries is a more complicated operation which needs more detailed coordination.

Now ,why i say all these?

First i try to explain that battallions assign missions and areas of operations to companies.

Second, movement inside these areas is responsibility and decision of the local comapny commander

Third, the company has flexibility in firing outside of its area of responsibility

Fourth, the company does not have the flexibility to maneuver outside of its area of responsibility without seeking a coordination and approval by higher and adjastent commands-which means DELAY IN EXECUTION OF SUCH MANEUVER.

With all these in mind ,i find that the proposition by some members regarding command zones are towards the right direction.

From what i understand ,the idea is that a player as a battalion commander during the pregame turn ,assigns areas of responsibility to subordinate commands -companies.

While the game is developing, any movement of a company or a platton of it, inside this zone should be treated as a"company commander decision".

Any movement which ends outside of that zone should be treated as a battalion decision and a certain delay should apply in order to "shape" a new zone which will include the end point of the new movement and will simulate the new mission-maneuver of the company received by the battallion command.

it might be also possible to have an option for a player to execute a new movement outside this zone without delay.In this case a certain type of risk must apply.

It would be nice if this risk included possible friendly fire-although i do not know if this is too advanced for the new engine to handle. It might include some other type of handicap,like instant reduction of certain command characteristics of the leader who violates the boundary (simulating less effective leadership due to the execution of a new mission without adequate preparation).

It is possible in this way that Germans who have better troops will be more eager to perform this type of operation which involves more inititative from small leaders and less directions from higher echelons.

In any case, companies should be able to initiate fire either in or outside their zones without any command delay.

I understand that this concept of zones might need some trimming or careful examination.

.

Some other consideration have to do with more details.

For example, it seems that these command zones can not overlap.

On the other hand there is the issue of reserves which must have the ability to pass through the front units.

So some tweaking and calibration of these rules regarding the overlaping of zones of command must be considered.

For example a unit which does not have an area of resposibility or command zone -(in other words reserves),should be able to move inside command zones of other formations.

On the other hand ,we can not allow the hole of having a player designate all formations as reserves with the intention to have the flexibility to mix them very easily

Anyway, i post my commments as food for thought and comments.

I would be interested to know if this concept of command zones is workable or if it can fit isnide the new game engine.

It is possible that the foundations of the new engine have already developed to such a level that it is not possible at this stage to alter the whole philosophy of the program in order to include command zones.

[ February 04, 2005, 01:27 PM: Message edited by: pamak1970 ]

February 3, 2005

I think another issue that SL handled better was the grazing effect of machine guns which rewarded a fire plan designed to have MGs acheive flanking fire and crossfire in front of a defensive position.

I had an interesting discussion regarding this issue and possible ways to introduce it in the new engine.

See

here

You can also read additional information regarding the importance of this concept in real life in the followinglink

See especially the section about MG tactical employment on page 14 of the pdf document.

January 28, 2005

Originally posted by tar:
Well, if squads presented a volume as a target rather than a single point, this would be easier to deal with, since one would only need to know to what extent the LOF passed through the squad volume. This would deal with the problem of the LOF passing overhead. It would need to pass through the target volume.

The fly in the ointment is that LOS is much easier to compute than LOF, since we are dealing with guns and not lasers. The LOS is a staight line and a lot of work has been done on things like ray-tracing algorithms. That means relatively fast techniques for computing that exist. LOF on the other hand has an arc due to gravity. Of course, balancing that is the fact that you might only need to do LOF computations when you actually shoot, rather than the much more numerous LOS tests to see if you can actually see anything.

Unfortunately, one would probably need to have the computer do more LOF computations as part of the target selection routines, since picking a line of fire that crosses many targets is possibly a better choice than one that hits a single, more exposed target. This would have to be one of the considerations that goes into the TacAI when it is choosing where to shoot.

I am sorry tar but i did not understand some things you say in your post.

For example you say that LOF is different from LOS since we deal with guns and gravity.

This is true regarding real life but i thought that the current engine does not calculate things like arcs or gravity or trajectory in the game.

Are you sure about that?

For example when i see that on map artillery can not fire indirectly ,i doubt that the engine treats LOF differently than LOS when it calculates machine gun fire.

Mortars is another example where the engine is using again a very simple model requiring LOS from a HQ before initiating indirect fire .

I do understand your point that if we want TAcAI to exploit grazing fire , then it will be nessesary to let it check various directions of LOF.

I do not know if current AI technology can accomplish a relative fast and proper selection of targets for effective MG grazing fire or if it is nessesary to let this on the player.

January 28, 2005

Originally posted by tar:
Well, if squads presented a volume as a target rather than a single point, this would be easier to deal with, since one would only need to know to what extent the LOF passed through the squad volume. This would deal with the problem of the LOF passing overhead. It would need to pass through the target volume.

The fly in the ointment is that LOS is much easier to compute than LOF, since we are dealing with guns and not lasers. The LOS is a staight line and a lot of work has been done on things like ray-tracing algorithms. That means relatively fast techniques for computing that exist. LOF on the other hand has an arc due to gravity. Of course, balancing that is the fact that you might only need to do LOF computations when you actually shoot, rather than the much more numerous LOS tests to see if you can actually see anything.

Unfortunately, one would probably need to have the computer do more LOF computations as part of the target selection routines, since picking a line of fire that crosses many targets is possibly a better choice than one that hits a single, more exposed target. This would have to be one of the considerations that goes into the TacAI when it is choosing where to shoot.

I am sorry tar but i did not understand some things you say in your post.

For example you say that LOF is different from LOS since we deal with guns and gravity.

This is true regarding real life but i thought that the current engine does not calculate things like arcs or gravity or trajectory in the game.

Are you sure about that?

For example when i see that on map artillery can not fire indirectly ,i doubt that the engine treats LOF differently than LOS when it calculates machine gun fire.

Mortars is another example where the engine is using again a very simple model requiring LOS from a HQ before initiating indirect fire .

I do understand your point that if we want TAcAI to exploit grazing fire , then it will be nessesary to let it check various directions of LOF.

I do not know if current AI technology can accomplish a relative fast and proper selection of targets for effective MG grazing fire or if it is nessesary to let this on the player.

January 27, 2005

Originally posted by YankeeDog:
That's a good start, simply constructing an exception for buildings is a bit too limiting -- you need to expand it to also include natural elevation differences created by crests and depressions.

The rule you present also artificially removes the grazing effects at times when it shouldn't -- for example, when the MG is in a building, but the building is in a depression, so most of the potential targets are actually at about the same elevation.

As a a very simple you might do something like this:

Let E(MG) = the elevation, in meters of the MG, including any effects of being mounted on an AFV, or on the 2nd upper floor of a building

Let E(T1) = elevation of MG's primary target

Let D(T1) = distance MG to primary target.

Let L = Flag attached to MG unit for your "LOF Rule". A value of L = 1 = LOF Rule ON. A value of L = 0 MG firepower potentially applies to other units along the MG's LOF.

Using the above variables, which I think the game already tracks and records, it should be possible to code some relatively simple exceptions to you "LOF Rule". For example:

LET L = 1

IF [E(MG) - E(T1)]/D(T1) > .05 THEN L= 0

Basically, this logical routine would assume that the "LOF Rule" is usually ON, but would turn it to OFF if the gradient of the line of fire was 5% or greater, representing the fact that fire with such a steep incoming angle is going to have a limited beaten zone. There would presumably still be a more limited "area of effect" for secondary targets close to the primary target, just as there is in CM now.

But I'm a hack amateur programmer who never got beyond Visual Basic. Someone with more programming and mathematics experience than myself could probably create more elegant exceptions subroutines. For example, you could probably use the difference in the gradients between the target and any potential secondary targets to also create useful exceptions rules.

Since such calculations would only have to be done for potential enemy targets along the LOF, and they don't reqire a complete calculation of the MG fire's parabolic trajectory or anything like that, the processor load shouldn't be too intense.

If would, however, take a lot of testing to make sure you got all the variables right, and that you didn't end up with wierd interactions with other other routines. That's what Beta testers are for, though. . .

Cheers,

YD

Agreed 100%.

Just adding that D(T1) should also be less than a certain value (depending on how big is the grazing range of a MG).

Meaning that if distance is too far, the MG can not aim directly the target like in the triangle you envision .

It has to fire the burst at a certain angle,which means that although it is possible to have E(MG) - E(T1)]/D(T1)<.05 ,the angle of fall of the bullets when they reach the target, will still be higher than .05 by nessesity and should not produce grazing effect.

I tried to avoid this type of calculation cause i do not know if this is too much for the game engine.

On the other hand i am thinking that since the engine does decide LOS ,it has somehow to track the elevation between a MG and a potential target.

It is just that i do not have any knowledge regarding programming to be positive about that.

January 27, 2005

Originally posted by YankeeDog:
That's a good start, simply constructing an exception for buildings is a bit too limiting -- you need to expand it to also include natural elevation differences created by crests and depressions.

The rule you present also artificially removes the grazing effects at times when it shouldn't -- for example, when the MG is in a building, but the building is in a depression, so most of the potential targets are actually at about the same elevation.

As a a very simple you might do something like this:

Let E(MG) = the elevation, in meters of the MG, including any effects of being mounted on an AFV, or on the 2nd upper floor of a building

Let E(T1) = elevation of MG's primary target

Let D(T1) = distance MG to primary target.

Let L = Flag attached to MG unit for your "LOF Rule". A value of L = 1 = LOF Rule ON. A value of L = 0 MG firepower potentially applies to other units along the MG's LOF.

Using the above variables, which I think the game already tracks and records, it should be possible to code some relatively simple exceptions to you "LOF Rule". For example:

LET L = 1

IF [E(MG) - E(T1)]/D(T1) > .05 THEN L= 0

Basically, this logical routine would assume that the "LOF Rule" is usually ON, but would turn it to OFF if the gradient of the line of fire was 5% or greater, representing the fact that fire with such a steep incoming angle is going to have a limited beaten zone. There would presumably still be a more limited "area of effect" for secondary targets close to the primary target, just as there is in CM now.

But I'm a hack amateur programmer who never got beyond Visual Basic. Someone with more programming and mathematics experience than myself could probably create more elegant exceptions subroutines. For example, you could probably use the difference in the gradients between the target and any potential secondary targets to also create useful exceptions rules.

Since such calculations would only have to be done for potential enemy targets along the LOF, and they don't reqire a complete calculation of the MG fire's parabolic trajectory or anything like that, the processor load shouldn't be too intense.

If would, however, take a lot of testing to make sure you got all the variables right, and that you didn't end up with wierd interactions with other other routines. That's what Beta testers are for, though. . .

Cheers,

YD

Agreed 100%.

Just adding that D(T1) should also be less than a certain value (depending on how big is the grazing range of a MG).

Meaning that if distance is too far, the MG can not aim directly the target like in the triangle you envision .

It has to fire the burst at a certain angle,which means that although it is possible to have E(MG) - E(T1)]/D(T1)<.05 ,the angle of fall of the bullets when they reach the target, will still be higher than .05 by nessesity and should not produce grazing effect.

I tried to avoid this type of calculation cause i do not know if this is too much for the game engine.

On the other hand i am thinking that since the engine does decide LOS ,it has somehow to track the elevation between a MG and a potential target.

It is just that i do not have any knowledge regarding programming to be positive about that.

January 27, 2005

Originally posted by YankeeDog:
Actually, for me this kind of situation shows up all the time -- sometimes in urban combat, but much more often in semi-built up terrain like small towns where there are some buildings around, and also in rougher, more varied terrain with lots of crests and depressions.

On average, I would say at least once a game.

For example, when defending a town, I often start the game with HMG and/or LMG teams in the upper floors of large buildings to disrupt enemy infantry formations as they approach the town. Placing the MGs on high floors gives them a wide field of fire, and the MGs will often have multiple infantry targets in their observation zone -- I've had a single MGs taking entire companies under fire at times. Of course, the result is that my opponent usually brings enemy armor forward blow up the buildings my MGs are in, but that's part of my strategy -- force my opponent to show his hand by revealing his expensive armored units for the possible cost of a few, relatively inexpensive MGs.

Pretty much all outgoing fire from such a high-elevation placement is plunging fire, and therefore would be unlikely to affect "more than one enemy unit in LOS," even if said units are along the same LOF, unless the units were quite close togther.

Also, MGs on AFVs -- remember, the co-axial on most tanks is already 1.5m or so above the the US M3A1 Halftrack, or the M7 Priest, or the flexible MG on some tanks, the MG is considerably higher than that. Put a tank on a small 4-5m high rise (only 2 elevation levels in most CM games, a very common placement for an overwatching tank), and suddenly almost all MG fire from that tank is plunging fire.

So, at least in the games I play, situations and deployments that create plunging-type fire with a small beaten zone are not uncommon in CM. In such situations, MG fire affecting more than one unit along one LOF is unlikely, unless said units are very close together.

Cheers,

YD

Probable solution.Just an idea

If either a MG or a target is located in any kind of upper building, there will be not grazing effect.

I think this type of decision is easy for the engine to handle.

In other words, if either the base of LOF or its end is located inside an upper building ,the MG will "lose" its grazing effect.

This might also lead players to use more proper real life tactics.

For example ,under these conditions it might be better to have MG on the ground, trading somewhat less observation with the ability to have grazing fire in front of your defence.

I say this ,cause i find it strange in real life to place every MG on the upper floors to resist an attack against the village from the outside.

I can understand it in some cases when you need to cover dead space but i can not see it as a general tactic for the deployment

I understand your point of your tactics in CM.It makes sense there to put even all your MGs in upper floors to have a better observation .

The way the system is now, leads players to use the tactic you describe without a lot of hesitation (at least regarding the effectiveness of the fire plan itself).

In real life ,it is not easy for a commander to decide to form a fire plan which does not produce grazing fire in front of his position.

[ January 27, 2005, 01:39 PM: Message edited by: pamak1970 ]

January 27, 2005