Jump to content

Infantry spotting


Recommended Posts

Not according to Steve, he very firmly told me that the 1-1 is actually abstracted and that the infantry arent necesarily where I see them on screen. With infantry Im led to believe therefore that its not true 1-1.

Do you have any source for this claim?

We know that HE effects are toned down to compensate for tight troop placement, and that building interiors are abstracted, but everything else is certainly 1:1 (including spotting).

Best regards,

Thomm

Link to comment
Share on other sites

This is what Steve said earlier:

Each friendly soldier (or vehicle) within seeks to establish LOS with each enemy soldier (or vehicle) that is within the Action Spots it has theoretical LOS/LOF to. This is where Spotting, an entirely separate concept from LOS/LOF, comes into play. The system decides, based on the exact and specific conditions (including unit based stuff like terrain, Experience, Morale, equipment, speed, etc.) if a soldier (or vehicle) can see some or all of the enemy in a particular "authorized" enemy occupied Action Spot.

So from that LOS/LOF is 1:1 and Spotting is a 'more coarse' filter.

Link to comment
Share on other sites

I think it's the other way around, Ron.

Per Steve, LOS is "coarser" and based on Action Spots, rather than individual soldiers. Once the Action Spot LOS check determines whether there is a chance of spotting happening *at all,* Spotting checks are then done 1:1, individually for each and every soldier.

AIUI, the purpose for this is that the Action Spot LOS checks work as a filter, eliminating a large number of situations where spotting is definitely not possible, so that the engine doesn't need to run as many spotting checks.

IME, as the game is now in 1.21, this system seems to work very well and only very rarely do a see a spotting (or non-spotting) situation that seems out of joint with reality.

In the initial release, before the "Enhanced LOS" was added, there were definitely problems.

Link to comment
Share on other sites

There is only one remaining spotting/firing situation remaining, outlined here: http://www.battlefront.com/community/showthread.php?t=83773&page=2

My abject apologies if this has been changed. I don't believe it has been.

The situation can exist when the action spots are out of LOS of each other, but a unit may not be entirely located at the center of the action spot. (Infantry in a building,or a vehicle poking around a corner.) In that case, there is no LOF until the unit in question gains LOS. From the thread I linked to, regarding an enemy soldier looking out a window from the corner of the building, at the moment he pops up, the two way nature of LOS/LOF exists. The two way nature ends once he drops down. The game result is quite limited; since he cannot see out (he is below the window), the enemy unit has no LOS/LOF. However, even though you KNOW he is there (by the corner of the building under the window) you also don't have LOS/LOF. Physically, you can see that portion of the building; you may want to area fire on that part of the building, but cannot.

To emphasize, this is limited: you will gain LOS/LOF to the enemy soldier once he pops up in the window; you are blocked from gaining LOF to that spot while the soldier is out of LOS.

This is a rare occurance. The same situation occurs with partial vehicle spotting (poking around a corner).

To summarize; LOS exists from you to the building corner (fringe of an action spot). LOF does NOT exist until an enemy unit appears in the LOS.

Ken

Link to comment
Share on other sites

Yankee Dog,

AIUI, the purpose for this is that the Action Spot LOS checks work as a filter, eliminating a large number of situations where spotting is definitely not possible, so that the engine doesn't need to run as many spotting checks.

Yup. A good way to think of this is to picture a piece of blank paper and a piece of graph paper. Pick a random point on each piece of paper and put an X, then randomly put an O somewhere else. On the graph paper, trace a random line from the X and trace square by square to see if it intersects a square with the O in it. After you check a square write down the result. Now take the blank piece of blank paper and do the same thing, but this time check every millimeter to see if it intersects any physical portion of the O you drew. Note the result for every mm checked. Now, repeat this process with the graph paper until a line is drawn that intersects each square. Then go to the blank paper and draw lines until every single portion of the paper is covered by pencil/ink.

This simulates, crudely, a single LOS check from a single soldier for a single instance of checking. If you have 500 soldiers under your command, a single minutes' worth of combat would require this process to be repeated roughly 6,000 times if each check is 5 seconds apart. Then multiply this by 5 to check each of the possible heights within each Action Spot (prone, kneeling, standing/vehicle height 1, vehicle height 2, vehicle height 3). That's now up to 30,000 times each 5 seconds. Then multiply this by the size of the map and the terrain density... yup, it just gets better and better :)

This is a crude analogy, but it does give you an idea why we will always have something like Action Spots. They may get smaller, as they did between CMx1 and CMx2 (20x20m reduced to 8x8m), however I don't see computers ever being able to handle this sort of massive checking. Not in my lifetime anyway.

IME, as the game is now in 1.21, this system seems to work very well and only very rarely do a see a spotting (or non-spotting) situation that seems out of joint with reality.

That's definitely the way we see it.

In the initial release, before the "Enhanced LOS" was added, there were definitely problems.

Absolutely. While CMx1 (and almost all other games) work fine with a single fixed height for doing LOS/LOF checks, with CMx2's more complex terrain environment there was a mis-match in the fidelity of the two systems.

Steve

Link to comment
Share on other sites

I think often the consusion here arises from the fact that weve gone 1-1 visually but as stated before the program doesnt really see the infantry as 1-1. I know that Steve has explained all this before and I fully accept the reasoning why it is the case. But it still leads to confusion as my brain wants to shout out that one of my soldiers can see the fecker in front of him, while my CPU is saying no.

Not quite correct. LOS is not 1:1, that's true. However, LOF is 1:1. That means the exact location of a soldier does matter in terms of shooting and being shot. There is some abstractions for cover and concealment because we can not possibly simulate, graphically, the true chaotic and varied nature of cover as found in the real world.

Spotting is completely separate from LOS and LOF in terms of how it is calculated, but it does use the LOF's 1:1 routines. Otherwise if you saw one guy in one Action Spot then you'd see all his buddies instantly at the same time all the time. Obviously that's not how the game works. To restate:

A LOS check is made between Action Spot to Action Spot. This is the primary filter. If you can not draw LOS from Action Spot to Action Spot then no other calculations are done. Period. This is where we save millions of calculations over the course of the average game.

If a LOS check passes then there is a Spotting check, which is 1:1 with the aforementioned abstractions for cover and concealment. It is also limited to specified heights (Enhanced LOS) and not based on where polygonal eyeballs actually are. If an enemy guy is lying prone behind a wall, and your spotting guy is not in a position to look over that wall, Spotting will likely fail because you can't actually "see" the enemy despite being able to see that Action Spot. Although not true 1:1, in that we don't initiate the starting point where a set of polygonal eyeballs actually is, it is pretty damned close and is generally inline with the graphical fidelity of the terrain.

Now, if something is Spotted then it's fair game for a LOF check. This is true 1:1 in that it goes from the polygonal location of the weapon to the polygonal location of the target.

If a shot is fired then the shot is traced 1:1, polygonally, and it intersects whatever polygons it intersects. What happens based on those intersections is based on whatever is being shot and whatever is being intersected.

Damage assessment is based on the intersections, but it can be fudged up/down based on very specific circumstances. For example, as stated before cover has a degree of abstraction based on the type of terrain the target is in. So there is a chance that a direct polygonal hit will result in little to no damage. The primary example of this is close in HE fragmentation hits, which is dumbed down a bit depending on range to simulate slightly better real world spacing than what CM is capable of simulating.

So there you have it... most everything important is 1:1 already.

One day we may have true 1-1 with each individual soldier being tracked as you see it and no need for the action spots!!!!

Not going to happen in my lifetime of making games, so I wouldn't hold your breath :D

The situation can exist when the action spots are out of LOS of each other, but a unit may not be entirely located at the center of the action spot. (Infantry in a building,or a vehicle poking around a corner.)

These are two different situations. The vehicle one is the easiest to visualize.

An individual entity (soldier, vehicle, heavy weapon, etc) can only ever be in one Action Spot at one time. NOTE THAT I DID NOT SAY UNIT!! A Squad of 12 men can, in theory, be in 12 Action Spots at the same time. With most things in most situations this doesn't matter. If a .50cal MG barrel is poking into another Action Spot, graphically, it really doesn't matter in game terms 99.9% of the time. Or a soldier model's elbow, for that matter.

But with large, horizontally orientated entities, this can produce some odd results where something like a tank is in one Action Spot 50.00000001% and 49.00000009% in another. So it could be that the vehicle really looks like it is around a corner when, according to the game, it's still behind a building. The 49.00000009% of the vehicle that is visually, but not actually, in that second Action Spot is as if it isn't there as far as the game is concerned. Nothing can spot it, nothing can shoot at it. Likewise, the vehicle can't spot it or shoot as if it were in that Action Spot either. Therefore, from a game standpoint there's a visual inconsistency but no game inconsistency.

With infantry in buildings things are a bit more complicated. Buildings, unfortunately, break a lot of the Action Spot rules because they often do not conform to the overall Action Spot grid. That's how we can have buildings which are diagonally orientated, slightly overlap Action Spots, not quite fill Action Spots, etc.

The system compensates for these exceptions as best as it can, but sometimes the visuals don't really lineup well. However, the game mechanics work the same for both players. When a unit is in a building it has the same LOS/LOF checks from Action Spot to Action Spot that a unit outside of the building has. It's 100% reciprocal two way.

Where things get a little dodgy is that Area Fire is trickier to handle going into a building than going out from it. That's because the nature of the terrain, not that there is something different going on under the hood. When two units are in open terrain, and LOS is temporarily blocked due to going prone, each can target the other to the same effective degree because there is nothing blocking your LOF to the center of each other's Action Spots. But as terrain becomes more complex things can get a little more difficult in terms of game results.

Specifically, when you go to area fire at a building there is a chance, depending on orientation, that you can't target the center of that building, which is necessary for Area Fire to be assured a hit on the specific portion of the building you're aiming for. In open terrain targeting an adjacent Action Spot with Area Fire usually has just about the same results as targeting the Action Spot you're interested in aiming for. So while there is 100% parity from a LOS/LOF point of view, it can be more difficult to establish effective Area Fire one way than another or both (as happens sometimes).

There's no easy fix for this other than making buildings consist of much smaller mini-Action Spots. Eventually I think we will be able to do this.

Steve

Link to comment
Share on other sites

Elmar,

could you clarify spotting from vehicles? Who is doing the spotting, from where and with what? I thought spotting was done from the individual vehicle stations with the appropriate equipment for that station.

At present there is only ONE check from each vehicle in terms of LOS. This is based on center mass and which of the Enhanced LOS heights makes sense to use. Spotting, on the other hand, is influenced by how many "eyes" the vehicle has, the type of equipment is available, and a host of other factors.

In order to have each "station" within a vehicle spot on its own we would have to draw specific LOS from AND TO each individual portion of that vehicle. That in turn would mean vehicles would be able to be in more than one Action Spot at a time, which is presently not the case. This would complicate how vehicles work by a decent amount. However, it is definitely an area we will expand into sooner rather than later because the horsepower to do this isn't such a big deal. It's the coding and testing that is the bugger here (as stated a few posts ago, horsepower isn't the only practical limitation).

Steve

Link to comment
Share on other sites

Steve,

Thanks for your very informative post above. Still being a new player but having an old brain, I wonder if I could ask for a further bit of clarification.

I know my enemy, a one or two man team, is at the right-hand corner of a substantial (i.e. more than one spot building). When I try and fire my MG at the spot where the unit was last seen I get a clear, bright blue line. However when I click on the fire button the target line snaps to the middle of the building. The crucial question is will my rounds have any chance of hitting the enemy unit?

Link to comment
Share on other sites

This is indeed the problem with Area Fire against "oblique angled" buildings. The chances of hitting are directly proportional to how close your target line passes near where the guys actually are. With Area Fire the outgoing fire is spread out specifically to compensate for this sort of situation. So if the line is really close there will be no effective difference. The further away the less optimal. Of course if you are using a tank or MGS it might not matter at all ;)

Steve

Link to comment
Share on other sites

This is indeed the problem with Area Fire against "oblique angled" buildings. The chances of hitting are directly proportional to how close your target line passes near where the guys actually are. With Area Fire the outgoing fire is spread out specifically to compensate for this sort of situation. So if the line is really close there will be no effective difference. The further away the less optimal. Of course if you are using a tank or MGS it might not matter at all ;)

Steve

Thank you, I think I understand now some of the results I have had. No big deal from my point of view but it is nice to know.

Link to comment
Share on other sites

Not quite correct. LOS is not 1:1, that's true. However, LOF is 1:1. That means the exact location of a soldier does matter in terms of shooting and being shot. There is some abstractions for cover and concealment because we can not possibly simulate, graphically, the true chaotic and varied nature of cover as found in the real world.

Spotting is completely separate from LOS and LOF in terms of how it is calculated, but it does use the LOF's 1:1 routines. Otherwise if you saw one guy in one Action Spot then you'd see all his buddies instantly at the same time all the time. Obviously that's not how the game works. To restate:

A LOS check is made between Action Spot to Action Spot. This is the primary filter. If you can not draw LOS from Action Spot to Action Spot then no other calculations are done. Period. This is where we save millions of calculations over the course of the average game.

You see, for me, this is the biggest, probably the most important failing of CMSF.

Infantry doesnt get a fair crack at the battlefield simulation. Infantry is just too tightly packed into a small area. This leads to several things in the game for me:

1. An an infantry squad is far too bunched up on most occasions. This leads to incoming fire causing far too many casualties when compared to real life.

2. In real life, to stop this happening infantry spread out and use various formations, which means that if they come under concentrated fire from a certain direction, not all of them become casualties.

3. Again in real life, infantry squads have Standard Operating Procedures (SOPs) for when they come into contact with the enemy. For instance, on contacting the enemy to the front soldiers should automatically adopt a procedure for this, which includes putting an immediate fire upon the enemy, seeking cover, even withdrawing if necessary.

My whole point I suppose is fundamentally this;

"If the infantry in the game are exactly where I see them, then the infantry model is flawed, because they are too densely packed into too small an area".

Of course this doesnt mean I cant enjoy the game, because I do, but it means I play infantry very differently from real life experience. I tend to always split the squads down and move them farther apart for example.

Link to comment
Share on other sites

Gee, if that's the biggest failing then the game's doing pretty darned good. CMx1 had 12 men packed into a single point! :D

Gee, Im not comparing it to CMBO mate..... Im looking at it for what it is. CMBO wasnt 1-1.

I wonder why it is that when anyone says anything about an aspect of Sf that they dont like someone comes along and compares it to CMBB etc?

Also, I think you may be slightly wrong there too, I dont think CMBO packed 12 men standing on their heads all together in some sort of circus act. As I understood it, that single point was abstractly representing a 12 man squad spread out over a wide area.

Maybe Im wrong though? I often am.

Link to comment
Share on other sites

The nerf relating to explosions has been known pretty much from day one, I believe. Less well known is that a bullet crossing paths with a soldier will not necessarily result in an injury/fatality.

So:

1. How is this calculated, what is my chance in a given situation of being hit, and

2. If this is written into the game mechanics, then why isnt it explained so that I can take it into account when moving infantry around and fighting with them?

Id like to know my percentage chance of having my men killed or injured in a given situation, how do the die roll?

Link to comment
Share on other sites

I think the small arms in the game are less accurate in real life to reflect abstracted cover as well as the fact that the men are more bunched up.

I am pretty sure the only reason why you wouldn't get an injury when the bullet intersects a soldier is when they are wearing body armour. When they arn't you get a lot more people going down. In fact, I have recently even seen one 5.45 round take down 2 people! (no armour)

Just my 2 cents worth though, I might be wrong.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...