With regard to spawning, the biggest problem forgers are faced with is the radius of the PE Primary on the forge canvases and the limitations it has on map sizes. A map can only be designed so small to a point past which it becomes a mathematical impossibility to ensure safe spawning. Carefully planned geometry can play a role in reducing this size barrier further, and a design pattern is offered to help forge smaller maps successfully.
Safe spawning is defined as spawning a significant distance from an enemy, or spawning on the other side of a wall (or other LOS obstacle) from an enemy. Safe Distance is a fluid term, but for our purposes we will use the prevailing PE Primary radius of 9.5 units as an indicator of what 343 Industry feels is a typical safe distance (which is the purpose of the PE Primary in the first place).
The root of the problem is the size of the PE Primary, which is to ensure that a player spawns at least that distance from every enemy player (that is, spawn outside of every PE Primary region). On all forge canvases, the radii is 9.5 units in size. Compare that with Haven’s 7.0 and Adrift’s 5.5, you begin to see that the forge canvases were designed for medium to large size maps, and that small maps will have the complication of the possibility that every Respawn Point on the map could be simultaneously covered by a PE Primary, leaving no way to determine which is the safest to use.
What this means is that if the map is small enough and the Respawn Points are in a statistical typical layout (see figure 1), then four enemy players can cover every Respawn Point with their PE Primary. Now on the surface most people don’t understand the implications of this problem. And in fact it will throw you a bit to learn that if all the weights were equally dropped by -20, then it would be like none of them were dropped at all. Why is this?
When all the Respawn Points exhibit the same weight, then it doesn’t matter what that weight is. What is important is the difference in weights. And when they are all the same, then only the tiny random variation that is added to each will be used to select the Respawn Point to respawn a player on. This will have the affect of making the spawning appear random.
As we see in figure 1, some of the Respawn Points have two PE regions covering them, making their weights twice as low at a hefty -40. That is okay, take all those out of the picture, you still have quite a number of Respawn Points at -20 and again the selection of a Respawn Point to spawn with from those sitting at -20 will appear random.
figure 1 typical spawn layout, typical enemy movement
Why is this random selection amongst all the Respawn Points important? Because unlike previous Halo titles, Halo 4′s spawn engine has been given extra importance to spawning away from enemies. This mechanism in the spawn engine requires that a Respawn Point exist far enough away from all enemy players that it will stand out with a much higher weight. In other words, the Halo 4 spawn engine is built to work properly when the enemy team cannot possibly cover most of the Respawn Points on the map at one time.
Going back to figure 1, the problem is that none of the Respawn Points can be located far enough away from another that at least some will always be outside a PE primary. In a small map, you will want to locate Respawn Points along the outer most edges to get the Respawn Points as far from each other as you can. But those Respawn Points must not be in the open either – the edges of the map must provide safe spawning as much as possible.
But while this can be achieved in figure 2, again the map size is simply too small to support proper spawning behavior within Halo 4. In figure 2, the playable area is about 900 square units, the PE primary is 9.5 units in radius.
figure 2 – more extreme Respawn Point layout, but it isn’t enough
Ultimately the consequences of having all the Respawn Points within the PE primary is that the Respawn Point to be selected for spawning a player by the spawn engine will guarantee to be within the PE primary of an enemy. What does that mean? The spawning player will spawn close to an enemy. Sometimes it will be say 9 units away, other times it will be the Respawn Point only 1 unit away.
The PE primary is both very wide and very strong for the purpose of preventing such a spawning experience. Yet, with small maps, it is inevitable. As players move around the map, regardless of walls or other obstacles, players will spawn on any Respawn Point when they are mostly -20 and some -40, ensuring that some spawns will be really bad. It becomes a mathematical impossibility to solve this problem without being able to change the parameters, such as the PE primary radii. (I have asked 343 Industries if they could provide forgers with a trait zone property or some other means to adjust the PE primary radii to tailor the behavior to the size of the forged map. They never responded to me, privately or publicly.)
Finally, to be complete, a word about the PE secondary. In figure 1, you will see some Respawn Points outside one PE primary and inside another. In my conversation I said that they would all be -20. But I did not mention the PE secondary. It drifts to 0 as you move further from the enemy, much like the PA does. This would have some impact on which Respawn Point is selected. But I didn’t mention it, because it is randomly offset by the PA that is moving about the map.
If your map is so small that the possibility of covering every Respawn Point with a PE Primary exists, then there may be hope for your map yet. Given the proper geometry, it may be possible to spawn players in safe areas based upon whether every Respawn Point is covered by a PE Primary or not at the moment of spawn. To do this, an assumption must be made – that when every Respawn Point is inside a PE Primary that some Respawn Points are preferred (via Respawn Zone) and are out of sight of the enemy, even if the enemy is just on the other side of a wall.
Let’s take a look at the map in figure 3 below. The map dimensions are less than 27 units along any side. This makes the map no more than 750 square units – quite small compared to Haven’s 1000 or Abandon’s 1200 square units of play area. In this first example, we see that all four enemies cover all eight spawn points with their PE Primary, which means that they are all dropped to -20 in weight. But those in the center are also covered with a Respawn Zone, so they are raised by +14, and have a total weight of -6. When you compare a Respawn Point within the center to that along the far corners, you see that those in the center are going to be selected by the spawn engine. That is, C will be chosen over A because it is +14 higher in weight.
figure 3 – All four enemies cover all eight spawn points
The Spawn Points are purposefully pushed to the most outer edges of the map for a very specific purpose. Let’s take a look at figure 4 and let’s say that one of the enemies wants to go into the center structure, near Respawn Point C. The weight at C won’t change, but the weight at B will, and B will now be chosen over A, C, or D by the spawn engine. By positioning B in the corner, we are forcing the enemy to choose between whether those in the center structure are safe spawns or not, and when not an alternative is provided by the movement of the enemy away from the corner of the map.
figure 4 – Same spawn layout, but an enemy moved into the structure and away from a corner
In essence, we are creating on a very small map a set of conditions by which either a corner spawn will be out of a PE Primary and essentially the furthest from any enemy, or the central structure will provide a safe structure to spawn under cover and preferred when all have an equally -20 weight by the presence of a Respawn Zone, even though the Respawn Points within the central structure are just one or two forge units away from an enemy.
In other words, if the enemy does not shut down the corner spawns, then the corner spawns may be considered safest (highest weight). But if all four corner spawns are weighted down to -20, then we know with certainty that there are no enemies within the center and we want to spawn in there over any spawn exterior to the center.
This strategy doesn’t require a center structure as outlined in either of these figures, but it does require that the enemy cannot see any of the Respawn Points in the center if they are covering all the corners with their PEs.
Let’s take this concept further and drill down to the essence of what we are trying to achieve. Take a look at figure 5 below. Notice that the only walls are positioned such that the only time a corner spawn cannot be used is when the players are either within the pairs of walls or on the outer edges of the walls. The inner walls thus block LOS to the center Respawn Points for safer spawning. Yet, if any enemy is in the middle are of the map, a corner Respawn Point will be used and again is out of sight of the enemy. In other words, figure 3 shows the beginning of a design pattern one can use to calculate the minimum LOS blockers for small square maps to yield spawning out of sight of an enemy.
figure 5 – Design pattern for safe spawning on a small map at its core
Halo 4 relied upon ensuring that there would always be some Respawn Point outside of the PE Primary to ensure safe spawning, which in turn required maps to be of a minimum size based upon the size of the PE Primary radii.
For maps that are too small to ensure at least one Respawn Point is outside the PE Primary, geometry can be forged to ensure that some Respawn Points are preferred by the additional weight of a Respawn Zone and guaranteed out of sight of the enemy for those moments when every Respawn Point is in a PE Primary.
Such small map solutions are far more dependent upon the geometry, and will not work for every map.