Week 11
- michael71510
- Aug 28
- 2 min read
This week I managed to get the helicopter to spawn in and fly to the specific point where it starts attacking the player (Figure 16). Getting this to work was a huge success for me. Having the helicopter move into position is far more effective than the fixed location I had last week, because it makes the attacks feel more threatening and dynamic. This puts a pressure on the player which helps to better sell the sensation of being hunted down.

Figure 16: Screenshot showing the helicopter as its flies in to take its attack position
I purposefully decided to not give the player powerful enough weapons to destroy the helicopter, to make it feel like something that they cannot defeat. This helps to create my intended user experience that the player is almost outmatched in this ambush, and that their only options are to be resourceful and escape.

Figure 17: Screenshot showing the updated elevation in the sidewalk
I have also continued to iterate on the "lack of depth" comments from my playtesting session last Tuesday. One way I have added more elevation is by raising the sidewalks more than the road (See Figure 17). While certainly small change, I think it is important in making the level feel more believable. It is also something which I have seen in the actual HL2 levels. While I am focusing on designing pacing, scripted events and combat encounters, having this sense of believability in the environment, will help me to keep the players immersed in the user experience, rather than on missing details or inconsistencies in the environment.
Another significant step forward this week was fixing my inability to compile my level using the default configuration. I initially found that an application called VVIS, Hammers compiling application (VVIS, 2025), was unable to compile my level. After carrying out extensive research on the Value Developer Community and Reddit forums, I discovered that the issue was caused by me having a high Numportal count. Numportals are effectively the number of visible surfaces in the level (VVIS, 2025). This forced me to change many of the objects in my scene to func_details which lowered the amount of numportals (Func_detail, 2025). The high amount of numportals was particularly caused by the size of my skybox.
This experience has taught me the value in using func_details to optimise things from the start of a project, and particularly not making the skybox bigger than it needs to be. Going forward, this will save me substantial amounts of development time, as I won’t need to spend hours troubleshooting.
Comments