Timeline for Kart Racing Attack AI
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Jun 10, 2015 at 9:39 | history | suggested | Alex | CC BY-SA 3.0 | Spelling and grammar |
| Jun 10, 2015 at 9:20 | review | Suggested edits | |||
| S Jun 10, 2015 at 9:39 | |||||
| Jun 9, 2015 at 18:10 | vote | accept | idurvesh | ||
| Jun 9, 2015 at 18:08 | comment | added | idurvesh | @Byte56 actually race tracks will have turns and multiple short and long cuts side by side, so increasing size of Physics.Overlap may go over other short tracks or long tracks n result of that it will provide wrong data, so was heisted with Overlap approach and changed. | |
| Jun 9, 2015 at 18:07 | comment | added | idurvesh | @Superdoggy yes you are right,I realized it after you guys pointed out.thanks for input. | |
| Jun 9, 2015 at 17:10 | answer | added | DMGregory♦ | timeline score: 2 | |
| S Jun 9, 2015 at 14:33 | history | suggested | newton1212 | CC BY-SA 3.0 | Fixed grammar, improved format. |
| Jun 9, 2015 at 14:09 | review | Suggested edits | |||
| S Jun 9, 2015 at 14:33 | |||||
| Jun 8, 2015 at 23:35 | comment | added | House | Sure seems like you're progressing backwards. Moving a collider will be less efficient and more buggy than Physics.Overlap or a raycast. Considering raycasts can be infinite, maybe you can explain more how the range is limited? | |
| Jun 8, 2015 at 23:32 | comment | added | Superdoggy | Distance calculation is basically sqrt((x1 - x2)^2 + (y1 - y2)^2). To a computer calculating that is practically nothing, especially compared to 2D (or 3D) collision detection. | |
| Jun 8, 2015 at 22:48 | comment | added | idurvesh | thats very good points....sounds reasonable too,the only reason I was avoiding Distance calculation as I read its too slow.....coming to your points ,when user clicks attack button, let random value get pass to enemy after which enemy will perform its attck so it will feel natural too? m I on right path?.....Feel free to expand this into an answer with lil more spotlights...thanks | |
| Jun 8, 2015 at 22:38 | comment | added | DMGregory♦ | It's only 8x8 if everyone is attacking in the same frame. If you only do the check when a character tries to attack, and stagger AI attack attempts, the typical hit will be far less. Also consider that each character doing a large-radius sphere check may be more expensive than checking squared distance to a small number of other characters. If you have a system that tracks race position, you can even limit the checks to the first 2-3 racers ahead of the character. If that sounds reasonable I can expand this into an answer. | |
| Jun 8, 2015 at 22:32 | comment | added | idurvesh | It depends on level type from 4-12.You have point but isnt it will add over head ,suppose level has 8 characters,now each character, (enemy+player) iterating over each other so that would be 8*8 = 64 iterations per frame... | |
| Jun 8, 2015 at 22:27 | comment | added | DMGregory♦ | How many characters do you expect to have? If it's on the order of dozens or less, you could probably just iterate over every possible target character and check if they're in range, without feeling any performance hit, and without distractions from other physics objects in the scene. | |
| Jun 8, 2015 at 22:20 | history | asked | idurvesh | CC BY-SA 3.0 |