Results 1 to 2 of 2

Thread: AI Philosophy

  1. #1

    AI Philosophy

    First of all, at some tracks, I find the AI in NTG11 pretty good. However at some tracks you can indeed pass about 20 cars in a single straight, without any car setup improvements. And at some tracks the AI speed varies very little from the guy in 15th to the guy in 1st.

    Now for the point of this thread. I am trying to figure out why AI (specifically AI speed, not talking about behavior) has been pretty slow on console games in general.

    Is it because they might be thinking that in a typical default 6 lap race (or so), a "very hard" setting means a really good driver should be able to go from 43 to 1st in 6 laps (or 15th to 1st in games with fewer cars)? And on an "easy" setting an average driver should be able to go from 43 to 1st in 6 laps. This seems like it could be the result of a philosophy or a calculation trying to condense how a good driver would progress.. 50 real laps = 3 laps in game, so a good driver can pass 25 cars in 50 laps in real life, so make a good driver pass 25 cars in 3 laps in the game? Are they trying to avoid the scenario where a really good driver, on very hard that doesn't qualify, only gains about 10-15 positions in 6 laps, out of the fear the consumer won't like it, becuase they only finished 30th after starting 43rd?...

    Not based on any real data, but years of watching nascar in my opinion:

    My guess for a real world setting:
    Really really good driver, nailed the setup, experienced at the track, moving through the field in cup:
    (not counting talladega/daytona).
    Pit crew could really impact this, so equalize pit crew across all drivers.
    Lap 0-25 ... 43 to 28.
    Lap 25-50 ... 28 to 18.
    Lap 50-75... 18 to 10.
    Lap 75-100... 10 to 4.

    What I think would be good for a game vs AI (max attainable gain by good driver):
    (not counting talladega/daytona).
    No qual, 10 lap race:
    Lap 1-3 ... 43 to 28
    Lap 4-6 ... 28 to 20
    Lap 7-10 ... 20 to 15.

    A good driver, no matter how great their setup is, should not be able to get into the top 15 on the very hardest setting in 10 laps... (without qualifying, meaning started last).

    What do you guys think for a 10 lap race in a game?
    (not counting talladega/daytona).
    Max positions gained by a really good driver, on the very hardest setting?
    Lap 1-3 ...
    Lap 4-6 ...
    Lap 7-10 ...
    Last edited by godavidR; 06-01-2012 at 06:52 AM.

  2. #2
    i see your point but you also have to account that the tail end of the field is generally the go or go homers and field fillers who dont have the greatest equipment...the top 30-35 is where it could be fairly decent teams that just missed it on qualifying but have decent cars on raceday...then you have the guys who finish well but qualify poorly inside the top 30 to low 20s and from there its pretty much the hot lappers on qualifying day like Ryan Newman back in his Penske days couldnt finish a race but could land on the pole almost every week...so after figuring the abilities of the drivers and teams that a good driver would pass in a ten lap race from dead last i would say should very depending on how the ai cars are spread out...if at the back of the pack you have Johnson, Gordon, Edwards, Harvick, Biffle, Stewart,...etc. than maybe not as many as you predicted....also depends on type of track Martinsville i say your lucky to make top 30 if the game is realistic nowhere to go and wrecking ai will bring out cautions so well that could be for later
    so here are my thoughts
    Lap1-3....at most top 35 (picking up 7 spots)
    Lap4-6....at most top 30 (picking up 13 spots)
    Lap7-10...at most inside top 25 (picking up 18 spots)

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •