F1 2018 Review: Perhaps One of the Best Racers Ever

Collapse

Recommended Videos

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • GaloVingador
    Pro
    • Mar 2018
    • 564

    #136
    Re: F1 2018 Review: Perhaps One of the Best Racers Ever

    Originally posted by GoDucks1224
    I find the most enjoyable careers are the ones where you start as a bottom 3 team with the difficulty on 100. You should lose. A lot. But after one season I’ll usually jump to a mid pack team and work on really bringing that team toward the top with upgrades.
    You're much better than me. I'd be finishing rock bottom every race if I played on 100.

    Comment

    • GaloVingador
      Pro
      • Mar 2018
      • 564

      #137
      Re: F1 2018 Review: Perhaps One of the Best Racers Ever

      Well, I think I found a way to objectively assess this.

      Instead of gauging it through my own performance, I decided to compare CPU pole laps in clear midday short quali sessions to the Steam TT leaderboards (adding exactly 5.7% to the 199th fastest dry lap with modern cars in every track and seeing how 78 AI compared to it):

      Australia: just right
      Bahrain: 0.8 sec faster
      China: 0.2 sec slower
      Azerbaijan: 0.9 sec faster
      Brazil: just right

      So it looks like I suck indeed in Australia (managed to shave 1.9s from my PB after my first post and I'm still behind the pace), but yeah yeah, there are still some AI performance differences between tracks. Anyway, this makes it possible to set individual difficulty levels for each track to keep AI performance consistent (I'll go with 71 at Baku and Bahrain and 80 in China).

      I'll repeat this experience in all the other tracks and share the findings. Even though I'm using a noobie difficulty as baseline, I think this is useful information.
      Last edited by GaloVingador; 03-28-2019, 10:54 PM.

      Comment

      • GaloVingador
        Pro
        • Mar 2018
        • 564

        #138
        Re: F1 2018 Review: Perhaps One of the Best Racers Ever

        Codies made a good job in this game, it's hard to see a racing game with consistent AI. And it looks pretty consistent between tracks, with the exception of a handful.


        - Track where the AI is slowest (+0.9 seconds):
        Austria

        - Tracks where the AI is fastest (-0.6/-0.8 seconds):
        Bahrain
        Azerbaijan
        Russia
        Abu Dhabi


        - Tracks where the AI is slightly faster (-0.2/-0.4 seconds):
        Britain
        USA

        - Tracks where the AI is within +- 0.2 seconds from where it should be:
        Australia
        China
        Spain
        Monaco
        Canada
        France
        Hungary
        Belgium
        Italy
        Singapore
        Japan
        Mexico
        Brazil


        I'm not 100% sure about this, as people usually complain about Canada being easy and Monaco being freakin hard. Maybe Monaco is super popular in TT and more competitive because of that? Or perhaps it's just because it's so easy to crash then people drive more carefully and this makes them slower when not hotlapping, idk.

        I'd advocate setting difficulty some 5 points down in the "fastest" tracks and increasing difficulty a lot in Austria.

        But this is purely maths. I haven't even raced in most of these tracks. Yeah, I'm a nerd.

        Comment

        • GoDucks1224
          MVP
          • Jul 2012
          • 2623

          #139
          Re: F1 2018 Review: Perhaps One of the Best Racers Ever

          Originally posted by GaloVingador
          Codies made a good job in this game, it's hard to see a racing game with consistent AI. And it looks pretty consistent between tracks, with the exception of a handful.





          - Track where the AI is slowest (+0.9 seconds):

          Austria



          - Tracks where the AI is fastest (-0.6/-0.8 seconds):

          Bahrain

          Azerbaijan

          Russia

          Abu Dhabi





          - Tracks where the AI is slightly faster (-0.2/-0.4 seconds):

          Britain

          USA



          - Tracks where the AI is within +- 0.2 seconds from where it should be:

          Australia

          China

          Spain

          Monaco

          Canada

          France

          Hungary

          Belgium

          Italy

          Singapore

          Japan

          Mexico

          Brazil





          I'm not 100% sure about this, as people usually complain about Canada being easy and Monaco being freakin hard. Maybe Monaco is super popular in TT and more competitive because of that? Or perhaps it's just because it's so easy to crash then people drive more carefully and this makes them slower when not hotlapping, idk.



          I'd advocate setting difficulty some 5 points down in the "fastest" tracks and increasing difficulty a lot in Austria.



          But this is purely maths. I haven't even raced in most of these tracks. Yeah, I'm a nerd.


          Part of why people might think Canada is easy is because a lot of people get away with significant corner cutting on Canada. I know I used to before I started driving more realistically and strict lol. But it’s definitely one of the easiest tracks to cut corners and gain big advantages on the AI.

          Comment

          • GaloVingador
            Pro
            • Mar 2018
            • 564

            #140
            Re: F1 2018 Review: Perhaps One of the Best Racers Ever

            I'm loving this so much I bought a G29 only for this game. Able to shave some half a second of my times using the wheel, btw.

            I'm racing on Austria now and the difficulty adjustments above worked very well to me so far, except for Monaco, where I'm 2 seconds behind the pace, this track is a nightmare. I'm using a baseline difficulty of 80 at qualis and 82 at races.

            I've been running 25% races but found that to be very short, no time for cars to spread out and things to develop, it's far too rushed. I've been contemplating a change to 50, but I'm worried with some things:

            1. At 50%, 1 stop strategies are far too op. Unlike in 100% races, there's no way to recover the time lost in an extra pit stop. But I noticed it's usual for the AI to stop twice anyway, especially front runners. Can this be mitigated somehow? Maybe if I play short quali instead of full so everyone starts with the same tyres this could be more balanced?

            2. I heard the AI doesn't take advantage of the safety car to make it to the pits. I'm playing with SC off just because of this, as winning a race taking advantage of these circunstamces would be very unrewarding. Is this still the case on the current state of the game?

            Many thanks!

            Comment

            • GaloVingador
              Pro
              • Mar 2018
              • 564

              #141
              Re: F1 2018 Review: Perhaps One of the Best Racers Ever

              Originally posted by GaloVingador
              1. At 50%, 1 stop strategies are far too op. Unlike in 100% races, there's no way to recover the time lost in an extra pit stop. But I noticed it's usual for the AI to stop twice anyway, especially front runners. Can this be mitigated somehow? Maybe if I play short quali instead of full so everyone starts with the same tyres this could be more balanced?
              I just realized only the top 10 are obligated to start the race with quali tyres after short quali as well. Never noticed that. So it seeems if I want balanced pit performance, my only options would be 25% so everyone pits only once or 100% for different strategies to be viable, right? Such a bummer...

              Comment

              • Godgers12
                MVP
                • Dec 2012
                • 2265

                #142
                Re: F1 2018 Review: Perhaps One of the Best Racers Ever

                Originally posted by GaloVingador
                I just realized only the top 10 are obligated to start the race with quali tyres after short quali as well. Never noticed that. So it seeems if I want balanced pit performance, my only options would be 25% so everyone pits only once or 100% for different strategies to be viable, right? Such a bummer...
                I think that tire wear is calibrated to race length. So I don't think it would really matter.

                Sent from my SM-G965U using Operation Sports mobile app
                Green Bay PackersSeattle MarinersNew York Rangers
                Syracuse Orange

                If walls could talk to spill the lies, we'd see the world through devils eyes
                -M. Shadows

                Comment

                • GaloVingador
                  Pro
                  • Mar 2018
                  • 564

                  #143
                  Re: F1 2018 Review: Perhaps One of the Best Racers Ever

                  From my understanding, wear is doubled for 50% races, but 25% races have the same wear as 50%, this is why every 25% race is an one-stopper.

                  The problem is, while tyre wear is scaled at 50% distance, the time lost in a pit stop is not, making any strategy with an extra stop included.a very clear loser, especially on circuits where it's hard to overtake

                  Comment

                  Working...