OSUPiper's CH2K8 CPUvCPU Experience

Collapse

Recommended Videos

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • OSUPiper
    MVP
    • Nov 2005
    • 1099

    #121
    Re: OSUPiper's CH2K8 CPUvCPU Experience

    Just a note on V6.0:

    Defensive Pressure at '100' in Coach Settings - I tried to avoid having a static setting for any of the coach options. Since I play CPUvCPU, Coach Settings are one of the few places you can tweak things during the season. In fact, I have a good PG that gets into foul trouble a lot and it is probably due to this setting.

    I may back it off to 95, but I can tell you that disparities in this setting will lead to much different shot selection, and what you will find is that games are being predetermined based on a Coach's D setting rather than by the players. This setting is too valuable for spacing and consistency in offensive pace and shot selection. When paired with the sliders for Awareness and Help Defense/Recovery, the variety of plays to the basket make up for the static setting IMO.

    Also, teams that cannot play defense still get punished, but they get punished in the halfcourt in individual matchups, which to me is a big plus for immersion into the game.

    In the end, I hope you guys see what I am seeing in my games and enjoy the sliders. I know I keep tweaking things, but I cannot see any reason I would move on from this set.

    v6.0 Gameplay
    As I have stated, my sliders are supposed to create gameplay that fits reasonably within the boundaries of simmed stats but have the look and feel of college hoops IRL. If you are looking for eye candy and fun-and-gun games in legacy mode (absolutely nothing wrong with that, my friends!), try version 4.x.


    Known Issues
    1. Still tweaking FG% overall. Might be low, but it is pretty close.
    2. Turnovers are too low. This is a game engine issue, as I have described in other places on the forum. I will probably not reduce Ballhandling any more than it is. Lowering BH will lead to live ball TOs, which means more run-out baskets (layups and dunks) which raises FGAs and FG%. In turn, we have to ratchet down pace of play in the halfcourt and at some point every game looks the same. Not worth messing with.
    3. Fouls are slightly low. Like with turnovers, it is probably not worth messing with to get an extra 2 fouls each game on average. The problem here is that some games are foul festivals already. Raising fouls tendencies will ultimately mean that you have a couple of starters out for most of the game every third game or so. IRL fouls balance out over the course of the game because, frankly, the refs are watching the scoreboard too, and there is an innate sense of "fairness" which causes us all to expect foul numbers to be even numerically. (In reality, this is stupid, because odds are that one team just fouls more than the other... college coaches take advantage of this by being extra physical, because coaches know that refs simply will not foul everyone out.)
    Last edited by OSUPiper; 12-30-2017, 11:14 PM.

    Comment

    • OSUPiper
      MVP
      • Nov 2005
      • 1099

      #122
      Re: OSUPiper's CH2K8 CPUvCPU Experience

      - Edited Page 1 to reflect V6.0 changes: Legacy Edits
      - Edited Page 1 to reflect V6.0 changes: Pre-Game and In-Game Edits

      Comment

      • rahrahcarolina
        Rookie
        • Sep 2017
        • 23

        #123
        Re: OSUPiper's CH2K8 CPUvCPU Experience

        The 19 min have always blown me away. Does 1 min really makes a difference when comparing cpu vs cpu and simulation? I have been watching cpu games for about 10 years now and been using piper'sliders the last couple of years. I am a purest also. I don't believe in adjusting other Coaches settings. 2k8 is the best college basketball game there is.

        I get playing Madden or NCAA football 8 min qts to equate to the stats of the 12 min simulated stats.
        Last edited by rahrahcarolina; 12-31-2017, 08:18 AM.

        Comment

        • OSUPiper
          MVP
          • Nov 2005
          • 1099

          #124
          Re: OSUPiper's CH2K8 CPUvCPU Experience

          Originally posted by rahrahcarolina
          The 19 min have always blown me away. Does 1 min really makes a difference when comparing cpu vs cpu and simulation? I have been watching cpu games for about 10 years now and been using piper'sliders the last couple of years. I am a purest also. I don't believe in adjusting other Coaches settings. 2k8 is the best college basketball game there is.

          I get playing Madden or NCAA football 8 min qts to equate to the stats of the 12 min simulated stats.
          At the outset, I need to note that I actually run 18-min halves in legacy, so the sliders I posted are in error. Now to your points...

          I probably have not addressed this before, but I can give you my theory behind half length. It is simple.

          First, I looked at sim stats at the end of a season. I was focusing on indicators for possessions per game, so I compared things like points per game, rebounds, turnovers, etc. The emphasis (IMO) must be on scoring average first, and this is also the best simple estimator of possessions at equal shooting percentages.

          The best representation of NCAA season stats (PPG) across the distribution of teams by PPG in simmed seasons occurred at a 19 minute half length.

          Also, consider that a reduction of one minute of half length is (in theory) a five percent (5%) reduction in every statistic that is averaged per game (points, rebounds, etc.). So, if you get 60 shots per game with a slider set, a 1-minute reduction gets you to 57 shots. In reality, the game engine shifts priorities given certain game states (e.g., a team trailing with under 10mins to go in the game will hustle the ball up the court, etc.), so the effect of half length is not exactly linear. It works at high numbers though.

          Personally, I have no problem setting played game half length at 15 and sim game half length at 20, as an example. It was my own slider "design decision" to see if I could keep half length equal but still generate comparable stats.

          So, next, I looked at played games stats and adjusted sliders to be in line with simmed stats. Any variation in half length would be the result of my inability to get possessions per game in line or a decision not to compromise good gameplay for a stat fix.

          Finally, your comparison to Madden/NCAA football raises a very interesting point. The theories are similar, because in essence you are trying to reproduce apples and apples, that is possessions per game in played and sim games. Of course, this should yield proper stats all the way around. Note that one very important distinction is that football has a play clock while basketball has a shot clock. A shot clock is a possession deadline, not a playcall deadline. Thus, quarter length in football is actually a function of number of plays rather than possessions.

          Football gameplay is very clock-driven. Even though playclocks serve the same purpose in each sport, the clock is a much bigger part of strategy in football than in college hoops. Many football plays begin just before the play clock expires. Importantly, an expired play clock does not end a possession.

          In basketball, the clock is really a deadline but you do not see many possessions in a game where teams are running up against the shot clock. Many possessions, particularly in transition, end after 10-15 seconds.

          In CH2K8, I can (and did) use sliders and coach settings to establish parameters for gameplay that embraces a variety of strategies, gameplay options, roster builds, and tempos, but still stays within a realistic statistical range over a 35-40 game campaign, season after season. In NCAA football, I used quarter length to control plays per game, but used sliders to direct run/pass playcalling and player impact.

          Thus, though these are similar concepts, they have different impacts in the different games.
          Last edited by OSUPiper; 01-03-2018, 03:24 PM. Reason: clarity

          Comment

          • OSUPiper
            MVP
            • Nov 2005
            • 1099

            #125
            Re: OSUPiper's CH2K8 CPUvCPU Experience

            HALF LENGTH IN PLAYED GAMES SHOULD BE 18 MINS.


            Corrected in original Version 6.0 post on previous page. Sorry for that confusion.

            Comment

            • OSUPiper
              MVP
              • Nov 2005
              • 1099

              #126
              Re: OSUPiper's CH2K8 CPUvCPU Experience

              V6.0 NOTES
              I was concerned that FG% might be dipping too low, but my games are trending upward now. I will continue watching this.

              I did lower ballhandling to test the impact.

              Finally, it is so difficult to generate off-ball fouls. Reaching is really the only way to do it. The issue is that high-rated PGs seem to get into foul trouble very early half the time....mine plays about 10 minutes in those games.

              Correct foul distribution is probably more important than correct number of FTAs, but if you average 75-80% on FTs, getting to the line is a serious advantage. This is true in real life, too, but in real life teams can adjust for how a game is being called. CH2K8 really does not adjust. So, I am not sure how I might address this.

              Comment

              • cubsfan203
                All Star
                • Jun 2004
                • 6689

                #127
                Re: OSUPiper's CH2K8 CPUvCPU Experience

                This thread is full of great information. I'll have to try these sliders out!
                Fan of....
                Memphis Tigers - Texas Rangers - Dallas Mavericks - Dallas Cowboys

                Coaching Career of Chris Matthews (CH2k8)
                The Legend of Ocho Ocho

                Comment

                • OSUPiper
                  MVP
                  • Nov 2005
                  • 1099

                  #128
                  Re: OSUPiper's CH2K8 CPUvCPU Experience

                  Version 6.1
                  Changes from V6.0 are:

                  RULES
                  Reaching 76 (78)
                  Shooting Foul 78 (73)

                  Tendencies
                  Mid Tend 5 (4)
                  Drive 27 (29)

                  Attributes
                  Ballhandling 27 (28)
                  <table width="430" dir="LTR" border="0" cellspacing="0" cellpadding="2"><tbody><tr><td width="25%" height="20" valign="BOTTOM"></td><td width="25%" height="20" valign="BOTTOM"></td></tr></tbody></table><table width="430" dir="LTR" border="0" cellspacing="0" cellpadding="2"><tbody><tr><td width="25%" height="20" valign="BOTTOM"></td><td width="25%" height="20" valign="BOTTOM"></td></tr></tbody></table>
                  <table width="858" dir="LTR" border="0" cellspacing="0" cellpadding="2"><tbody><tr><td width="13%" height="20" valign="BOTTOM"></td><td width="12%" height="20" valign="BOTTOM">
                  </td></tr></tbody></table><table width="858" dir="LTR" border="0" cellspacing="0" cellpadding="2"><tbody><tr><td width="13%" height="20" valign="BOTTOM"></td><td width="12%" height="20" valign="BOTTOM"></td></tr></tbody></table><table width="858" dir="LTR" border="0" cellspacing="0" cellpadding="2"><tbody><tr><td width="12%" height="20" valign="BOTTOM"></td><td width="12%" height="20" valign="BOTTOM"></td></tr></tbody></table>

                  Comment

                  • OSUPiper
                    MVP
                    • Nov 2005
                    • 1099

                    #129
                    Re: OSUPiper's CH2K8 CPUvCPU Experience

                    Testing:
                    3pt(%) at 45
                    Ballhandling at 26
                    Reaching Foul at 75
                    Shooting Foul at 80


                    So far, very slight improvement to foul distribution, but I think this is gonna be an "either-or" issue absent player edits. No real difference in turnovers per game, although sloppiness seemed to be more evident. Did not see much difference in 3pt% so far.

                    Comment

                    • BRASKA15
                      Pro
                      • Feb 2013
                      • 513

                      #130
                      Re: OSUPiper's CH2K8 CPUvCPU Experience

                      Editing a roster I think is the best way for optimal gameplay and aim stats. I want to create a one time roster from one year, in which the players are created similar to 2k and their archetypes and player models they have. Once a roster is complete, it will be easier to get sim and good gameplay as well as edit recruits coming in.


                      Sent from my iPhone using Operation Sports
                      I PLAY FOR NEBRASKA

                      Comment

                      • OSUPiper
                        MVP
                        • Nov 2005
                        • 1099

                        #131
                        Re: OSUPiper's CH2K8 CPUvCPU Experience

                        Originally posted by BRASKA15
                        Editing a roster I think is the best way for optimal gameplay and aim stats. I want to create a one time roster from one year, in which the players are created similar to 2k and their archetypes and player models they have. Once a roster is complete, it will be easier to get sim and good gameplay as well as edit recruits coming in.
                        I hope you are correct. I am afraid that the issue will be the sim engine producing different results from a range of values as compared to the game engine with the same range. We see this now with free throw percentage.

                        Comment

                        • BRASKA15
                          Pro
                          • Feb 2013
                          • 513

                          #132
                          Re: OSUPiper's CH2K8 CPUvCPU Experience

                          I think you have to make the sliders catered towards the roster. Make the sliders help replicate the in sim engine stats. I am going to keep looking into it.


                          Sent from my iPhone using Operation Sports
                          I PLAY FOR NEBRASKA

                          Comment

                          • OSUPiper
                            MVP
                            • Nov 2005
                            • 1099

                            #133
                            Re: OSUPiper's CH2K8 CPUvCPU Experience

                            Originally posted by BRASKA15
                            I think you have to make the sliders catered towards the roster. Make the sliders help replicate the in sim engine stats. I am going to keep looking into it.
                            Just so I am clear, I think you are proposing that you would edit rosters using CH2K8 general templates, but adjust player ratings to achieve better sim results. Then, we set up sliders to get in-game results in line with the new sim results.

                            My concern is when Player A with a 3p rating of 60 shoots 28% in sim games and Player B with a 3p rating of 95 shoots 40% in sim games, but in played games Player A averages 20% and Player B averages 50%. This is what I am experiencing, which is why my edits all tighten up the ranges at certain positions or shot tendencies. FG ratings and Foul rating are specific examples of this.

                            Comment

                            • BRASKA15
                              Pro
                              • Feb 2013
                              • 513

                              #134
                              Re: OSUPiper's CH2K8 CPUvCPU Experience

                              I haven’t got that. Usually if I set up the player ratings right, I’ll get a pretty good base to set up sliders. I’ll make a team and let you test it out with the set I am using.


                              Sent from my iPhone using Operation Sports
                              I PLAY FOR NEBRASKA

                              Comment

                              • OSUPiper
                                MVP
                                • Nov 2005
                                • 1099

                                #135
                                Re: OSUPiper's CH2K8 CPUvCPU Experience

                                If we can get this game to play well from 35 to 99, I am all in. Overall, that could actually reduce the edits from year to year.

                                Comment

                                Working...