Breaking Down Sliders... How They Really Work

Collapse

Recommended Videos

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • shadox
    Rookie
    • Jul 2009
    • 307

    #46
    Re: Breaking Down Sliders... How They Really Work

    Allright, I am going to test these now.

    Will play 3 Games PIT vs. Ravens, 12 Mins, 25 sec. run-off, 65 Treshhold, 60/60 fatigue/injury, equal sliders for Human+CPU.

    I will edit my results within this post.
    Soccer: Werder Bremen, FC Barcelona,
    NBA: Celtics, Bulls, Supersonics
    NHL: Flames
    NFL: Steelers

    Comment

    • bill2451
      Banned
      • Jan 2009
      • 1123

      #47
      Re: Breaking Down Sliders... How They Really Work

      Pitt vs Ravens huh... i would have tried to use two more balanced teams for testing lol... should be interesting though, and thanks for helping out!!!!

      just curious which team you'll be...

      Comment

      • bill2451
        Banned
        • Jan 2009
        • 1123

        #48
        Re: Breaking Down Sliders... How They Really Work

        Originally posted by Dice
        This is some great work. Would slider modifications remediate(or at least tone down) the passes to the flats issue? Thanks.
        that I do not know... I havent really tested everything together yet, just been simply isolating sliders to see their effect

        so idk what will happen when everything comes together

        that and I also havent tested every slider extensively either... for example WR Catching had some VERY unexpected results (read my long post on page 2 if you havent yet), so i'm wondering if I test things like INT's tonight what may happen (maybe low INT settings, which is what i've been playing with, causes the CPU to play more aggressively to make up for their poor interception skills, and then when i move INT's to 100 different things may happen that are unexpected... idk, i must test to see what strange things pop up)

        so as of right now, nothing... it seems to largely be an AI problem though (you never know though) so trying to get settings to make up for it could cause the rest of the AI to be unrealistic

        Comment

        • bill2451
          Banned
          • Jan 2009
          • 1123

          #49
          Re: Breaking Down Sliders... How They Really Work

          I wouldnt be surprised if the cpu is able to get to the qb too quickly either because the human o-line cant hold their blocks very long

          w/ no hard numbers to go on, it seems to me that cpu AI doesnt need as high a Pass Rush slider to get to the qb as Human AI does... which they would just make them even

          thats not tested or anything (although it should be evident when shadox posts his results), but i'd wager thats the case
          Last edited by bill2451; 08-19-2009, 02:32 PM.

          Comment

          • g2thecore
            MVP
            • Jun 2008
            • 1818

            #50
            Re: Breaking Down Sliders... How They Really Work

            It's killing me right now that I'm at work. I'd love to test these sliders out. Can't wait to hear the results. Hurry!!!! lol

            So Bill these are for both Human and CPU?
            "I'll die before I lose, cuz I was born to win."- Drake

            Comment

            • bill2451
              Banned
              • Jan 2009
              • 1123

              #51
              Re: Breaking Down Sliders... How They Really Work

              lol yeah, but they're not based on any actual gameplay ive had, so the tuning could be totally off... so don't get your hopes up yet

              plus, there are sliders i know shouldnt be human and cpu specific (like pass rush... or at least i think lol) but for this experiment i felt like keeping it simple

              i'm at work too, and just threw these together based on what seems like it'd be in the ballpark of the direction I think the sliders should go

              Comment

              • JkA3
                Strategy
                • Jul 2008
                • 858

                #52
                Re: Breaking Down Sliders... How They Really Work

                great post man. thanks for the info.
                STRATEGYHUSTLERESULTS

                Comment

                • pbz06
                  Banned
                  • Oct 2008
                  • 1504

                  #53
                  Re: Breaking Down Sliders... How They Really Work

                  I love how we're all at work discussing sliders on the internet, hahaha.

                  We should all call in sick tomorrow and start testing

                  Comment

                  • shadox
                    Rookie
                    • Jul 2009
                    • 307

                    #54
                    Re: Breaking Down Sliders... How They Really Work

                    Allright...have already SEEN enough. Just embarrassing...I tried mySkill for a half a week and adjusted sliders for another half week...you haven't tried to acutally play these sliders and they are so f*cking great...at least concerning smoothness *lol*

                    Btw...sorry if anything is hard to read/understand....english is only my second language And I LOVE typos

                    The Settings:
                    12 Mins, 25 sec. run-off, very slow, 65 threshold, 60 fatigue/injury, fight for fumble off,rewind off,backtracker on.

                    Steelers vs. Steelers (rivalry games sucks to compare slider outcome...) Just played a qtr untill now, but the results are amazing.

                    After every play, I watched the replay. I started with a slow overview for the whole play, then watched every player indivudaly and took notes (because that took quiete a while, I only played 1 qtr. so far).

                    On offense, I called my most loved plays and on D, I always controlled the SS in order to let the CPU do their job.

                    I only played to see the animations/what happens on the field.
                    I will do a playtest for the stats and the overall gameplay later on.

                    Here is, what I wrote down (take it as notes, maybe I can sort them tomorrow):

                    OL/DL Interaction
                    - really good and smooth blocking. NO suction blocks at all in any of my 30 plays!
                    - You can not get any pressure, if the number of Dlineman< Olineman...simply blocking outmatch. E.g. Shotgun vs. 3-4 Cover 1
                    - Blitzing is very effective if you can overload or if the blitzer found wholes in the blocking.
                    - Blitzers actually FIND wholes in the Oline, slip through and rush the QB!! => The Olineman doesn't magically block the blitzer anymore.
                    - Blitzer need wholes. They can't run through a blocking package (Olineman vs. Dlineman)
                    - Dlineman needs swim/etc. move to break the block -> Ratings count! Except: Ballcarrier is around, then Dlineman tries to get rid of Oline blocker (unfortunatly, no animation for this, that's why is seems a bit weird sometimes...they use some kind of "stiff arm" animation)
                    - Momentum+positioning counts!! If Olineman tries to block a blitzer who is on the run...he has nearly no chance except the blitzer comes directly from the front. Blockers can't stop anybody who is in full motion if they come from the side.

                    Pocket:
                    - Pocket haven't collapsed (counting only Olineman vs. Dlineman)
                    - Without Blitz, HUMAN and CPU have up to 10 secs.
                    - With Blitz, HUMAN and CPU really have to rush, if blitz is succesfull (~2-3 secs depending on plays). Even if it's not succesfull, you will feel the pressure (~3-7 secs)
                    - Blitzer can slip through 2 active Olineman blockers (active=they already block somebody), if there is any space to run through. Else, they will try to get around the "block package"

                    Blocking/Tackling
                    - Blocks and Tackles are only triggered if player is within range (~< 1 yrd)
                    - Momentum+positioning counts! Tackles will barely (but not never) be succesfull if the tackler is in the wrong position or has less momentum/motion (e.g. tackles from the side while the ballcarrier is running -> 8/10 broken tackles..depending on the ratings!)

                    Throwing
                    - CPU over-/underthrows or simply miss their target (even without pressure) occasionally (not fully tested -> stat test follows)
                    - HUMAN hasn't overthrown (I am just too good )
                    Rushing
                    - no issues. Simple thing: Good playcall -> gain. Bad Playcall -> loss
                    - CPU wasn't succesfull at rushing (but a stat test will show the real result...I called the right plays)
                    - Lead blocking was a bit off, but that's not up to the sliders

                    Defense
                    - Zone D vs. Pass : you cann see how the D watches the Receivers, backpaddle to their zone and have a visible reaction time after the throw.
                    - Zone D vs. Pass: D players are a bit slow on their reactions if a receiver enters their zone (but that seems to be more like a general Madden problem)
                    - Man vs. Pass : Ratings count! Examples:

                    Sweed vs. Townsend
                    Sweed was on a slant route, runs, did small steps, footplant, turn, run again. (smooth and natural animation)
                    Townsend backpaddles, gets slower as Sweed got slower, follows Sweed motion, turns and stays close (within 1 yrd.)

                    Holmes vs. Burnett
                    Holmes on a curl route. Runs towards Burnett, shifts his momentum to the left and straifes past Burnett and follows his route.
                    Burnett backpaddles, see Holmes doing a slightly leftturn, takes a split second to react, turns itself and runs about 2-3 yrds. behind Holmes untill the catch.

                    Ward vs. Madison
                    Ward on Streak route. runs towards Madison, very small step left (visible but without a real momentum shift) and follows streak route.
                    Madison backpaddles, sees Ward keep sprinting, start turning when Ward was 2-3 yrds. in front of him, turn is finished when they are about even and follows ward within 1 yrds.
                    (Nevertheless, Ward would have been open for a lob...but I think Madison would have disturbed the catch)

                    Wards vs. ?
                    Ward on a fake slant (I don't know if that's right...fake left, then go 90° right)....you can really see, how the defender follows the fake step, shifts into the wrong direction and need some time to turn and follow ward in the right direction. Ward was already 4-5 yrds. ahead.


                    Issues
                    Barely found some...
                    - On another 90° route, Polalamu (CPU controlled due to safety swap) anticipated the WR route...he was acutally knowing (or guessing) how the WR will run. But as this happens on the same play like the Sweed vs. Townsend situation I think its due to Troys ratings and/or a general Madden issue. Was the only "superman" D route.
                    - I got one issue of a "railslide" of an Olineman, where he "slides" about 1/2 yrds to the left to fit in the blocking animation. But as the situation was natural, I think that really was an animation issue. The block itself was logical and natural
                    - D seems to have an disadvantage as they always stay behind a little bit. Realistic though, but CAN make the game way to easy/allow to many big plays. I will closely watch this.
                    - Many people are annoyed by D "superman" catches where a D player following a Receiver tries to catch without even looking at the ball. I figured out, that this happens due to a missing animation. The WR looks behind, and then the catching animation is triggered. The D doesn't look behind but triggers the exact SAME animation. That's why it looks weird...lack of animation at this part.

                    Timing
                    - Perfectly! I haven't seen any magical catches, any miracle blocking or any morphing at all. You can only tackle (trigger any tackle animation) if you are really close to the player...same goes for blocking and swated passes.

                    I really love the way this game looks now.
                    Even if you don't see much of this on real gamespeed...but somehow, it just looks right...haven't seen any weird moment.

                    I think here we have the classic game dev. problem...animation triggering vs. rating mechanic vs. game logic/algorhytm....NBA 2k9 anyone?

                    Anyhow..I will now try to really play the game and show some stats/gameplay impressions.

                    OUTSTANDING job so far,bil!
                    Last edited by shadox; 08-19-2009, 03:15 PM.
                    Soccer: Werder Bremen, FC Barcelona,
                    NBA: Celtics, Bulls, Supersonics
                    NHL: Flames
                    NFL: Steelers

                    Comment

                    • bill2451
                      Banned
                      • Jan 2009
                      • 1123

                      #55
                      Re: Breaking Down Sliders... How They Really Work

                      nice overview shadox

                      this is pretty much what i expected... the animation should look much more natural and play better (appear to factor in momentum, etc much more), but the tuning of the other ratings is probably off because i was just spit balling and havent actually played with them, which means stats prolly wont be right w/ these exact settings...

                      but thats ok, becuase the blocking and db/wr interactions are what i'm really after in this phase... if those are working more true to real life then that is the first step

                      one thing I'll ask shadox is that next time you see the db "guess the wr route" you look at a couple things... 1) If he was pressing the receiver, my guess is yes... 2) If his press animation was the one where the db puts a hand up, the WR sort of hand checks/pushes the db's hand away, and the two sort of bump a little and slow down... 3) Watch their feet/momentum... did the WR take a couple small steps and plant to turn while the db just super turned w/out any momentum?

                      if your still in that same game just go to the highlights from the pause menu and if you can find that play... or if you notice it happen again where the db seems to know the WR route before the WR does, check those 3 things and get back to me w/ what you saw
                      Last edited by bill2451; 08-19-2009, 03:32 PM.

                      Comment

                      • stewaat

                        #56
                        Re: Breaking Down Sliders... How They Really Work

                        Wow, awesome thread. Why can't we get new guys like this all the time?

                        Man I hate all the other new people!

                        Comment

                        • g2thecore
                          MVP
                          • Jun 2008
                          • 1818

                          #57
                          Re: Breaking Down Sliders... How They Really Work

                          Originally posted by stewaat
                          Wow, awesome thread. Why can't we get new guys like this all the time?

                          Man I hate all the other new people!
                          +100
                          "I'll die before I lose, cuz I was born to win."- Drake

                          Comment

                          • Justin Clark
                            Rookie
                            • Aug 2008
                            • 87

                            #58
                            Re: Breaking Down Sliders... How They Really Work

                            Originally posted by stewaat
                            Wow, awesome thread. Why can't we get new guys like this all the time?

                            Man I hate all the other new people!
                            I resent that statement! j/k i totally understand

                            Comment

                            • cable guy
                              MVP
                              • Jul 2005
                              • 3288

                              #59
                              Re: Breaking Down Sliders... How They Really Work

                              Great thread. I want to reread the whole thing again.

                              Props to the OP!

                              Comment

                              • bravesfan1984
                                MVP
                                • Mar 2008
                                • 2808

                                #60
                                Re: Breaking Down Sliders... How They Really Work

                                i have to agree with shadox...i gave the sliders you posted a go and here were my stats and some notes:

                                12 Min Q's, 15 sec runoff

                                DAL (Me) vs. SEA

                                Romo 14-21 191yds 2TD, INT
                                Hasselbeck 14-32 152 yds, 1TD, 4INT

                                Barber 25 carries for 123 yards 1TD
                                Julius Jones 20 carries for 125 yards, 1TD

                                Tatupa 7 tackles, 4 TFL
                                Ware MVP with a pick 6

                                Total Offense
                                DAL 311 yards, SEA 304 yards

                                TOP:
                                DAL 32:35, SEA 15:25

                                I thought this game played out great...definitely one of the better ones i've had so far...everything felt great in all honesty...nothing seemed cheap, all the INT were legit...i even went back into replay and checked out the DL/OL interaction and gotta say it looked great. With the accuracy knocked down you could tell the difference between Hasselbeck and Wallace (Hasselbeck was knocked out for a few plays). The running game was great (I use the OL shifts which help out in the run game). Great work Bill and i can't wait for a finalized version.
                                Braves | Cowboys | ND Football | UNC Basketball | 4-Kevin Harvick


                                Comment

                                Working...