Relationship between TCRM and hitrate

Jump to navigation Jump to search
Revision as of 10 July 2019 at 06:49.
The highlighted comment was created in this revision.

Relationship between TCRM and hitrate

According to my stats, a tcrm battle on average contains 1194.5213333333336 shots, ignoring 1.0 bullets, and given the hitrate as x, you get tcrm score = 1194.5213333333336 * 16 / 35 * x = 546.0668952380953x

for 100 tcrm score, the expected hitrate is 0.18312774656738373 and for 0.001 improvement in hitrate, you get 0.001 * 546.0668952380953 = 0.5460668952380954 improvement in tcrm score.

If the relationship is mostly linear, then you can just expect that, 0.55 diff in tcrm indicates only 0.001 in hitrate.

Question: is that worth it to work on tcrm scores so hard for only 0.001 hitrate?

    Xor (talk)01:15, 9 July 2019

    At first glance I would say no.

    And doing some calculations, if the average points scored in a 35 round battle is ~5,000, and the average amount of bullets fired by you in a match is ~2,000, a gain of 0.001 in hitrate would mean that you hit 0.001 * 2,000 = 2 extra bullets per match.

    The bullet damage done would go up to 20, assuming you are firing at a firepower of 2.

    So if the prior score was 2500 / 5000 = 50%, a 0.001 hitrate gain would mean your score would go up to 2520 / 5020 = 50.4%.

    But I took a lot of assumptions with my math and it's probably not super accurate.

      Slugzilla (talk)05:20, 9 July 2019

      And from practical perspective, my answer is no again. I have different guns with different tcrm score, however the best gun in tcrm generally performs even worse in rumble

      The difference in tcrm score of those guns are ~0.5, sounds quite a lot, however after calc I know it’s only 0.001 However the best gun in rumble could score ~1.0 more APS than the worst (which, ironically, is the best gun in tcrm score) So I think it isn’t wise to improve tcrm score any more once the result is already among the best.

      Maybe for gun challenges, we should create something more relevant to rumble score then.

        Xor (talk)11:44, 9 July 2019

        Hmm, we could make a Targeting Challenge 2k19

          Slugzilla (talk)15:42, 9 July 2019

          I would vote for the targets to fire a 1 power bullet away from you each time you fire. Then it would be more sensitive to small changes even at higher levels.

            Skilgannon (talk)16:07, 9 July 2019

            I am not sure about that. You get a bit more difference in bulletscore, but less difference in survival, at least against the good movers (which are the most interesting)

              GrubbmGait (talk)16:20, 9 July 2019

              Well it will affect both. Basically then we actually have a direct measure of the hitrate, and not just a measure of the amount by which we run out of energy.

              The score wouldn't go to 100 anymore though so we would have to scale it.

                Skilgannon (talk)17:18, 9 July 2019

                That makes sense.

                We could calculate the scores by our bullet damage / a constant, or by our bullet damage / the total score

                  Slugzilla (talk)19:27, 9 July 2019
                   
                   

                  I think this is better than a non-firing target, as hit-rate is not the only thing that affects score.

                  If choose to fire only when a guaranteed hit is possible, hit-rate will be 100%, however you merely earn score this way.

                    Xor (talk)07:49, 10 July 2019
                     

                    I think the reverse of mc2k7 could be used

                    The reference bots will be using RaikoGun.java from mc2k7 for radar, gun and energy management

                    And the challengers use their own gun which fires 3.0 bullets where possible, and the movement should be a modified version of RaikoMicro’s, which allows prediction of one’s own movement. The appearance of the movement should be kept the same though.

                    Although bullet damage doesn’t scale to 100 in this case, since the only variable is your gun, all reference bots are given the same weight actually, scaling isn’t necessary.

                      Xor (talk)07:33, 10 July 2019
                       

                      Another option would be to use a snapshot of the entire rumble However in this case we can’t control the guns

                      So that each opponent maybe given different weights, where normalization should be used to compensate

                      I think this approach should be the most relevant one to rumble score, however it makes relationship between hitrate and score unclear, if different opponents aren’t normalized well

                        Xor (talk)07:41, 10 July 2019

                        Or we could just plug in a component to calculate raw hit-rate, and maybe even patching roborunner to report this result directly in this case

                          Xor (talk)07:45, 10 July 2019