kabill Posted May 7, 2014 Share Posted May 7, 2014 (edited) I have done some testing and this hypothesis seems to fit what I have seen. Somehow the bullet pathing is messed up. With the new formula for drawing the "best" line to target, on top of the new bullet deviation system that changes where "misses" go something is definitely wonky with the hit percentages. Just to back this up, I did a quick test, modding accuracy up high to give 95% shots. This was the shot: This was a 95% to-hit shot, and you can see that the line of fire is drawn to the side of the target rather than it's centre. After five shots, this was the result: One destroyed tree, no hits. Obviously, it's only a small test, but what happened was basically the same as the impression I've had from playing generally. EDIT: Did it again using the debugger. Here's some example output: [3520] >>> Begin calc. projectile trajectory <<<[3520] Target tile = (27, 59)[3520] Unmodified accuracy calculation:[3520] range penalty (1.000000) = 1 - ( max(0, dist. to target (14.317822) - weapon range (30)) / weapon range (30) )[3520] shoot accuracy from props: 0.660000[3520] shoot accuracy (0.660000) = shoot accuracy (0.660000) - max(0, weapon recoil (0) / 100 - strength (0.650000) )[3520] weaponAccuracy (315.000000) = weaponAccuracy (105.000000) * accuracyModifier (3.000000)[3520] unmodified accuracy (207.900009) = shoot accuracy (0.660000) * range penalty (1.000000) * weapon accuracy (315.000000)[3520] Damage for unit:[3520] damage mul. (1.000000) = difficult mul. (1.000000) * ai race damage bonus (1.000000)[3520] normal damage (33) = weapon damage (35) * random bonus/penalty (0.960000) * damage mul. (1.000000)[3520] stun damage (0) = weapon stun damage (0) * random bonus/penalty (1.480000) * damage mul. (1.000000)[3520] emp damage (0) = weapon emp damage (0) * random bonus/penalty (1.010000) * damage mul. (1.000000)[3520] Terrain bullet score:[3520] obstacle: level = 0, cell = (85, 177), bulletScore = 0.000[3520] terrain bullet score = 0.000000[3520] chanceToHitTargetTile (0.000) = (unitAccuracy (207.900) + shortRangeHitBonus (0.000)) * terrainAccuracy (0.000)[3520] Random number (96) is > chanceToHitTargetTile (0.000) => miss shot[3520] Hit an obstacle on tile (28, 59) with stopping chance 100.000 As in the example Chris posted above, it's returning a 0% terrain modifier, reducing the chance of hitting to 0 and shooting the prop instead. EDIT 2: And it's doing the same for non-100% cover props. Edited May 7, 2014 by kabill Quote Link to comment Share on other sites More sharing options...
Chris Posted May 8, 2014 Share Posted May 8, 2014 I never said this one was fixed in the hotfix - we only fixed it last night. It'll be in the next update. Quote Link to comment Share on other sites More sharing options...
UmberSkies Posted May 8, 2014 Share Posted May 8, 2014 You don't realize just how bad this bug is until your Scimitar tank shoots its Plasma Cannon at its treads, blowing up itself and the 4 soldiers sheltering behind it -.- Quote Link to comment Share on other sites More sharing options...
Chris Posted May 8, 2014 Share Posted May 8, 2014 That's actually a different bug, but that's also fixed in the next build. Quote Link to comment Share on other sites More sharing options...
UmberSkies Posted May 8, 2014 Share Posted May 8, 2014 Oh! Okay, well that works then! Thanks for all the quick responses Chris, its good to see a developer that is in touch with the community. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.