rabs 10 Posted October 16, 2009 AI problem. Ok a little back ground. I open the difficulty settings and set friends units to novice and enemies to expert. Save. Open up the editor load UTES, USMC squad on one end RU squad on the other. Move way points to each other. Play as USMC rifleman. I lay still and roll around a little bit. USMC squad annihilates RU squad with 1 loss. I play as a Russian marksman, I lay still and roll around a little bit. RU squad destroys USMC squad with no losses. What gives? This is stupid. it removes any challenge from the game, because your presence on a particular side seems to make them super men. Share this post Link to post Share on other sites
Second 0 Posted October 16, 2009 (edited) Based on demo difficulty setting has little to do with anything. What ever the difficulty setting is and in which ever side player is results seems to be always same. Bigger squad (more firepower) wins. One idea springs to mind. You don't advance so your squad won't advance, as a result enemy moves and your squads just fires. I'm not familiar with current .FSM, but best i can guess at the moment. This is basic and effective winning tactic from days of OFP. And simple too: can be issued with three keys pressed in split second. As a hint: leave your self out of both squads and watch results from some distant place with binoc or something. Or just don't give them waypoint that way they don't need to move anywhere. However leaving your self out of it is more reliable way, less room for errors (which have potential to be many). Edited October 16, 2009 by Second Share this post Link to post Share on other sites
Bulldogs 10 Posted October 17, 2009 (edited) Odd, from running the test myself, many times the team I was on lost. That said, there's a lot of factors that go into it. Terrain, AI skill level, etc, starting position and direction, etc. I should mention that I did that test a long time ago (patch 1.02) just ran it again about 10 times and regardless of position or side I was on the US always won with 3-9 men remaining. But that was just a close range, no waypoint test. Need to run some more tests with distance and maneuver factors. Edited October 17, 2009 by Bulldogs Share this post Link to post Share on other sites
fabrizio_t 58 Posted October 17, 2009 AI problem.Ok a little back ground. I open the difficulty settings and set friends units to novice and enemies to expert. Save. Open up the editor load UTES, USMC squad on one end RU squad on the other. Move way points to each other. Play as USMC rifleman. I lay still and roll around a little bit. USMC squad annihilates RU squad with 1 loss. I play as a Russian marksman, I lay still and roll around a little bit. RU squad destroys USMC squad with no losses. What gives? This is stupid. it removes any challenge from the game, because your presence on a particular side seems to make them super men. It sounds like you're being as a targets "spotter" for your group (as long as you are "embedded" into it). AI has big problems with threats detection over distance. A player into a AI group gives its units an unfair advantage (especially when auto-reveal is active). See for reference: AI units can't spot standing enemy at 250m. with clear line of sight http://dev-heaven.net/issues/show/2209 and AI shooting accuracy: unit SKILL value has negligible effect on shooting accuracy http://dev-heaven.net/issues/show/4483 Share this post Link to post Share on other sites
Second 0 Posted October 17, 2009 (edited) It sounds like you're being as a targets "spotter" for your group (as long as you are "embedded" into it). AI has big problems with threats detection over distance. A player into a AI group gives its units an unfair advantage (especially when auto-reveal is active). See for reference: AI units can't spot standing enemy at 250m. with clear line of sight http://dev-heaven.net/issues/show/2209 and AI shooting accuracy: unit SKILL value has negligible effect on shooting accuracy http://dev-heaven.net/issues/show/4483 Yeah. Player will always have expert skill, little to do in other ways but with autoreport it has to do a lot. There's just too much variables at work. I dont' get much effect from autoreport in my tests, but that probably is because at start sides are just barely over 100 meters away from each of other. Autoreport has little to do in that case as they pretty much see each of other anyways. Yet. Difficulty setting seems to have little effect. Firepower has much more to do with it. My test scenario is kinda no-brainer. Put two squad and leave 100 meters of space between them, no waypoints. They will start killing each of other immediatly at start of scenario with their maximum capacity. I'm outsider and observing action. Results: Similar squads with similar firepower (i used CDF, another as red) is where you see difficulty setting's effects very clearly, but more there's difference in firepower more it dictates the results. Novices seems to be able to beat experts even with small edge in terms of firepower. That is because it doesn't seem to affect the accuracy of shooting (as has been said), so even if experts have edge on other sides it still piles down to accuracy of fire, in which difficulty setting has little to do. In more complex situation things can be different. Hard to say as one doesn't know what effects difficulty level has. Do they have better/worse morale, do they see better and break tunnel vision, are they faster on trigger, do they follow orders faster, do leaders issue more commands etc. Edited October 17, 2009 by Second Share this post Link to post Share on other sites
