Pud Posted October 21, 2003 Share Posted October 21, 2003 Just wondering has the routing been tweaked to overcome the tendancy of units in cover when routed to run towards enemy , across open terrain rather than away from fire to find more cover behind their current position? 0 Quote Link to comment Share on other sites More sharing options...
Berlichtingen Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by Pud: Just wondering has the routing been tweaked to overcome the tendancy of units in cover when routed to run towards enemy , across open terrain rather than away from fire to find more cover behind their current position? If enemy units are routing toward you, it means you are between them and their friendly map edge. Sign that you did well (hit the enemy from the rear), or a badly set up scenario 0 Quote Link to comment Share on other sites More sharing options...
Michael Emrys Posted October 21, 2003 Share Posted October 21, 2003 I wonder if we will see more infantry surrendering in AK in situations where they would have routed before, simply due to the lack of places to run in the desert. Michael 0 Quote Link to comment Share on other sites More sharing options...
Pud Posted October 21, 2003 Author Share Posted October 21, 2003 Thanks Berli, I would like to say it was my brilliant tactics but in all truth probably scenario error 0 Quote Link to comment Share on other sites More sharing options...
Berlichtingen Posted October 21, 2003 Share Posted October 21, 2003 The default direction that units rout in is toward their friendly map edge. It is a setting that is easy to overlook in scenario design. Recently I was playtesting a scenario by Andreas and managed to get around behind the enemy. I proceeded to get my ass kicked and my troops routed through the enemy defenses! I'd call that a bug, but not something I'd expect to see addressed until the engine rewrite 0 Quote Link to comment Share on other sites More sharing options...
JasonC Posted October 21, 2003 Share Posted October 21, 2003 The other time I've seen men rout the wrong way is when they go toward a particular bit of cover, but when they get there find it is already occupied and they don't fit (an "overstacked" building, an occupied foxhole, etc). I've seen them head for the enemy after that, and back into fire they just ran out of. As though the "where is cover?" routine is measuring from the location of the original break, rather than the current location. A big part of the problem seems to be that Tac AI cover thinking doesn't know LOS from anatomy. It only looks at the terrain type. Reaching dead ground only happens "empirically", by getting up and moving again if still getting shot at. 0 Quote Link to comment Share on other sites More sharing options...
CMplayer Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by Berlichtingen: The default direction that units rout in is toward their friendly map edge. It is a setting that is easy to overlook in scenario design. Recently I was playtesting a scenario by Andreas and managed to get around behind the enemy. I proceeded to get my ass kicked and my troops routed through the enemy defenses! I'd call that a bug, but not something I'd expect to see addressed until the engine rewrite Yes I've been asking for them to get rid of the whole idea of 'friendly edges' for as long as I can remember. It penalizes any successful attack that shifts axis of advance (either by getting behind the enemy, or by breaking into their line and 'rolling them up' laterally). The whole system of how infantry choose routes when they move on their own needs some clever solutions. Run for A: nearest cover as the crow flies, or B: Friendly edge, produces all kinds of strange results in practice. Many other people have noticed this as well, and one thing that seems to help is to design scenarios diagonally, where each side gets 2 adjacent friendly edges. [ October 21, 2003, 03:02 AM: Message edited by: CMplayer ] 0 Quote Link to comment Share on other sites More sharing options...
CMplayer Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by JasonC: A big part of the problem seems to be that Tac AI cover thinking doesn't know LOS from anatomy. It only looks at the terrain type. Reaching dead ground only happens "empirically", by getting up and moving again if still getting shot at. One relevant question might be how would it sit with you to have longer turn-computing times, in exchange for better tac-AI. For instance, if every infantry squad which plots its own movement under fire were to do a realistic 'threat analysis' in order to run the way real guys would feel was towards 'safety'... and would also understand that breaking LOS to the firer=safe, and would plot the real distance to where it's going, not the 'as the crow flies' distance, would you be willing to accept twice as long computing times for the turns? 0 Quote Link to comment Share on other sites More sharing options...
Pud Posted October 21, 2003 Author Share Posted October 21, 2003 I would gladly trade some computing power for accuracy. Given when I first started playing CM a few years ago it was on a 500mhz machine (+16mb video) and now its on a 2.4gig (+128mb video). Any delay would soon evaporate with the next PC upgrade. Mind you I cant recall any computing time going longer than 15-20 seconds, Im not that impatient 0 Quote Link to comment Share on other sites More sharing options...
Michael Emrys Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by JasonC: I've seen them head for the enemy after that, and back into fire they just ran out of. As though the "where is cover?" routine is measuring from the location of the original break, rather than the current location.And then there is the infamous "Dance of Death", which is either hilarious or pathetic if it is happening to the other guy or infuriating if it involves your own troops. This occurs when a unit gets shot at in the open and breaks for cover, gets shot at again and reverses direction only to get shot at yet again and reverse direction again, etc., etc., all the while taking casualties and becoming increasingly exhausted until it is quite dead. Realistically, all they would have to do in most cases to survive is to continue on toward the first cover, or even just plop down in a fold of the ground to get out of the LoF. Almost anything would be better than spinning around drunkenly like in a tarantella. This seems to have improved somewhat in BB, as troops are now more likely to go to ground and sneak, but it still crops up now and then. Michael 0 Quote Link to comment Share on other sites More sharing options...
Michael Emrys Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by Pud: Mind you I cant recall any computing time going longer than 15-20 seconds, Im not that impatientI take it you've never played "To the Volga" then? Michael 0 Quote Link to comment Share on other sites More sharing options...
Pud Posted October 21, 2003 Author Share Posted October 21, 2003 Originally posted by Michael Emrys: I take it you've never played "To the Volga" then? Michael Yep, but Im waiting for my Cray to arrive in the post. Im not THAT patient 0 Quote Link to comment Share on other sites More sharing options...
Michael Emrys Posted October 21, 2003 Share Posted October 21, 2003 Originally posted by Pud: </font><blockquote>quote:</font><hr />Originally posted by Michael Emrys: I take it you've never played "To the Volga" then? Michael Yep, but Im waiting for my Cray to arrive in the post. Im not THAT patient </font> 0 Quote Link to comment Share on other sites More sharing options...
JasonC Posted October 21, 2003 Share Posted October 21, 2003 "This seems to have improved somewhat in BB, as troops are now more likely to go to ground and sneak" I don't consider it an improvement. Sneak only makes sense if there is good cover within 10m. Otherwise it is a recipe for (1) exhaustion (2) exposure forever and (3) never firing back. 9 times out of 10 they'd be better off halting where they are - shooting back if possible, waiting for the rest of the force to free them by reply fire otherwise. The routines treat being shot at in the open as a national disaster, when often (e.g. on open steppe) it is par for the course, because long range is the only real cover available. They go into cover panic even when the range is already long. It is a western Europe, cover abundant and ranges short, expectation, poorly adapted to open steppe fighting. In steppe, shot in the open means go to ground and wait for supports. If they are too scared to even consider sitting still, then they should move - away - at a rate likely to actually make it to cover or at least open the range - like "advance". 0 Quote Link to comment Share on other sites More sharing options...
Michael Emrys Posted October 22, 2003 Share Posted October 22, 2003 Originally posted by JasonC: "This seems to have improved somewhat in BB, as troops are now more likely to go to ground and sneak" I don't consider it an improvement. Sneak only makes sense if there is good cover within 10m. Otherwise it is a recipe for (1) exhaustion (2) exposure forever and (3) never firing back. 9 times out of 10 they'd be better off halting where they are - shooting back if possible, waiting for the rest of the force to free them by reply fire otherwise.Well sure, no argument there. But even Sneak lets them live a little longer than the "Dance of Death". If they are in steppe and the range is anything like long, the enemy will often lose sight of them and stop firing, which is a distinct improvement. But yes, I'd like to see further improvements along the line you describe. Michael 0 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.