TheVulture

A colleague of mine once got the fun job of trying to create an algorithm for a computer to automatically fit sloped roof shapes to buildings given their outlines. It actually worked surprisingly well as long as the building shape was pretty regular. But give it something a little more complex.... one roof it came up with, for what ought to have been a reasonably simple case, had several hundred separate sloping surfaces. It was a thing of beauty - and you could have a good platoon-sized game of cat and mouse just on the roof of that one building Bet we don't see that in CM:WWII

Yeah, i have that problem too. Now I feel inspired to go create a Wonderpets scenario - just have to create a background sound mod of the theme tune. Platoon commanders: Linny, Tuck and Ming-Ming.

Lol. Not got to that one yet. Do you have a "Boots the Monkey" task force and a platoon of explorer stars to help you on your way? Or just a baby Jaguar stuck in a tree?

The only limitation on that approach (which is the only way to do it in WeGo) is that you have to move at least one action spot, and that isn't always feasible. Maybe you could use 'hunt' - not sure if, in the presence of enemy, they'd even change direction before cancelling the command though - but that might stop them before they actually move anywhere out of cover and get them throwing smoke the right way.

I always assumed that red dots equals longer delay in the mission - if they have just finished a mission and have lots of red dots, it takes them longer to get reorganised for another shoot. I have no idea whether that is right though As for the others, all good ideas. But as always, coding it so that the player can use it is only part of the problem. Having the AI able to use it sensibly can be a bigger problem. I'd personally love to see move speed (slow, move, quick, fast) and move logic separated (just move, move to contact and stop, move to contact and continue when contact is lost, move until LOS to a specified point is found). You could create complex behaviour with that kind of system. But. Can the AI use it to plot reasonable move orders in a the available (real-)time? Can the UI accomodate it without making plotting even a simple move require twice as many clicks? (bit of a killer in real-time games). I suspect it is not practical, as much as it would be very powerful in the hands of the player. But yeah, fix the damn hummer tachometer already And add the odometer - I want to know how far my humvee has driven in a scenario...

You should note that the important point for what gets blasted is how close to the wall the blast move starts. You can put the destination waypoint half a mile away - they will just try and blow up the first wall with 16 meters that they come to in that direction. This can be useful to know to avoid sticking a blast target waypoint just on the other side of a wall, followed by a move into building command, and have the guys mill around in their 'waiting for everyone at the waypoint' state. As an example of three variations on a theme (yellow is a 'quick' move, red is a 'blast one). The black line is a wall, the white box is a house. Top-most move will go up to the wall, blast a hole, and then pile straight through the hole into the house via the nearest door. The middle move will go up to the wall, blast a hole, move through, reassemble at the waypoint, and then move into the house via the nearest door. The lowest move will blow up the wall, reassemble, move up to the house, blow a hole in that, and then enter (probably via mix of the breach and doors, depending on positions). Entering / leaving houses via blast can be complex, particularly for MOUT squads with 3 teams, since the teams will tend to take different paths. 1st team plots its move through the breach, 2nd team sees 1st team blocking it and plots a move via a nearby door, and so on.

Here's how it works as far as I have been able to determine: For a blast path, the unit will stop at the waypoint at the start of the blast command (which might be their current position of course). They will blast the first wall (within 15-20 meters) that lies in the direction of the blast move. Then they will perform a quick move to their destination, which may or may not involve going through the blasted hole. If there is no wall close enough to their blasting position, they will wait a while doing nothing, and then do their quick move. Does that cover it?

'cept they pop it at the one after that...

Okay, did some quick tests on infantry smoke. If you have a target arc, an infantry unit won't lob smoke. I haven't found any way to make a stationary unit pop smoke in a specific direction aside from the obvious face command, and then smoke the turn after (in WeGo mode). If you have a little freedom of movement you can control smoke direction. BUT the smoke command won't trigger at the waypoint where the command is issued, but at the one after. If you put a smoke command on the final waypoint, it will be ignored (not 100% sure of this - I might have used a face command which cancels it, but I think I am right). Smoke commands only show up when issued to the unit in its current position (no waypoints selected), when it shows lines at every waypoint but actually triggers at the first one. Since this all seems rather convoluted, I can only assume it isn't working as intended, even though it is possible to get what you want. But the obvious solution is for a smoke command at a waypoint to generate it at that waypoint, and to be able to specify a direction (for infantry units at least) for the grenades to be thrown in.

I suspect that the troops present in the second mission are more or less those you didn't kill in the first one, with a little bit of redeploying. I may be wrong though.

Depends on the scenario designer I believe. If you give the AI troops in setup zones it won't always place them the same way (not 100% sure of this, but I know that in playtesting a scenario I was making enemy units turned up in different locations, although I can't swear that I didn't change something else, like the position of some zones, some of the terrain or enemy forces). If you don't give the AI setup zones then its units are locked into the positions the designer placed them in at the start.

Has anyone tried using a very short movement order in the direction you want the smoke to go, and then firing smoke from the end waypoint. Don't know whether it works or not, but it might. Also, changing facing with cover arcs might be different to changing using the face command. It used to be in CMx1 in certain circumstances - changing facing in a building manually could get you spotted due to the movement, while changing it with a cover arc was 'free' and didn't get you spotted.

The TacAI is much better these days. Infantry are pretty good at doing sensible things and making decent choices about returning fire, seeking cover, avoiding the Crawl of Death. There are always going to be some cases where it doesn't make the right choice, but it's pretty solid and you can tell it is doing well because you don't notice it (rather than being annoyed about it making obviously bad choices). Infantry pathfinding is fine. As always, there are occasional things that seem strange, but it is usually a matter of incorrect expectation. Once you understand that infantry will tend to stay out of houses unless ordered in (and so won't go *though* a house on the way from A to B in the absence of a waypoint in the house), or that in narrow spaces US squads with multiple teams can do odd things since one team seems its route blocked by another team (you can micromanage around it, or just use assault...), then you can be pretty confident what the effect of an order will be, and that is all you need to avoid frustration with pathfinding really. Vehicle pathfinding is much better than 1.0, although seems to have taken a slight step back from 1.08, with a few random bits of maneouvering around piles of invisible sand. Overall enemy AI is more a matter for scenario designers than programming, and unsurprisingly improves as the designers get more used to how to use the AI plans in the editor to achieve what they want. Camera panning is about the same as it was. I don't have any problem with it now, although it took me a while to get used to it. But if it actively annoyed you before, it probably still will. Issuing commands it much smoother. You can define hotkeys for all the commands so that you can use them without paying attention to the tabs. And there is a pop-up menu to issue commands - if you have a unit selected you can hit the space bar to bring up a menu with all the available commands for that unit in it. No more playing around with the tabs. I never really got into the CM:SF campaign, despite trying it several times. They just weren't the kinds of battles that interested me for one reason or another. Only just started the marines campaign, but I am loving it so far. It feels much more like CMx1 in terms of immersion and playability. I've only tried one scenario off the disk (Day at the Beach - only about 10 minutes in but it has been awesome so far). There have been some great scenarios and campaigns produced for the stock game, and playing those in 1.08 was fine. 1.10 is even better. Can't tell you anything about multiplayer. I stick to single player or PBEM games, playing WeGo, not real time. They all work fine for me.

Oh me, me I know this one! Easy answer: doing a new module is primarily creating data files, models, artwork, and so on, with (in an ideal world) only a little but of coding for special cases thrown up by new units - plus the usual bug fixes. Producing the WWII game on the other hand requires great deal of coding. ISTR Steve saying that Charles had pretty much started working entirely on the code for WWII some months ago because he just wasn't needed for anything in the modules. Meanwhile the artists, model makers, sound bunnies have been working on the Brit module for a while, and at the same time the testers and scenario designers have been playing around with the Marines module getting it ready for release. So I think you'll find that once CM:SF was got into a decent state, the first stirrings of work on CM:Normandy began, and there simply isn't any way of getting it out the door sooner than it is going to be: it takes as long as it takes for it to be coded and tested. Meanwhile other people can work on producing the content of the modules without slowing down the production of CM:Normandy at all because it is different work being done by different people. You don't get a choice between Brit module in 6 months, CM:Normandy in 12 months vs CM:Normandy in 6 months and the CMSF:Brits in 12 months (I'm totally making these numbers up BTW). It is rather a choice between Brits in 6 month, Normandy in 12 vs Normandy in 12 (maybe 11), Brits in 18 (plus lots of artists sitting on their hands with nothing to do waiting for code to be finalised ).

Surely you mean that it's role is being spotted at very long ranges, rather than spotting, I swear I just had one taken out first shot by a BMP-2 at long range when it was the least exposed of any of my Bradleys. But somehow the BMP spies it and blows it up. Maybe I'm just bitter

Turns out it works with vehicles too:

Your 3 step guide to rope use: Well, it is one way of making terrain that is accessilble to infantry and not to vehicles (actually, not tested it with a vehicle.... will get on to that). The trick, in case it isn't obvious, is simply to put a trench running up the cliff, and the troops ability to regard the trench as passable terrain apparently overrides the impassibility of the cliff.

There is a hotkey to show all movement paths: alt-p I think. Was put in a few patches ago. No way to show all fire orders though.

Well if I haven't got hold of the Marines module by tomorrow lunchtime, I'll have something to occupy me for another hour of testing

Assault into buildings specifically, or in general? And are we talking MOUT units or normal stryker ones here? The AT guy isn't a team leader in a regular squad. I did some testing for a normal stryker squad using 'quick' to move over open ground from one trench to another while getting shot at by a Syrian rifle squad. There was no discernable preference for anyone one person getting hit more than anyone else. I did tests with and without javelins being carried, and it again made no difference to the AT guys survivability. The AT guy was in fact hit less than anyone else in the squad. The guy who got hit most often was the squad marksman, but it wasn't by a significant amount. Whether there is some additional effect like the guy humping the javelin around being more likely to be spotted in marginal conditions and draw fire I don't know. Not tested building entering, but it wouldn't surprise me if certain squad members were more likely than other to be first through the door and come off worst. (See what I am reduced to doing in my lunch hour while waiting for Marines to be released... Have pity, Battlefront)

Not least it prevents perfectly legitimate uses of area fire on potential or suspected enemy positions while advancing. Any time you want to cover an advance by shooting up the few areas where enemy units might be waiting for you, you are suddenly slapped with long delays and ineffective fire. As a rule, you don't want to go around blocking legitimate tactics to prevent occasional misuse. While that would be no bad thing, BFC do seem to have a general policy of making us all play the same game as far as possible, on the basis that the more options you allow the player, the more you fragment the head-to-head player base who have their own favourite rule sets, which in turn reduces your number of potential PBEM / TCPIP opponents and arguably hurts the appeal of the game for people who enjoy playing human opponents. Or I gather that is more or less the reason. (c.f. Civilization multiplayer games which are usually preceeded by extended wrangling over which of the many, many rules options in the games are turned on, never mind which player-created rules mods are to be used). For all that it is nice to be able to customise a game to play how you want it to, the multiplayer community benefit greatly from everyone having to play by the same ruleset all the time. (Maybe there is a system that would work which everyone will like, but I don't personally think that penalising area fire depending on what the area-firing unit knows relative to what some other unit knows is the way to go. I find it is far too common that I want to shoot the hell out of a few buildings just to be safe, whether or not some stray AT team happens to have picked up a sound contact in the area).

Well then, I am very pleased to be wrong. Got to love the battlefront guys.

All the new AFVs are in marine formations only as I understand it, and aren't available for regular army (CM:SF base) formations one way or the other. The only change of what is available to existing formations that I noticed is the introduction of the t-90 for Syrian republican guard formations, and for those there is a new 'excellent' equipment level that doesn't exist in the base game. So existing scenarios won't include them since they won't have the appropriate equipment level selected. (It's also the only reasonable way to do it without causing major incompatibility bugs and problems I suspect).

Does that mean people who *didn't* pre-order it can go and download it now?

Done a little testing, and I can now make it happen or not fairly reliably in my test scenario. You *can* have back-to-back blast orders quite happily, so my first guess was wrong It depends on the spacing of the houses and position of the blast waypoints. The first rule of blast club is that they will go stationary at the start of the blast move, wherever it is, and try to blow up the nearest wall in the direction they want to move. (So you can have two houses, put a single blast move across the first house and into the second, and they will blow up the near wall of the first house and then run around the outside into the second - which is prefectly reasonable behaviour to my mind. I just mention it so that people know what to expect from a blast). The second rule of blast club is that if there isn't a wall in range (range is around 16 meters AFAICT) they will go prone, wait for the 10 seconds, and when no blast comes get up and do a 'quick' (it looks like) move to their destination. So one explanation is that the blast command path started too far from the house they were to enter, so they changed it to a quick after the wait. How close together were the houses, and where did they waypoints go?