BletchleyGeek

Let me second those dreams... and the advice at the beginning, especially with the Germans, as the MG remains with the squad leader. As Harry says, the whole German tactical system pivoted around those three guys (the leader and MG crew).

Sorry mate - I missed the adjective "basic" in the OP.

I made similar questions and observations on the thread discussing the hard coded range limitation on some weapons, a couple of days ago. I do think some changes could be done on the effect of such weapons at longer ranges - which would indeed result in "nerfing" lethality, and possibly "buffing" suppression.

It's an official CW module scenario.

Amedeo, just an observation on the data, informed by my experience dealing with that kind of data while working on another game Always take that kind of figures as the "best that weapon can do". Full stop. These figures were collected from field trials, with the weapons being operated by technicians or marksmen on perfect conditions (a firing range with props that allow to measure exactly dispersion). The British Army Ordnance Service conducted a more "realistic" testing using recruits and the differences one can see between the technical specs and the actual outcomes achieved with highly reputed weapons as the Sten or the Enfield, well, let's say that the technical specs of the weapon predict very badly how they will actually perform in the hands of your average recruit, under psychological stress and against targets that try to close the range quickly or are using cover in an intelligent way. From a more "operational" point of view: when you feed this data, taking it as the baseline probability of killing or maiming, on any reasonable model of infantry combat by fire, you'll get massive casualty rates. Massive as in totally ahistorical. The data is "wrong" in the sense that either it needs to be toned down to set the "any given Sunday" to a more reasonable level, which is difficult, since the definition of "any given Sunday" is a moving target, or the combat model (and the AI if it is empowered to manage ammunition) has to provide with the elements to "modulate" these figures, starting from the assumption that it is an absolute upper bound on what that weapon can do. That data integration and curation job is very hard. BFC call seems to be to have the AI to disregard the PPSh as something that can be fired above certain ranges... and I think it makes all the sense in the world. Yet as ASL Veteran and others point out, there seems to be something a bit off with the lethality of these weapons on the latest versions of the engine. I do tend to agree with those observations. But I don't think the data BFC is using is wrong, their models are buggy or the AI is being unreasonable: I think it is more a question of design of effects and interactions with other aspects of the design of the CMx2 engine. Regarding design, I - personally and subjectively - would like to see the volume of fire these things can put out at ranges beyond 60 meters to result in more suppression over a larger area, rather than generating killing or incapacitating hits. As for interactions with other parts of the CMx2 design, the tighter-than-in-real-life packing of our pixel truppen due to action spots constraining their deployment that increases lethality. If you have five guys standing in a 8 square meters area and something like 100 rounds of ordnance shot fly through the volume encompassing the action spot and the men, chances that all of them are hit will be quite high, as they will be physically occupying a significant proportion of that volume. And that assuming that the rounds trajectories are uniformly and randomly distributed, for even loosely aimed fire, those chances can become almost a certainty. Observations about action spots and burst fire extreme lethality have been made in the past, I am not sure what is BFC opinion on that.

Hi Mark, I am not a great fan of the series - as noted above it's like Close Combat but without the info and clarity of presentation of unit status, etc. Regarding your 4 bullet points, GTOS scores full marks on the first one, 5/10 on the second one (it has an operational layer that looks very cool but it just area based as in Close Combat, with the same kind of shenanigans you could expect, like not being able to have two companies occupying the same 1 square km area etc), 6/10 on the third one (the AI sometimes is impressive, and sometimes is hilariously bad) and 0 on the fourth one. You may have some enjoyment watching the evolutions of your units and the pretty graphics. But as soon as you try to implement your plan, it is a bit like the game says to you "get your dirty hands off me, do you want me to take that hill, I will do that myself" and then bunches the troops into a mob that charges uphill on the open.

This is quite interesting. Harry is showing us, just mixing and matching animation files, how one can obtain out of the engine more - realistic? sensible? - behaviours out of our pixeltruppen. This indeed should inspire BFC on how to achieve significant changes in the performance of the AI with no programming effort - just playing around with existing resources. I have no idea of what is the architecture of CMx2, but I don't think it is too different from the layered approach used in F.E.A.R. http://aigamedev.com/open/article/fear-sdk/ Harry is changing CMx2 AI just by changing some of the actions (implicitly by changing how the action is performed in the game world by switching animations) the simulated soldiers are doing in the game world. I have in the past have several interactions with you over the Internet, Harry, and I must say that this is the second or third time that your dedication to understand the workings of game systems leaves me really with a wide open mouth :-) On the other hand, Ian concerns do deserve some consideration. My guess - and it is a guess - is that either in WEGO, one of the two player adopts the role of "server" and the other that of "client", where the "server" is the one that does the actual number crunching. In WEGO, my guess is that the "server" is the second player - one cannot resolve the orders given by each player until you have all of them, and that happens when the second player presses the end turn button. On the other hand, in RT Multiplayer, I wonder if the role of "server" is distributed or is passed around between the two players. If it is distributed, then Ian concerns do hold - probably the game would go out of sync as soon as a changed animation kicks in. In WEGO, turn resolution is fixed - so if the "client" player has shuffled around animation files, he gets results as if he hadn't changed his files, and if the "server" player changes animations, the results would stand for both players. While I do think Ian concerns regarding stability etc. should be weighed in by anyone wanting to try this modding in multiplayer, I don't see any harm being done when one plays the game against the AI. And in multiplayer, the "harm" would be quite relative (game tanking due to players going out of sync). So I don't think it follows automatically that Harry's "research" is a bad thing at all.

DasMorbo, Regarding your pathing issues it seems to me that you need to keep in mind that path planning in this engine tries to find the "most efficient" routes for your units. And "most efficient" means "fastest". Not "cleverest" or "safest " or "insert here whatever you fancy". No more, and no less. To get what I think you wanted to achieve in the first screenshot, assuming your platoon started in column formation: 1) convert formation from column to line, either wheeling on its right or left flank. You might want to move your troops in column formation, in order to get to an area big enough to deploy in line. 2) issue to each element of your platoon a Move command forward. Note that 1) is just not about finding routes, there's a non trivial spatial reasoning puzzle to solve. That's up to you to be solved with the, quite limited, tools at your disposal.

This sounds very good, Jon, I am already looking forward to this Battlepack. Even if I was half expecting that the Lorraine campaign would get this treatment first. Cheers, matey, 25 sqr kms is fair bit of mapping!

I'm currently playing this scenario H2H as well... and hasn't played anything like yours Stonecutter. Hats off to Jon Martina (?) for this one - plenty of options and room to get creative with your forces :-)

I will just observe that refactoring vehicles simulation, AI and animations code so that tank riders are introduced on the humongous Western Allies vehicle park, has been mentioned in the past as a major undertaking. Everyone, myself included, have their own wishlist for CMX2. All of them more or less loony Probably there will be other not so important new features added (or important, yet invisible, like improved/calibrating HE models to account better for trees splintering, smarter TacAI, etc.). BFC never show their cards so early. I am pretty sure that Charles and Phil find some aspects of the discussion bemusing.

As a practitioner, I can't other than agree with Stagler. The technology is there ready to be used provided that 1) you have the money to pay for the skills, and 2) you have the money (or time, or willpower) to eventually and incrementally rework your codebase in order to integrate those algorithms and techniques. Note the and - both things need to be true at the same time - and the eventually - a project manager, who needs to focus on the short term - paying the salaries and running costs every month, etc. - won't be very keen about such major surgery operations.

Leaving aside the discussion about SOPs - where I agree with Ian to a certain extent, especially if we're talking at the Platoon level rather than the squad or team/vehicle. There is indeed some SOPs in the game, associated to the basic 'Move' command. You can issue a move order towards a certain point specifying the speed/caution - Slow, Quick or Fast - whether you want it to be bounded - Assault - or to be a "move to contact" kind of thing - Hunt. You can "implement" SOPs by combining this basic commands (and their pre-canned SOPs). Take a look at Bil Hardenbergers' tutorial (linked at the top of the Normandy game forums) for inspiration. But going back to the actual point of the discussion: what Jason says, really. "Active" recon that is, actively seeking the enemy will entail some casualties more often than not. You learn where the minefields are by stumbling on them not using the magic ESP powers Recon vehicles have. You learn where the enemy infantry is by being fired upon (and maybe having the vehicle commander killed or wounded). You learn where the enemy Pak Front is when a high-velocity AT round goes through the leading vehicle and hits the one immediately behind. You learn where the enemy armour is in a similar way. In CMx2 you'll find usually light recon armour in two ways. One, as some kind of infantry support vehicle (which isn't completely ahistorical), deployed on relatively small maps. The other, as the actual thing, with little or no infantry besides some scout or FO teams on really big maps (George MC has made some scenarios where you can use Recon to perform actual recon, like Huzzah or Studienka), and victory conditions requiring you to identify enemy units and retreat most of your recon assets safely. Regarding tactics... if you don't know where the enemy is, you'll have to bit the bullet and really probe like a blind man until you get fired upon. If you suspect where the enemy is, then masking your movement with the terrain features is a must, rather than fancy footwork.

Judging from the trend it does indeed seem that the Black Sea is just going through the roof. I don't know how well do forum statistics correlate with sales, though. But I would say - this is a wild ass guess - that it's probably already selling better than CMFI and RT. We'll know for sure if the next release happens to be a Black Sea module

I would be willing to enter. I have got plenty of stuff half finished and this could be a great motivation to get some of that stuff across the line.

Green CMx2 units behave very much as I would expect: poor fire discipline, target overkill happening all the time, etc. Panicking and fleeing might well be influenced - it is a matter of the "cohesion" of the small unit, besides leadership and morale. "Quality factors" from other games are hard to translate into CMx2... mostly because often those factors make little historical or practical sense, since they're highly influenced by the expectations of the game designer and the playtesters. In other words, are the combination of historical or contemporary sources of varying quality and "balancing". Not to say that CMx2 isn't influenced in a similar way, mind you I just think that here these "soft factors" are more straightforward and make more sense than those I find elsewhere.

I mostly agree with that. But I need to say that "superweapons" are as "super" as the brain guiding the hand that moves those assets knows what they're good for and what are they weak spots. The Royal Tiger has a sweet lethality, but its gun is best not used as a shotgun of sorts, and it's not like you can maneuver very well with it. The IS-2 is fast, reasonably armored (and awesomely designed turret), has a hot gun... but compared to any German tank it will take longer, to spot and engage enemies. And that can be very bad if the opposing German armour doesn't do silly stuff, like charging forward and exposing its flanks or closing in the range below 500 meters. It's also obvious that these weapons, in tight quarters are both extremely exposed to infantry AT weapons. With the AA capability that came in with RT and without the 'WW2 laser designation secret tech' that came in CMBN and CMFI - I think that close air support makes more sense tactically. But only on the bigger scenarios. Add a IL-2 to a platoon level engagement and probably you'll be killing the scenario. Adding a couple IL-2 to a Bn level scenario creates the kind of variability that might want people to replay the scenario (i.e. deploy your AFV's under the cover of trees, rather than on those perfect hull down positions which are so conspicuously exposed from the air). As you say: it does require design. But it's not like we're trying to figure out the nature of dark matter here, either.

Wow man. Are we talking about a QB or an scenario? 4 Shturmoviks can be very rough...

Do you find this effectiveness to be affected by map size? On the biggest CMRT maps - and with AAA on the map - I see that they're not the "Hand of God" they used to be.

I use the word "contrived" to refer to certain designs which are meant to be a very static puzzle: there's one or two solutions, and you need "search" for them by trial and error. That doesn't make them bad scenarios necessarily. Making scenarios that pose several, dynamic challenges requires experience, skill and time. What Kohlenkau mentions above about "blocked terrain" used to be fairly common on some of the first scenarios made for CMBN. People went a bit overboard with the bocage: bocage fields too small or with illogical entries preventing any kind of meaningful use of the land, access to the fields designed so you had to charge at a HMG/AT Gun/Inf Gun frontally, etc. There's more subtle stuff to put you on rails. Most people don't realize it, but if you look at things, you'll tend to find stuff like maps where two-storey (or higher) buildings with windows looking on a particular direction are hard to come by or heavily defended, a treeline can be found in a most curious alignment and much thicker than any other nearby covered areas, etc. This kind of "contrived" map making I think is fair - as usually there is some way to work around that obstacle/challenge/problem rather than figuring out one very particular sequence of arty support/smoke/bounding assault.

What they got right was the theatre of operations, the rest... well, now in 2015 sounds extremely contrived. Maybe in 2003 or 2004, made more sense (the Assads had limited nuclear capabilities until the Israelis managed to pin point the location of a North Korean nuclear reactor and blow it into smithereens in 2006 or 2008, funnily with a very "mild" Syrian protest). BUT certainly nobody saw coming that the Assad dynasty would be challenged in the way has been. The latest trouble in Syria dated back to 1981 or 1982, when Assad Senior troops made an example of Hama for the Sunni to mull over. In any case makes more sense than Arma 3 setting. The politics and strategy of the whole thing are ludicrous.

I got it and I am liking it much more than I expected.

Let me say that I also feel a bit like John - without this AAR or Chris' videos I'd have just passed on CMBS, at least for a while.

This was great, indeed. Well done Ian! I didn't find the engagement with the Russian infantry to be a turkey shooting - got a couple casualties from long range shooting. And my right flank stumbled upon a Russian squad which was patrolling the perimeter... That did not end stellarly. As I got my second SAM vehicle, a very lucky Russian marksman fragged the Platoon CO, and I decided that discretion is the better part of valour. BTW when the platoon co fell, the HQ support team became the Platoon Hq - changing the icon and all. I think this is new, either the engine is showing us something it didn't up to CMRT 1.03 or this was brand new feature.

I am not into crying as a coping strategy, but I can't help feeling a bit sad. One of the saddest memories I have was in the A Temple to Mars scenario in CMFI. Far too many GI's lied strewn on those snowy slopes.