Pericles
-
Posts
170 -
Joined
-
Last visited
-
Days Won
1
Posts posted by Pericles
-
-
I have been witnessing all kinds of irrational panic behavior in the SF2 demo Wilcox scenario, where panicking pixeltruppen run towards the enemy rather than away. I am in the process of uploading three examples to Youtube that conclusively prove that this is the case. Generally, others in the forum have agreed that there is a pathfinding "problem" with the SF2 demo.
If we are to accept that there is a problem when panicking units commonly run towards the enemy, how can this be solved by changing the code? I have never seen any suggestions by the developers as to how this might be improved.
I have a suggestion that I thought I'd share, perhaps it has been recommended before: another movement command option called "Flee" (or something like that). What it would allow you to do is to plot a point to where a squad or fire team is to run if they panic. It seems like the obvious solution. When the squad/fire team panics, it will run to the panic waypoint, no matter what. If the squad/fire team does not end up panicking during the turn, the "Panic" waypoint will not take effect.
My guess is that this sort of feature is not possible given how the game is coded.
Thoughts?
-
5 hours ago, IanL said:
So, what does happen on replay? Do they withdraw on their own every time? In other words I am curious if you have seen multiple self preservation withdraw actions or if there was only one.
One thing that I would like to check is the relationship to friendly map edge. That scenario setting plays heavily into the decision so if it's wrong... So, what direction did they actually go when the made the poor decision you witnessed (talking compass points here). Then I can compare that to what the scenario setting is.
Having said all that this is a probabilistic thing. The troop's reaction to the bad event will not be the same every time and the decision on where to withdraw to will not be the same every time either. Which is to point out soldiers making a bad decision from your point of view is most likely not a bug. It could be but it is not the most likely reason.
IanL - Do you have a response to my question? I asked whether you agree that this is a bug now that I provided video evidence. I'm interested to know your thoughts on this given your experience with CM (e.g. you have over 12,000 posts on the forum).
-
No, I only witnessed the unrealistic behavior once. I re-played the turn and the fire team did not take a casualty and so I accepted it and moved on.
In the scenario, your forces start in the South and attack Northwards. When the fire team panicked, they ran North (towards enemy forces) not South (towards friendly forces).
22 minutes ago, IanL said:Which is to point out soldiers making a bad decision from your point of view is most likely not a bug. It could be but it is not the most likely reason.
Luckily I made a video of the behavior I witnessed, in the event that my description was not enough. Please have a look at the video below (I will delete it in 24 hours). Given that the fire team in question was ordered to "Hunt" one action spot away, to the outside of the door , and given that there were no other commands issued for that fire team, and given that they chose to run towards the enemy forces/unexplored area of the map/North when they panicked, this is conclusively a bug.
IanL - Would you agree that this is a bug?
-
LongLeftFlank - very funny
db_zero - your comments regarding "Hunt" into buildings do not have any relevance to the identified coding problem.
-
3 hours ago, peter thomas said:
Mostly, when they panic now they seem to lie flat where they are, no? So there is certainly, I think, some progress there, perhaps?
Agreed.
3 hours ago, peter thomas said:It's easy to cure it, I've found, but setting a path to right outside the door you want them to use, then a path into the building, instead of just a path into the building, allowing them to choose.
1 hour ago, 37mm said:I agree with Peter that "wrong door" issue's are usually (indeed I think entirely) caused by my own, usually rushed, orders.
These statements do not apply to my case. I issued a "Hunt" order to one action spot away from where the fire team was. This put them outside a door. I did not plot another waypoint inside the building. I would describe myself as "very patient" in terms of how I issue orders. They then took a single casualty from a grenade. They then fled, and instead of running back towards friendly forces, they ran directly towards the unexplored area of the map. They were then eliminated by a group of enemies hiding in the next building that had yet to be cleared.
"Unrealistic" is how this behavior can be described, and :"coding problems" are what caused it in this particular case.
-
I just witnessed a fire team do something completely unrealistic. Their leader became wounded from a grenade in close quarters combat (Wilcox scenario). The orders were for the fire team to advance a single square in "Hunt" mode. Once the casualty was inflicted, the rest of the fire team became nervous and decided to run away - they were no longer under fire. I'm ok with this - they panicked and fled, seems like something that might happen under the circumstances (even to Veterans).
The problem is how they fled. They were sandwiched between two buildings so they essentially had two options: (1) run back towards friendly forces (the side of the map from whence they came) or (2) run towards enemy forces (the side of the map that had yet to be explored). They chose option (2) and became pixelcorpses.
Concerned with this because I want CM to be perfect. I feel like this is a "bug" that can never really be fixed given that it still isn't fixed. I understand, it's complicated given the complexity of the code. But it sucks. Good thing that it is relatively rare (I've logged at least 10 hours with the new demo and this is the first time I've come across it).
That being said, re-playing the turn does not yield the same unrealistic behavior so I'll forget about it for days, weeks, or months until it happens again.
-
Very glad you enjoyed it. Like most others are saying, this is the freshest CM scenario I've ever played. I tend to steer away from using the term "innovation", but this is truly innovative.
I decided not to go with the air assault - transporting my men by helicopter at 1 pm in the afternoon in a Separatist invested area seemed like a bad idea.
I wasn't sure about the affected radius around the tires so I went with the river crossing (once and only once).
The intel is definitely understandable and useful.
Stay tuned for part 2, the Power Plant remains hot and I'm about to attempt to extract a couple defectors from the North Outpost.
-
Here is part 1 of a cinematic AAR of the Tactical Operations Center scenario. A bit on the hasty side but hopefully it does the scenario some justice.
-
Ok, ok.
I recently witnessed the change in walking ("move") animations that occur when soldiers move up or down sharp slopes... very impressive.
-
I love everything about CM, except [insert grinning troll face here]:
-Lack of variety in sound effects/missing sound effects (e.g. there are perhaps three background sound tracks available throughout the whole CM series; no sound when vehicles are stuck moving their wheels)
-Poor voice acting and lack of variety in phrases (e.g. "burn you mother")
-Frequent lackluster scenario design
-Graphical effects and graphical modeling could be improved/be more detailed (e.g. flame, flame casualties, more variety in soldiers' appearances)
-Campaign format lacks continuity (does not feel like missions are connected in most cases, just a series of scenarios strung together with a little dialogue)
-
Yes, and there you have it. If most customers of CM have a strict preference for historical simulation, then perhaps the first CM3 release should include two titles, one that is futuristic and marketed so as to expand the customer base, and another that is historical (or "what-if" historical) that caters to the existing base (or those out there who are history buffs but who have yet to learn of CM).
As for your gold coin rain analogy, it is far more likely that a future futuristic CM scenario would be done right. You will agree if you are sane.
-
Never go full robot.
-
1 hour ago, domfluff said:
I'm suggesting that attempting to model something entirely speculative using an explicit model might be the wrong kind of solution - if you have to make everything up, you're not simulating anything.
I believe that one can use the term "simulate" in the context of the development of a game that models future military units.
1 hour ago, domfluff said:Decent simulations of high levels of speculation are generally better done with models which are less concerned with details.
I don't think this is true, at least as far as video games are concerned. It seems that you're essentially saying that video games seeking to create immersive imaginary futures ought to remain coarse like obscure board games. Your evidence for this is that you can't model things that don't exist. Of course, you can. This is done all the time in research and development, prototypes, etc. etc.
1 hour ago, domfluff said:This is even true in military exercises - making assumptions that tech will exist, without being too particular about the whys and wherefores, but generally modelled in abstract, broad terms rather than specifics.
If CM: 2067 went so far as to model laser weaponry, I'm sure a believable model could be created using the CM frame. In my opinion you are needlessly complicating things here. You claim that it would be better to use a different type of game to "simulate" or "model" future combat. I am trying, but it just seems to be an illogical argument. None of your points have substantive bearing on the issue of developing "CM: 2067".
-
1 hour ago, domfluff said:
That's an opinion (not an invalid one, but opinion nonetheless) - you could take exception with a number of unit parameters, and others do.
My opinion is that you are not seeing the bigger picture here: the Combat Mission series is one of the few games available that attempts to model military conflicts in such detail and fidelity. Of course, some people take exception with some of the unit parameters. The bigger picture is that CM does the job better than most. If your opinion is to the contrary, I'd like to know which titles do it better (with such detail).
1 hour ago, domfluff said:The lack of detail firmly works in it's favour when it's used to model things that traditional wargames find veyr difficult - hidden information, random events and complex socio-political leanings. They are therefore a powerful tool for modelling speculative or hypotheticals, since the fine details are unimportant.
I still don't understand. For example, in all seriousness, what does modeling of "hidden information, random events and complex socio-political leanings" have to do with what we are talking about here?
-
12 hours ago, Michael Emrys said:
If you have no reliable history to check your game against, it tends to fly off in all directions with no validity whatsoever.
By writing "tends", you imply that there is a chance that BF or some other group could make a solid, believable military simulation set in the not-so-distant future (e.g. 2067). So I contend that if BF really put their heart into a title like this, the resulting product would not "fly off in all directions with no validity whatsoever". Further, I think they do an excellent job modelling current military capabilities... describing the modeling as "borderline" is essentially synonymous with saying that it isn't done well.
6 hours ago, domfluff said:Even with Black Sea, there's a worry about over or under-modelling certain aspects, which just aren't an issue with the WW2 data.
Again, we have concerns over the ability to "model" reality. This is a misplaced concern when considering the accuracy and believability of CMBS. For me, the modeling in CMBS is believable - BF's speculations seem reasonable. Any problems I have with the game have tended to relate to scenario design and "AI" behavior, which have nothing to do with unit parameters.
6 hours ago, domfluff said:In general with wargaming, you can do speculative or counter-factual stuff, but it usually requires being set at a less granular level to be plausible - something that an Engle Matrix game or TEWT can do really well, for example, but that a detailled counting-bullets sim will struggle to keep plausible.
I don't understand.
6 hours ago, domfluff said:Would I like to see a CM-like serious take on something speculative? Sure. I think it would be unlikely to end up all that similar to Combat Mission though.
Combat Mission is essentially the following: realistic model scales; WeGo; full 3D camera freedom; ability to pause, replay; realistic/believable modeling. Successfully applying this formula to a hypothetical TO&E for future military forces 50 years out is more likely than not in my mind, so long as there is genuine interest on the part of the developers. If done right, the final product would feel exactly like Combat Mission, only with different units, environments, and scenarios.
-
44 minutes ago, DerKommissar said:
I kid you not, the theatre was empty -- 8 people including us
Yes, I'm aware that the movie "flopped" in the theatres and ended up losing money. I still think that if packaged properly with really nice graphics, graphical effects, sound effects, and marketing, CM: 2067 could double the customer base.
47 minutes ago, DerKommissar said:The success of this game would depend largely on the quality of this imaginary future. The world has to be built to be consistent, interesting and lived-in.
Agreed.
As for the potential "lore" of such a title, it wouldn't be overly complicated. Developing a world consistent with a CM: 2067 vision requires only that one ask the question: What would modern military forces look like in 50 years? What are the current development trajectories of modern military forces? These are exciting questions. Add a relatively coarse "what-if" military situation/storyline (China-Russia invasion of North America) and much better marketing (better trailers, BF presence at soul-less functions like E3) and it could work.
-
2 hours ago, DerKommissar said:
Fantasy units would take away a lot of the appeal of Combat Mission. Arma 3 did this to attract a larger customer base, and it made vanilla a joke.
Futuristic units are not necessarily "fantasy units". The appeal of Combat Mission can be boiled down to its realism (ballistics, unit fatigue, and pretty much everything else) and gameplay mechanics (WeGo, full 3D freedom of camera movement). Neither would be sacrificed in a futuristic title. The units that end up being included in the title would be carefully configured so as to be "realistic" in the sense that they are entirely consistent with the imaginary future in which the combat takes place (ballistics, fatigue, and damage modeling would remain believable).
If done carefully, this sort of title has the potential to become a masterpiece due to the creative license afforded. Essentially it would be the video game equivalent of Blade Runner 2049 if done right. And the engine would be readily amenable to other more realistic what-if scenarios like the Fulga Dap.
-
For me, the ultimate CM3 release would be sci-fi involving futuristic warfare (e.g. Combat Mission: 2067). Robotic units, yes (e.g. squad-level drones, mech suits maybe). Alien races, probably not. Could be a "Russia-Chinese force invades western USA-Canada' scenario or something like that. This would be amazing, not only because of its futuristic scenario and aesthetic (including much improved graphics on top of the stellar CM gameplay mechanic) but because it would bring in a much larger customer base if done correctly. That would translate into more sales, company growth, and ultimately faster development of other CM3 games (e.g. CM3 in obscure WW2 settings like Finland or Cold War settings like the "Fulda Gap', whatever the hell that is).
-
Good to know, thanks for the Intel.
-
Is it possible to add ammo to supply trucks from supply dumps?
Playing the TOC scenario and wanted to add Ukranian ammo to a LMTV supply truck from the supply dumps at the TOC in an effort to re-supply northern Ukranian forces. Unable to "Acquire" from truck. Dismounting driver and passenger to "Acquire" and putting them back in the truck does not allow for ammo sharing with passengers (ammo is not added to truck's inventory, but rather is fixed with the driver and passenger.
My guess is that this is not possible.
-
Why is it that infantry units often spot other infantry units moving through forests before spotting armored vehicles moving through forests?
The reason I ask is because I just witnessed a Russian BMP move into a forest about 10 meters from one of my Ukranian infantry units. Russian infantry units dismounted from the BMP - these were promptly spotted by the Ukranian infantry unit (basically right after they dismounted). But the BMP was not spotted even though it is closer. I suppose they'll notice it next turn.
I guess I'm supposed to imagine that foliage and brush not graphically modeled in the game is hiding their view of the behemoth 10 m next to them?
All I'm saying is that it doesn't seem right to for them to be able spot infantry and not the BMP in this case. I have witnessed this spotting behavior once before.
-
-
Can anyone explain why enemy units commonly face away from the front line? I encounter this often in CMBN and CMRT scenarios. For example, I'm currently investing time in the CMRT scenario "Getting Ugly". Enemy tanks moved forward (towards my forces) and then ended up in positions facing away from my forces towards the direction they drove from. My guess is that this is an artifact of the Upgrade 4.
-
Good question. I added the music in an attempt (futile?) to add a suspenseful-emotional dimension to the movie... not so sure it had its intended effect. You'll notice that it tended to coincide with suspenseful moments (e.g. spontaneous enemy advances, Russian tanks). It was worth a try at least.
Concerned over rare pathfinding problems in SF2 demo
in Combat Mission Shock Force 2
Posted
I have uploaded three videos to Youtube regarding the pathfinding problems. Links are pasted below.
I have also started a thread that might get us some idea as to whether there is actually a coding solution to this problem.