Jump to content

More CMx2 Wishlist .


Paul AU

Recommended Posts

I have NO doubt that this is NOT easy at all.

BUT I like this idea:

"AND it is fun to test whose AI fights best! "

My AI vs your AI and just sit back and watch!

this is like a two scenario designers planning the whole battle

and just letting the AI duke it out.

BUT

I don't think that is the direction they are going in.....

oh well

-tom w

Originally posted by jep:

I would like to see scriptable AI as poor (and soulless) artificial Intelligence prevents me playing anything but multiplayer games. Perhaps versatile C-like syntax with triggers.

There should also be an option to import AI modules.

It should be quite easy to create AI-script-language as every key elements already exist in game. AND it is fun to test whose AI figth best!

Link to comment
Share on other sites

  • Replies 57
  • Created
  • Last Reply

Top Posters In This Topic

My 2¢:

One thing about playing against the AI is after awhile you tend to know the game's 'tricks' (not meant as a criticism - name a game that isn't like that!).

As a partial solution I'd like to see several random/rotating AIs, each with its own 'style' of play. I don't mean completely rewriting totally new AIs, I mean the game's ability to swap out one or two small functions that would have the cumulative effect of altering the AI's 'personality'.

One pitfall of random AI personalities is invariably people are going to like playing against one more than the other. Then we'd start seeing posts like "How can I turn off AI-'B' and only play against AI-'A'?"

Link to comment
Share on other sites

I agree completely..

Variable random AI game personallities would be a dream come true.

I have NO idea how one would program random and variable AI "personalities" or agreesiveness..... BUT I would say that kind of feature would be a HUGE bonus for the game if the AI was NOT in anyway predictable from game to game or scenario to scenario.

Even it if just meant the player could adjust a few sliders or scales or imput some variable (like an agression scale or better yet a "shrewd" scale or something smile.gif ) that might give the AI some variability.

BUT I have NO idea what I am talking about when I suggest anything for the AI so I will just take their word for it that it will be way too hard and not worth the effort to make it any better or more variable or random than we see in CMxx....

:(

-tom w

Originally posted by MikeyD:

My 2¢:

One thing about playing against the AI is after awhile you tend to know the game's 'tricks' (not meant as a criticism - name a game that isn't like that!).

As a partial solution I'd like to see several random/rotating AIs, each with its own 'style' of play. I don't mean completely rewriting totally new AIs, I mean the game's ability to swap out one or two small functions that would have the cumulative effect of altering the AI's 'personality'.

One pitfall of random AI personalities is invariably people are going to like playing against one more than the other. Then we'd start seeing posts like "How can I turn off AI-'B' and only play against AI-'A'?"

[ February 09, 2005, 12:02 PM: Message edited by: aka_tom_w ]

Link to comment
Share on other sites

I saw a comment about the tile system going away, and the idea struck me...without tiles it is no longer necessary to have rectangular maps.

Imagine playing on a circular map; no more corners and following the map edge means taking the long way around.

Would this be advantageous? Is it likely to happen?

Link to comment
Share on other sites

One thing that has been nagging me of late is that engagement ranges seem a little close, and ammunition supply seems a little scanty.

These two concepts are to some extent related. When playing CMBB I tend to give my infantry covered arcs that wouldn't be out of place in the nineteenth century. Why? Because my troops burn through ammunition so fast I don't want them wasting precious bullets on something 500 meters away. So that makes me want to fight battles with modern weapons at ranges that would do Wellington proud. And that probably has a lot to do with why CM battles are so quick and bloody.

Now there's probably a feature in CMBB that I haven't found yet which reduces the rate of consumption based on range to the target. (That might be a nice feature if it doesn't already exist).

But I'm really wondering if troops normally went into battle carrying less ammo than they could squeeze off in seven and a half minutes.

How many minutes of ammunition should they be carrying? And shouldn't it vary depending on whether they're taking pot-shots at targets 500 meters away, or fighting for their lives at short hand grenade range?

And doesn't a squad tend to have a little more ammo available to it than just what the men are carrying for their own personal weapons? And isn't this even more so the case with a platoon, and a company, and a battalion?

I guess what I'm saying is that it doesn't seem like it should take the two or three hours that can elapse between battles in an operation for the Islandhwana memorial munitions sergeant to dole out a few extra clips of ammo in the heat of battle, one Jagged Alliance-style bullet at a time.

I'm not groggy enough to know the extent (or even existance) of the small internal ammunition reserve intrinsic to each level of unit ( i.e. the extra machine gun ammo belts strapped around the company clerks shoulders), or what is a reasonable lapse of time and/or what level of being in command should be required to have access to this. I'm assuming that a squad should have ready access to any extra ammunition that it was already carrying ( = why units low on ammo keep firing), but that for each higher level of ammo it needs to draw from there should be an increasing delay. And at some point some sergeant is going to have to run around carrying boxes of ammo from the back of the horse-drawn cart that gets used to move the company'e extra ammo supply. Or something like that.

Notice that implicit in all the above is a need for maintaining a chain of command all the way up and down the line. No command, no staff sergeant throwing an extra box of goodies in your foxhole in the middle of a firefight.

If you add to that some of my strange ideas about the need for evacuating the wounded, you could really begin to simulate the logistical tail without having to resort to taking global morale hits for losing the favorite sheep or mobile soup kitchen.

The Aid station idea, by the way, is that certain hq levels have to designate an aid station for their command at a specific spot. As they take casualties, there would be a tendency for the loss of units from the line to be slightly greater than what was actually suffered, to simulate the wounded being carried back to the aid station. The extra loss would reverse itself when the 'stretcher-bearers' got back to their unit. Units with higher morale would be less likely to detach 'stretcher-bearers', and the length of time the bearers would be absent would depend on the distance from the unit taking losses to the (immobile) aid station. There would be no need to represent any of this with actual figures while it was going on because it is really an abstraction (maybe a medic playing with his shoelace at the aid station when he gets busy), and there should be some kind of penalty (akin to losing the field kitchen or the favorite sheep) if the aid station got hit by a large shell, or run over by a truck.

Speaking of field kitchens, I've got some great black and white shots of German military mobile bakeries and wurst wagons, and I'm itching to do a mod of one of them. I'm also itching to mod mobile field brothels (ever wonder what the title of the opera "La Fille du Regiment" really means?), but that's another story.

Link to comment
Share on other sites

Originally posted by Ichthys:

I saw a comment about the tile system going away, and the idea struck me...without tiles it is no longer necessary to have rectangular maps.

Imagine playing on a circular map; no more corners and following the map edge means taking the long way around.

Would this be advantageous? Is it likely to happen?

To paraphrase aka_tom_w GREAT thinking smile.gif

There was discussion about map edge effects in on of the "what do we want in the new CM game" threads, but discussion of such an important issue has been lacking.

Map edge effects and issues like unrealistically knowing where the setup zones are in a QB are high on my list of fixes for the current CM engine. BFC talks of increasing uncertainty, well uncertainty based on where you are on the map is most important. A ROUND map seems to have some advantages going for it.

How many minutes of ammunition should they be carrying? And shouldn't it vary depending on whether they're taking pot-shots at targets 500 meters away, or fighting for their lives at short hand grenade range?
I agree, this is one of the most unrealistic parts of large and/or long battles. Fire discipline and ammo conservation was pretty tight in most WW2 armies, so units should fire less often for better effect if they are well trained. Also, I can't imagine an entire squad being out of ammo without some steps taken to requisition more as soon as possible, whether fighting was still going on or not.
Link to comment
Share on other sites

Originally posted by Hoolaman:

...this is one of the most unrealistic parts of large and/or long battles. Fire discipline and ammo conservation was pretty tight in most WW2 armies, so units should fire less often for better effect if they are well trained.

I agree. Troops in CM tend to be pretty trigger happy, especially when it comes to targets outside 100 meters, say.

Also, I can't imagine an entire squad being out of ammo without some steps taken to requisition more as soon as possible, whether fighting was still going on or not.
I think the catch here is time. The amount of time it would take for all this to happen might well exceed the anticipated length of a CM battle. I think what Steve and Charles had in mind in laying out the original design parameters of the game, was to depict the 20-30 minute firefight when opposing companies came into contact. As players have expanded both the size of the forces in play and the length of time they are engaged, it has altered the entire nature of the game. If you want to depict a regimental attack that goes on all day, then yes, resupply and casualty clearance become serious concerns.

Personally, I am of the opinion that it is madness to try to model an engagement of that size at the CM scale, but if players desire it, and BFC is willing to undertake it, it'll probably happen.

Michael

Link to comment
Share on other sites

Originally posted by Battlefront.com:

Yes, an anywhere LOS tool was much requested when CMBO came out. The majority making the request were Steel Panthers players and boardgames, since both allowed this sort of thing. Like others have said, this is unrealistic.

Ok, I have to say something since this is a pet peeve:

I agree it is unrealistic, but that's irrelevant.

The reason is obvious: You can already go to an arbitrary spot on the map and look around; a floating LOS tool would just make that easier.

Now, some may balk at moving the camera to arbitrary locations, but I have a feeling this is the way the game is played by virtually all players and further, it is the way the game is intended to be played (or else this would have been limited already); Since the orders menu is relatively sparse and the troop intelligence is relatively simple, it is necessary for the user to plot covered routes for troops (there is no way to simply tell them to go somewhere, but stay in cover. The user must guide them.) The only way to have an idea if one is in cover (or hull-down, has LOS) is to go to level 1 to plot the endpoints. Thus, I conclude it is necessary and intended for playing anything but a very clumsy game to go to level 1 and move around the map to look around.

If one is able to do that, then the floating LOS tool becomes nothing more than a time-saving device.

Also, since the user is not a person on the map (the user is not the company commander nor a squad grunt, etc.--BFC has always claimed that CM will never become a "command game" so I don't think this will change), to claim that this is more information than a commander would have is also a bit nonsensical. In real life, a commander could only see from 1 location at a time and would have to physically move somewhere to observe; otherwise, they'd be looking at a map.

In other words, as long as the game is not a "command game" (or command simulation), it is useless to talk about "gameyness", except in terms of overall capabilities of force on force. I mean, it doesn't matter that a real-life commander wouldn't have that ability (because we already have WAY more god-like abilities anyway); it only matters if it can cause the player to use his forces in an unrealistic manner. And I don't see that happening; at least no more than now. Heck, I bet the AI already has an internal floating LOS so that it can plot its moves.

In real life, a commander would probably have told some of his men to scout, for example, a covered route that is out of LOS of, for example, a town they are attacking. Once they came back and reported the lay of the land and figured out their plan, they could proceed through cover and out of LOS. Similarly, a commander could order his men to take up covered positions overlooking some key feature that they were to cover. They'd probably know well enough that they should have LOS to the area and not take up blind positions.

Originally posted by Battlefront.com:

Although we do admit that there is a bit of problem in that if a unit ends its move just a hair out of LOS of its intended target that the position can't be adjusted until the next turn after going trough a command delay. In some cases this is perfectly reasonable counter balance to the player putting that unit into too smart a spot in the first place, but in some cases it is not realistic. We don't think there is a reasonable compromise to be had.

Actually, I think as long as the game will remain similar to what we have now (meaning not become a "command game" or somesuch), the floating LOS tool is a good compromise.

However, other options would be to expand the orders menu and the AI considerably so that the men could find their own covered paths (within limits defined by the user) and you could order them to take up positions in an area that have LOS to a location.

That would be a much harder option to do programmatically, though, and in reality the floating LOS tool is really not any gamier than the game already is by virtue of allowing unrestricted placement of the camera around the map (so I'd say it is a more practical option.)

It certainly is much more unrealistic that I can't order my men to move to the edge of the woods to cover an area and they end up stopping just short of having LOS to the area. Real soldiers would just move forward another meter.

I would also recommend that in CM2X, we have an LOS that we can change the endpoint elevation on (so that we can find LOS to, say, the second story of a building.)

Originally posted by Battlefront.com:

The computational requirements of a "show all LOS" is too big. Remember that when you are moving around the LOS tool the CPU is only saying "can I see from this one specific point to that specific point". Asking the game to show "all points from one point" is asking it to do thousands of times the work all at once. It would be horrible. I remember even in Steel Panthers the "show all LOS" took time on my computer of the day.

Steve

Well, since we're talking about during the orders phase, I don't think it would be that bad, but it is probably inferior to the floating LOS tool and would surely take awhile to calculate (and would be the more difficult option to code.)
Link to comment
Share on other sites

Quick answers... 1:1 spotting is computationally out of the question. LOS checks are one of the worst things to compute, so multiplying the checks by 100+ is obviously out of the range of capabilities of even the best comsumer computers. Well, unless you want turn times to take hours to crunch :D

If scriptable AIs were "easy" all games would have them since AI coding is a thankless task that everybody always winds up bitching about. Since no game I can think of (commercial type anyway) has scriptable AI in the way described, I think everybody here knows why. Hint... it ain't easy!

AI personalities are also not easy to do. All effort is spent on making the AI do things the "right" way as best as possible. And of course it doesn't come even close to it considering "right" is what a good Human player would do. Alternative personalities would mean coming up with different methods for doing things "right", and that is just "wrong" ;) It's not something that can happen easily for a CM type tactical game, though for strategic or FPS games it is actually not too hard to accomplish.

Yes, CM was designed for about a 1/2 hour firefight. Therefore "in game" resupply and casualty treatment were never part of the game design. The current time limit for CMx1 games was something we added into the game AFTER initial release IIRC. Users requested it, we advised against it, but since it was easy to add we did. But we can't be held accountable for shortcomings of pushing the game to the extreme. If we are to be held accountable, then the solution is for us to patch the game right now to limit games to 1/2 hour time periods tongue.gif

Steve

Link to comment
Share on other sites

(And this).

"Ruthless" said it better than I could.

Battlefront.com, thanks for seeing and replying. It was worth a goat.

(And CM's A.I. is pretty darn good, BTW)

One last request... don't make CMx2 too different, 'cause it's near perfect as it is.

[ February 10, 2005, 09:39 AM: Message edited by: Paul AU ]

Link to comment
Share on other sites

Originally posted by Ichthys:

I saw a comment about the tile system going away, and the idea struck me...without tiles it is no longer necessary to have rectangular maps.

Imagine playing on a circular map; no more corners and following the map edge means taking the long way around.

Would this be advantageous? Is it likely to happen?

To an extent, this can already be done in CMx1. I know of several scenarios where the designer has oriented the square map so the action is directed along the diagonal. This makes "hugging the map edge" even longer than on an equivalent circular map. For examples, see Schrullenhaft's scenario Pavlov's House and Andreas' Gefechtsaufklaerung (both on the CD).

For a truly circular map, you could always take a square map and fill in the corners with water.

Link to comment
Share on other sites

Originally posted by Hoolaman:

Also, I can't imagine an entire squad being out of ammo without some steps taken to requisition more as soon as possible, whether fighting was still going on or not.

"Requisition"? Like, filling out a written request....??

The company sergeant major (or equivalent in the US, Germany, et al) would set up an ammo point behind the front and involve himself in moving ammunition forward. In the CW that was done most likely by universal carrier, the Germans likely used horse carts, the US jeeps. If necessary, men, mules, horses would carry the stuff forward without mechanical assistance. The ammo point would be located somewhere near company HQ. Resupply would be done where and when possible.

The RSM of a battalion would keep ammunition going forward to the companies.

Where possible, a full load of ammunition would be issued before any major engagement, and ammunition supply would be part of the operations order. The ammo dump would be marked on the company commanders' map, or at the least reference would be made of the map reference where it could be found, likely the platoon commanders would make note of the reference also, along with the location of the casualty collection point, Regimental Aid Post, etc. This was done before a major battle, and of course when in defensive positions.

Things were more confused if contact was made while the unit was on the march - a situation that would be interesting to see simulated in CM, actually - if ammo resupply was to become part of the simulation. Even if an abstracted version would be of interest.

Link to comment
Share on other sites

1:1 is here to stay heh? OK, but how about fixing a few things along with it?

1. How about the giving us the ability to move assault boats on land?

2. How about ending those strange weather combinations like snow in Hot weather?

3. How about allowing vehicles to block LOS? (Actually, I think this was already said to be included. I could be wrong.)

4. How about allowing infrantry to fire from vehicles?

5. How about an image on the ground for a fighterbomber that has been shot down, when it is shot down? Smoking wreckage would be cool.

6. How about using more than one CD so we won't hear, "Sorry, there wasn't any room left on the CD for......" ?

I know that others can give more examples. Please don't get me wrong. There is MUCH to this game that is right.

1. The 3D environment.

2. WE-GO.

3. The dedication of BFC to getting it right (as best that they can.)

4. Unlimited re-playability.

Are just a few. Do you have any that you would care to add?

Link to comment
Share on other sites

Originally posted by Battlefront.com:

Quick answers... 1:1 spotting is computationally out of the question.

Will this lead to situations where soldiers virtually stare into the enemy's eyes and do not react because the LOS between the centers of mass of the squads is broken?

And ... do you plan to accept such situations out of necessity or do you have a workaround?

Best regards,

Thomm

Link to comment
Share on other sites

One can not compare a 2D, hex/tile based wargame with very limited terrain types and spotting rules to CMx1, not to mention CMx2. What SSI did with Steel Panthers is child's play compared to what would be required of it in CMx2, even considering the exponential increase in CPU speed and other subsystems capabilities. This is just simply fact guys, not something that you can debate with uninformed theory and irrelevant examples. End of story. Period.

Now, what will this mean for the game? We have several plans that will largely give the desired behavior without burning out the CPU. We should be able to have individual LOF, which is far more important than individual LOS because the latter can be fudged a lot easier and with less undesirable side effects than LOF can.

As for the specifics of what we plan on doing to simulate many eyes looking for targets simultaneously, we'll let you know when we have something in place and working. Until then there really isn't much more that I can say since too many things are unknown at this point.

Steve

Link to comment
Share on other sites

Tom W,

In response to your initial post… “grogs” who want to “simulate battle field conditions as realistically as possible without actually getting shot at” should only be able to “see” the canvas walls of their CP tent, and nothing else.

A “realistic” company or battalion level command game would be a text-based one, with the occasional hysterical field-telephone conversation thrown in.

Now… why wouldn’t that be a good “game”?

Like I said, I’m only as “gamey” as the average gamer.

Link to comment
Share on other sites

One thing I can say about the LOS tool in CMx2 is that it will be more functional than the one in CMx1 (which IMHO is done better than other games). The best thing about CMx2 is that with less abstracted terrain and units, things will be far more intuitive without the LOS tool. But once used it will certainly do more than the CMx1 tool. Distance read out, better refinement of what it is pointing to, etc. should all make it in.

Steve

Link to comment
Share on other sites

Originally posted by Michael Dorosh:

...Things were more confused if contact was made while the unit was on the march

or in a hasty defense perhaps?

Originally posted by Michael Dorosh:

....a situation that would be interesting to see simulated in CM, actually - if ammo resupply was to become part of the simulation. Even if an abstracted version would be of interest.

I agree.

Perhaps supply tied to the company HQ location where ammo is distributed out form there. All units start with X ammo points, and as time progress ammo points begin to increase radiating outward from there, to reach some limit then of course decrease as ammo is expended.

Just thinking out loud, have no idea if this is realistic or possible to do in CMX2.

Link to comment
Share on other sites

Originally posted by Battlefront.com:

Yes, an anywhere LOS tool was much requested when CMBO came out. The majority making the request were Steel Panthers players and boardgames, since both allowed this sort of thing. Like others have said, this is unrealistic. Although we do admit that there is a bit of problem in that if a unit ends its move just a hair out of LOS of its intended target that the position can't be adjusted until the next turn after going trough a command delay. In some cases this is perfectly reasonable counter balance to the player putting that unit into too smart a spot in the first place, but in some cases it is not realistic. We don't think there is a reasonable compromise to be had.

The computational requirements of a "show all LOS" is too big. Remember that when you are moving around the LOS tool the CPU is only saying "can I see from this one specific point to that specific point". Asking the game to show "all points from one point" is asking it to do thousands of times the work all at once. It would be horrible. I remember even in Steel Panthers the "show all LOS" took time on my computer of the day.

Steve

You throw "LOS check from anywhere" and "LOS fan"/"show all LOS from here" into one basket here. These are different concepts.

The concern that a unit will stop one meter short of LOS is valid and fixing it fixes unrealism. It is a very common thing to tell a soldier or a tank to "go there until you can spot the enemy line". It is not overly fine-tuning the command mechanism, it is an integral part, much more basic than e.g. control when to run and when to walk.

Link to comment
Share on other sites

No, I never did and never would put a "Check Anywhere" feature in with "Check Everywhere". Earlier I addressed the Check Anywhere as being unrealistic, and threfore undesirable. Check Everywhere is computationally impossible, and tehrefore not possible for that reason.

The solution for both is much more realistic terrain and unit portrayal. Less abstraction, in other words. That is not the total solution, but it is a big part of it. Combined with other features I expect there to be no reason for a Check Anywhere feature.

Steve

Link to comment
Share on other sites

Originally posted by Redwolf:

Yes, less abstrated terrain where the player's view would actually judge LOS would make a "check LOS from any point" unneccessary.

That assumes the trees aren't abstracted. Maybe in CMx2 they are, maybe they aren't, but it's that issue in CMx1 that caused me the most annoyance wrt eyeballing LOS.
Link to comment
Share on other sites

Originally posted by Le Tondu:

6. How about using more than one CD so we won't hear, "Sorry, there wasn't any room left on the CD for......" ?

If they switch from using CDs to DVDs, there will be plenty of room for everything on one disk. But I don't know how the economics of production would be effected by the change.

Michael

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...