Reinforcement Systems (Scenario/Campaign)

[Guest Big Time Software]

OK, this is a continuation of the big debate about user controlled vs. scenario/campaign designer reinforcements. The thread has caused me, at least, to evaluate our campaign's reinforcement system prior to Beta. We were planning on waiting for feedback to give us some direction on it, but the lively discussion made us dust off some old plans sooner rather than later. So here is the current plan for the final thing...

----

SCENARIOS

Absolutely no change from the current system. It works really well, is historically accurate, and has a huge positive impact on gameplay. So far our Alpha testers Fionn and Martin have given it four thumbs up (Patrick I think has piped in with thumbs up too just now). That is 12 for 12 thumbs [NOTE: Rhet just pointed out a math error with this "12" figure, but I think the guys were using their Big Toes too, which explains the higher than normal count ]

There are five reinforcement "Slots", each of which can contain any number of different troops (from a squad to a fighter bomber). The designer specifies which gets what. Each Slot has three attributes:

Reinforcement Turn - the EARLIEST turn for the reinforcement to come into play

Reinforcement Chance - the chance that the reinforcements will come in at the specified turn.

Location - a specific spot on the map for the reinforcements to show up. There are no restrictions for where this marker can be placed.

A 50% chance on Turn 6 means that there is a one in two chance that they will show up on Turn 6. If they don't come in, then there is a 50% chance that they will come in on Turn 7. So on and so on until they come into play or the game ends. When they do grace the battlefield they go to their Slot's designated reinforcement area.

As for knowing and not knowing about what the player will get and when, this is left up to the scenario designer. Each scenario has a pre-mission "orders" screen for each player, which are written up by the designer. THIS is where the player learns, or doesn't learn, about pending reinforcements. If someone wants to see an example of what I am talking about, check out the "orders" that Fionn and Martin received. They are the first postings in the Battles section. In each I say what I want them to know (I could lie too!) and that is what the get for info. I could have said "you get x on y turn with a z chance" but it would have been more than they should have known.

In real warfare a Battalion commander, who is likely to be a lowly Major, might know that he is supposed to get a platoon of tanks and some infantry, but the "when" is generally quite variable. I mean, do you REALLY think that a Battalion commander should be able to set his watch by the arrival times of his troops? No way Plus, maybe he was expecting a platoon of tanks and instead got only 2. To do this, just lie when writing the report! "Major, expect a Company of Jagdtigers in 10 minutes" and instead give him a Kübelwagen with a drunk Hauptman 20 minutes into the scenario So instead of making some tedious gaming experience and lots of code, we leave it up to each individual scenario designer to deal with this stuff since no two battles have the same requirements for knowledge.

Replayability isn't an issue with any of this, even if reinforcements are assigned turn numbers with 100% chance of entry. Hard to explain why it isn't an issue to anybody who hasn't played CM for real, but it just isn't Plus, with all the options for free and random scenarios, why play the same scenario more than once or twice? Variety is the spice of life!

----

CAMPAIGNS

Here is where we are going to make a change. The idea has not been implemented, and probably won't for the initial Beta test, but will be soon. It goes something like this...

At the beginning of each battle, starting with the second, you MIGHT get some reinforcements based on your need, mission, and availability. This is determined, in large part, by the scenario designer and in smaller part by CM. The player has NO direct control over this at all.

The designer allocates forces for each side to have as reserves and puts them into up to five Slots (like with Scenarios). However, instead of specifying a turn or chance, the designer specifies their strategic level:

Battalion - units readily available to be used as reinforcements.

Regimental - units that can be used in a pinch, but generally aren't going to be released.

Divisional - you are going to have to be in a world of hurt before you get anything from Division!!

The designer can designate each slot as any of these things. To make reinforcements VERY easy to get, make all slots Battalion reserves. Want the side to really get the snot beaten out of it before getting help? Make 'em all Divisional. Obviously you can do any mix of designations you wish, but the slots will be called upon in order (i.e. a Battalion reserve is released, the first available slot containing such are given to the player).

The above only determines how likely the player will get reserves when they are actually "requested". So the other important part to have is something figure out the "when" part. Again, it will NOT be the user who decides this. Instead, CM will crunch some numbers and figure out if your force could use some help. If so, it will then see what reserves are available and decide if you should get them (Bat/Reg/Div decision mostly). If you do get them they will appear during the Setup Phase and will be restricted to whatever friendly setupzones you might have.

If you start up a battle, and think that there is no way you can carry forward, you can request to "skip" the current battle. If BOTH sides decide to skip, the battle is ended and the process of determining reinforcements is done again. A battle can also be skipped if neither side is doing much, if any, significant action after a variable length of time (don't worry, it takes moving into account, so this doesn't mean you have to blaze away to prevent the game ending). This allows, for example, the attacker to hold off and wait for more reserves. A battle ended in this way will likely increase the chance it gets reserves since it shows that both sides aren't feeling particularly happy with their ability to fight.

There you have it. Flexibility, designer control, variable chance, logical availability, historical realism, and SIMPLICITY. Should be a lot of fun too

Steve

[This message has been edited by Big Time Software (edited 09-03-99).]

[Fionn]

So, it looks like they get what they wanted BUT get it at the appropriate time (in between battles and not during battles).

One point Steve made which is very good is that usage of reinforcements etc is UP TO THE SCENARIO DESIGNER. If the designer is very poor then having all these options is of no use. A good scenario designer can weave a great scenario.

In the end what determines scenario quality isn't all the options and choices available to a scenario designer but, rather, their innate ability levels.

------------------

___________

Fionn Kelly

Manager of Historical Research,

The Gamers Net - Gaming for Gamers

[Mike D]

Steve,

Ahhh, it was a long "sec", but now I see what you meant. As for the other thread being locked down I did actually manage to put a post in after your message saying it was going to be locked down. I'm not sure if this was because I had an open Netscape window sitting here on my computer for much of the night ready to paste my latest in from MS word which I later used to submit my reply about an hour after your close down message, or what. But I don't think it is locked down right now (i.e. I don't see the lock symbol on it.)

Mike

aka Mikester

[Mike D]

This will be one of the shorter posts in recent Mikester history here on the board

I like it. I can live with the way reinforcements are being handled in the individual battles. Only thing I would still like to see here (and I can very well live without it) is to have the capability as the scenario designer to set varying probabilites for entry of a reinforcement slot allocation from turn to turn vs. the current same probability for every turn.

I like the fact that for the campaigns what you are proposing does have a certain degree of what I was looking for in that it captures much of the essence of what I thought should be in the game without giving the player direct control over such matters (yes you guys convinced me that battalion commanders didn't have much control over their reinforcemnt/reserve situation). Was quite surprised to actually have you put in all three reserve pools and even consider regimental and divisional type of reserves, but all the better. Will the scenario design guidelines give some general feel for us as to what the rough chances are of regimental and divisional reserves appearing in the game as determined by the AI so that we, as scenario designers, can make good decisions as to what units and how many we should put into these pools?? Overall, sounds like a very good way to handle things for the campaign reinforcements. I look forward to getting to try it out first hand.

Thanks for putting something like this into the game. I think what you guys have come up with will make the game a lot more interesting.

Regards,

Mike D

aka Mikester

[This message has been edited by Mike D (edited 09-02-99).]

[Guest Big Time Software]

Hehe... actually, I forgot to lock it down after posting Already read your post and locked it for real. [and you just slipped in another one well, I already wrote mine, so I will let it stand in case anybody else has doubts]

I think the problem you are having Mike is with REALISTIC knowledge, timeframe, and position of reinforcements. In real war commanders, no matter what their rank or experience, don't have some sort of train station timetable of what is coming in and where. Sure, sometimes they do, but generally they do not. And when they do, more often than not the "schedule" is worth more as toliet papper The battlefield is a chaotic place that spits in the eye of anybody that tries to have things work exactly as planed. Therefore, stuff just happens

As battlefield commander Martin, he knew that his forces were fighting through the town, or around the town (see the Orders I gave him as the designer of the game. They are posted on TGN). He also knew ROUGHLY when they were going to arrive (I left this out in Martin's report by mistake, but told him in an email to make up for it. Fionn was told roughly 15min I think). This is probably more info than he would know in real life.

These forces of Martin's weren't sitting on their butts waiting for the grace of Martin's command to put them into action. They were, in fact, in combat off map. Combat is a variable thing. Maybe they got clobbered and never come into play. Maybe 1/2 the men that were expected actually make it. By telling him something like "you are going to get reinforcement of roughly this and that around such and such time" I can make things REALLY realitic for him. And, more fun too. He was like a little kid getting a visit from St. Nick when he got his last batch

Lying (or simulating realistic problems) works too. I could have told him to expect a platoon of Shermans and in fact only give him 2. What happened to the other 3? Dunno. Breakdowns, enemy ambush, got lost, you name it. THIS is what we want to simulate in CM. Realism. That means NOT knowing what is happening 100% of the time.

The reserves in CM's scenarios are meant to be forces that could be available for the battle, but aren't quite there yet. This happens ALL the time in war. The attack starts at 0500, rain or shine. If a couple of TDs aren't in position, it is unlikely the attack would be held up unless they were the central part of the attack (i.e. the spearhead). War can not sit around and wait for everybody to get their act together, so we certainly aren't going to have CM model a perfect world that doesn't exist.

For campaigns, reinforcements are those elements that are released for combat because of some need or future plan. As stated above, we have this system laid out really nicely. But again, the battlefield commander only knows what he is told. If Division want's to release forces, it isn't as if the Battalion commander knows 4 hours in advance what he might have or not have. This simply isn't realistic. So the campaign designer could write something like "Battalion has a company of infantry in reserve and some assets from divison on the way". This would be perfectly normal to find in a report. Why? Because very often the higher ups don't know what the heck they are going to send either. Sometimes they SAY they are going to send stuff, and in fact never do (Hitler LOVED to play these kinds of "jokes" on his commanders).

We feel we have CM's realism AND gameplay bases covered. Also really easy to use too, which is never a bad thing!

Steve

[This message has been edited by Big Time Software (edited 09-02-99).]

[Doug Beman]

Okay, a question going back to your description of

"A 50% chance on Turn 6 means that there is a one in two chance that they will show up on Turn 6. If they don't come in, then there is a 50% chance that they will come in on Turn 7."

Too much to ask to have a system of increasing percentage? I think this might simulate reinforcements that are trying to get to the scene and WILL there for sure by some certain time, but COULD get there earlier (simulating they had an easier time getting to the scene)

If that's too much coding, or my logic for having it included is flawed, then the set percentage will suffice.

DjB

[Pixman]

I like it. Good job sticking to your guns Mikester. I admire you because you are the only one here who makes my posts seem brief and to the point, lol.

Steve, I have to chime in with Mike and Doug on the variable arrival probabilities, with all of the caveats that Doug put in at the end of his post.

To hit on one of Mike's biggest themes to date -- don't slow down the project for scope creap (paraphrasing) -- how much of a release date hit are we going to have because of these enhancements? Don't get me wrong, I think they are worth doing. Just curious as to the trade-off.

Once again, great job of listening to your public and keeping their best interests at the forefront.

------------------

The enchanter may confuse the outcome, but the effort remains sublime.

[Dar]

I disagree with the need for varying percentages with each turn. If you set the percentage of reinforcements arriving on a given turn to something low like 25%, then obviously, there is a 25% turn they will arrive on that turn. If not, there is a 25% chance they will arrive the next turn, or the next turn if they fail again, right?

However, if you look at it in sequence, there is actually a 25% turn they'll arrive on turn X, a 44% chance they'll arrive on turn X+1, a 58% chance on turn X+2, 69% on turn X+3, etc. (Take 100% - (the chance of reinforcements NOT appearing) ^ (the number of turns failed)) In other words, (1.00 - (.75)^1) for turn X, (1.00 - (.75)^2) for turn X+1, etc.

Therefore, you could say that reinforcements will most likely show up on turn X+2 (>50% chance), although they may show up earlier or later, as would seem quite realistic to me.

Really, a lower variable of 5% might be more "fun": 5% on turn X, 10% on X+1, 14% on X+2, 19% on X+3, etc. Then you can really keep 'em guessing!

Dar Steckelberg

[tss]

I'd like to add a new angle to the talk about campaign reinforcements. Most posts have centered on giving reinforcements to the side that is losing battles, but in some occasions it would be realistic to add them also to the winning side. For example, if the attacker manages to decisively crush a defence line the high command could then allocate a tank batallion or two from reserves to exploit the breakthrough.

At least Russians practiced a lot of "violent recons" (I don't know the correct English term, this is a direct translation from Finnish "väkivaltainen tiedustelu") where a lot of company or batallion sized attacks were directed against German lines to probe weaknesses. In few cases the recon probes actually resulted in breakthroughs and then Soviets were usually quite quick in sending all available T-34s through the resulting gap.

-Tommi

[Mike D]

Dar,

I agree w/ you on your probability anaylsis. However, I think what I and the others are looking for is some better control on what I'll call the "ramping factor" of the probability of the reinforcements showing up as the game progresses. For example. Say starting turn 10 I want to give a 5% chance on that turn and the following turns through turn 15 of a given set of reinforcements showing up. For turns 16 and 17 I want it to change to 15%. On turns 18 - 20 I want to go up to 40%. On every turn thereafter I want to increase it by 10% each turn so that if they still havn't shown up by turn 26 I'm guaranteed that they will show up by that turn since the probability is then 100%. This allows for two things that the current system won't. It gives the scenario designer the ability to have probabilities for the guys showing up, but it also provides the ability to be able to guarantee that they actaully will show up by a certain turn as well. Secondly, it gives the designer a greater degree of control of the cumulative probability of the guys showing up on any given turn vs. the existing system. If you run the numbers on the example I gave and then compare them to just assigning a single probability value of say, 5% for turns 10-26, I think you will agree you are going to get some different results doing it one way vs. the other. Such a system would give you the ability to set the probability really low over a period of turns to "keep them guessing as you say", but then be able to ramp the probabilities up to ensure that the poor bloke will more than likely actually get the reinforcements eventually; just a little later than he would have otherwise unless he got lucky and hit one of the low probabilities on one of the earlier turns and they came in. Such a system would also allow you to do some other interesting things like actually decrease the percentages of the guys showing up if they didn't make it in earlier. Not sure why one would want to do this, but it might be an interesting thing to try out.

Regards,

Mike D

Aka Mikester

[This message has been edited by Mike D (edited 09-03-99).]

[Dar]

Ah, well, to each his own. Personally I like the idea of no guarantee that reinforcements will show up. I'm just that chaos-loving kind of guy!

Dar Steckelberg

[Mike D]

Dar,

As you say to each his own. Great, if you want to not be able to guarantee a given set of reinforcements to show up just set the probabilities to never have a turn w/ 100%.

As for me, in a given scenario design I might want to do what you propose, but then in another one, I might want to be able to guarantee that certain units would show up by turn x if they hadn't previously come in. What we are looking for here is some additional flexability to the system. If you don't want to use it, then fine, you don't have to. But for myself, and at least several others, I think we would like to have these kind of tools available to decide for ourselves how we are going to control the reinforcements. If I'm going to play "God" as the scenario designer then gosh darnit I want more power! The game would be fine "as is" without this, I just think it would be nice to have the capability is all.

Mike

[Pixman]

To add my two cents in support of Mike here (gulp, did I say that?), I think the key use of variable arrival probabilities for me in designing a scenario would be when tweaking a bit of "what if?" into a historical battle. For example, in a purely historical battle, if the designer has good enough info., he can totally prescribe the arrival of forces using 100% throughout. However, I might want the historical to be the worst case scenario, with a chance that some forces arrived a little early. In that case, I might use some 20%, 30%, 40%, etc. on earlier turns and plunk down the 100% on the historical turn.

No great shakes, just trying to point a good use for this flexibility, without which we cannot do what I am suggesting.

------------------

The enchanter may confuse the outcome, but the effort remains sublime.

[Moon]

Mike, since the other thread is locked and I was away yesterday, I'll use this one for a short reply.

<BLOCKQUOTE>quote:</font><HR>Martin,

One thing that bothers me for some reason is one of your comments (don't take this the wrong way, I'm just puzzled is all). In remarking on the system that I'd proposed you said that if you had forces only 10 minutes away from the battlefield that you would have had them called up and ready for the start of the game in real life. Martin, if that is the case then what do the reinforcements that are showing up 5, 10, and 15 minutes after the start of a CM game showing up for under the current reinforcement system doing there???? I don't see the difference here between what the game already does and what I had proposed in this regard?????<HR></BLOCKQUOTE>

First, I am not taking it the wrong way at all. I trust that you don't, either. In fact, I guess it's my fault for not making clear what I ment.

There is no problem with reinforcements arriving 10 minutes after the start of a game. Or even 1 minute. The problem is with the time between calling for reinforcements and getting them. In CMs current system, the reinforcements that arrive 10 minutes after the game started, might have been called two days ago. This is even true if there has been no sign for an attack yet - somebody somewhere simply decided that this are here needed some extra punch.

I hope that makes it clearer what I ment.

[This message has been edited by Moon (edited 09-03-99).]

[CoolColJ]

Hi all

I haven't said anything on this particular thread, so I'll now happily air my voice!

I agree 100 % with Mike and PIxman's previous posts, having scenario designer variable percentage abilty, will allow historically accurate campaigns, in which the arrival of certain units were vital to the flavour of the battle being fought. This shouldn't be difficult to do, all we need is a percentage field - something as simple as: Start turn no.; Start turn %; End turn no.; End Turn %. The computer can then ramp the % chance between the start and end points in equal increments.

This is the simple method I wouldn't mind seeing, or we could use: Start turn no. / %, Median turn no. / %, and End turn no. / %.

The Median turn no. could be the highest % point turn the unit should enter historicly, if it doesn't enter then, then the computer can keep checking until the End turn / % parameter is reached which may either have a lower or higher % than the Median turn no. The computer interpolating the % chance smoothly between the 3 % points.

This could also be applied to battle no. in conjunction with turns within a battle of a campaign

Phew

CCJ

[This message has been edited by CoolColJ (edited 09-03-99).]

[This message has been edited by CoolColJ (edited 09-03-99).]

[This message has been edited by CoolColJ (edited 09-03-99).]

[Mike D]

Moon,

I'm not offended at all by any of your comments. I was simply puzzled by that one is all. And it probably had as much to do with my interpretation of what I thought you meant as anything else. In any event, now that you've explained in a little more detail what you meant I understand perfectly. Thanks.

Regards,

Mike D

aka Mikester

[This message has been edited by Mike D (edited 09-03-99).]

[ChromaTick]

BTS,

Here is another vote for the sliding percentage ability..if it's not too difficult to code in. Otherwise, keep up the great work!!

ChromaTick

[Guest Big Time Software]

No to increasing variable percentages. From a realism standpoint we aren't trying to simulate sure bets, but rather highly variable ones. Yes, we COULD add in extra features to allow both, but good Lord, how many kitchen sinks does one game need?

As Dar stated (using math that makes my head spin ) you CAN make sure that statistically the chance of arrival is 100%. Just don't make the percatage really low and late in the game, or give a low percentage and set the first possible turn very soon into the game (like Turn 1, 10% chance). The fact is that CM already gives you the chance to have stuff be totally predicatable. Either give it 100% and pick the turn, or pick a mid game turn and make the percentage more on the high side, or pick an early game turn and make it low. It is all there for you.

The problem with variable increasing rates is that it limits total wildcard reinforcements. If I want something to pop or possible not, a n increasing percentage ruins this. As it is now I can have a platoon have a decent chance of never arriving. Personally, we think that is very cool. If you don't, then set your scenarios up with higher percentages.

And guess what guys... since we are talking random chance, if we did it with sliding percentages, how would you know the difference? You would need to know all variables and results from probably 100 or more games, for each system, to even have a hope of telling the difference. But one system allows total wildcards easily, and the other doesn't.

Tommmi, the system does work for both attacker and defender. If you want the attacker to get LOTS of reinforcements easily, just fill up the slots and make 'em Battalion level.

Steve

[fatherof4]

Good for you Steve. If people in favor of "varying probabilites" can post more than once, those of us who think it's needless chrome can do the same.

PLEASE keep this game on track. Otherwise, we wouldn't be able to play it until after Christmas.

[Fionn]

One thing about feature creep is that it is not a noticeable thing.

You add 1 minor feature in one area taking 4 hours, another in another area taking another 4 hours.

After just a few days of creep game release is pushed back a week or two...

FWIW I've used the reinforcement system in CM and I like it.. Having battalion, regimental and divisional-level reinforcements adds spice to post-battle reinforcing and it would actually slot in beautifully into a meta campaign system. THANKS guys.

However, I just don't see the HUGE advantage variable reinforcements would have to have to justify all the coding etc.. What HUGE advantage does it have over the present system?

------------------

___________

Fionn Kelly

Manager of Historical Research,

The Gamers Net - Gaming for Gamers

[Mike D]

The feature creep thing can be a big problem. I hope it doesn't get to be here. Steve already well knows (and anyone else that can remember) that long ago I'd written a number of post regarding the evil that lurks in the shadows. I'm sure these guys know better than most what they are doing though. However, like I told Steve back then, at some point you have to lock down the configuration of the game (i.e. no more changes), play test it, fix what needs to be fixed, and then get it out the door. You can't keep fiddling w/ it forever, making this little change, then that one, and so on, and so on, and so on. It will kill you in terms of ever getting a release out the door. I have a feeling that we are very quickly approaching the point where Steve and Charles will decide to do this (i.e. Beta is coming up real soon). So I'm not too worried about the progress of the game and imagine they will get it done and in our hot little hands soon enough (God, did I really say that???).

Regards,

Mike D

aka Mikester

[Guest Bobb]

I am not so sure but what I have grown

fond of waiting. It sort of creeps up

on you. One day all the anticipation

will be gone. Then what? Getting

kicked by the AI or in PBEM. By the way

maybe Steve et al should considering

selling tickets to battles as being now

waged by Fionne and Moon. This has been

as involving as playing it myself. Perhaps

replace pro football. That would be a

worthy side effect alone. (dodging brickbats with screen shakes!!!)

[Fionn]

Bobb,

One thing that I think is going to be popular is going to be downloading saved game files from either BF.C or TGN which you can plug into your copy of CM and then watch the ENTIRE battle unfold.

Imagine checking out your next opponents playing style in a tournament by watching the "tapes" of his last three battles or of watching the finalists in a CM tourney battling it out in a 90 minute-long battle on your monitor.

Hell, a little recording equipment for voiceovers and you need NEVER turn on the TV again

I have lots of ideas for the TGN CM site as you can see but I'm not going to really float them until the demo comes out and people can seriously commit to taking part in the various things.

------------------

___________

Fionn Kelly

Manager of Historical Research,

The Gamers Net - Gaming for Gamers

[Greg Deych]

Re: Reinforcement slots.

Would 5 slots be sufficient for a campaign?

[Guest Big Time Software]

Hopefully it will be, since an expansion of the slots would be difficult for us to do at this point. Not impossible, but ugh

5 Slots is about one reinforcement per battle (we figure average will be around 6 battles). With the new variable reserves there is a good chance that you won't get them all the time, so it should be OK.

Steve