Home Store Products Research Design Strategy Support News

21 June 2006

 Products Napoleon’s Triumph Design Diary Romancing the Stone

Napoleon’s Triumph Design Diary: Romancing the Stone

Map Board
Click on the image above to open in its own window
Click on the image above for the latest rules draft

In the last diary entry, I said to expect the next entry soon. And for once, by soon, I actually meant what a reasonable person would expect. (“That word you keep using; I do not think it means what you think it means.”) The purpose of this diary entry is to review the various changes that have been made to the game since *cough* December of 2005. There really haven’t been any major changes; the design process has been largely focused on refining and smoothing the game design, not on fundamentally re-working it.

An examination of the mapboard may not at first reveal that there have been any changes at all, but in fact close examination will reveal otherwise. The most important differences between this map and the previous one are the many changes to the objective locales. The changes to the objectives on and behind the Allies are particularly extensive. There are now more objective locales, they now come in three colors rather than two, and objective locales have been added to the Allied right. These changes were made for three reasons: (1) to open up the game more as the previous objectives tended to encourage a rather unvarying line of Allied play once they switched to the defensive, and (2) to force the Allies to pay more attention to protecting their rear, and (3) as part of a set of changes to improve play balance. As for the other map changes, readers so inclined can amuse themselves by comparing the current map with that of the December 2005 diary entry and seeing how many differences they can spot.

The rules were in pretty good shape even six months ago, but any text can always be improved by the expenditure of more labor and it was no different here. My biggest concern with the previous version was the 20-step combat resolution procedure. It wasn’t that it was complex: all the steps were really pretty simple and I had no doubt that people would be able to understand it. The problem was that it defied memorization. I knew I was in trouble when I found it desirable to add a play aid to the map that listed the steps of combat resolution; the very existence of the play aid amounted to an admission of failure in an important part of the design. I don’t think I ever articulated this, but I keep it as a general design goal that experienced players should not have to consult the rules during play; it should be possible to internalize everything in them. If even I was having trouble keeping track of the combat procedure without a cheat sheet, that meant that this design goal had certainly not been met.

I felt that the problem with combat resolution was mainly one of presentation. The number of steps needed to be reduced, and they needed to be presented in such a way that players could understand the logic driving the sequence. To accomplish this, it was necessary to collapse steps together where it made sense to do so, even though that introduced a certain amount of redundency to the rules. The current procedure is as follows:

  1. Attack Threat
  2. Retreat Option
  3. Feint Option
  4. Defense Declaration
  5. Attack Declaration
  6. Initial Result
  7. Counter-Attack
  8. Final Result
  9. Attacker Losses
  10. Defender Losses
  11. Completion

Steps 1 to 3 are the preliminaries: the attack is initiated and both sides are given a chance to end it early. If both sides elect to continue, then the heart of the procedure follows with steps 4 to 10, and the attack ends with step 11. The internal logic can be seen in more detail by color-coding the steps according to which is the “active” side in each step:

  1. Attack Threat (Attacker)
  2. Retreat Option (Defender)
  3. Feint Option (Attacker)
  4. Defense Declaration (Defender)
  5. Attack Declaration (Attacker)
  6. Initial Result (Calculation)
  7. Counter-Attack (Defender)
  8. Final Result (Calculation)
  9. Attacker Losses (Attacker)
  10. Defender Losses (Defender)
  11. Completion (Both)

With the color-coding, the back-and-forth symmetrical structure becomes clearer. After the “Attack Threat” step, there are two “Option” steps, one for each side, followed by two “Declaration” steps, one for each side. The “Counter-Attack” step actually has a conceptual symmetry with the “Attack Threat” step, even though they are not sequential, which may become more evident when you consider them in their relation to the pair of “Result” steps: “Initial Result”, which is the result of the “Attack Threat”, and “Final Result”, which is the result of the “Counter-Attack”. There procedure winds down with a pair of “Losses” steps, one for each side, before the final “Completion” step. The “Completion” step is not symmetrical with any other step in the procedure, but it has an internal symmetry in that it contains two procedures within it: one for an attacker win and one for a defender win.

I don’t expect players to necessarily be consciously aware of this structure, but when experienced it has a natural “you-go-I-go” flow that I believe will enable players to be able to internalize it without difficulty, even if they could not say why it is that they are able to do so. This is a major playability advance over the December rules, even though there isn’t any actual reduction in the number of rules covering combat.

There was a second combat problem that while not critical, did bother me, and that was the relationship between artillery and high ground. I had improved on the Marengo rules with the December draft of the Napoleon’s Triumph rules, but still thought that it could benefit from some more work. The current rules contain additional revisions in that direction. The artillery rules allow artillery to lead attacks, but they can lead them every turn from a hill locale, but only every other turn anywhere else. In addition to this change, the penalties for artillery attacks into hill locales have been largely removed as they tended to make every army like the British army at Waterloo: using reverse slopes to protect themselves from enemy artillery fire, which is not really desirable.

The scenario shake-out is the next substantial rules change worth discussing. In December there were three scenarios left, each with a French Attack and Allied Attack variant. The French Attack variants have been dropped. (The remaining scenarios are based on the original Allied Attack variant, though they no longer have that name.) Also, the 1 December 11:00AM scenario has been dropped, leaving just two scenarios, 7:00AM 1 December and 7:00AM 2 December. As it happened, the 11:00AM scenario just didn’t add that much in the way of play interest that wasn’t better presented in the 7:00AM scenarios for 1 and 2 December. The two scenarios that are left, however, I think are well differentiated and have more than enough interest to justify their inclusion in the game.

The end-game design is another area which has received a significant change. A problem with Bonaparte at Marengo as a game was its tendency to result in low-key endgames. This is partly because of my dubious decision to keep the game going after one side was demoralized even though they no longer had any real chance of winning the game, but it also was partly due to the fact that experienced players would tend to avoid attacks where the loss differential would be greater than one (if the defender felt he was going to lose more, he would withdraw into reserve and take a one point loss when forced to retreat, while if the attacker felt he was going to lose more, he would limit his attack to a two-step unit to inflict a one-step loss). Although the game system permits attacks to have much more dramatic outcomes (and players with any significant number of games under their belts will tend to have such attacks permanently etched on their memories), such attacks are much more the exception than the rule in Bonaparte at Marengo.

The potential for drama in Napoleon’s Triumph has been heightened in part by the more violent and unpredictable combat system, but in the end-game in particular it is heightened by the fact that an army’s morale level cannot be reduced below 1 in an attack if it is the winner in that attack. Suppose for example, that I have a morale level of 2 and my opponent has a morale level of 6. I can afford to make an attack that costs us both 3 points of losses, so long as I win it. Such an attack would only reduce my morale to 1 (I win, so my morale cannot go below 1), but it would reduce his morale to 3. If I can make a second such attack, again inflicting 3 points of losses on both of us, and win again, I can demoralize him by driving his morale to 0 while my morale remains at 1. This feature of the game combined with the presence of units on both sides with heavy offensive striking power (Guard infantry and heavy cavalry are particularly important in this regard), adds a lot of drama of the end-game, and does so in a way that I think makes historical sense.

The Guard unit rules have always been a source of trouble in the design. They have never been a big problem, but I have constantly found myself thinking as to whether the Guard rules really work the way I want them to. There were three models of Guard usage that I really didn’t want to see in the game, and which I constantly worried would be effective due to bad design on my part:

  1. The Guard as WWII Armor. In this model, the Guard is committed at the start of the battle to break through a weakpoint in the enemy lines, and spread havoc by leading a drive into the enemy rear.
  2. The Guard as Viet Cong. In this model, the Guard lies in wait next to expected enemy lines of advance, and then springs out in ambush to take the unsuspecting enemy in the flank.
  3. The Guard as Vultures. In this model, the Guard infantry roams the battlefield, looking for units so weak that they can be picked off without risk of loss to the Guard.

Rules that strengthen the Guard against strong units tend to produce the first model (Guard as WWII Armor). Rules that make the Guard easy to conceal tend to produce the second model (Guard as Viet Cong). Rules that strengthen the Guard against weak units tend to produce the third model (Guard as Vultures). My fear has never been that I thought that there was no solution to the problem of these three tactics, but rather that in trying to keep the number of special rules for the Guard to a minimum, I might make a mistake and one of these three models would be the result. The December rules were one effort to get the results I wanted, but it was hard to be happy with those rules: they were not very historical, they were middling complicated, and they posed trust issues. The current rules represent a new combination of other rules that I had tried before and discarded in various iterations of the design. I think the new rules will do the job, and do it without the issues of the December rules.

The territorial objectives for marginal victory conditions are unchanged in basic concept (two sets of territorial conditions based on whether or not the French bring on reinforcements) but differ substantially in their particulars. There is quite a lot that could be said about this, above and beyond the brief overview of them made in the context of map changes, but I feel that I should avoid going into this subject as it leads me down the path of discussing game strategies for the two sides, and as I much prefer not to do that, I will just note for you that the changes exist, and move on to the next topic.

Optional rules are something entirely new to this draft. They had been considered many times during the design, but they never actually made it into the rules. One reason I finally decided to go ahead with them is for the simple reason that there was a blank page left in the rules book. Because of the way booklets are bound, they pretty much have to have a page count that is an even multiple of 4 (4 pages, 8 pages, 12 pages, etc.) I ran out of rules on page 9, leaving pages 10-12 available for other uses. Page 12 is the back cover, which I prefer to leave blank as it tends to get scuffed. Page 11 was intended for use by the design notes, but page 10 was up in the air. There were several things I could have done with it (expand the design notes, provide historical material, add player’s notes, etc.) but I thought I had some interesting ideas for optional rules and so I decided to go with those.

The Game Balance optional rules include an auction system that will be instantly familiar to the participants in the online Bonaparte at Marengo ladder or tournament games. While I hope that game is balanced without this option, the fact is that a system that allows players to change the balance of the game without wrecking it (as messing with the territorial objectives, for example, is likely to do) is extremely useful. Adjusting the game balance has been a frequent subject of on-line discussion on Bonaparte at Marengo, and I thought it a great advantage to include information on this subject in the printed rules so that even players who never go online can use it (any rule also has a greater feeling of subjective “official-ness” when it is actually in the printed rules).

The Santon special rules have been in and out of the rule book almost since the beginning of the design. Every history of the battle comments on this terrain feature, but actually it mattered very little in the historical battle. There was a minor Allied flank incursion that was driven by the French forces on the Santon, but the French might well have accomplished the same thing if the hill hadn’t even been there. The way to split the difference between the fame of the feature and its general importance in the game was to include Santon rules, but to make them optional, and thus here they are.

The Team Play rules were, without a doubt, some of the most entertaining rules I’ve ever written. Their length makes it seem that they add more complexity to the game than they really do, but an examination of them will reveal that they are mostly just a codification of what would be pretty obviously needed changes to make the game work for team play. Bonaparte at Marengo may have a very solid solitaire version, but it is a game that adapts extremely poorly to team play. In contrast, Napoleon’s Triumph is a game that (so far) has no solitaire version, but which is remarkably well-suited to team play. What really makes the team play rules go are the communication restrictions between players on the same team, and the ability of the more “senior” players on a team to make their junior team-mates lives hell (with or without meaning to). Readers will note that the communications clamp-down is actually stronger than the needs of historical simulation might seem to require, but this is an intentional over-compensation to deal with the fact that players in the game know more about what is happening in distant parts of the battlefield than their historical counter-parts did. Without such strong communications limits, there would be nothing to prevent one player from micro-managing the moves of another player, in a manner that is not only ahistorical but which also destroys the entire point of team play. As it is, the players on a team really do depend on each others’ judgment in a way that I think really makes team play a genuine team experience.

Design notes are the last new addition to the rules. While I doubt there is much to them that is really new to the readers of this diary, they do provide a high-level perspective on the design as a whole, without all the detail of the design diary, which can by its nature obscure what I think is really important by the sheer verbiage devoted to lesser (though hopefully interesting) topics. In this diary entry, I have the interesting opportunity to offer a meta-commentary by discussing not the design, but the notes on the design, and so I’ll take advantage of it.

One thing I try to do in game design notes is to keep the discussion at a pretty high level. I don’t want to be dragged into topics like the details of the order of battle, the terrain, or the historical course of the battle. It isn’t because I don’t care about these things (I obviously do) but because that is only one part of the larger design picture; there is a conceptual design hierarchy here: “game” / “wargame” / “Napoleonic game” / “Austerlitz game”. By keeping the discussion on a higher level, the design notes are able to address the problems that arise at all of these levels, which they could not do if they over-obsessed about the design problems peculiar to Austerlitz.

One question that did come up for the design notes was the matter of whether they should repeat or build on the notes for Bonaparte at Marengo. I opted to build on rather than repeat, because I was worried that repeating the earlier notes would bore those who had read those notes already and leave myself with almost no space to discuss anything new. There is some risk in this decision, in that readers who never read the notes for the earlier game could get an overly-narrow view of my design priorities, but I hope that the repeated references to Bonaparte at Marengo in the notes will clue readers in that this game is a second step in a larger design process, and that the earlier game can be profitably studied by those who would like to know more about the design of Napoleon’s Triumph.

In closing, I would like to say that I am confident that the version of the game you see in this update is very close to what will be the final published version. As I have in the past, however, I invite anyone who cares to do so to proof-read the rules and map. Feedback is welcome (comments can be sent by email to or posted to the Simmons Games thread at talk.consimworld.com). Although I do not generally write personal responses to proofing feedback, please do believe that I read it all, and that many of the changes that have gone into prior drafts of the game materials have been the direct result of what you all have had to say. I thank everyone for their help, and I also thank them for their patience. I am as aware as I could be that it has been a longer process than everyone has hoped for, but I think the result will be a very good game indeed.