Duelmasters: Hoser Report 10 From: leeme@cnsvax.uwec.edu Date: Tue, 01 Jun 1993 18:13:13 +0000 The following is an issue of the Hoser Report. Jeff Morgan ran the Hose Machine in Duelmasters a long time ago. I was lucky enough to know him and learn correctly how to design characters after playing DM for a couple of months. Please keep in mind this is 1987 material, stat/skill tables were just being started and no one really knew yet that skills were lost if raising stats before getting 20 skills in a skill area. Jeff Morgan dropped out of DM right when I entered ADM and he supposedly was going to start up his own PBM game/company (Eudaemonic Enterprises). --Mike Lee THE HOSER REPORT The strategy newsletter for Duelmasters c 1987 Eudaemonic Enterprises All rights reserved #10 October 6, 1987 The Managers Corner The Not-So Perfect Warrior (Gee Honey, not Leftovers Again) I think everyone is in agreement as to what makes the super warrior, the future Duelmaster and Tower Guardian. However, that is like knowing (believing) that the Jaguar XKE is the ultimate car; most people end up driving $5,000 imports. What do we do with the vast majority of the downright awful roll-ups? The usual response is the Dark Arena... and for those "must win at all cost" types, that is probably the only answer. But I started playing Duelmasters because I liked the concept and for the brief amount of time I'm reading the fight results, I'm in a dark smelly underground catacomb, sharpening my sword or gritting my teeth as a teammate sets a broken bone. It goes against the feel of the game to DA every below average player! When I received my first team roll-ups, I also got a "terrific" roll-ups guide (golly TWO whole pages!!) which offered such insights as "... consider the type of character Large and Strong, or Small and Fast... attributes will affect your characters ability to fight... ability to carry weapons and armor...". Blah-de-blah. "Big help!", I thought. But wait a minute, lets look at those statements in connection with the statistics and the fighting styles. The first is ok if all you want is a basher or a wall of steel, and the second makes a great parry riposte or total parry (old rules). But what about Large and Fast or Strong and Small? I think that all too often, we get into a "tunnel vision" mode of character roll-up. Just got to get that wit/will combo or he's no good syndrome. Why not look at the character and see what can be enhanced rather than forcing a fit. Small ST? Only need a 5 for the EP or SH. And if 12 is an average score (84/7) then you have 7 points to go elsewhere... like WT or DF. And then pick the style that will take advantage of the low ST (PRP maybe?). He won't ever knock anyone down or bash through a shield, but he just might be able to put his little sword into his opponents eye with the speed of a striking snake! One final thought, an example of what I've been rambling about. My first Duelmaster was destined for the DA after his first fight (yep, I'm one of those w-a-a-c types). Here are his stats: ST CN SZ WT WL SP DF Basher 16-6-0 11 7(1) 14 17 9 17(1) 12(1) He won 14 of the 16 fights in the 1st minute, the other two in the 2nd. He lost two of the 6 in the first minute, the other 4 in anywhere from 2-7 minutes. Gave him just enough ST to handle a decent weapon and the rest to WT so he'd out-learn his opponents fight for fight. I hope this has stirred the old feelings of excitement and challenge in character design. The arenas need more unusual character types out there. Mike LaPlante HOSE KNOWS Questions and Answers Q: You mention in #6 about a luck number (0, 5, 10, or 15) being part of character generation. I wasn't sure if you meant that the number was computer generated or assigned by the operator. Also, do you know if it can be changed later? A: The number is computer generated. I do not know if it changes, my guess would be that it doesn't. If my guess that it adds 0,1,2, or 3 skills to each skill area is correct, then a change in this number could be detected. If it decreased, you may notice a character achieve the same expert (or adv. exp) status twice. Q: Do certain styles tend to learn at different rates (all else being equal)? A: All else being equal, there is no difference in learning rates between styles. Note the key phrase "all else being equal". As you know, some styles tend to have longer fights than others. Fight length is the most important factor in learning that the manager can (usually) control. Q: Is a SP 9 lunger who just achieved expert rating in initiative just as fast as a SP 11 lunger receiving the same (all other relative attributes being identical)? What I'm asking is this: Does an attribute such as SP really make a warrior faster, or does it make related skills such as riposte and initiative easier to learn, and thus being able to get more skills in those areas or getting to advance expert faster? A: What? I think you may be comparing apples to oranges. Starting at the top, I assume you're defining "fast" in terms of initiative since we're talking lungers. Yes, the SP 9 lunger is currently as fast as the SP 11 lunger since they both reached the same level at the same time (the maximum potential of each may be different). Remember, SP is just one factor that influences initiative. Suppose the SP 9 lunger next turn learns another another initiative skill, while the SP 11 gets no skills. They are no longer equally fast, SP 9 lunger is faster by one skill. SP will make a warrior faster, but so will SZ, WT, and style. The only thing that make skills easier to learn is WT (and maybe WL). Q: You mentioned WL should be increased to an even number. Why? A: Because, if you increase to an odd number, you may hit a break point. Gladiators can learn up to 20 skills in each area (used to be more before th program was fixed), unless the gladiator in question has been raising attributes. If you have a very experienced fighter in AD, you probably have already noticed this. The more attributes you raise, the less skills are possible. When Gladiators became Duelmasters (getting back to ancient history now) it was Ed's (the game designer) intention that a gladiator could overcome his 20 skill "maximum" by raising attributes after reaching 20. However, raise before 20 and it just counts as normal skill value (which is why raises sometimes trigger expert ratings). When I was working at RSI I asked Chuck Kraver about this on many occasions, and the learning program was to be programmed to do this. Ex: Two identical lungers, 11 WT. Lunger A never raises attributes. Eventually lunger A tops out at 20 attack skills, and is 100% better than originally (20 * 5). Lunger B raises WT to 13, then tops out at 19 skills. Since raising to 13 is equivalent to 1 attack skill, lunger B is 100% better (19 * 5 + 1 attribute increase). What is the difference? Lunger A can still go to 13, and will be 105% better. Lunger B has already "wasted" that increase. Note that B may be able to get to WT 15, or raise ST, WL, or DF. But so can A. And A will always be 5% better in attack. I don't know if this actually go programmed, we've all heard about enhancements that never came about. Keep in mind you can go for years in real time before a gladiator can accumulate 20 skills in a single area. Is being "perfect" worth the wait? HOW DOES IT ALL FIT IN? Since the beginning I have written about the learning system and character design. I could probably deal with the fine points of skills for another 10 issues, and still have more to say. A well designed character with a good number of skills can get by with almost anything on the strategy sheet. Yet the ability to read fights, and the strategy sheet have been neglected. Next issue we'll shift gears and take a closer look at thes factors. Lets start at the very beginning, with the replacement roll-up. How exactly are the initial skill levels determined? First, we must have a place to begin. Lets assume for now that the character is a total zero. Zero attack %, riposte %, everything. How does the roll-up program transform this into a new gladiator? Consider the following: 0 + (style modifications) + (attribute indications) = new gladiator Zero is our arbitrary starting point. We may wish to go back and change it to fit observed facts, but it does seem a logical place to begin. Style modifications. As you may have guessed, each style has certain bonuses and penalties. For example, the PRP gets a substantial boost to riposte %. The TPS gets a parry boost. While much of this is intuative, we do not know for certain what declaring a particular style does to a roll-up. However, by setting it up in terms of an equation, we can solve for it when we know the other values. Attribute modifications. This we have the ability to learn by using the golden rule of 4 so thoughtfully provided by RSI (see HR #7). Imagine having a table for each attribute. Essentially, this is what the roll-up program has. It looks at each attribute, determines the effect on each skill area, and presto. By using the golden rule of 4 we can reproduce the information the roll-up program has. Consider the following hypothetical tables for WT: 3 5 7 9 11 13 15 17 19 21 ATK +5% +5% +5% +5% +5% +5% +5% +5% +5% +5% DEC +5% +5% +5% +5% +5% +5% DEF +5% +5% +5% +5% +5% +5% +5% +5% +5% INI +5% +5% +5% +5% +5% +5% +5% +5% +5% PAR +5% +5% +5% +5% +5% +5% +5% +5% RIP +5% +5% +5% +5% +5% +5% 3 5 7 9 11 13 15 17 19 21 ATK -10% -5% -5% -5% +5% +5% +5% +5% +5% +10% DEC -5% -5% -5% +5% +5% +5% +5% DEF -5% -5% -5% +5% +5% +5% +5% INI -5% -5% -5% -5% +5% +5% +5% +5% +5% PAR -5% -5% -5% -5% +5% +5% +5% +5% RIP +5% +5% +5% +5% +5% +5% Which do you think is more workable? Lets suppose that we are interested in maximizing the attack percentage of a new basher. The natural WT is 9. Using the first table, the contribution of a 9 WT would be 20% (4 skills). Now, assume we dump 6 attribute points to WT. The contribution would rise to 35%. For WT alone. Don't forget that ST, WL, and DF will add to attack (because everything is a plus in this model!). And, the fact that he is a basher should give him another attack bonus. So, using the upper table it is quite possible to have even an average basher with an initial attack percentage of say, 150%. Sounds a tad high to me. That brings us to a minor obstacle, that of knowing "how much" percent is right. I think the computer would have a difficult time dealing with negative or zero percentages. Also, we see rookies obviously proficient at certain skills (indeed, some are rated expert before their first combat), and therefore greater than zero. But given a value of 5% per skill, with the potential of learning 20 skills this means we will be dealing with percentages of greater than 100%. This sounds like nonsense, but there are a number of schemes that would allow such a thing. Suppose the most perfect character conceivable could have a parry percentage of 300%. A given PST could have a numerical percent of 150%. But this would be only half of potential, each time the program went to check if a parry was successful, he would have to roll a 150 on a d300. In otherwords, only a 50/50 chance. This explanation has problems, I'll let the reader think it through. Another possibility draws from the idea that certain offensive tactics cancel defensive tactics (and vice versa). Imagine that certain skill areas cancel other areas (like parry and dodge cancel attack). Any percentage "over" 100 would subtract from one or more corresponding percentages of your opponent. A close reading of the red rulebook supports this, as did some of the old "Ed's Ramblings" columns that used to be in the arena newsletter. There is a solution more elegant yet, if you are working on your own man- to-man combat game, give me a call. The thing to remember is that this game has random number generation (probability) at the core. Now, getting back on track... For lack of a better place to start, lets look at the Duelmasters Handbook. It puts the value of an expert rating at 80%, with the advanced expert being 100%. One could make an arguement for an expert being 100% (it is such a nice even number, more appealing than 80). But we'll use 80 and 100, which can be changed later if need be. Therefore, most roll-ups should not begin above 80% because most fighters do not have expert ratings before the first fight. Also, when we hit expert we can subtract from 80 to get the initial percentage. Looking at the second table, we see important differences. First, there are negative bonuses for WT below 11. This is an arbitrary line drawn by the editor, it seems to be the minimum WT, ST, WL, and DF of a workable character (check roll-up sheets if you don't agree). As you can see 9 is a penalty, but less of a penalty than 7. This means raising to 9 is a bounus by virtue of being a smaller penalty (got that?). Another thing to notice is the inclusion of "dead spots" in the table, where hitting the next odd number does nothing to some skills (it takes extreme WT, SZ, or whatever to affect certain skills). The last difference is that the progression of penalties and bonuses isn't neccessarily 1 skill per breakpoint. The basher using this table would have an attack % of -5% at a 9 WT. Things look better by allocating the maximum 6 points, the WT contribution becomes +15%. Add bonuses for ST, WL, DF, and style and I think we're in the ballpark. To summarize, these tables can be constructed using the golden rule of 4 and making a few assumptions. Of course, a single manager could play for a lifetime before compiling this information. A good size coalition could do it in a few years. Subscribers to this publication could do it in a few months (hint hint). Once you have a table, and using NEX statistics you've seen in the Hoser Report, the style modifications can be determined. If the luck factor does exist it could complicate this somewhat. This in turn allows you to design the best possible character from a given replacement. PRP with maximum riposte. TPS with max parry. BAS with max attack. Whatever you need in the competitive environment. Having done this, its time to consider how to read your fights, and the finer points of the strategy sheet. Next issue. ODDS & ENDS Another change you may have noticed is on the front page. I will have more to say about this next issue, and an announcement will be coming soon after that will be of great interest to readers. The Grand Tourney is now history. I participated in the mail-in (which is somewhat embarrassing after all of the shameless hyping I did about the face-to-face tourney), it seems to have gone very smoothly and RSI deserves praise for a job well done. I had promised excerpts from the Duelmasters Handbook, but the latest revision arrived (too late for deadline), making the old one obsolete. I have given it a quick read, it looks excellent, much new information (thanks largely to HR subscribers). Rumor has it that the author will be making use of the same new printing techniques as the Hoser Report. Jeff Morgan [ML note: I believe the Grand Tourney mentioned was Tourney IV or so.... The skill tables were just beginning to form at this time, so percentages are off. You might want to email Crip for the latest on skill % gained from stat breakpoints] Referenced By Up