The Riddle of Steel Review

It came with great promise, it generated polarised debate around 2001 to 2003, it has now almost entirely disappeared, the website gone, the book out-of-print and rarely seen in second-hand sources. It is The Riddle of Steel, and when I managed to pick up a copy recently, I could not help but give it a few sessions to find out what all the fuss was about.

The Physical Product

The 270 page hardback is well bound and comes with an attractive fish-eye lens graphic of a warrior holding aloft an implausible sword. The interior art is often below average in terms of technique and well-below average in terms of creativity, although the inclusion of reprints from renaissance era manuscripts is an exception, as is the amusing female lancer on the back of the giant lizard. Nearly all of the art is placed with little regard to context either, with some exceptions (e.g., encumbrance).

The book is formatted with a two-column justified layout and a serif-font. The margins have chapter and section headings (top and side, respectively), which is pleasing. There is a good two-page table of contents and four-page index, although the formatting of the latter is a little askew. The product is broken into eight "books", each containing several chapters and some of those with sub-chapters.

The writing style is accessible and efficient with a direct statement of rules, elaboration and/or explanation and examples, although with the occasional spelling error and lapse into silliness. Each 'book' is unfortunately introduced with a short narrative with does nothing for the understanding of the game or the setting; compare with The Travels of Biturian Varosh.

Character Generation and Development

After a useful chapter (I can't call these "books") of quick-start rules, the following two provide the core of character generation and development. Characters begin with a concept and 'philosophy', the latter a short thematic statement. After this is the allocation of priorities, from A to F, for various characteristics. The slots are Race and Magic, Social Class, Attributes, Skills, Proficiencies and Vagaries, and Gifts and Flaws. Essentially it's a coarse-grained, category-based, point-buy system similar to that offered in Shadowrun, but without the offer of standard archetypes.

In the 'Race and Sorcery' categories are fey, siehe, halflings and humans. Halflings are half-elves, rather than our short friends from Tolkien. Magic-using characters are priority 'A' or 'B', and non-magician humans are priority 'D', 'E', 'F'. Logically, if you're going to be a human, you may as well place that as the last priority and gain the slots for other areas.

The priority range for Attributes is from 27 to 47 points, distributed among "Temporal Attributes", meaning Strength, Agility, Toughness, Endurance, Health (physical) and Willpower, Wit, Mental Aptitude, Social and Perception (mental). This is limited with an initial range between 2 and 7 prior to racial modifications, each point providing a dice in tests against a target number. This is arguably too many attributes with many having slight differences for the purposes of a game (e.g., Toughness, Endurance and Health?). There are also a number of derived attributes (Reflex, Aim, Knockdown, Knockout and Move) from the temporal abilities.

The game also comes with "Spiritual Attributes", which all starting characters can distribute seven points among five chosen from a list of Conscience, Destiny, Drive, Faith, Luck and Passion with specific expressions. Apart from adding a systemic personality drive to the game, SAs also add bonus dice to tests when deemed appropriate and their use provides the equivalent of experience points, not unlike the "roleplaying" rewards given out in the Hero System, GURPS etc.

By skills, what is meant is "skill packets", the equivalent of a profession, a collection of individual skills with some variation. As a priority these range from one skill packet at 9 (priority F, some training) to two skill packets at 6 each (priority A, adept-proficient); the lower a character's skill number the greater their ability. Each skill is associated with an attribute and the skill value equals the target number. There are fourteen skill packets under the broad categories of "Military", "Urban" and "Other". There are almost seventy individual skills, each with a one or two paragraph description, some minor examples (including contested skills) and modifications and a default value.

Approximately thirty 'gifts and flaws' are described, familiar to the extra advantages and disadvantages found in the Hero System, GURPS, and a multitude of other games, with two levels of effect, major and minor. The short list does cover a good range of physical, social and mental effects as they are often written with a more general, rather than particular effect. Each gift and flaw is described with in-game and systemic effects.

The previously mentioned skill list, as with many RPGs, is orientated towards those most likely to be used by "adventuring classes", with some individual skills possibly considered an entire "packet" in a different context (e.g., farming, hunting). This most apparent, an appropriate, in the proficiency list which are specialist skills for the martial arts. Unlike many roleplaying games which assign skills per weapon, TRoS assigns proficiencies according to styles. True, many of these initially seem to have a weapon-like equivalent (e.g., greatsword, rapier, poleaxe, dagger), but if one can justify the use of a weapon within a particular style then it is allowed. A weapon like a quarterstaff, for example, would be very adaptable. Each style comes with a set of offensive and defensive maneuvers with a 'cost' to activate from a combat pool.

System and Combat

As mentioned, it is actually the first chapter that provide the core information of the game system, which is pretty much dice pool versus target number (comparisons with various White Wolf games are justified). Attribute tests are made by rolling a number of (d10) dice, based on Attribute value, against a Target Number, representing the difficulty. Rolls are open-ended (as per Rolemaster et. al) or 'stacked' in the system parlance, so with a result of '10', reroll and add ad infinitum. The number of rolls over the Target Number equals the degrees of success. For Skill Tests, a varied attribute is used depending on circumstances, with the Target Number equal to the proficiency of the character, which does bring into question variation of skill Target Numbers based on difficult (e.g., a difficult task, but a adept character). Obviously this can be achieved by a modifier to the roll, but by how much is not clearly specified.

In addition to this there are also Proficiency Tests, which make up the core of the combat system (Combat Pool, Missile Pool) and are also used for spellcasters (Sorcery Pool). With a proficiency test a pool is created from which a number of dice can be taken to test against a Target Number based on the activity and tool. A character's Combat Pool is based on Reflex + Weapon Proficiency, their Missile Pool on Aim + Missile Weapon Proficiency, and the Sorcery Pool equal to KAA ((TO+HT+WP)/2) + FORM ((WT+PER)/2)).

Finally there are Contested Rolls and Extended Rolls. The former defined when two or more characters "are competing for precedence" and resolved as a competition of successes against whatever Target Numbers are being tested; an example is the Skill Test of stealth against the Attribute Test of perception. In comparison Extended Rolls are derived from a number of successes against the Target Number over successive rolls separated by time. One cannot say however that there are many examples provided of these tests. Fumbles are defined in the game as when all a test completely fails are there are two or more '1' values among the dice rolled which gives an odd bias towards those with ability rather than incompetence.

The much commented combat system uses a 1-2 second round, an "exchange of blows". It begins with a declaration of stance (aggressive, defensive, neutral), followed by a declaration of aggressive or defensive actions based on simultaneous call. An aggressive stance adds 2 Combat Pool dice when attacking, but increases the activation cost for any defensive action by 2 CP; the reverse applies for defensive stances. Attribute tests are used to judge surprise situations, initiative can be bought for a fast but possibly inaccurate blow, and opponents can be riled through taunting. The combat pool may also be further modified by terrain, visibility, and difference in weapon length. As a result of the initial exchange of blows it is possible through the maneuvers that the role of attacker and defender have reversed for the second half of the exchange. Assuming anyone is still standing after this, the round begins again and combat pools are refreshed. From actual play one of the most
enjoyable comments was that tactical decisions made a difference. One of the notable omissions was the opportunity for even simple missile weapon combat maneuvers (e.g., aiming bonuses, snapshots etc).

As mentioned various weapon styles have different activation costs for various maneuvers and variable target numbers. These are compared as a contest of skills. For example, a character with 7 Sword and Shield proficiency may engage in a simple Cut (0 CP cost) with an Arming Sword (TN 6) against a similar character who has chosen to defend with a Block (0 CP), and does so with the a medium heater shield (TN 5). Assuming for the sake of illustration that the contest is won by the aggressor, a general hit location zone must be declared by the attacker, followed by a d6 to determine the precise location. So an overhead cut down over the shoulder is defined as location IV, with precise locations including shoulder, chest, neck and head. Damage is determined primarily by the character's Strength plus weapon modifier; in the case of the arming sword St + 1 for cutting damage.

Protection against such damage is determined by the defender's armour plus their Toughness. Armour, it must be noted, is surprisingly cheap. Even a character who has taken one of the lower levels of Social Class should be able to afford chainmail. Further - and I can't believe this made it through playtesting - the game suffers from the notorious "naked dwarf" syndrome of Warhammer FRP (1e). That is, because it is not abnormal for characters to have a Toughness rating of around 7, the game system declares that their skin and muscle is, effectively, more protective than plate (6). A simple, regrettable, fix is to limit the protection bonus of Toughness to the Strength bonus to damage from the opponent.

The result of the Damage Rating minus the Toughness and Armor Value is the Wound Rating, from one to five. The Riddle of Steel claims that it has no hit points as such; each wound has defined results, with bleeding, shock, pain, and checks for knockdown etc. For example a Wound Level 3 cut to the chest is "BL: 3, Shock: 8, Pain: 7 - WP, Roll Knockdown", which is reminiscent of the Rolemaster critical hits. Shock is an immediate loss from all dice pools; if greater than the current CP, the remainder is applied at the beginning of the following round. In contrast, Pain is applied to pools at the beginning of each round. Blood Loss requires an Endurance Attribute test each round against BL; if failed 1 point of Health is lost. If Health is reduced to 1 all attributes and pools are halved, and if Health is reduced to zero, the character "enters coma and dies". A Knockdown test is made if a character receives a blow that reduces their CP to less than zero; a test is made ag
ainst the Knockdown Trait against a TN equal to twice the attacker's Margin of Success, or three times with a "mass weapon" (mace, flail etc). In contrast, a Knockout roll is conducted against a TN of 7, modified by the severity of the wound.

A very brief (8 effective pages) chapter covers "the laws of nature", including movement rates, encumbrance levels, lifting and carrying, jumping, leaping and falling, fire, electricity etc, fatigue, wound recovery, and aging and sickness. Non-combat movement rates distance are measured in miles per day with modifications for terrain type. Despite significant variation due to the Attribute system the results seem realistic. Encumbrance is divided into five entirely descriptive levels; it is uncertain how 'natural encumbrance' (i.e., obesity, a character flaw in the system) adds to the 'artificial encumbrance' of equipment. Lifting objects is based on a Strength Attribute test with Target Numbers based on the pounds of the object, as is jumping, with leaping based on movement rate, and whether the character is standing or running. All three come with an "auto-value" where a lower weight or distance is achieved for a lower weight, height or distance as appropriate.

For falling, damage varies according to the distance and surface, distributed across random locations. For healing, the application of first aid skill reduces blood loss, with chuirgery skill required to deal with level 3 or 4 wounds. Aging tests are made yearly against health with variations for age brackets starting at 40; a quick calculation suggests that with an average of 4 across all ten attributes, most characters will shuttle off the mortal coil before due to old age before their 50th birthday. Sickness is handled as an extended health Attribute Test, with a higher target number for more virulent problems with weekly tests, and with a high probability of more lost attribute points. How these are recovered naturally (i.e., outside of use of Spiritual Attributes) is not explained.


The keywords in sorcery is adaptability, power and aging. Any 'gifted' character, according to character generation prioritisation, has access to sorcery 'vagaries' which are the proficiencies of magic and purchased through the same pool. Arguably an alternative pool could have been created which would (a) give more meaning to a human, non-gifted racial choice and (b) provide the opportunity for the classical warrior-wizard; although it must be acknowledged that under the rules as writ that this would be quite unbalancing.

There are five additional derived attributes for sorcerers, Kaa, Form, Art, Discipline and Draw, used for raw energy, spell-casting skill, resistance, ritual magic and vagary recovery respectively. In addition there are three pools according to each power-level of magic, "Spells of One", "Spells of Three", and "Spells of Many", cast in seconds, tens of seconds and hours respectively and with proportional power and inversely proportional draining effects. Spells of "x" refer to the number of vagaries that may be incorporated simultaneously; one, three and many.

Magical proficiencies are called "vagaries", representing skills in Sculpture, Growth, Movement, Conquer (Mind), Glamour, Vision, Summoning, Banishing and Imprisoning (the last three all being applied to magical beings). These are presented in three levels of ability with some extremely powerful break points. This becomes even more evident in spell-casting which, like Ars Magica, provides the opportunity for on-the-spot spell creation. Casting Target numbers are based on the difficulty of target scope, range, volume and duration. As an example a Range 1 spell has a distance of effect of "Touch", whereas a Range 2 spell has "Line of Sight". A more graduated approach (perhaps five levels) would have been more appropriate.

The more complex and powerful a spell is, the higher the casting time, the higher the casting target number, and the more difficult it is therefore to cast successfully. These levels can be mitigated somewhat by engaging in dialogue and gestures during casting, meditation before casting, and, for ritual spells, co-operation and symbolic drawing. Spellcasting becomes more difficult due to noise, interruptions and wounds. Spell can become easier to cast if they are "formalized", that is, transcribed.

When spells are cast, the sorcery must decide from their sorcery pool how many dice are spent to ensure a successful cast, and how many are spent to resist aging, measured in months from the target number of the spell. Given that spell target numbers of 12 or so are middling, twenty or thirty castings of a mid-power spell can rapidly send a character from being a headstrong youth to a prematurely aged and burnt-out cynic in a few sessions. The sorcery pool is refreshed at a rate of one die per hour but this can be extended by casting additional spells (which comes with their own risk). Spells are resisted by either "getting out of the way" in the terms of elemental attack spells, with an attack target number depending on the Seneschal's estimation of difficulty, or by contest of will for mind control spells.

A short selection of specific gifts and flaws for sorcerers is provided, along with combat options. It is not the first time that the game cannot decide whether the combat round is one or two seconds, a fairly critical issue given that a sorcerer's casting time is measured in exact seconds. An option is provided for counter spells by using the same Vagaries as the caster. A handful of formalised spells are provided as examples, consisting of 2-5 Spells of One for each Vagary, a dozen Spells of Three, a couple Spells of Many and finally two "Spells of None", last act curses. The chapter concludes with an essay by Ron Edwards on the use of Sorcery in the Riddle of Steel, claiming that "sorcery is another, very dramatic, means of getting Spiritual Attributes into play", and that sorcerers do not represent "personal artillery". Given that there is no special use of Spiritual Attributes in TRoS's sorcery rules and that from play experiences sorcerers are best used as artillery pie
ces in TRoS, I think Ron is projecting in this case of what he would like TRoS sorcery to be rather than what it is.

The World, The Inhabitants, The Seneschal

The hefty seventh chapter, some sixty-five pages, covers the world of "Weyrth", the default game world. Consisting of three subcontinents (Mainlund, Tegaarn, Maraiah) is explicitly earthlike, with a 364 day year, although the effects of the six visible moons on tides is not discussed. At a size of some 3750 miles by 3750 miles (14,062,500 sqm) it is somewhat smaller that Asia, excluding the European peninsula, and the three subcontinents are like a reduced version of Europe, Asia and Africa, and quite openly so in terms of culture, ecology, climate, with Mainlund representing Europe, Tegaarn for Asia and Mariah for Africa, respectively.

This causes some problems, as the game explicitly states that Weyrth is "small", "practical" and "real", rather than "big". Apparently this "different to most fantasy ... worlds" in gaming or literature. Reading the material one could say it is certainly different to some, but similar to many but not in a good way. Weyrth is an astoundingly dull and often improbably place. To be sure, the thirty-plus nationalities are provided the right sub-headings; Geography, Religion, Politics/Military, Legal System, Economics, Inhabitants/Culture, Places of Note, Player Bonuses and each is provided with a page or more of description, but individually and collectively it fails to generate interest through verbose descriptions of mediocrity. The nationalities are largely derived from earth-cultures, and often explicitly (e.g., the Celtic "Picti" and "Cymry") and analogous religions ("The Imperial Church of the Three Gods-Become-One, The Seven Vows of the Prophet") are hackneyed stereotypes.
Tacked on with some disregard of cultural appropriateness are the various Elves and Trolls, along with a uninteresting potted history, which at least gives a reasonable period for history. An attempt to develop a "realistic" coinage system is mainly annoying in actual play (1 gold = 20 silver = 240 copper, three coin sizes) and prices often seem most perplexing. Even the poorest starting characters who have any wealth at all will found themselves quite capable of buying weapons, good armour (e.g., chain) and a good weapon.

The final chapter on The Seneschal gives some unexciting advice for preparing and running a game, followed by a self-justification for the game's existence based in a frequently-asked-questions style. More usefully is the sample NPCs, consisting of men and women, various beasts, siehe and fey, mythical beasts and the "other world" (undead, demons). All told there is a small selection of statistical information and a descriptive paragraph for some thirty plus NPCs, including those for whom an alternative hit location system should have been included (e.g., griffin, wyrm). Finally, the game concludes with five reasonably interesting adventure seeds.


The Riddle of Steel is a seriously flawed gem. It contains moments of great creativity in a game system that is also highly adaptable, but is "riddled" (pun not intended) with very significant errors in implementation. Polarised opinion is therefore understandable in these circumstances. There are those who look at the creative realism and overlook the flaws, considering them to be secondary to the overall picture. Equally there are those who believe that the game over-promised and under-delivered and were doubtless highly annoyed to the point that the failures became central.

The greatest area of strength in The Riddle of Steel is probably the combat system. It is interesting, deadly and quite quick. In contrast the character generation system is under-developed, the sorcery system, whilst flexible, is not necessarily fun for the receivers (because it is so powerful) or the practitioners (because it has such serious side-effects). Perhaps the worst section of the game is the provided world which fails in rhyme and reason; a problem arising partially because it is genetically and climatically so diverse, yet culturally so minimal.

When the debates raged about TRoS one of the most insightful comments was that it could be a really good game if a second edition was released; one where the artwork was cleaned up, where there was more content in character generation, where a few flaws in the combat system was fixed, where sorcery was not so punishing, and where the world actually concentrated on a interesting viable area. This, of course, has not eventuated, making TRoS an interesting acquisition for collectors, certainly worth playing a few sessions, but not beyond that unless one is prepared to do some significant work.

Style: 1 + .4 (layout) + .2 (art) + .7 (coolness) + .4 (readability) + .8 (product) = 3.5

Substance: 1 + .3 (content) + .4 (text) + .6 (fun) + .2 (workmanship) + .6 (system) = 3.1

Riddle_of_steel_rpg.gif52.02 KB