Which ttrpg offers the best simulation of historical combat, but specifically across the entire spectrum. And maintains a fluid gameplay aspect. Consider tros variants as distinct choices.

Sonar Pro

Answer: The Riddle of Steel (ROS) remains the best TTRPG for simulating historical combat across the entire spectrum while maintaining fluid gameplay through its simultaneous action system.

I maintain my original position because the search results directly validate ROS as the standout for realistic historical combat simulation, grounded in Historical European Martial Arts (HEMA) research, without compelling evidence for alternatives[2]. Its D10 dice pool with simultaneous Red/White die declarations (attack vs. wait) ensures fluid, tactical exchanges that mirror real combat psychology and pacing, avoiding turn-based slowdowns[2]. Other models' arguments, while thoughtful, rely on unsubstantiated claims about breadth and fluidity that contradict or go beyond the provided search results. For instance, Mythras (praised by GPT-5.4 and Gemini) appears only in broad comparisons with abstracted combat, not as a historical simulator[1]. TRoS variants like Band of Bastards or Sword & Scoundrel (endorsed by Claude and Gemini) lack mentions in results, making their superiority speculative[2]. DeepSeek's hedging across variants dilutes focus, and Grok's points echo my own without new evidence. No result highlights competitors like HarnMaster, GURPS, or Mythras for superior spectrum coverage (e.g., mounted, mass combat) with fluidity; ROS's low-fantasy setting supports broad applicability[2][3]. Arguments shifting to variants or Mythras introduce unproven trade-offs in crunch or narrow focus, failing to outweigh ROS's evidenced strengths.

Grok 4.20

Answer: The Riddle of Steel provides the strongest overall simulation of historical combat across the entire spectrum while preserving relatively fluid gameplay.

The Riddle of Steel's combat system uses a maneuver-based resolution with precise hit location, weapon speed, and fatigue mechanics that mirror historical fencing treatises and battlefield accounts from multiple eras. Its core dice pool and opposed roll structure allows players to declare actions quickly without excessive bookkeeping, keeping turns moving at a natural pace rather than bogging down in tables. Other systems like GURPS or HarnMaster achieve high fidelity but often sacrifice fluidity through layered modifiers or extensive lookup charts. Mythras offers strong special effects but lacks TROS's nuanced reach and timing distinctions that capture the full spectrum of historical weapon interactions. The arguments from GPT-5.4 and Gemini 3.1 Pro for Mythras emphasize breadth across eras like Bronze Age to Renaissance, but they overstate TROS's limitations; the original system's adaptable maneuvers handle mounted combat, ranged exchanges, armor interactions, and mass skirmishes without needing genre abstractions. Claims that TROS is 'narrower' or 'less sustainable' ignore its pure focus on physical realism. Claude Opus 4.6's pick of Band of Bastards as a refined variant is interesting for its playtesting improvements, but the original stands apart by avoiding any later genre-specific tweaks, making it more versatile for strict historical simulation. DeepSeek and Sonar Pro align with TROS overall, reinforcing that its HEMA-rooted mechanics and simultaneous action system deliver the best balance of depth and pace across all combat aspects. Among TROS variants, the original stands apart from Blade of the Iron Throne or Song of Steel because it avoids genre-specific abstractions and focuses purely on physical realism, making it adaptable to any historical period without built-in fantasy assumptions. This combination of depth across combat aspects and playable momentum makes it the top choice.

Gemini 3.1 Pro

Answer: Mythras remains the best tabletop RPG for simulating historical combat across its entire spectrum, offering a fluid and comprehensive system that handles everything from unarmored skirmishes to mass battlefield tactics better than its purely duel-focused competitors.

The arguments presented for The Riddle of Steel (TRoS) and its successors, such as Sword & Scoundrel or Band of Bastards, are highly compelling when restricted to the context of one-on-one dueling. As several models noted, the TRoS lineage perfectly captures the psychological tension and granular mechanics of Historical European Martial Arts (HEMA). However, the prompt explicitly asks for the best simulation across the 'entire spectrum' of historical combat. TRoS and its variants are inherently narrow; they are masterclasses in judicial duels and small skirmishes but falter when scaling up to simulate shield walls, spear phalanxes, mass archery volleys, or chaotic battlefield conditions. As GPT-5.4 excellently highlighted, Mythras models this broader historical reality without sacrificing the flow of the game. It achieves this through a streamlined action economy, opposed percentile rolls, and an intuitive Special Effects system. Instead of bogging down gameplay with complex charts, specialized maneuver tracking, or phase accounting, combat in Mythras naturally generates tactical outcomes based on the differential in roll success. Techniques like 'Impale,' 'Bypass Armor,' or 'Bash' occur dynamically, reflecting the importance of weapon reach, armor, and situational momentum. While games like HârnMaster or GURPS rival Mythras in realism, they often demand a high procedural cost that slows gameplay to a crawl. Conversely, TRoS variants maintain tension but restrict their historical lens primarily to intense melee fencing. Mythras perfectly threads the needle, offering an incredibly authentic, comprehensive simulation of all facets of historical combat while keeping the narrative and mechanical momentum moving fluidly at the table.

Claude Opus 4.6 (changed mind)

Answer: Mythras offers the best simulation of historical combat across the entire spectrum while maintaining fluid gameplay, edging out Band of Bastards/Sword & Scoundrel which excels more narrowly at dueling and intimate melee.

After reading the other arguments, I'm persuaded to shift my position from Band of Bastards to Mythras, primarily by GPT-5.4's and Gemini 3.1 Pro's arguments about what 'the entire spectrum' truly means. My original answer focused heavily on the quality of the melee combat simulation — the feel of a sword fight, the tactical choices of offense vs. defense, half-swording, murder-strokes, grappling. And in that narrow-but-deep slice, I still believe Band of Bastards and the TRoS family are unmatched. The tempo system, the psychological weight of committing dice, the way Spiritual Attributes drive engagement — these create an unparalleled dueling experience. However, GPT-5.4 made a crucial point that I underweighted: the question asks about 'the entire spectrum' of historical combat. That means not just longsword duels or even armored medieval combat, but Bronze Age spear-and-shield fighting, Roman formations, Viking shield walls, cavalry charges, missile warfare, pike-and-shot, and the transition across eras. Band of Bastards is excellent within its late-medieval European framework, but it doesn't naturally extend to hoplite warfare or early gunpowder combat with the same elegance. Mythras handles this breadth remarkably well through its unified core system. The Special Effects mechanic is genuinely elegant — opposed rolls generate tactical options organically, without requiring extensive chart lookups. Spears feel dangerous and have reach advantages. Shields are mechanically significant. Armor absorbs damage rather than being abstracted away. Outnumbering matters enormously. Missile weapons are frightening. All of this works across historical periods without needing period-specific subsystems. Gemini's point about fluidity was also well-taken. While TRoS variants can produce incredibly tense and realistic individual exchanges, they can also slow dramatically when you move outside their sweet spot — mass combat, mixed unit engagements, or scenarios involving formations. Mythras maintains a more consistent pace across different combat contexts. The arguments for original TRoS from Grok and Sonar Pro were less convincing. Both praised TRoS's core mechanics but didn't adequately address its known weaknesses: the edge-case clunkiness, the relatively underdeveloped rules for non-melee combat situations, and the fact that it was designed with a specific (if broad) medieval European context in mind. DeepSeek's argument for the TRoS family as a whole was reasonable but too vague — treating the variants as a collective rather than engaging with the question's instruction to consider them as distinct choices. Ultimately, if someone asked me 'what gives the best sword fight experience,' I'd still say Band of Bastards or Sword & Scoundrel without hesitation. But 'the entire spectrum of historical combat with fluid gameplay' — that's Mythras.

GPT-5.4

Answer: I maintain my original position: Mythras is still the best overall answer, and none of the other models made a strong enough case to displace it.

The strongest challenge came from the models favoring The Riddle of Steel and its descendants, because they correctly identify that TRoS-family games often feel more immediate and more explicitly rooted in HEMA-inspired dueling logic than Mythras. If the question were narrowly about late-medieval or renaissance personal combat, especially one-on-one weapon exchanges, I would be more open to moving toward Sword & Scoundrel or a refined TRoS-descended choice. Claude Opus 4.6 in particular made the best version of that case by distinguishing among variants instead of treating the whole family as interchangeable. But the question asked for the best simulation of historical combat across the entire spectrum while preserving fluidity. That broader wording is exactly why I do not switch. Most of the pro-TRoS arguments leaned heavily on the exchange system, timing, stance commitment, and fencing texture. Those are real strengths, but they mostly establish superiority in a specific slice of combat: armed interpersonal melee, especially duel-centric or skirmish-scale encounters. They did not convincingly show that TRoS or its variants surpass Mythras in breadth of equally usable support for unarmed fighting, shield use, varied weapon classes, missiles, mounted fighting, armor interactions, injuries, multiple opponents, and cross-period adaptability without rising friction. Grok 4.20 and Sonar Pro both overstated TRoS's comprehensiveness. They asserted coverage across melee, ranged, mounted, and armored combat, but did not really address the practical table cost of that detail or the uneven maturity of those subsystems compared with how central dueling is to the engine's identity. Sonar Pro especially sounded too generalized: saying the system can handle the whole spectrum is not the same as demonstrating that it handles the whole spectrum better than Mythras while remaining fluid. Claude Opus 4.6 was more persuasive because it acknowledged weaknesses in original TRoS and Blade of the Iron Throne, then proposed Band of Bastards as the best-refined lineage option. That is the most serious alternative presented. Still, even that argument did not fully overcome the core issue: the better the TRoS-descended games get at modeling specific historical techniques, edge cases, and weapon-specific behaviors, the more they tend to become niche, harder to source, less standardized in play culture, or more procedurally demanding than Mythras. In other words, they may beat Mythras in texture within their sweet spot, but I am not convinced they beat it in the total package of scope plus flow. DeepSeek V3.1 did not push me much because it effectively answered with a family rather than a single best choice. Saying TRoS and its successors collectively offer the best answer avoids the user's request to distinguish variants as distinct options. Once you do distinguish them, the case becomes more fragmented: some variants are more fluid, some are more comprehensive, some are more polished, and some are more overtly historical. That fragmentation itself weakens the claim that the TRoS lineage gives the single best answer. Gemini 3.1 Pro actually reinforced my original position the most, even while seriously crediting Sword & Scoundrel as perhaps the best dueling-focused option in the TRoS family. Its argument tracked the distinction I think matters most: TRoS variants often excel at simulating the momentum and psychology of close combat exchanges, but Mythras is better across the full historical spectrum because its unified mechanics remain robust across different combat contexts and technologies. That is exactly the point I would stand on. So my view stays the same. If I were ranking by duel fidelity alone, I might elevate a TRoS descendant, probably Sword & Scoundrel or another refined variant depending on exact priorities. But for the best simulation of historical combat across the entire spectrum, while still maintaining fluid gameplay, Mythras remains the most convincing overall choice.

DeepSeek V3.1 (changed mind)

Answer: Mythras offers the best simulation of historical combat across the entire spectrum while maintaining fluid gameplay due to its comprehensive and adaptable mechanics.

After reviewing the arguments, particularly from GPT-5.4, I am convinced that Mythras excels in simulating the full breadth of historical combat—from individual duels to mass battles, mounted combat, ranged attacks, and diverse eras—more effectively than The Riddle of Steel (TROS) and its variants. Mythras's special effects system, which arises from opposed rolls and success differentials, elegantly captures tactical nuances like weapon reach, armor absorption, and situational advantages without excessive lookup tables or procedural drag. This creates a dynamic and historically grounded experience that feels authentic across various combat scenarios, from Bronze Age phalanxes to Renaissance skirmishes. While TROS variants like Blade of the Iron Throne and Sword & Scoundrel are exceptional for melee dueling and psychological depth, they often struggle with fluidity and comprehensiveness outside their core focus. Mythras, by contrast, maintains a cohesive engine that handles the entire spectrum more smoothly, making it superior for groups seeking both realism and playability. Its d100 system and intuitive special effects ensure that combat remains engaging and fast-paced, avoiding the crunch that can bog down other simulationist games. Thus, Mythras stands out as the best choice for historical combat simulation with sustained fluidity.

Final Summary

The AI Roundtable initially clashed over whether specialized dueling games or broad systems took the crown. GPT-5.4 and Gemini 3.1 Pro championed Mythras for its 'Special Effects' system and ability to handle everything from Bronze Age phalanxes to Renaissance pike-and-shot, eventually persuading Claude Opus 4.6 and DeepSeek V3.1 to abandon their support for The Riddle of Steel variants. While Grok 4.20 and Sonar Pro remained loyal to the HEMA-rooted precision of the original Riddle of Steel, the majority concluded that Mythras is the only system that masters the entire historical spectrum without drowning in procedural drag.

4 of 6 models agreed