Optimization and preparation

From An apology to some min-maxers:

In fact, if anything, my sub-optimal character builds were me being lazy, and not doing the homework other players were doing to build more optimized characters.

At the same time, as GM you should min-max the opposition to the same level as you have allowed for your players. Dig into the system and look for how to optimize NPC’s and monsters. That is likely going to be more work for you, as you may need to take stock stat blocks and beef them up, but it will create challenging opponents for the players, making encounters more exciting.

This concisely captures why I gravitate away from games that support or require significant optimization. It’s just making more work for everyone involved, and not in a way that adds to the play experience. It’s work inflation: everyone needs to spend more time in order to get the same result. And the result is the same, assuming that competition has no value (which it does not, for me, in RPGs).

Compare this to working more on setting detail or even character personality. Such preparation can consume as much time as you want, but it can also qualitatively improve the play experience. Optimization and what I am going to call (for lack of a better term) preparation can thus be seen as two independent dimensions of out-of-game work required by RPGs, leading to four rough categories of game: high-op/high-prep, high-op/low-prep, and so forth.

optimization and preparation bw

A railroad is a way of constraining preparation requirements. There may be some overall conservation law operating such that sustainable games on average tend to be low in either or both dimensions. This may be why it is difficult for me to find an example of a game that is naturally high in both categories. Games associated with high optimization (Pathfinder, 4E D&D) can be played using sandboxes or with more extensive world building, but doing so ends up being a higher-maintenance activity.

As an observation it seems that many games in the focused design tradition, especially Forge and Story Games, seem to prize minimization of both kinds of work. Games like Lady Blackbird and Apocalypse World, for example, put few barriers before getting started (AW requires the MC to create some fronts, but the instructions seem geared to avoid potential scope creep). Minimizing out-of-game work a way to increase approachability by decreasing cost. If the reliability of play experience varies with the prep time, that is seen as a flaw in this tradition. The system is seen as “not working” if play is inconsistent, as it will surely be if preparation is required given that different groups or players will invest in different amounts of out-of-game work. Other traditions seem to focus more on the other side of the equation: increasing value (such as the intricate setting and art of Warhammer, or atmosphere of the World of Darkness, separate from amount of work required). Obviously, the two approaches are not exclusive, but there still seem to be trends toward one or the other.

And yeah I may have been working on a data analysis assignment before writing this post…

Deflective shields

My first Dark Souls dude, with a shield

My first Dark Souls dude, with a shield

These are the current shield rules (approximately the third revision) for The Final Castle. To make sense, I have included preliminarily a few general combat rules as well. Combat Tests are d20, roll high, aiming to meet or exceed an enemy threat level (similar to the probably familiar armor class or difficulty class). Hopefully the fragmentary nature is not too hard to understand.

“Unbalanced” is a state, something like a temporary condition (in 3E terms) that persists until addressed by the combatant. Deflection is a reaction that can be taken in response to an enemy attack.


Exceeding the target number of an Ability Test by at least 4 or rolling a 20 prior to any modifiers is an Overkill. For Combat Tests, Overkill adds 1d6 damage and may have additional effects in other contexts.


Unbalanced combatants may only Melee, Shoot, Flee, or Recover and may not deflect attacks withs Shields. Recover balance with a maneuver.


To deflect an attack, deploy a shield. Deflection must be declared before rolling dice to resolve an attack. Deploying a shield Unbalances a combatant but does not require an Action. Shields may not be deployed when Unbalanced. Bypass Maneuvers cannot be deflected with a shield. Different kinds of shields offer other benefits. See the Shields entry in the Weapons section for details.


Resolve Maneuvers with Melee, Shoot, or Throw Actions depending on the character of the desired effect, substituting Maneuver effects for damage. Overkill applies to Maneuvers also. Thus, Maneuver Overkills cause the Maneuver effect and also 1d6 damage.

Recover (Maneuver)

Recover from being Unbalanced, stand up from prone, or escape a Grapple.


To deflect an attack, deploy a shield. Deflection must be declared before rolling dice to resolve an attack. Deploying a shield Unbalances a combatant but does not require an Action. Shields may not be deployed when Unbalanced. Bypass Maneuvers cannot be deflected with a shield. Shields come in three varieties:

  • Bucklers grant +4 to parry maneuvers, but are useless against ranged attacks and great weapons.
  • Medium shields grant +2 defense against small missiles.
  • Tower shields deflect all small missiles but are useless against standard or smaller weapons.

Putting this all together, it means that PCs with a shield can deflect (that is, totally nullify) one melee attack per combat (not per round) essentially for free, though the deflection must be “used” before rolling Defense (the equivalent in the system of an enemy attack roll). After a shield has been used for deflection, combat options narrow generally due to becoming Unbalanced, and specifically the shield may not be used again until balance is recovered.

Since maneuvers work like attack rolls, but substituting effects for standard damage and only inflicting any damage upon Overkill results (exceeding target numbers by 4 or rolling a natural 20), the effect is that a skilled fighter attempting a Recovery Maneuver is still fighting (not potentially “wasting” a turn), just at a disadvantage (approximately -4) if they wish to earn another use of their shield during the current combat.

(Don’t worry about Parry Maneuvers; they are beyond the scope of this post.)

The Buried Giant

Image from Wikipedia

Image from Wikipedia

Set in the British dark ages one generation after Arthur, Kazuo Ishiguro’s novel The Buried Giant mostly follows the adventures of an elderly couple on a journey to visit their son. A mist of forgetfulness blankets the land, so the story unfolds shudderingly as if through a clouded or scorched lens. This both evokes a wonderful mood and allows Ishiguro to narrate without explaining too much too quickly. Though there are several underlying ideas, the novel is rewarding even if read as a straightforward fantasy, so I will not burden you with my thematic interpretation.

The depiction of myth-historical Britain and the way combat is described are worth focusing on in particular. The ogres and monsters and fairies and dragons are none of them cliches, and this is particularly impressive because Ishiguro describes them all in detail rather than resorting to the common strategy of leaving the imagination to the reader. For one example that I think does not risk spoiling anything overly much:

They might have been gazing at a large skinned animal: an opaque membrane, like the lining of a sheep’s stomach, was stretched tightly over the sinews and joints. Swathed as it was now in moonshadow, the beast appeared roughly the size and shape of a bull, but its head was distinctly wolf-like and of a darker hue—though even here the impression was of blackening by flames rather than of naturally dark fur or flesh. The jaws were massive, the eyes reptilian.

And the warrens that the Britons live in would be a great backstory for a series of mega-dungeons. It would not stretch the imagination to have many of these connected together in one great subterranean sprawl, partially taken over by restless dead, ogres, and other haunts while being active “towns” in other areas.

For warmth and protection, the villagers lived in shelters, many of them dug deep into the hillside, connecting one to the other by underground passages and covered corridors. Our elderly couple lived within one such sprawling warren—“building” would be too grand a word—with roughly sixty other villagers. If you came out of their warren and walked for twenty minutes around the hill, you would have reached the next settlement, and to your eyes, this one would have seemed identical to the first. But to the inhabitants themselves, there would have been many distinguishing details of which they would have been proud or ashamed.

The technique used to describe combat is singularly effective, with many paragraphs of positioning and perhaps some exchange of words prior to any actual attacks. The closest thing I can think of is the high noon contest of nerves in the wild west, but this is not quite that. Finally, the conflict is concluded swiftly with decisive action, with the loving detail focused on the slow buildup rather than a balletic interplay of punches, stabs, and blocks as is usually done in cinematic or adventurous combats. It is probably unlikely that any game would be able to capture this sense of combat, but it might be worth trying. A single scene ends up having many paragraphs like this:

Then the two men became fixed in their new positions, and to an innocent spectator, they may have looked, in relation to one another, practically unchanged from before. Yet Axl could sense that these new positions had a different significance. It had been a long time since he had had to consider combat in such detail, and there remained a frustrating sense that he was failing to see half of what was unfolding before him. But he knew somehow the contest had reached a critical point; that things could not be held like this for long without one or the other combatant being forced to commit himself.

Ishiguro’s Gawain is also perhaps one of the most memorable characters I have read in a long while. Here is a taste.

What do you say, sir? I’m a mortal man, I don’t deny it, but I’m a knight well trained and nurtured for long years of my youth by the great Arthur, who taught me to face all manner of challenge with gladness, even when fear seeps to the marrow, for if we’re mortal let us at least shine handsomely in God’s eyes while we walk this earth! Like all who stood with Arthur, sir, I’ve faced beelzebubs and monsters as well as the darkest intents of men, and always upheld my great king’s example even in the midst of ferocious conflict. What is it you suggest, sir? How dare you? Were you there? I was there, sir, and saw all with these same eyes that fix you now! But what of it, what of it, friends, this is a discussion for some other time. Forgive me, we have other matters to attend to, of course we have. What is it you asked, sir? Ah yes, this beast, yes, I understand it’s monstrous fierce but no demon or spirit and this sword is good enough to slay it.

A successful writer of literary fiction can afford to dedicate five or more years to the writing of a single novel, and the level of resulting craft is clear. Though doing so is a career risk, I hope that more serious writers experiment in this way, because The Buried Giant is a masterpiece. I devoured it over the course of a few days despite my current huge grad school workload, and I plan to read it a second time soon. Recommended without reservations. Apologies for the long quotes, but I felt like I needed to let the novel speak with its own voice.

Jupiter Ascending

Jupiter Ascending delivers fantastic visuals, including lepidopterous attack ships, galactic palaces, refinery cities inside gas giant planets, and gothic cathedral styled battle cruisers. The last I suspect were a conscious homage to 40K, especially given that at one point the protagonists need to “break through a field of war hammers” (some sort of geometric space mine). Warning, there may be some mild spoilers below, though I do not think they would really spoil the movie.

The plot is somewhat ridiculous, though not nearly as bad as some reviews I read made it out to be. The Wachowskis retain their fixation with harvesting human resources as seen in The Matrix. The universe is apparently ruled by giant space cartel noble families that live for millenia due to youth serum harvested from subject planets after they have marinated in DNA diversification for long enough. I actually enjoyed the over the top performances of the antagonist Abrasax siblings. Eddie Redmayne’s Balem Abrasax in particular was suitably Harkonnen with his decadent rasp, though his final behavior was not really believable for a semi-immortal space magnate.

Channing Tatum, unfortunately, does not sell the space wolf character at all. Someone with more believable sense of violence and threat was required, and the primary impression I got from his performance was tranquilized. Further, he looked soft, clearly not having trained very hard for the role. Chris Evans, Henry Cavill, and Hugh Jackman as action heroes have raised the bar and Tatum just does not cut it here. This is especially clear in his shirtless farmhouse fight scene where the camera seems to recoil from ever focusing on his midsection. It does not help that he has zero chemistry with Mila Kunis, who also does not seem particularly committed to her role as Russian immigrant housekeeper turned space princess (yes, you read that correctly). Her character is not particularly competent or clever and seems to be largely carried along with little agency of her own.

The major plot arc is essentially the navigation of galactic inheritance law. The process of claiming her title requires Jupiter to proceed through a host of kafkaesque and ridiculous official requirements. The source of this bureaucracy is unclear, but the sequence itself is enjoyable enough, and it culminates with Terry Gilliam himself playing some sort of heraldic functionary, in a clear homage to Brazil. There is a capitalist exploitation theme under all the action, with the Abrasax heirs being primarily concerned with the profit from their youth serum business.

The combat, both in terms of style and power level, looked on screen like how Numenera reads, especially the gravity parkour boots and materializing energy shield. The gravity surfing is not something I think I have seen on the screen before, and it was genuinely kind of exhilarating to watch, especially since much of the action happened in my city, Chicago. I think I could even see my apartment building in a few of the scenes.

Overall, the parts are more interesting than the whole, but we do not get many big budget space operas. The vistas are probably worth it alone, though occasionally the style overwhelms the substance so much as to be distracting.

Oh, and there are dragonborn (which, surprisingly, I can not really find any good pictures of).

jupiter-ascending-005-970x646-cjupiter-ascending-0030-970x646-cjupiter-ascending-00390-970x646-c(Images from here.)

Training bonuses

dark_souls_swords_by_bringess-d7bebkw copy

Dark Souls swords by Bringess

This is an idea for weapon training that I had which is probably too fiddly for online play, but might work in person. To gain more than a +1 to attack or damage (whether from attack bonus, strength, or wherever bonuses come from in your system of choice), a combatant must train with a given weapon. Each weapon is rated with minimum stat requirements and maximum bonus potential. Mundane weapons might be, for example, min +0 and max +3/+3, meaning that anyone without a penalty can use them and they can support at most a +3 to attack and damage. Insisting on using a weapon without the minimum stat imposes some large penalty (-4 or 5E style disadvantage).

The constraints would need to be tracked for each weapon along with training level, which could default to starting scores for starting weapons initially (representing background training). For example, consider a first level Labyrinth Lord fighter with 16 strength, which grants a +2 to attack and damage. A first level fighter in this system also has the equivalent of a +1 bonus to attack. Assume further that this character starts with a battle axe and dagger. Under skills, the player would write battle axe +3/+2 and dagger +3/+2. This is the character’s initial training. If a sword is picked up, it will be wielded with +0/+0 until the character can train with it.

How does training work? As a downtime action, the character can pay for training in a weapon. This costs some set amount, maybe based on bonus to be unlocked, say 500 GP per target plus (so, moving from +1 to +2 would cost 1000 GP). This raises either the attack or the damage by one point, assuming that is supported by class attack bonus or ability scores. Thus, that example first level character above can gain no more bonus points in battle axe until ability scores or attack bonus increase.

The benefits are that it gives fighters something to do during downtime actions, somewhat restrains bonus progression, makes special weapons more valuable without needing to resort to magic weapons as treasure all the time, and allows fighters to ease in to using crazy weapons like the Berserk dragon slayer sword or the asylum demon’s great hammer which might make their use feel a bit more special (and also more fictionally justified). It would also allow weapons to be numerically defined on dimensions of finesse and brutality. For example, a big club might be +1/+4 in potential, making it a good choice for a character with low skill but high strength (assuming such is possible in the base system).

This would would particularly well with a system that has regular stat increases, such as this adventurer class or Green Ronin’s Dragon Age, but should also be functional with a regular attack bonus.


Pits & Perils divergences

Because I am tentatively planning to run a Google Plus hangout game of Pits & Perils this friday, I was reviewing the rules. While doing this, it occurred to me that it might be useful to summarize the aspects that differ from “how things are usually done.”

As a player familiar with traditional fantasy roleplaying games but new to Pits & Perils, I think these are the most obvious differences.

  • PCs have 1 or 2 abilities. As in, you “have strength” (no numbers).
  • Abilities constrain what you can do.
  • To attack, roll 2d6. 9-11 = 1 damage. 12 = 2 damage.
  • Fighters get +1 to attack. Magic weapons grant bonuses here also.
  • Saving throws succeed on 7+ rolling 2d6.
  • Using an ability (for example, dexterity) works like a saving throw.
  • Encumbrance is armor & shield + 10 more items (hard limit).
  • HP is static (for example, a 3rd level fighter has 14 HP).
  • Armor works by adding a small amount of HP.
  • Casting a spell costs one spell point (no preparation).

Image from Pits & Perils title page

Image from Pits & Perils title page

There are a number of other specific rules, but if you already know, for example, B/X D&D, I think this will get you most of the way there.

Also, here’s a list of the core spells with numbers, in case you want to determine starting spells randomly (break out that d24). Or click here for a popup.

  1. Bolt
  2. Call
  3. Calm
  4. Cure
  5. Fade
  6. Fear
  7. Find
  8. Foil
  9. Gaze
  10. Glow
  11. Heal
  12. Hide
  13. Know
  14. Link
  15. Load
  16. Mend
  17. Mute
  18. Null
  19. Pass
  20. Rise
  21. Ruin
  22. Send
  23. Stun
  24. Ward

Dispell as counter-spell

To counter a spell, expend one prepared spell and follow the formula presented for dispell magic:

Dispell Magic: Unless countered, this spell will be effective in dispelling enchantments of most kinds (referee’s option), except those on magical items and the like. This is modified by the following formula. The success of a Dispell Magic spell is a ratio of the dispeller over the original spell caster, so if a 5th level Magic-User attempts to dispell the spell of a 10th level Magic-User there is a 50% chance of success. Duration: 1 turn. Range: 12″.

Source: Men & Magic, page 25.

That is, any prepared spell can be “converted” into an antagonistic dispell magic. One could without issue probably extend that to full dispell magic functionality (thereby doing away with the need to find, learn, or prepare dispell magic as a distinct spell), given that removing enchantments is a relatively core aspect of wizarding, though I could understand wanting to keep the full dispell magic a separate thing.

Note that OD&D does this differently than AD&D, which works like the BRP resistance table (50% success modified by level differential). To make the difference clear, in OD&D a 9th level magic-user has a 90% chance of successfully dispelling an 10th level magic-user’s enchantment (9/10), while in AD&D the chance would be only 45% (50% -5% for having one less level).

Alternatively, compare spell levels rather than class levels using the OD&D formula. So, expending a second level spell in an attempt to counter a third level spell would have a 2/3 (66%) chance of success. This method might be preferred if you see the countering process as an opposition of specific spell energies rather than a contest between the overall skill of the two magic-users.

The burn spells paradigm is becoming increasingly attractive to me. There is some danger of complexity creep, so the possibilities should be limited to a small number of effects. That said, having several default options frees up magic-users to prepare more obscure or interesting spells, which otherwise might not get as much use, much as 3E clerics were able to convert any spells into cures. Further, requiring the expenditure of prepared spells to power such effects retains some degree of resource constraint, unlike make other unlimited or at-will approaches, and doesn’t require tracking any additional information, since spell slots are already managed.

The options so far that I have thought about are maleficence, magical defense, and now dispell magic. I could see adding read magic to that list perhaps, though I am also experimenting with replacing read magic with a “skill” type d6 roll that comes at the cost of an exploration turn.

(This idea came to me when I repurposed dispell magic for banishing summoned creatures.)

Slumbering Ursine Dunes

SUD cover from RPGNow

SUD cover from RPGNow

Chris K. of Hill Cantons is one of the procedural innovators of the OSR blog scene, and his innovations are all the more valuable for arising from solutions to problems experienced in actual play. For examples, see posts on point crawls, ruin crawls, and the chaos index. So, I was excited to see how Chris would realize these ideas in a more formal product, such as Slumbering Ursine Dunes (original kickstarter, current RPGNow PDF).

For a quick overview, the atmosphere of SUD is one part Slavic mythology, one part Moorcock. The major points of the content are a small scale, keyed wilderness, two medium sized “dungeon” adventure sites, a collection of monsters, faction details, and a modular subsystem (the chaos index) which could be used generally. The dunes and adventure sites are both flavorful and easily integrated, I suspect, into most exploration-focused hex crawl type campaigns, though see the note on Slavic cultural specificity below. The tone is light, and though it violates my general preference for setting as straight man (players generally add enough humor on their own), the numerous jokes and site gags here work. For example, the sloth variation on another classic monster, the Terminax, etc.

The basic structure of SUD is a master class in faction design, and for me this was both a complete surprise and its strongest component. Adding any factions at all to an adventure site increases the richness of potential interaction, but the distinct character of each faction here seems like it would drive dramatic conflict particularly well. There is one faction of each major alignment quadrant (lawful good, chaotic good, lawful evil, chaotic evil). Thus, one faction is just interested in carnage and watching the world burn, but all could be useful by enterprising and creative players. When I bother with alignment at all, I generally gravitate toward cosmic interpretations that equate chaos and magic, such as forwarded by Carcosa and LotFP generally, but this setup makes the nine point alignment system shine in a way that I have not seen done nearly as effectively before.

The Slavic cultural inspiration has both upsides and downsides. It makes the texture of the setting more distinctive, but it also makes the dunes slightly harder to drop into a campaigns without modification. The names in particular seem like they may need to be replaced, though that is no big labor. That said, I suspect this distinct cultural flavor will be a draw for anyone that has been saturated with the English-Germanic pastiche implied by many D&D settings and modules. The Moorcockian influences are, for me, uniformly wonderful. For example:

The Eld presence in the Dunes precedes even that of the Master, the extradimensional elves having wrestled the Glittering Tower from the Hyperboreans back when their necromancer king-led states began to crumble more than a 1,500 years ago and then lost it to Medved when he ascended to divinity in the area.

I also enjoyed the immanent divinity of the many minor gods. This is not just a point of taste, though it is that as well. Making deities immanent means that players can interact with them in play. It transforms a campaign element from either info dump or character build option to active encounter, so engaging all players, not just those that do their homework.

The area of the dunes is intended to be governed by mystical, dreamlike logic rather than staid ecological assumptions. In Chris’s home campaign, this is manifested as shift from rational, civilized areas to impressionistic, wild areas called the Weird. It is easy to say this, but standard causal assumptions can easily reassert themselves unless the Weird manifests in play, either through continuous referee engagement or some game system. The chaos index is such a game system. It consists of three a5 pages detailing a simple state machine and several random tables. The referee (secretly?) maintains a number representing the irrationality of a campaign area. (Pause a moment to absorb the irony of managing chaos with a numerical, predicable framework.) Depending on the chaos index level, the referee rolls on various tables which include outcomes like a tesseract opening, the arrival of Eld Bubbleships, and shadowy illusions. The chaos index by default rises 1d4-1 points each session (thus, the expected rise is small but positive) and is also affected by player actions. The system is light enough that I do not think it would be a drag to use, and the tables could be repurposed for encounters or magical catastrophes even if you do not want to engage in the rather minimal bookkeeping.

The layout is unfortunately not as useful as it could be. Objects such as tables are often spread between two pages or even between two spreads. Examples. Pages 7, 8, and 9. Pages 41, 42, and 43. Zombastodon stats are on page 54 and description on page 55 (which are two different spreads). And so forth. Maps diagrams and key entries do not generally share a spread, requiring either page flipping or printing out maps (one more thing to shuffle). Descriptors are also often nestled within several paragraphs of prose, which means I either need to read and take notes, scramble to reread continuously during play, or miss details. Though this remains the standard mode of module writing, and is good at communicating atmosphere during a linear read-through, I find it difficult to use at the table. These are not huge flaws, but I highlight them to bring them to the attention of future creators. A table of contents (both textual and as PDF metadata) would also have been appreciated, to allow quick access to modular tools such as the chaos index.

The art in SUD proper (example) is also not exactly to my taste (the texture looks too much like Photoshop brush). It adequately illustrates many included situations, but does not excite me. I am looking forward to the art in the related and upcoming Misty Isles of the Eld supplement by Luka Rejec.

Despite the layout (objective) and art (subjective) critiques, this is a great little module with several new ideas and fun, interesting details. The content could easily be used in an existing hex crawl, many of the tools are modular, and it is particularly worth picking up to see the faction presentation and how those affiliations permeate all aspects of the scenario.

You can buy the PDF or print on demand perfect-bound paperback at RPGNow.

OD&D summoning

Mateo had the (brilliant) idea to use the retainer and morale system as the base for a summoner class. His approach adjusts the loyalty rolls based on summoner level, monster HD, and other factors. This allows higher level summoners to control more powerful creatures (on average).

Here is another take on the same basic idea, but using the retainer system literally. That is, summoned monsters just occupy retainer slots. Reaction, negotiation, and loyalty rolls are made exactly as if the monster was encountered in the wild. Base rules are 3 LBB OD&D, but I suspect a similar approach would work with other systems. See pages 12 and 13 in Men & Magic and also this old post I wrote about OD&D loyalty and morale.

Monsters defeated in combat may be subdued and brought into service following the rules in Men & Magic. Now, we could pretty much just end there and have a relatively comprehensive system. Magic-users would need to find minions during adventures and convince them to serve or employ charms. However, I like the idea of magic-users being able to gate in creatures directly, and the only even remotely similar spell in Men & Magic is conjure elemental. The monster summoning spells introduced later are not to my taste. So it seems like at least one new rule is required, in the form of a spell that any magic-user could employ if learned:

Summon Monster, level 1 magic-user spell

In a puff of smoke, a monster appears. Use the dungeon random encounter tables, determining first dungeon level* then monster with dice. Establish reaction normally, adjusting for incentives (2d6: 5- hostile, 9+ friendly). Only one monster is summoned. Do not use the number appearing value.

I like the minions that Mateo crafted, but also find the idea of just leveraging everything in the Monster Manuals compelling.

Beyond the new spell, the core of the summoner’s art lies in the skillful application of other standard spells.

Summoned creatures cannot cross a properly prepared circle of protection (requires salt and casting the spell protection from evil). Creatures may be summoned into or outside of such a circle, at the summoner’s discretion.

Other spells that a summoner may want to master: charm person, charm monster.

The dispel magic spell may be used to banish summoned creatures. The “original spell caster level” is the higher of summoner level and monster HD.

The polymorph others spell may be used in conjunction with flesh to stone (using special techniques, they may be cast together, though both must be prepared separately) to transmute minions into figurines, holding them in stasis until recalled using dispel magic (no check required, given that you are dispelling your own enchantment). Such figurines are significant items, and creatures that resist are permitted a single saving throw versus the combined polymorph/stone effect.

Thus, higher level magic-users become better summoners because they are more likely to have all the other spells and be able to set up the summon, ward, compel “combo” more easily and reliably than a lower level conjurer.

In OD&D, I allow magic-users of any level to create scrolls of any spell possessed, even if it is too high level to prepare normally, at the cost of 100 GP per spell level and one downtime action. This rule is based on the scroll creation system in Holmes. It essentially allows ritual casting of higher level spells given time and sufficient GP. Thus, with proper access to spell texts and resources, even a low level magic-user could perform a summoning, but it would be a long, expensive, arduous undertaking, and still entail significant risk.

* Roll 1d6 for dungeon level if using The Underworld & Wilderness Adventures (page 10). The AD&D Monster Manual II probably has the best dungeon encounter tables (page 133), though, so I might use that instead, in which case roll 1d10 for dungeon level. The MMII provides the chance of randomly getting, for example, a duke of hell.