Fixing the GURPS Magic Framework

Fixing GURPS Magic is a two-part process and this post is merely going to offer ideas for the first step: fixing the general framework. A comprehensive bug-fix for GURPS Magic would have to review every single spell, many of which need more than just a quick once-over.

Cooking It up

There are three main problem areas in GURPS Magic and they’re all interdependent:

1) IQ + Magery beats investing points in spells
Unless you want to build a one-spell-pony, the rules-as-written don’t encourage you to invest more than a single  point per spell. IQ and a level of Magery may cost 30 points, but the number of interesting spells is great enough that most mages will have at least twenty spells in two or three colleges. Add in the general versatility of IQ and in-depth study of a spell will never make sense. Official builds from 3rd Edition supported this method with rare exceptions and 4th Edition’s Dungeon Fantasy follows that trend. Not only does this demote spells from real skills to quasi-perks, it invalidates a lot of interesting archetypes.

2) Generalists beat specialists
Magery is one of the traits with the greatest number of canon special limitations, but the one limitation we often see in fiction or the-other-game sees relatively little use on character sheets: One-College-Only. At -40% it offers a decent discount (and can be combined with other limitations), but reduces versatility by something fierce. Going from 800+ to on average 40 spells is not the main problem. The problem is lacking access to the main utility spells (Recover Energy, Magesight, Armor etc.) offering only measly 4 points / level in exchange. Tying maximum Magery to points invested instead of level alleviates this somewhat, but even then other limitations are the better option.

3) Everybody just aims for skill 15/20/25…
Arguably the biggest problem with GURPS Magic are the “free maintenance” break-points. While it is a cute idea to have an extremely skilled mage maintain spells for free, it does turn the game into “Buffing 101”. Combined with the fact that all spells (barring very hard ones) are likely to be at the same skill level this just makes generalists even more powerful − and boring.

Before I search for possible solutions, let’s look at what GURPS Magic does right − something that’s not done very often, but if it was a whole lot of garbage, why even try to fix it?

1) Spells are skills
This fits a lot of tropes in fantasy fiction. While some wizards are masters of improvisation, many do indeed use spells and invocations learned by rote. Even if they can be varied, that often has strict limits. What could and should be changed, though, is that spells are all based on the same attribute and can never be of easy or average difficulty. Some existing spells could also be changed into techniques based on similar spells.

2) You’ve got to learn to walk before you can run
Prerequisite trees are logical in a way that immediately resonates with players. Of course, you need to learn how to conjure little flames before you start throwing around fireballs. The individual trees might not always be completely logical (or easy to follow), but the concept is sound. This comes to the fore when some spells (especially the Weather college) need solid grounding in two or more colleges. This could be expanded even further

3) Great talent can overcome restrictions
One of the few changes 4th Edition introduced what that a high level of Magery can overcome built-in restrictions of spells − and also limits your damage output. This idea is basically sound, but suffers from the fact that virtually all PC mages will have the maximum allowable Magery level.

4) Competence has its perks
While they weren’t added until many years after the release of GURPS Magic, the addition of Magical styles and style perks offer the GM great tools to make their casters unique.

Tying these together this brings a couple of changes to mind:

1) Fixing attribute costs is a good start.

2) Next is disassociating Magery from its talent bonus. Ten points for +1 to all spells was never balanced and if we make IQ more costly, it’s even worse. Instead Magery costs a flat 10 points per level (including level 0) and does only do two things: It allows you exceed levelled or energy-based limits on spells and it allows you to learn more powerful spells.

3) This new Magery is complemented by Talents for single colleges, which cost either 5 to 15 points per level. Talents act as Magery for their respective colleges, but also give a skill bonus in addition. The combined number of Magery levels and Talent levels has a maximum. You can either have broad access to all spells and little talent or loads of specialised talent and a little access to more powerful spells outside that sphere.

4) Skill level has no influence on ritual requirements and maintenance costs. Instead both are tied to relative level and voluntary penalties. With a relative level of attribute −2 you simply cannot cast a spell without words or gestures, with attribute −1 you can try either at a −4 penalty each. Similar, you can try to save on maintenance costs by taking a −4 to effective skill per point of energy − a chancy thing for resisted spells.

5) Casting costs are reduced according to Margin of Success.

The Finished Dish

GURPS Magic is a complicated beast and therefore the finished rule tweaks are a bit more elaborate than usual.

Magery and Magic Talent

Magery costs ten points per level (including level 0). It allows you to exceed limits on effect as per GURPS Magic p. 9. A certain level of Magery is also part the prerequisites of most advanced spells. Without Magery 0 or a corresponding Magic Talent you cannot cast spells. Magery might or might not allow you to recognise magic items on sight or touch as per B 66, but the GM can treat this as a setting switch. Magery never gives a skill bonus or reduces studying time to learn spells.

Magic Talent costs five to fifteen points per level (excluding level 0) and gives all the benefits of the Magery within its respective college. A mage with Magery 0 does not need to buy any zero-level Magic Talents, but it is possible for a mage to have no Magery whatsoever and only rely on Magic Talents. In that case each Magic Talent +0 costs 4 points. While this is rarely cost-effective for more than two levels it opens up some ways to get more Magic Talent than usual (see below).

Magic Talent Cost by College
Air 10 Light/Darkness 5
Animal 5 Making/Breaking 5
Body Control 15 Meta 5
Communication/Empathy 5 Mind Control 15
Earth 5 Movement 10
Enchantment 10 Necromancy 10
Fire 5 Plant 15
Food 5 Protection/Warning 10
Gate 5 Sound 5
Healing 5 Technology 10
Illusion/Creation 5 Water 15
Knowledge 10 Weather 5

In each campaign there is a maximum cost for combined levels of Magery and Magic Talent. In a typical fantasy campaign like Banestorm this might 50 points. That’s enough to buy for example Magery +3 and Magic Talent: Fire +2. Alternatively it could buy Magery +1, Magic Talent: Food +4 and Magic Talent: Movement +1. Or it could buy Magic Talent: Mind Control +2, Magic Talent: Healing +2 and Magic Talent: Knowledge +1 without any Magery whatsoever.

Limitations and Enhancements affect these costs. All special modifiers for Magery work for both advantages except for Limited Colleges, which is only available for Magery, and One-College Only, which is now superfluous.

Higher and lower maximum costs are possible, but the system starts to breaking down at 80 to 90 points unless you have a considerable number of spells requiring Magery/Magic Talent +4. It’s generally a bad idea to allow any Talent give more than +6 bonus so that sets another limit on how high you want to go. At 20 to 40 points the system still works fine, but requires mages to make some hard choices. Less than 20 points are probably not a good idea.

Ritual, Energy Cost and Casting Time

All three of these are tied to both relative level in a skill and how well he can cast it. Ritual grows less elaborate with higher relative level, but does never vanish unless the mage makes a conscious effort and takes a penalty. Maintenance costs can be lowered by taking a penalty, but only if the mage has a high enough relative level. Casting time can be lowered in the same way. Casting costs are dependent on Margin of Success.

Relative level Standard Ritual1 Reduced Ritual2 Cost Reduction Casting time5
Casting3 Maintenance4
Attribute −3 Extremely Elaborate −1 per 5 MoS
Attribute −2 Elaborate −1 per 4 MoS
Attribute −1 Normal No Words or No Gestures −1 per 4 MoS −5 per energy −5 per second
Attribute +0 Normal No Words and No Gestures −1 per 3 MoS −5 per energy −4 per second
Attribute +1 Subtle No Words and No Gestures −1 per 3 MoS −4 per energy −3 per second
Attribute +2 Subtle No Words and No Gestures −1 per 2 MoS −4 per energy −2 per second

1 The words and movements normally required for casting the spell:
Extremely elaborate: Requires sweeping movements of both arms and legs − both hands must be free − and shouted words (base hearing distance: 6 m), which give a clear indication of the spell being cast. Base casting time is multiplied by five.
Elaborate:
Requires movement of both arms and some body full body movement − one hand and both legs must be free − and loudly spoken words (base hearing distance: 4 m), which give some indication of the spell being cast. Base casting time is doubled.
Normal:
Requires gestures with one hand and clearly spoken words (base hearing distance: 1 m), that give those with Thaumatology an indication of the spell being cast. Normal casting time.
Subtle:
Requires subtle gestures with one hand and whispered words (base hearing distance: ½ m), that give those with Thaumatology an indication of the spell being cast. Normal casting time.
2 Eliminating either ritual words or ritual gestures gives a skill penalty of −4, eliminating both gives a penalty of −8.
3 Every full multiple of the given margin of success reduces casting (but not maintenance) costs by 1 energy. MoS is always figured from effective skill, including penalties from reduced ritual, maintenance or time.
4 Maintenance costs can be reduced by 1 energy for every multiple of the penalty taken.
5 Casting time can be reduced either by one second or 10%, whichever is better. Casting time can also be increased to get a bonus to effective skill: x5 gives +1, x20 gives +2, x60 gives +3.

The Leftovers

That’s the basic framework. You still need to look at every spell and decide whether it should be easy, average or hard − very hard spells can stay that way − and whether it wouldn’t be better served by using Will, Perception, Dexterity, Health or even 10+High Manual Dexterity as the controlling attribute. This will be a long and tedious process, but the results should be worth it. While doing that, spells can be balanced against each other, made compliant with 4th Edition concepts (e.g. no absolutes), tagged with keywords, checked for incongruent prerequisites and maybe put into different colleges. It’s probably also a good idea to reduce the number of spells by declaring some to be techniques based on similar spells, too.

As for tweaking the presented rules further, you could fine-tune the costs of Magic Talent per college to cost one point per 5/6/7 spells (minimum: 5 pts.). Or, going into the opposite direction, even declare all 15-point colleges to cost only 10 points. As it is, they are a bit less attractive at the moment, but be aware that a mage with Magery +0 and Magic Talent: Mind Control +4 can be pretty darn effective.


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy

Real-World Stat Conversion: The Stroking Chancellor

By popular demand (as in a lot of people being pissed off at Angelina for being a general nuisance and making refugee children cry) I present a first: a political GURPS post. Stats for our “beloved” chancellor Angela Merkel. For this one I’m using standard by-the-book point costs.

Name: Angela Dorothea Merkel
Age: 61 as of today (don’t confuse this post with a birthday gift)
Occupation: Chancellor of the Federal Republic of Germany
Attributes [0]: ST 8 [-20]; DX 9 [-20]; IQ 12 [40]; HT 10 [0]
Secondary Characteristics [2]: Damage: 1d-2/1d-3; Basic Lift 6.5 kgs; HP 9 [2], WILL 13 [5], PER 11 [-5]; FP 9 [0]; Basic Speed 4.75 [0]; Basic Move 4
Advantages [125]: Administrative Rank +8 (Chancellor) [40], Reputation +2 vs. Power-Loving Bastards: Ruthless (Small class, always) [4], Reputation +1 vs. Ordinary Germans Who Should Know Better: “Mutti” (Large class, always) [3], Status (Head of State) +7 [15, three levels free from Rank, one free from Wealth], Wealth: Multi-Millionaire +0.5 (Net worth ca. $ 11 Mio.) [63]
Disadvantages [-48]: Appearance: Unattractive [-4], Callous [-5], Enemies (9): Various, mostly less powerful (Rivals) [-5], Hidebound [-5], Odious Personal Habit -2: Completely out of Touch with the Common Man (Accessibility: Only in interaction with Status +0 and lower, -40%) [-6], Pacifism: Reluctant Killer [-5], Reputation -3 vs. Various Left-Leaning People (including almost all of Greece): The Boring Axe Woman (Large class, always) [-8], Stubborness [-5], Unfit [-5]
Quirks [-2]: Likes Power [-1], Obsession: Outdo Kohl’s Sixteen Years in Office [-1]
Cultural Familiarities [1]: Western [0], Eastern European [1]
Languages [4]: German (Native) [0], Russian (Fluent) [2], English (Fluent) [2]
Skills and Techniques [43]: Administration-13 [4]; Area Knowledge: Germany-12 [1], Area Knowledge: World-12 [1], Chemistry (TL 7)-12 [4], Computer Operation (TL8)-12 [1], Current Affairs: Business-12 [1], Current Affairs: Politics-15 [8], Diplomacy-10 [1], Electronics Operation: Scientific (TL 7)-11 [1], Intimidation-11 [1], Mathematics: Pure (TL 7)-10 [1], Physics: Quantum Physics (TL7)-9 [1], Politics-16 [16], Public Speaking-11 [1], Skiing-4 [Default], Writing-11 [1]

At 127 points Merkel comes in at many more points than many of her detractors would think. Keep in mind, though that she comes in at 2 points if you strip away her social advantages. Also note that her politics skill doesn’t mean she’s good at solving problems, merely at staying in power. Also having a black heart isn’t a GURPS disadvantage if it doesn’t inconvenience you in any way.

Now, I hope you can work with this base and apply whatever modifications you want. May I suggest some of the templates from GURPS Horror? The Android, Alien Invader and Lilitu seem most appropriate.

Lest It Become a Hereditary Affliction

House rule articles contain a short intro, a rambling section on how to come up with a solution to a problem called “Cooking It up“, just the plain rules in a section called “The Finished Dish” and some musings about what else you could do with that in the final section: The Leftovers“.

One of the things most people agree on is that Affliction is overpriced in the current rules-as-written. That is, what we actually agree on is that effective Afflictions are overpriced. The first level gives a straight HT roll to resist. For that price it’s probably fair that the average Joe can shake it off with a 50/50 chance. The second level costs the same as the first… and makes that a chance of 37.5% – for the average Joe. In short, there’s a low chance to take out important adversaries or monsters with a straight-up Affliction. It’s clearly one of the half-dozen or so things that would be changed, should we ever see another edition of GURPS. There are some ways around this, though.

Cooking It up

The rules-based approach around this problem is taking Malediction 1, which turns the resistance roll into a quick contest between your WILL and the target’s HT +1 minus the level of the Affliction. Unless your Affliction is so heavily limited it costs less than 5 points there is never a reason to buy a second level of this – and even with an end cost of 3 points it would be a debatable investment.

Most official treatments of this, like Psionic Powers, have combined this with a “Skills for Everyone” approach (Powers p. 162). This is extremely advisable if the PCs are expected to have more than one Affliction and even if they don’t. Basically it replaces your WILL roll with a unique hard skill roll. You might even waive the requirement that the Affliction must be part of a power.

Sean Punch suggests two ways to change the costs here:
1) 10 points for the first level + 3 points for the following ones.
2) Give each level 1d6 virtual damage dice that need to overcome armour in the normal way and penalise resistance rolls according to how much damage gets through.

Number 1) solves most problems, but has limits where Afflictions have lots of enhancements and not enough limitations to bring the cost of the later levels down to 3 points. Also an Affliction could be a Malediction 1/2/3 for another 10/15/20 points minus limitations and therefore liable to a much greater discount in cases of high WILL or skill.

Number 2) is a good alternative for most Afflictions that should realistically interact with armour.

What other rules-compliant options are there? You could use Follow-Up for an Affliction that needs less levels to succeed because it ignores armour. You could also choose not to use an Affliction. If you don’t mind doing damage at the same time Side Effect and Symptoms do add explicit Affliction states for a reasonable cost that is not measured in actual levels of Affliction. Indeed, Side Effect is already very similar to the second of Kromm’s suggestions.

You could also rebuild Affliction as follows: It has only ever one level, but you can take a new special enhancement called Hard to Resist that is priced at +40% for each -1 it gives to the target’s resistance roll (or +1 for beneficial afflictions). This works both for normal Afflictions and for those modified by Malediction, though in the latter case it might be more cost-effective to just raise your WILL/skill. You’ll note that this is close to Kromm’s suggestion no. 1 above, but making the lowered resistance an enhancement takes away the danger of inflated costs for the more extreme Afflictions.

The Finished Dish

Rules-compliant ways to effectively use Afflictions:

  • Modify it with Malediction and use a unique skill.
  • Modify it with Malediction and use straight WILL. This is likely too powerful!
  • Modify it with Armour Divisor to get rid of part of the bonus armour provides.
  • Use a carrier attack and modify the Affliction with Follow-Up to do damage and ignore armour if the carrier attack overcomes it.
  • Modify an Innate Attack with Side Effect to afflict negative states on a target – at a penalty according to penetrating damage.
  • Modify an Innate Attack with Symptoms to automatically afflict negative states on a target once damage exceeds certain thresholds.

Variants that change the rules-as-written:

  • Kromm’s two ideas.
  • Affliction only ever has one level. A special levelled enhancement called Hard to Resist gives a -1  penalty (+1 bonus for beneficial Afflictions) per level to resist the Affliction. It costs +40% per level and works with Malediction.

The Leftovers

All the old rules-compliant ways to make your Affliction cost-effective still work with the new Affliction pricing. Kromm’s second suggestion works as an alternative. In that case Afflictions retain their old cost structure.

Now, what if you think it’s too cheap to make a heavily modified killing curse? After all,  Affliction (Heart Attack, +300%; Hard to Resist +8, +320%; Costs 8 FP, -40%;  Limited Use: 1 time / day, -40%) is only 15 points if you use multiplicative modifiers. The easy solution is to make the different effects change the base cost instead of being enhancements. So Attribute Penalty ST -2 would only cost 12 points as base, while Heart Attack would be 40 points. The aforementioned ability would then cost 26 points, which seems appropriate for its limitations.


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy

Skills vs. Attributes

House rule articles contain a short intro, a rambling section on how to come up with a solution to a problem called “Cooking It up“, just the plain rules in a section called “The Finished Dish” and some musings about what else you could do with that in the final section: The Leftovers“.

One problem I’ve often faced when it comes to character creation is the fact that it’s just more effective to take yet another level in a controlling attribute than raising a skill above attribute level. This happens most often with mages who raise their IQ sky-high after taking however much Magery the GM allows.

It’s less common with fighters and other characters who need more than one attribute to fulfil most of their functions, but the fact remains: There’s little point to raise more than one or two skills above attribute level. Most often these are the one main combat skill and another to get into position (Driving, Riding etc.).

Cooking It up

There are many who say that the attribute/skill pricing is working as designed and that a highly-skilled character should have high attributes to show this, but I think that over-simplifies things. Skills also represent experience in a subject and can float to other attributes and even a flat base 10. Having DX 16 doesn’t help you maintain your gun, care for your horse or remember to buy fuel.

That’s ancillary to the two main points, however. The first is character concept: There are many times where you want to have a character that is brilliant in three or even four unrelated fields without being an overall genius. That’s a legitimate and realistic concept, but in GURPS you are forced to accept that this is going to waste a lot of points. The GM could make up a special 5-point Talent just for you, but that’s it, as far as options go. The second point is niche-protection. With high DX and IQ (HT, PER and WILL are less of a problem) characters easily encroach on each others’ terrain. The face man has been taken out? No, problem: Let the mage do it. He has two points in Diplomacy and one in Fast-Talk that gives him 16 and 15 respectively. You can mitigate that problem by not letting your characters buy certain skills, but then you’re in Dungeon Fantasy territory again and that’s a problem when the rest of your campaign follows a realistic pattern.

Just to make myself clear: I do not subscribe to the view that all characters should have attributes in the 10-12 range and high competence should be modelled by having half a dozen skills at attribute +5 level. We do, however, need a little more wriggling room.

Now, what to do about that? A solution for half of the main problem can be found among Reverend Pee Kitty’s house rules: PER and WILL are separate from IQ and IQ costs 20 points a level. Personally I’d adjust the price of IQ to 25 points a level. Sure that sounds like a lot, but keep in mind that you can always adjust your starting point level accordingly.

The other main half of the problem is, of course, DX. Instead of separating out both Basic Speed and Basic Move, there is a strong point to be made for keeping coordination (plain DX) tied to reaction speed (Basic Speed). Now running/flying/swimming fast is a completely different kind of beast and it should rightly be separated out. But there’s a third one, isn’t there? Tasks where High Manual Dexterity (HMD) comes into play still profit from a high base DX that represents mainly gross motor skills.

Separating HMD out is a bit of a problem. We don’t want to turn it into a full attribute, because the drawbacks of a low level (rightly represented by Ham-Fisted) are by far not as dramatic as the benefits of a high level. It’s probably best to keep in mind that in certain cases it makes sense to float the relevant skills to a flat base + HMD that makes sense. Keep in mind that a flat base can still be modified by task difficulty.

So, how much should the end product of DX + Basic Speed cost? I’d put it at an even 25 points per level. That makes it come out slightly ahead of IQ, whichlost both its secondary characteristics. However, there are no skills based on Basic Speed and there’s a lot more overlap in skills covered by DX (and relevant talents). Nobody needs more than at most ten combat skills, but even twenty IQ-based skills can cover wildly disparate subjects.

Now, there’s also the problem of HT. There the problem is less the existence of too many HT-based skills, but that the stat is darn useful overall. I hardly ever see an adventurer-type character with less than HT 11 (12 for fighters). Douglas H. Cole has covered this in a very readable article called “The Price of Fitness“. He comes up with a final price of 20-25 points per level. That is a bit high compared to my other attributes, so I suggest a final price of 15 points per level, but with FP separated out. The connection to Basic Speed stays as it is.

The last part of the puzzle is ST. While there are no ST-based skills and only two techniques based on it (Wrench (Limb) and Neck Snap), it still needs to be on par with the other attributes. Now does ST 20 pack the same punch as DX 14, IQ 14 or HT 16 and FP 13? Well, it does and then some. The base damage along with the increased carrying capacity is already more than enough. So, let’s at least separate out HP as a newly independent stat. After all, fat is not necessarily worse at absorbing damage than muscle. Indeed muscle might be more problematic since it fulfils an innate function. We end up with ST as the cheapest attribute at 10 points a level. Smarter minds than me might think about an appropriate way to rescale damage that makes this price a bit more reasonable. For now just keep in mind that – like HT – it’s relatively cheap compared to DX and IQ.

The Finished Dish

Attributes and secondary characteristics are changed as following:

ST (10 points/level): Does no longer affect HP. But HP are still limited to +/- 30% of ST. Damage and Basic Lift are unchanged.

DX (25 points/level): Does no longer affect Basic Move, which is now completely independent from DX and HT. Still has its normal effect on Basic Speed.

IQ (25 points/level): Does no longer affect PER and WILL at all, both are completely independent from IQ.

HT (15 points/level): Does no longer affect FP, which are still limited to +/-30% of HT (including HT bonuses from Fit and Very Fit).

HP (2 points/level): Are unaffected by ST, but limited to to +/- 30% of ST. The GM might rule that certain builds might modify that limit (+/- 10% for Skinny, +/- 40% for Overweight, +/- 50% for Fat and +/- 60% for Fat), but keep in mind that heavier builds generally have somewhat higher ST to compensate.

Basic Speed (5 points/0,25 levels): Unchanged from RAW.

Basic Move (5 points/level): Starts at 5 for native environment. All other rules referring to Basic Move or full move use the final level bought up or down from 5.

WILL (5 points/level): Is completely unrelated to IQ. Even mentally handicapped people might have great resistance to influence and genetically engineered slave races might have next to none.

PER (5 points/level): Is completely unrelated to IQ, but should rarely go below 7 for characters who are able to lead a relatively independent life.

FP (3 points/level): Are unaffected by HT, but still limited to +/-30% of HT.

Hard to Kill (4 points/level) and Hard to Subdue (4 points/level): No change apart from the costs. Are still included in HT.

Arm ST (4, 6, 9 points/level), Lifting ST (4 points/level) and Striking ST (6 points/level): No change apart from the costs. Lifting ST and Striking ST together literally are ST. Don’t buy both of them, simply buy ST!

The Leftovers

I haven’t yet said how this affects starting point values. Some character types are, of course, more affected than others. The all-rounder with 12 in all attributes and secondary characteristics comes in at 185 points in this system, compared to 120 in the rules-as-written. The brute with ST 18, DX 12, HT 13 and the secondary characteristics to match costs 205 points in the new system and 150 in the old. The genius with IQ 16 costs 210 points and 120 respectively. All in all, you should probably make sure to use a 30-50% higher starting point total if you want to make all these concepts possible. So your standard 150 point campaign should at least go up to 200 points now, possibly even to 225 points.

As a side-effect of this change characters built on raw physical strength and endurance become more viable compared to technical specialist fighters that always go for the eyes or vitals. Personally, I think this is a worthwhile outcome of the change. Also players might consider using 15-point talents now – especially ones that cover IQ-, PER- and WILL-based skills (Smooth Talent Cost from Power-Ups 5 is still advisable though).


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. Creations of other GURPS fans are clearly attributed. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy

Fine-Tuning Languages

House rule articles contain a short intro, a rambling section on how to come up with a solution to a problem called “Cooking It up“, just the plain rules in a section called “The Finished Dish” and some musings about what else you could do with that in the final section: The Leftovers“.

In GURPS problems with languages go two ways: how to handle having many languages and how to handle not being completely proficient. I’ve yet to deal with the former problem. I usually tell my players beforehand what languages might be useful and throw the occasional bone in the direction of players who took one that wasn’t usually cropping up. But proficiency was a problem we often had to deal with. Even having broken proficiency lets you communicate rather freely. Only in stressful situations does it require an IQ roll to understand someone or get your meaning across. Everybody who has ever visited Paris with nothing but their rusty school French can attest that this is overly optimistic – though admittedly most situations will become rather stressful fast.

Cooking It up

What’s needed are more finely-grained levels of proficiency and better rules for operating at that proficiency. Now, it makes little sense to inflate the price of languages any further. They’re already pricier than having decent skills. So, the intermediate levels could either use half-points (I know I start sounding like a half-trick-pony) or link spoken and written comprehension levels together. I am going to present both options below.

Now, there’s nothing really difficult about coming up with the rest of the rules, but I’d like to throw in a couple thoughts about language defaults. For most languages these should realistically be at a level below broken, but there are some exceptions concerning bona fide languages (e.g. Norwegian and Swedish). Having a smoother progression of proficiency levels helps in this regard.

The Finished Dish

Instead of the three comprehension levels for spoken and written language in Basic Set, there are six (not counting None).
All IQ rolls are modified by how stressful the situation is and how well the participants can perceive the statements made. The modifiers given on B24 for broken comprehension apply for spoken conversation. Written conversation seldom takes penalties here, though online chats can be stressful and writing can be defaced, smeared or hard to read. Also you won’t get immediate feedback during most written communication. Time Spent modifiers (B346) apply.If you are making use of half-points, use the rules as written. If not, either make sure the total sum of points invested into a language is not a fraction or pay another half point to round it up.

Spoken Comprehension Levels

Rudimentary (0.5 points): You are only able to express the most basic concepts (“I need food”, “I surrender”, “I have money” etc.) and even these require an IQ by both participants to get across.
Using skills dependent on language is mostly impossible or at least takes a -12 penalty (doubled for skills dependent on the beauty of the language).
Those predisposed to dislike the character react at an additional -5. Those intolerant towards the character’s nationality won’t react at a better level than bad.
This level is mainly useful for travellers who have a tiny chance of encountering native speakers, but a large chance of those being hostile. On the discworld Rincewind was known for being able to beg for mercy in dozens of languages.

Broken (1 point): You are able to express slightly more complex concepts (e.g. “I need medicine for my sick daughter. I have the money to pay for it.”), but your grammar, vocabulary and pronunciation are very bad. Communicating all but the simplest concepts requires you to make an IQ roll.
Using skills dependent on language takes a -6 penalty (doubled for skills dependent on the beauty of the language).
Those predisposed to dislike you react at an additional -3. Those intolerant towards your nationality won’t react at a better level than poor.
This is the level most pupils manage to reach after three years of foreign language education.

Limited (1.5 points): You are able to express moderately complex concepts in fields he is familiar with (e.g. “If you want to remove the blockage, you need to open the outflow valve an increase water pressure”). Treat everyday knowledge and the subject of every skill in which you have invested at least two points as a familiar field. Your grammar is simple, but correct more often than not. Your pronunciation is still moderately bad, but generally comprehensibly. Vocabulary varies according to subject matter. There is normally no IQ roll for communicating, but in stressful situations the GM might require one.
Using skills dependent on language takes a -3 penalty (doubled for skills dependent on the beauty of the language).
Those predisposed to dislike you react at an additional -2. Those intolerant towards your nationality won’t react at a better level than poor.
All but the worst pupils should reach the level after six years of being taught the language at school.

Fluent (2 points): You are able to express complex concepts and you generally use correct grammar, vocabulary and more-or-less correct pronunciation. False cognates are still a source of problems as is using words of the wrong register or with the wrong connotation. The GM might assess a penalty of -1 or (in extreme cases) of -2 to a skill roll where applicable.
Using skills dependent on language generally takes a -1 penalty (doubled for skills dependent on the beauty of the language).
Only those intolerant towards your nationality react at an additional -1.
This is the level usually expected of students who want to enrol for a major – at least in countries where this language is commonly taught in school.

Near-native (2.5 points): You are able to express even the most complex concepts. You make very few mistakes in grammar, vocabulary and pronunciation, but are usually not quite good enough to pass as a native speaker when talking to actual native speakers.
There is only a penalty of -1 for using skills dependent on the beauty of the language.
Even those intolerant towards your nationality won’t react worse than their intolerance dictates.
This level is acceptable for somebody who has completed a major degree in the language.

Native (3 points): Even if you aren’t a native speaker, a regular native speaker won’t be able to glean that fact just from hearing you talk. A master linguist might be able to tell after talking to you at length and making a PER-based Linguistics roll.
You can converse in the standard language and a dialect of your choice (free of charge). This can be a dialect that is very far from the standard language. If you choose such a dialect, be aware that others will treat you as if that were your native dialect – unless you keep it completely secret. They might congratulate you on learning the standard language or despise you for giving up your roots.
This level isn’t very common among mere students of a language. You normally need full-time immersion to get there and even then many non-native speakers never reach this level.

Written Comprehension Levels

Written comprehension levels work exactly as spoken ones, though your handwriting does not necessarily give you away if you have a higher proficiency in another language that uses the same script (e.g. Latin or Cyrillic Script). Self-study by using books works as follows.

Rudimentary (0.5 points): Self-study using books is impossible.

Broken (1 point): Self-study using books takes triple the usual amount of time (1200 hours for exclusively book-taught skills)  if it is possible at all.

Limited (1.5 points): Self-study using books takes double the usual amount of time (800 hours for exclusively book-taught skills) .

Fluent (2 points): Self-study using books takes 1.5 times the usual amount of time (600 hours for exclusively book-taught skills) .

Near-native (2.5 points): Self-study using books takes no penalty.

Native (3 points): As above, in addition you may acquire a handwriting style uniquely associated with this language at no extra cost.

Language Talent

With these rules Language Talent works slightly different. Instead of giving you comprehension one level higher it gives you comprehension two levels higher than what you pay for, e.g. Limited spoken proficiency at 0.5 points and Native spoken proficiency at 2 points.

Default Use of Languages

In most cases it’s convenient to treat most languages that show differences, but are still for the largest part mutually intelligible as dialects – especially when they share the same written form. That does not necessarily reflect a reality where Chinese speakers from the north cannot understand the southern speakers and East Frisians cannot make sense of speakers of Swizerdüütsch, but the line has to be drawn somewhere.

Genuinely different languages might, however, be closely enough related like German and Dutch, Swedish and Norwegian or Ukranian und Russian. These give a character proficient in one language an automatic default in the related language without having to pay points for that. Quite often conversations between two such speakers do take quite a while longer, though. The GM can require an IQ roll when time is of the essence and rule that the information takes up to thrice as long to convey if that roll fails.
They can then buy up their proficiency from that default without having to pay the points needed to reach that level (similar to skill defaults). Should a character buy the language without having the other one at a sufficiently high level to get a default, they still get points back if they ever reach the level needed to get that default. The following table gives the levels needed for defaults and the defaults provided:

Mutual Intelligibility Proficiency Level Needed Default Provided
Significant Fluent 2 levels lower
Partial Near-Native 3 levels lower
Limited Near-Native 4 levels lower

Examples: Norwegian and Swedish are significantly mutually intelligible. Native speakers of one language can converse with native speaker of the other as if they both had fluent comprehension. Mutual intelligibility of German and Dutch is limited. A near-native speaker of German will have only a rudimentary comprehension of Dutch and even a native speaker won’t have better than Broken comprehension.

Instead of listing all the languages with relevant levels of mutual intelligibility, I just link to the relevant wikipedia article, which completely incidentally uses the same terms for the three levels above. Unfortunately, it doesn’t give the direction of intelligibility when it is asymmetrical. You’ll have to do a bit more research in these cases.

The Leftovers

I haven’t yet touched on the subject of pidgins and creoles. The former are simplified languages often used as a means of communications between members of different languages. They don’t as a rule have any speakers who use them as their first language. The latter are complete languages that developed from a mixing of languages. Most creoles should be treated as regular languages, although the might have a rate of mutual intelligibility.

Pidgins are most often used as a means for facilitating communication. They don’t have a written language of their own (but can usually make use of one of the parent languages’ written system) and often cannot express all the concepts available in a complete language. The GM sets the highest proficiency level that exists in the spoken and written form of the pidgin. Often this is the fluent level for the spoken form and none for the written form. It is however perfectly possible for a pidgin to go up to near-native or end with broken – the latter will probably develop further if they remain in use.

Pidgins are easy to use and have a simplified structure. Whenever there’s the question of an IQ roll the GM assigns a bonus to that for every comprehension level missing at the top. For example a pidgin that goes up to fluent gives +2 to to all IQ rolls for comprehension at lower levels. Skill penalties are even reduced by twice that number. There are normally no reaction penalties associated with a low proficiency in a pidgin as it is nobody’s mother tongue. They may however be called for in special situations, e.g. when a person sent for important negotiations shows low proficiency in the pidgin that would be commonly used.

The GM should also keep in mind that pidgins might develop into full-blown creole languages and that might even happen within the space of a character’s life.


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy

Technique Pricing

House rule articles contain a short intro, a rambling section on how to come up with a solution to a problem called “Cooking It up“, just the plain rules in a section called “The Finished Dish” and some musings about what else you could do with that in the final section: The Leftovers“.

For highly-skilled characters I’ve frequently run into the problem that I can’t really justify putting points in more than one technique apiece and often these are very powerful hard techniques like Feint or Dual-Weapon Attack. In 3rd Edition with its 8 points a level for DX-based skills above DX+1 it made much more sense to use techniques (then called manoeuvres). I feel that this led to more colourful characters. So how can we achieve the same thing in 4th Edition?

Cooking It up

There’s no reason to go back to 3rd Edition’s skill pricing. DX and IQ are already much more useful than buying more than three skills based on each up to above the attribute level. So, the techniques themselves need to become cheaper.

There’s also no reason not to include easy techniques alongside average and hard ones. These would be for things you really don’t want to spend a lot of points for, like the Impersonate technique (B233). There’s also room for very hard techniques. These could be used as a catch-all category for skills currently classed as hard that you might want to make more expensive, because they are a good idea under most circumstances (Dual-Weapon Attack, Feint and many cinematic techniques). This way easy techniques will be very different price-wise from their more heavy-hitting brethren.

While you’re at it, ignore the advice on B229 about selecting the difficulty level. The rules in Martial Arts for building your own techniques don’t use it either. Difficulty is based on how difficult something would be in real life, how large a part of the skill it helps with and how useful it is. Generally nothing that on its own allows you to pursue a decent career (e.g. Motion-Picture Camera for Photography) should be less than an average technique. On the other hand, many easy skills mostly have easy and average techniques.

For the actual pricing it’s good to consider half-points or – as an alternative for those not using these – skipping over certain levels.

The Finished Dish

To make some very specialised and/or not so useful applications of a skill easier, introduce easy and very hard techniques. Use half-points to give smoother progressions or skip over every second level for easy, average and very hard skills.

Final Skill Level Easy Average Hard Very  Hard
Default 0 0 0 0
Default +1 0,5 1 1 2
Default +2 1 1,5 2 3,5
Default +3 1,5 2 3 5
Default +4 2 2,5 4 6,5
Default +5 2,5 3 5 8
Default +6 3 3,5 6 9,5
Further +1 +0,5 +0,5 +1 +1,5

Be sure to re-evaluate all existing techniques in your campaign. They might fit better in a neighbouring category with this scheme.

The Leftovers

Ideally, I’d also post an adjusted technique listing for all the techniques found in Basic and Martial Arts, but I’m not sure whether it’s legal to list all techniques and I don’t have the time to go over each of them at the moment.

If you want to encourage techniques even more, consider giving a character a free point in a technique once they reach attribute level +1 in a skill.

The less coarse-grained technique difficulties also invite you to think about a way to redo optional specialisations as techniques. Instead of shifting down the difficulty of a skill if a character knows only one special subject, buy a lower level of the skill normally and buy up a technique for the character’s speciality. Base the difficulty on how large a subset and how useful the speciality is. Zoology is probably an very hard speciality technique of Biology, while Entomology might just be hard and the study of domestic cats might get away with being average.

Everything that uses techniques is also affected by these changes. I’m especially fond of the way they impact adjustable spells, which were really too pricey in most cases. You might think up a scheme that makes use of the different difficulty classes, too. Maybe reduced energy costs for higher difficulties.


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy

The Humble Half-Point

House rule articles contain a short intro, a rambling section on how to come up with a solution to a problem called “Cooking It up“, just the plain rules in a section called “The Finished Dish” and some musings about what else you could do with that in the final section: The Leftovers“.

One of the things from 3rd Edition Revised I’ve been missing is the use of half-points for skills with little training. It was an easy way to give a character some background colour without making them spend too many points. Of course, that turned out to be the crux of the problem. In 3rd Edition spending half a point on a skill gave you a skill level of -1 below what you got for spending a whole point. That worked fine for a regular-guys range of attributes, but it quickly broke down for mages and martial artists – the former especially spent only 0.5 points on each of their hard spells and 1 point on their very hard ones. In short abolishing the half-point was better than keeping it unchanged.

Cooking it up

This section contains my thoughts on the matter at hand. It may be a bit rambling. Look at the next section if you just want the rules.

Now, in 4th Edition we have the Dabbler perk, which seems to offer a good way out of problems like that. I don’t like it for several reasons:

  • You basically need two or more skills you want to be slightly proficient in and whatever level you choose in one affects what level you can choose in the others. In practice it’s a bit messy. Do you write the skills and their levels with the perk or do you write them in your skill list with a special note? What happens if you learn one of the skills for real?
  • The maths is based not skill level as when learning a skill, but on default attribute level. That makes for weird cases like having a level of attribute-4 in Mathematics (a hard skill) as costly as having it in Physics (a very hard skill). That may be more of a problem of the defaults, but that’s how it works out in play.
  • That brings us to the worst offender so far: it’s darn cheap. The typical IQ 15-mage can get the classics Diplomacy, Mathematics, Naturalist and Tactics all at skill level 11 for a single point.
    Sure, you need to sacrifice a general perk slot, but I haven’t ever run afoul of that limit in 100-point and higher campaigns – in lower-point campaigns dabbler doesn’t do much anyway. It doesn’t count as a studied skill, but that’s the only drawback.
  • That drawback makes sense game-mechanically, but not story-wise. What does dabbler represent if not superficial learning?

Now, an easy solution is to reintroduce the half-point, but keep it at steady -2 to skill below the 1-point level. The mage in the example above would pay two points for his four skills. There would be no incentive to use dabbler for very hard skills with defaults of -6. Skills could still be listed in the regular skill section with a cost of 0.5. They could be used as defaults and so on. In order to prevent abuse they cannot be used to fulfil prerequisites.

The Finished Dish

This section gives the plain and simple houserules that can be used without further ado.

Skills can be bought for 0.5 points. That gives you a skill level of -2 compared to what 1 point in a skill gets you (e.g. attribute -4 for a hard skill). That skill is normal in every other regard. The rule for prerequisite skills is unchanged. You need at least a whole point in a skill for it to count as a prerequisite.
The GM is free to set a limit on the number of half-point skills a character can have if they feel the rule invites abuse. That may be a hard cap on the number of these skills – preferably an even one – or restricting them to mundane skills and requiring a one point minimum in esoteric skills, cinematic ones and spells.

The Leftovers

This section contains further musings on the topic that might be codified at some later point.

If half-points get out of hand in your campaign, insist that character’s have to spend the other half point, once they have used a skill often enough to justify some increase in proficiency.

Half-points could also be used for determining the final cost of abilities. This would often prevent the common problem of having a low point advantage with a low-point, but non-trivial limitation. It would however make it necessary to recalculate all the official abilities and templates. But for those using multiplicative modifiers that work has to be done anyway.

Another option would be spending half-points to turn a marginal failure by one into a success. Sure, it’s pricey compared to turning any failure into a success, but if you don’t need to succeed by more than 0 you might as well save that half-point.

Also giving out half-points (or even quarter-points) is a good way to reward clever players or moving role-playing without inflating points in essentially realistic campaigns. If the players get three points a sessions adding one whole point is a big deal. If you just give half a point extra it’s perfectly justifiable.

Or you might use half-points to give smoother progression of technique levels.


The material presented here is my original creation, intended for use with the GURPS system from Steve Jackson Games. This material is not official and is not endorsed by Steve Jackson Games.

GURPS is a registered trademark of Steve Jackson Games, and the art here is copyrighted by Steve Jackson Games. All rights are reserved by SJ Games. This material is used here in accordance with the SJ Games online policy