Nobody's perfect, not even GURPS. There are a few rules that I think are actually pretty bad as written and should really be revised. I'm not including optional rules that I don't ever use, or even obscure traits that I just haven't seen a use for (but acknowledge there might be). I'm also not going to cover rules that saw a significant revision in later works (e.g. GURPS High-Tech's treatment of hiking, GURPS Gun Fu's revision of Gunslinger, or GURPS Horror's rules for mental stun). Instead these are all general rules that I think just don't work as intended and don't have a good fix in print. I'll cover three of them here, but there are more, so I may revisit this topic in the future.
Animal Empathy (p. B40)
This advantage does two things. The first, an IQ roll to sense animal emotions, is fine. It's the second, the use of influence skills on animals, that causes two major problems.
By using the same influence skills as those to manipulate people it incentivizes characters that are good at both. In fact it's hard to make a character that's not good at manipulating people with useful levels of influence skills for animals. Since the usual fictional archetype of the beast tamer is a wilderness scout or feral person without much human socialization, this is a challenge for people wanting to play these types of characters. You can take traits like Animal Friend with disadvantages like Low Empathy, and also take (Animals) as an optional specialty on your influence skills, and get something close to a poorly socialized person that's able to soothe wild beasts but this is a very convoluted approach to what should be a straightforward build (and results in canceling the benefit of the first three levels of Animal Friend).
The other problem is that once you apply the influence skills to animals, they must work very differently than they do with humans. Your diplomat suddenly switches from a mastery of subtle negotiation to mastery of subtle threat and submission postures. Streetwise becomes a skill both about navigating the underworld and about communicating to lone wolves and other "outsider" animals. Sex appeal becomes a skill that you can use to trick animals into thinking a mating opportunity exists, but you still can't use it like this on people, that's Fast-Talk. This
post and this one
by PK explains the rules as intended here, and no, I still don't like it.
Note that Intimidation works across species boundaries, already, without any special traits.
Instead of making influence skills work on animals, Animal Empathy could make animal skills into influence skills:
Animal Handling, Falconry, (IQ-Based) Riding, and Teamster work on the appropriate specialization; success gives a "Very Good" reaction.
Mimicry (Animal Sounds or Bird Calls) works on animals whose vocalizations you can mimic. Can be used to enable specious intimidation, with the normal consequences for failure.
Naturalist works on all animals. As with Diplomacy, use the better of this or a regular reaction roll.
(IQ-Based) Survival works on animals native to the biome.
Improvement Through Study (pp. B293-B294) and Time Use Sheets (p. B499, p. B569)
First I hate homework. The course of my life has largely been defined by my difficulties with homework. Obviously other people like homework, so it's largely a matter of taste. However I think making homework a not optional rule was probably a bad design choice, but I acknowledge that I'm biased.
These rules have problems that go far beyond that though:
- Improvement through Study has no diminishing returns. 200 hours with a teacher (or 400 self-study, or 1600 on-the-job, or 100 hours of intensive training) gives a character point. It doesn't matter if it's your first or fiftieth. Not only does this not reflect pedological science, it also doesn't reflect common experience. After you've learned the basics of something continuing practice doesn't result in steady improvement. Practice has diminishing returns, as you find less and less opportunities to challenge your ability or expand your knowledge. This post on Gaming Ballistic goes into much more detail about this.
- Many, perhaps most, advantages and disadvantages should have effects on downtime (Status, Contacts, Addiction, Dependents, Duty...). Of special note are Single-Minded and Laziness, both of which should directly impact how efficiently you can use your free-time. There's no mention of this on B499, and Time Use Sheets aren't mentioned in any specific trait description, with only traits like Less Sleep and Slow Eater expressly modifying any downtime activity. Monasticism and Mysticism are exceptions; but it's unclear exactly how "75% of your time" and "Most of your time" actually translate into hours for "Religious Observances"; taken literally these traits would seem to preclude any downtime use whatsoever.
- It's not very generic, e.g the 40 hour GURPS workweek is largely a 20th century American construct, modification for other cultures or even species is left to the GM.
- It allows unrealistic amounts of time spent on studying a single skill. A unemployed person can spend 12 hours a day on a single skill, a part-time person can spend eight hours, a full-time person can spend four hours. There's no accounting for boredom, Single-Minded isn't a requirement, Laziness has no specified in-game effect here; anybody can just mindlessly spend half or more of their free waking hours alone doing math problems or sword drills or whatever, day after day for years on end if they want.
- The form on p. B569 only allots 2 hours and 51 minutes per day for "meals, personal care, etc." and only Slow Eater expressly modifies this. At minimum this is three half hour meals, leaving 71 minutes for "personal care, etc.". It's unclear what "etc." includes here other than that it excludes "Travel", "Entertainment", and "Religious Observances"; if it's meant to include everything that's missing (see the next item) it's a very rushed life, a GURPS character leads.
- Activities missing from this include:
- Maintenance — Characters with the Maintenance disadvantage require a specified time, but most characters also need to spend some time repairing or maintaining weapons, equipment, clothing, vehicles, dwellings, etc. Similarly health care also at least occasionally requires dedicated time.
- Sustenance and sundries — Whether you are a hunter-gatherer or a 21st century office drone at some point you'll need to go out and get food, clothing, tools, etc. This often can average more than the hour or so a day you'd have if this was included in "personal care".
- Exercise — Realistically, characters with traits like Fit or Very Fit, would lose these traits without regular exercise, and most characters living sedentary lives would eventually get Unfit. The (optional) rules for maintaining skills don't address advantages at all, but really there just should be time for exercise included, if it's not coming from a physical job or study of an athletic skill or advantage.
- Maintaining personal relationships — Most people who spend most of their waking hours doggedly buried in textbooks or doing katas would be very poor spouses, parents, and members of their communities. Real humans are social, and spend much of our time at maintaining our social connections.
- Illness and fatigue — Humans (and presumably most other biological characters without significant Immunities) don't always feel well. In times and places with poor public health, you may even spend a plurality of your time suffering symptoms of non-life-threatening diseases and syndromes.
- Wasted time — Nobody is 100% efficient, and we all waste time. Some traits should cause more wasted time than others, and really this includes most mental disadvantages. The form assumes completely efficient time management which is absurd.
- Delays — Even if you are somehow perfect at time management, the world you live in isn't. You will be stuck in traffic, or waiting on a bureaucratic process, or in a long line while shopping, etc.
- Hygiene and cosmetics — While some people can easily fit bathing, etc. into that 71 minutes per day, this is definitely not true for many other people. Some culture's standards of beauty require hours per day of maintenance, and this typically increases with both Status and Appearance in many cultures.
- Commuting — While this may be supposed to be included in Travel, it's not clear that it is. When the GURPS 40 hour week is literally the 20th-21st century American model, it also includes something like an hour daily commute on average. Other people in other times and places may spend more or less time getting to and from work.
The no homework solution:
Schools, boot camps etc. should just give a fixed amount of points in specific traits at the conclusion of the course of study (which needs to be appropriate to the character's skill level). Every two months of total immersion gives 1 point towards both Language and to Cultural Familiarity. Furthermore, at the GM's discretion the character may either work on a Long-Task or Study a new skill. If the latter they gain 1 point in the skill after six months (halved with a teacher) — traits like Single Mindedness or Laziness can reduce or increase this time by up to 50%.
The more homework solution:
Time use should include all the items mentioned above, and be tailored for cultural circumstances; GM's and players should consider all of these concerns before allocating time for Long Tasks or Study. A rule of thumb for traits that take up time (and don't otherwise specify) could be 1 hour per five points each day (this is ~56% less than Slow Eater; but this is all that Slow Eater does). Characters should make Will rolls to study the same skill for more than 2d+3 days straight, modified by Single Minded. On a failure they need to do something else instead for 1d days. Lazy characters should probably halve the maximum time available for study (just as they halve job income).
Either way, self-study and on-the-job training should be limited to just learning the basics. Languages gained through immersion should be limited to Accented without Language Talent. Skills should be limited to the first two points (or first four with a teacher); further improvement requires adventuring, formal higher education, or intensive training. Learnable advantages like Fit may be limited to the first level without a private couch, gym class, dojo, militia, etc.
Rated ST for Crossbows (p. B270)
The rules for crossbows allow any crossbow to have arbitrarily high rated ST. While this is also true of bows, crossbows can be loaded with mechanical aids, up to 3x effective ST. Increasing ST doesn't ever increase the weight or bulk of the weapon, which of course, is nonsensical. Rated ST is dependent on draw length, which means that at some point your crossbow is going to get larger and heavier, but the rules do not address this. In real life TL3 pistol crossbows were expensive toys, but in GURPS you can easily have one at ST 21, doing 2d+2 imp, weighing only 4 lbs. and with bulk -4; making it absurdly better than TL4 and early TL5 pistol firearms with similar loading times!
Furthermore "Bows, Crossbows, and Rated ST", GURPS Low-Tech p. 74 has steel bows which "can be built with much higher rated ST" at the cost of halving effective rated ST. However no limit to rated ST actually exists. Some have suggested the limit of 3x Min ST given on B270 for melee weapons also applies to bow and crossbow. This however still results in steel crossbows being largely pointless, e.g. a composite crossbow can be ST 24 and a steel military crossbow can be ST 36, but is treated as ST 18 for damage and range. So the supposedly "advanced" steel bow is twice as heavy, has -2 worse bulk, and requires more ST to load, but still does less damage with less range!
"Referencing a rule that doesn't exist" is a flaw you find in many poorly edited games. It's not generally one found in GURPS but here it is on Low-Tech p, 74. I find this kind of thing especially frustrating as it sends the reader on a wild goose chase, and ultimately leaves a mechanical hole in the system.
The solution with the least change to the current rules would be to create the rule that Low-Tech p. 74 references and actually give a limit on rated ST to crossbows (and probably for bows as well). This limit is almost certainly much less than 3xMin ST. It should be the point in which the statistics change enough that the table entry no longer reasonably describes the same weapon. This should definitely include any whole integer change to bulk, and also significant increase in weight (something like anything more than +25% heavier).
Another solution would be to rewrite the weapons table, treating crossbows more like firearms. Historically, the majority of crossbows were generally made in only a few specific draw weights. This is analogous to firearms; sure it was possible to have a gun made for any diameter shot and any powder load, but the weapon tables only include a few common types. The weapon tables could easily just have had e.g. "Composite Crossbow, 350lbs" and given fixed stats for this weapon and some other typical draw weights and constructions. There's really no need to have variable rated ST at all.
"The Deadly Spring" in Pyramid #3/33: Low-Tech (and the included spreadsheet) allows for the design of bows based on draw weight and construction, but is a very detailed design system and requires a lot of work. It could, however, be used to do either (or both) of the above solutions, which would result in much simpler and more playable results. I may come back to this in the future and see if I can do this.