I speculate that GURPS (and maybe The Fantasy Trip; I wasn’t around for that one) started with a fairly generic view of damage that incorporated penetration, blunt trauma, injury, and a certain amount of gamist fun. GURPS High Tech introduced the dependence on the square root of kinetic energy as a “damage” scale by giving the general rule of 20d = DR 70 = penetration of one inch of armor steel. I’m not sure from whence that one came, but I suspect David Pulver provided it to Mike Hurst

All of a sudden, you had two different progressions for damage, and some projectiles (bows) were on the ST-based scale, while the more-easily-quantified set (guns) were on another.

Sometime around the year 2000, I started fiddling with Excel’s solver. I thought that if I could find enough penetration data, I should be able to take a priori data about projectiles (specifically bullets) and turn that into GURPS stats. The result got me a mention on TV Tropes (look for Arbitrary Gun Power), and was published in 2002. But it worked fairly well, and got me noticed by a few other gamers

Anyway, THAT little adventure got me to notice that while firearms matched penetration of armor with known and quantifiable data – in this case the mass, velocity, and caliber of the bullet (from which you derive kinetic energy and momentum – that and cross-section are all you need to derive basic damage/penetration and a wound channel size modifier), weapons on the melee scale (things that you look up as “thr+2” or “swing+3” on the Damage Table on p. B16 of GURPS Basic Set, Fourth Edition) got pretty high penetration numbers pretty fast. Maybe too fast.

I once again scored a mention in TV Tropes by nearly breaking my brain (and Steven Marsh‘s sanity) figuring out how to do for bows and arrows what I did for guns. It was phenomenally more difficult. 

OK, so I’ve covered arrows and bullets. What about sling stones? Swords? Maces? Polearms? Pointed sticks? 

Um . . .

My previous post was an attempt to show how you could at least get the scaling of melee weapons right, where every doubling of ST also doubles penetration.

What you say?! GURPS ST measures applied force, and it goes as your ST score squared (Basic Lift is measured in pounds of force). Energy is force times distance, and if distance is more or less related to how far you can move your body (push-pull-twist-turn!), then doubling force should double deliverable energy, quadruple force is quadruple energy, etc. Then when you put this on the same sqrt(KE) progression as guns, you can arrive at a nice progression where ST 10 is about 1d, ST 20 is 2d, ST 40 is 4d, etc.

Hey, that’s damn playable. Even for me.

Well, that’s penetration, great. But injury? GURPS as-is is probably more right than wrong here when it comes to squishing meatbags. People are notoriously fragile (and strangely and unpredictably robust as well). 

So then you get into wounds, penetrating trauma, blunt trauma, and other things.

Fortunately, GURPS already has the tools required to handle this.

It has a basic damage number that should be used to represent penetration.

It has damage type and damage size modifiers that I’d personally apply a bit differently (maybe that’ll be Part 3), but basically are a “wound channel severity” modifier. 

It has armor divisors to represent something that penetrates better (or worse) than the raw energy might otherwise indicate.

With those things in mind, the only thing that’s really missing is a better way to represent blunt trauma and other non-penetrating injury.

Were I to start from scratch, I’d try and figure out:

Penetration rating, probably mainly based on energy. This would be GURPS’ “basic damage,” but possibly on a new scale. I like the “ST equivalent” number from the last post, since many weapons just act as lever arms, multiplying force. Multipliers are also nice because they allow the quite-simple “add +1 per die” type math, which is even easy in play.

Blunt Trauma rating, probably based on momentum and impact force. I think this is mostly what is done for hand weapons now, implicitly.

Armor Penetration modifier based on shape, hardness, or other factors. This is the source of my assertion that impaling weapons (a damage type I loathe) should probably get a (2) armor divisor, due to all their force being concentrated on a tiny, hard point.

Wound/Injury modifier based on the size and depth of damage. If I were really good, you’d incorporate some sort of “you have to achieve so much penetration before you can get to a Vitals type location.” That would help scale certain effects where you really NEED to go deep (Giganotosaurus hunting! Or a Frumious Bandersnatch)

I’d also have blunt trauma (including punches) be resisted by a HT roll, possibly influenced by skill. During the writing of The Last Gasp, one of my helpers ran a boxing match, and we decided that a HT roll to avoid actual HP of damage (but preserving the AP loss damage causes) from punches would make a LOT of sense, since even with boxing gloves, a typical 15-round boxing match as played in GURPS would kill both fighters. Extending this to all blunt trauma wouldn’t be too much of a leap, and if there were a penalty to HT rolls based on the rolled damage or something, that would keep maces and staffs scary. Alternately, padded hands might do the trick for bonuses to HT. People CAN get killed in fights, after all. They just usually don’t.

None of this is to say “GURPS is broken, wah!” It’s the best playable simulation out there that you can nod your head and say “yep, I get that” for most things. But when the tool of choice for penetrating DR 8 plate armor is “give me a sword” rather than “give me a spiked warhammer” (this one, not that one) or “grab me my gun,” then some head scratching might be in order!

The key in all this is to keep it fun and playable. All this crap needs to be subsumed into a table and calculated but once, during character generation or on obtaining the item the first time, and then never again. Even as ugly as The Deadly Spring was, the math is all in a spreadsheet (generously provided by Steven along with the issue of the magazine!) done up front.

Over at the GURPS Forums, there’s a really interesting (well, interesting to me) thread asking the question if basic melee weapon damage is too high (what you look up on p. 16 of the GURPS Basic Set).

I think the answer is “yes.” In fact, as part of my “Armor Revisited” article I had originally written about a page explaining why, how, and how I’d fix it. But I cut it out since it opened up a few cans of worms, potentially, that didn’t need opening.

Still, it was Just Sitting There, so I posted it on the forums, and now I post it here (this will make my musings in Part 2 of this topic a bit more cogent, I hope).

I note that JCurwen3 posted that he used it in his games, and finds that whatever cans were opened, they might have been more Whup-Ass than worms.


Rescaling Melee Weapons

While firearms are explicitly on an energy-based scale that doubles penetration for every quadrupling of energy, melee weapons do not scale this way, and the raw damage rises much faster, making high ST unrealistically effective when applied to hand weapons and muscle powered ranged weapons.

For an alternative take that scales the same way as firearms, consider rescaling thrust and swing damage. A cinematic scale would have thr equal to ST/10, and swing equal to ST/5; a more realistic one might have thrust as ST/20, and swing as ST/10.

Oddly Small Damage Increment Table

Roll one off-color die, and adjust the number rolled to the table value. The Notes entry shows where rules of thumb can be used to more quickly interpret the roll of the odd die. Each entry is scaled to give an average result consistent with a fractional d6.

Roll on 1d6
Dice Average Damage 1 2 3 4 5 6 Notes
0.1 0.35 0 0 0 0 1 1 1d/5, drop fractions
0.2 0.7 0 0 1 1 1 1 1d/5, round normally
0.3 1.05 0 1 1 1 1 2 1d/4, round normally
0.4 1.4 0 1 1 2 2 3 1d/2, drop fractions
0.5 1.75 1 1 1 2 2 3
0.6 2.1 1 1 2 2 3 3 1d/2, round normally
0.7 2.45 0 1 2 3 4 5 1d-1
0.8 2.8 1 2 2 3 4 5
0.9 3.15 1 2 2 4 4 6
1 3.5 1 2 3 4 5 6 1d6

Implications for Play

Using the “realisitc” scale with ST/10 for swing damage has many repercussions, which must at least be considered before adopting these rules.

Devalued ST

ST is “the attribute that you use to buy damage” in GURPS, with a healthy side-order of “lifting heavy stuff” and “more HP.” The realistic scale requires ST 20 to do 1d thr and 2d sw, and probably calls for a price break. HP are 2 points each; Lifting ST is 3 points per +1 ST, and Striking ST 5 per +1.

HP and lifting are as valuable as ever; reducing Striking ST to 2 per +1 ST in realistic campaigns, with overall ST at 7 points per +1 ST, seems appropriate.

Hand Weapon Damage

Weapon damage is lower, and thrust weapons in particular will have limited penetration capability. This takes what was the historically preferred method of punching through heavy armor and nerfs it further. To compensate, add a (2) armor divisor to weapons with a sharp, narrow penetrating surface. Consider optionally allowing a still more impressive construction, buying a fine (hardened) penetrating point that gives a (3) for ¥4 cost. This would not be available for arrows (see The Deadly Spring, Pyramid #33, for suggestions concerning bows and arrows).


With most hand weapons using the realistic scale, personal armor just got a whole lot more valuable at low DR, especially using the revised armor weights from GURPS Low Tech. This isn’t a bad thing; in fact, it might make a campaign world more closely resemble what we know about our historical real one . . . on didn’t need to lather on 80 lbs. of mail, leather, and plate to protect against basic cuts from swords, or long-range shots from flight or hunting arrows.

Character Differentiation

With thrusting damage at ST/20, and swing at ST/10, there will be very little differentiation between average, fairly strong, and even very strong characters. Weapons rated at sw+2 or sw+3 (maces, axes, and falchions, for example) have their damage almost all defined by the weapon, not the user.

Consider treating the “adds” as an indicator of the power of the lever arm being applied. Treat each +1 as adding +1/3 to a ST multiplier – a sw+2 weapon instead multiplies ST by 1.66. A ST 14 warrior swinging a falchion (sw+2) will do damage as ST 23.2, or 2.3 dice.

One would thrust using the same calculations: a short spear thrust by the same warrior for thr+1 imp would multiply ST by 1.33, for ST 18.6; damage would be ST/20 (2), or 0.9d (2) imp. Note that same spear using the Damage Table (p. B16) would do 1d+1 imp, an average of 4.5 points of penetration. Even the realistic scale here will do 6.3 points, thanks to the armor divisor.

Coup de Grace

The changes to melee damage on the realistic end would bring GURPS damage on the hand-to-hand scale more in line with firearms damage. For a character to do as much damage with a punch as with a 9mm pistol would now require about ST 52 . . . well into supers territory. Even a strong character with ST 20 (1d thr) will be out-penetrated by a .22 LR (1d+1 pi-). For those who want a more realistic scaling for penetration, this may be more satisfying.