There’s 2 kinds of decisions, interesting decisions, and uninteresting decisions (no-brainers). Both have value. Interesting decisions represent choice. No-brainers represent progression. The problem with fusion blaster arms is they’re found everywhere (I think this is a bug/typo, why would a guy with a fusion blaster for an arm get killed by a couple wolves?), and ammo is too easy to get once you’ve raided a lab or two and have 100+ energy and every means to recharge it. All CBMs are no-brainers (except, ironically, the fusion blaster which precludes using 2h weapons, but apparently you won’t need them); they represent progression, and the risk/choice involved is limited to acquiring them. This is fine, as long as they stay in the later part of the game where they belong and not in every wolf and ant cave. It might be sensible to change installation success chance to depend on how many bionics you’ve already installed making it a bit less of a no-brainer, so you don’t always end up as an invisible teleporting cyber-tank complete with fusion cannon in the end-game. Even positive mutations have the drawback of somewhat locking you into a mutation tree. A better solution would be to have bionics fit into a limited number of slots (eyes, head, skin, arms, etc.) and installing multiple bionics into the same slot would have an exponentially increasing difficulty. Also, the fusion blaster specifically should have some drawbacks other than “boohoo I can’t use any inferior weapons.” Having only one hand should at a minimum double all crafting and construction times, and perhaps skill checks required as well.
Now as for firearms, it’s mostly been covered already, but I’d like to see a system where you could dismantle a gun, get parts like “full-auto receiver, direct-impingement” (high burst value, low durability but field-serviceable, i.e. no need to fully disassemble to repair, lowering skill check and time required) or “5.56mm barrel, 18in” (increases accuracy, range, and damage, but increases mass and size, and thus firing time) or some such, and mix and match to rebuild a custom weapon. If a gun wasn’t “field-serviceable,” you might have to either find a workshop, or lug around specialized tools to repair it. The current guns would just be mass-produced weapons according to a template, the same way the default vehicles are implemented. I don’t know a whole lot about guns, but I’d still want that level of customization. I don’t know anything about cars, but I still appreciate the level of customization in the vehicle system, even if it does seem simplistic and overly abstracted sometimes. Ignorance isn’t a barrier as long as the part descriptions clearly state the benefits and drawbacks. They also don’t necessarily have to reflect real guns; they can be abstracted the way vehicles are by lumping parts together. There could also be valid no-brainer decisions like “7.62x51mm super-alloy barrel, lightweight and high durability” as long as the rarity is commensurate with the benefit.
Actually, I’d like to see materials basically become egos for all items (super-alloy katana, surgical-steel combat knife, iron crowbar, etc.) instead of being fixed. This is already partially implemented with armor, all we need is the rest of the items, and a system of material tracking and randomization. Items could also have egos indicating the general level of quality, e.g. “military-grade super-alloy double-barreled full-auto shotgun” or “hobo-grade plastic .22 smoothbore single-shot zipgun.” Crafting skill as well as individual part quality could affect this for crafted items. Certain parts could be crafted from materials, e.g. crafting barrels from sheet metal or even super-alloy sheets from disassembled CBMs, while parts like the receiver would presumably be too complex to craft by hand, analogous to vehicle engines which aren’t craftable either.
There’s a few changes the firearm system probably needs to balance this. Currently firing time depends only on the weapon skill used and your skill level. This should be changed to depend on weapon mass. High mass should increase firing time (at least for the first shot, perhaps mitigated by strength) while decreasing recoil. Large physical dimensions (long barrel, full-sized stock, etc.) should also increase initial firing time due to being unwieldy. The durability system also needs to be fleshed out, since currently it only affects misfire probability and there is no way to affect it (like barrel fouling or cleaning).
I’d like to see the “first shot penalty” concept rolled into a stance system. Non-combat stance would have a penalty to the first shot (drawing/unshouldering your weapon, etc.), and would automatically put you in combat stance after that. Combat stance would have a penalty to move speed dependent on weapon mass and size since you would be walking around with your gun pointed (or your sword raised, etc.), limiting your ability to run. It would also cause a move/turn penalty upon trying to perform any tasks unrelated to combat (i.e. requiring your hands) due to switching stances. There would be a configurable number of turns before automatically switching back to non-combat stance. All penalties would be related to the size and mass of the weapon and perhaps mitigated by strength, skill for smaller weapons and maybe new traits (quickdraw, battoujutsu, etc.). Automatic switching would ensure that newbies aren’t unable to play the game because they’re stuck in a stance, while manual switching could be used by the experienced. The stance concept could also affect NPC interaction. Nobody would trust you if you were in combat stance, but nobody would get the first shot against you either. It could also give you an idea of the general hostility level of the NPC. While what I described isn’t much different than just requiring you to unequip your weapon before doing anything unrelated to combat, it would streamline the process in a way that takes the hassle out of it.
Headshots should probably be looked at too. The whole “the gun doesn’t matter because you just press f and it dies” argument is probably predicated on 8x headshot damage. While in real life, a headshot would likely be instantly fatal or at least incapacitating, that doesn’t necessarily make it a good game mechanic. Either the probability (except against unaware targets) of headshot or the damage should be decreased, except perhaps against zombies since they presumably don’t understand enough to take cover. This could also be reworked into a “Fallout” style body-part targeting system, making headshots the payoff in a risk/reward scheme, rather than just a given with enough skill. In game mechanics terms, firearms skill affects critical chance (critical multiplier too, if you consider most games have 2x crit damage), accuracy, and firing rate. This is way too powerful for a single skill. That’s basically every component of aggregate dps except raw non-crit damage per shot. It even affects the aoe radius of burst fire (which has its own problems, chiefly being implemented as a collection of individual shots rather than “spray and pray”), meaning you could potentially do 8x20x[ammo damage] or more in ~1/3 of a turn (you could still do that if it didn’t affect aoe radius, but nothing has that much hp).
Sorry if I’m derailing the thread, I tried to address the topic because this is something I would really like to see, but as others have said, there are also a lot of external but related issues that need to be addressed to make a meaningful implementation of this.