Wednesday, February 17, 2010

Game Balance Analysis: Battletech 3025 Weapons

I love game balance, and when I study a game, I usually like to look at the game balance and figure out whether the various items in the game are balanced against each other, and which are too powerful or too weak compared to the others. I thought I'd go over my game balance analysis of an actual game.

The game I felt like analyzing was classic Battletech, the version in the 3025 Technical Readout, an old game book I have a great fondness for. I was thinking of Battletech recently, and started thinking of the game balance between the various weapons, so that is what I'm writing about.

I'm not really an expert on Battletech tactics, and it has been quite a while since I've actually played the game. But this is primarily an analysis based on mathematics and logic, supplemented by some play experience, the typical kind of analysis I like to do. And I'm using my old Battletech Rules of Warfare manual for the analysis, not the latest Battletech rules; this is an example of game balance analysis, not a commentary on the current state of the game.

The first thing to note about the weapons is the distinction between energy weapons and ammunition weapons.

1. Energy weapons have unlimited ammunition.
2. Ammunition will explode if it suffers a critical hit; this typically destroys the Mech.
3. Energy weapons are better at clearing woods and starting fires.

In addition to these advantages, energy weapons tend to be smaller but require more heat sinks. Spending tonnage on heat sinks rather than spending the same tonnage on more massive weapons has advantages of its own.

1. If you have a weapon that takes up fewer critical spaces but needs more heat sinks, then more of the mech's critical spaces are filled with heat sinks compared to weapons. This is good, because a hit on a single heat sink is no big deal, much preferrable to a hit on a weapon.
2. Weapons that require more heat sinks are more flexible. If you are in danger of overheating, a mech with heat-heat weapons and lots of heat sinks can just skip a high-heat weapon and cool down a lot. A mech with low-heat weapons and few heat sinks would have to lose a lot of firepower just to cool down a little. This could be important if the mech has taken engine criticals or needs to use jump jets.
3. If you build Mechs like they do in the classic 3025 Technical Readout, with more weapons than heat sinks to handle them, then the flexibility of high-heat weapons becomes really useful. You can choose to fire the high heat weapon, and accept the heat buildup, or not fire it, and cool down.

I should then mention some other important effects I didn't put into the ETR, since I don't feel I can calculate them, but they do factor into my balance considerations. The effectiveness of a weapon is not directly proportional to its damage. Weapons that do a lot of damage to a single area are especially good because they cause damage more unevenly, and can defeat the enemy mech without destroying all armor on all locations. Weapons that cause a smaller amount of damage have the advantage of being more efficient at causing critical hits, since you roll on the critical table for each hit, and the amount of damage doesn't matter. In general, I think the first effect is more important, particularly as the absolute damage gets larger. A 10 damage PPC is definitely more than twice as good as a 5 damage medium laser. And the devastating power of a 20 damage AC/20 is really extraordinarily useful. At damage 5 or less it isn't so clear, and certainly two 1 point weapons would be better than a 2 point weapon. It should also be noted that having two small weapons is slightly better than one weapon with twice the firepower, since one hit can only destroy one weapon, and you can choose to fire only one for heat purposes.

To analyze the weapons more conveniently, I'm going to make up a number that estimates how resource-intensive the weapon is. It will be the tonnage of the weapon, plus 90% of the heat, plus the weight of 15 shots of ammo. This total will then be increased by 5% for an ammo-based weapon (to account for the disadvantages). The numbers don't have to be super-exact, since I'm not interested in perfect balance calculations, only determining whether different weapons seem close enough that you could argue over which one is better.

I made some charts calculating the ETR, damage, and damage efficiency of each weapon. The tables didn't copy easily into the blog, you can see the version of this article with tables here.

Let's start out by examining the long-range weapons. They generally have a range band of 5, or 6 with minimum range 3, or 7 with minimum range 6. It is hard for me to make a statement on whether the longer ranges with higher minimum ranges are better or worse, so I'm just going to treat them as pretty close.

My baseline weapons are the PPC and the laser laser. These weapons are pretty similar to one another.

The AC/10 seems like a pretty similar ammunition-based weapon to these two. By my estimate, it is pretty close, just about as good as the other two. The ETR is very close to that of the PPC, which does the same damage with slightly different range characteristics. I'd prefer the PPC, but still, they are pretty close.

The AC/5 is clearly an inferior weapon compared to the PPC. It is significantly less efficient than the PPC in my calculations, and it does less concentrated damage. Even if you ignore the advantages of the PPC as a concentrated-damage heat-based energy weapon, and use a more favorable way to calculate ETR, you still have the fact that the AC/5 does half the damage of a PPC, yet consumes more than half the tonnage.

The AC/2, on the other hand, is just silly and useless. It requires about 75% of the resources of the AC/5, but does 40% of the damage. It does have the best range characteristics of any weapon, but the extra range is not worth this much! And the AC/5 is already a weak weapon to compare to. It is true that it has the very longest range in the game, so maybe you could use it for some sort of strategic siege. But the weapon is so weak that if you carry the standard amount of ammo for mechs in the 3025 readout, you won't be able to seriously hurt another mech at long range before exhausting your ammunition.

The LRM/5 feels a bit weak to me. While its damage/ETR is slightly higher than average, the damage is very non-concentrated. And although the range is very good, I feel the terrible accuracy at close range is a real hindrance. However, that being said, for balance purposes it is close enough to the PPC and large laser that I would consider it an equivalent weapon, one you could easily argue was as good or better.

The other LRM's are practically just scaled-up versions of the LRM/5. They use slightly more tonnage and fewer heat sinks, and cause slightly more concentrated damage. I consider the LRM/5 the best, but really they are so close it doesn't matter much.

Next, we move on to the short-range weapons, those with a range band of 3. They are easy to compare with each other, but hard to compare with the long-range weapons, because I can't say for certain how much better the longer range actually is.

My general feeling is that the SRM/6 is the weapon closest to being balanced with the PPC and Large Laser. It has slightly more than 50% more damage efficiency, which seems to be in the ballpark of being an equal exchange for the long range.

The SRM/4 and SRM/2 are just inferior versions of the SRM/6. This can be seen pretty clearly by the fact that the heat output is more per missile in the smaller launchers, while everything else scales about evenly. This assumes you aren't using the optional Inferno SRM/2 rules; that weapon is insanely potent for its size.

The medium laser has a much better ETR than the SRM/6, and seems like an extremely efficient weapon. This has always been my intuitive feel as well; the key is that the one ton weight is just amazingly low for a weapon with pretty good damage. I tend to think that the medium laser is more efficient than any of the long-range weapons, despite the shorter range.

The AC/20 is a queer weapon. The ETR is pretty low, rather less than the SRM/6. But the damage from this weapon is so amazingly concentrated, it is absolutely deadly. 20 points to one hit location is just incredibly frightening. Because of this, the power of the weapon is hard to judge. My feeling is that it is a very good weapon, much better than an SRM/6, but it is probably not as efficient as the medium laser.

Finally, the point blank weapons. Again, it is hard to say how these compare with longer-ranged weapons. The Mechs in the 3025 Readout carry so few of these that it is almost irrelevant how efficient they are, they just don't have much effect on the total effectiveness of the Mech. But when you use the Mech creation rules to design your own Mech, it is very tempting to load it with an enormous number of point blank weapons.

The small laser is what I would think of as the "standard" point blank weapon, although with so few point blank weapons, this isn't a very meaningful designation. If you look at my ETR, I estimate it is more than 50% more efficient than the medium laser. It seems plausible that 50% might be worth the extra range of the medium laser, so these weapons seem maybe equally efficient to me. Since the medium laser is unusually good, I would say the small laser is awfully good too.

However, if the small laser is good, then the machine gun is really good, at least according to my ETR. It has the big disadvantage that the tonnage I'm calculating doesn't "come" with a heat sink; most of the tonnage of a small laser is in the heat sink, which can be used for another purpose if you are out of range, but the machine gun just sits there taking up space if you don't get to point blank range. But if you are building a custom mech, the machine gun seems like the best weapon in the game. You can get 10 machine guns and a ton of ammo for only 6 tons!

Now, there are some caveats to this. Saying it is the best weapon in the game doesn't necessarily mean you would actually want an army of nothing but machine gun mechs; there might be some situations they can't handle. It just means that in a typical Mech battle, if you could replace one of the Mechs with a custom design, I'd guess you couldn't do better than to put on a horde of machine guns.

Another point to note is that some of the weapons are much less effective when used in the typical configuration of a mech in the 3025 readout. You can make a pretty evil mech with machine guns using the mech creation rules, but machine guns on the standard mechs are usually pretty awful. They often have 2 machine guns with a whole ton of ammo, which is way too much ammo. And the key to counteracting the poor range of a machine gun is to be totally focussed on getting to point blank range. With so few machine guns, the average 3025 Mech may have little incentive to close the distance, so the machine guns stay idle, and might better be replaced by a different weapon.

The final weapon is the flamer. Since a flamer is identical to a medium laser, except that it does 2 damage instead of 5, it is clearly worthless for any purpose other than setting fires. In my opinion, it is pretty worthless even for that purpose. I would rather have a medium laser to start fires with. Even though it has only 7/12 the chance to start a fire, it can start a fire from three times the range. The flamer is mostly good if you want to set your own position on fire!

A final note is that from a strategic point of view, energy weapons are clearly better than ammunition weapons. The odd thing is that, in general, they do not seem to pay anything for this ability. The average energy weapon is about as good as one of the better ammunition weapons, even without the extra advantages of being an energy weapon. I'm not sure why you would mess around with the logistics for ammunition when energy weapon based Mechs are at least as powerful and never need rearming.


  1. This post brings back memories. I'm surprised at how much of the weapon data I still knew.

    My take was that the 3 weapons that most dominate the game are the medium laser, small laser, and the machine gun.

    The issue with the medium laser is that it does so much damage that the long range weapons have difficulty effectively outranging it.Your chart has the medium laser at about twice the efficiency of a PPC, so a PPC mech exchanging fire with a medium laser mech at any range up through 9 doesn't really have an advantage. It's difficult to imagine a map that allows for a lot of long range sniping. An open map let's the mechs close quickly and maintain negative modifiers to hit, and terrain generally allows for cover.

    It is possible for short range weapons to outrange the point blank weapons, especially with the way the initiative system works. It's also easier for short range mechs to concentrate fire, so I don't think there's the same problem with short range vs point blank weapons.

    Small lasers are important because of the 10 free heat sinks. I looked up the costs, and if I did the math right, a 40 ton mech with a speed of 6 can spend 11.5 tons on weapons. With 10 small lasers, and 11 machine guns, it's more of a small laser mech than a machine gun mech.

    The standard source mechs in the sourcebook added a few machine guns for roleplaying purposes to be good against infantry, which is amusing.

    I don't think the limited ammunition is much of an issue in the strategic sense. After all, mechs are useless if they don't have armor replaced, so they might as well have ammo supplies as well. I always assumed that energy weapon mechs were more high tech and expensive.

    I always wished they would have done something more interesting with the missile weapons. In retrospect, I think it would be interesting to give them flatter range brackets, and maybe the ability to ignore the penalty from the attackers movement. After all, they are tracking missiles.

  2. Good to see that your experience agrees with my analysis. As you say, even if machine guns are best, you are obviously going to need to use up the 10 free heat sinks. And I do recall in my games that long-range sniping did indeed seem very difficult and didn't have much effect on the battle.

  3. Interesting. I played a lot of Battletech when I was younger. I have a few points.
    1: Those 3 range weapons. Consider every mech but one in 3025 can walk backwards 3 - meaning in 50% of turns it can put itself out of range , or at least into the long range bracket. And numbers above 7 (8 is the long range roll) on two dice are bad. So range is worth something.

    Range also allows overheating mechs to duck behind terrain and cool down.

    2. You aren't shooting every round, so high damage energy weapons are worth more. There are some rounds where your shot is impossible, or near impossible and it's not worth shooting. The optimal strategy seems to be to overheat a litle on favourable shots and cool down on less favourable ones. The AC 5 looks good when you can shoot 2/3 of the board every round for no heat, but often, you can't, because of terrain etc.

    3. This is an unobvious point when people dicuss Battletech weapons, but every mech can kick and punch for big damage. Minimum range becomes important, becasue you can stack you your punches and kicks with weapons, if you keep those weapons off the arms and legs, and without minimum range.

    In 3025, I found an excellent configuration was a 4/6 mech with max armour, average movement, and 2 large, and some medium lasers on the torsos.

    Provided you can avoid being jumped behind, the sheer brutality as you close in and then kick and punch your way to victory is enourmous. mechs that are overgunned can't shoot and punch in close, and have to slow down their fire for heat. ALL short range weapons are useless because a mech can keep out of 3 range. However, heat and terrain means you can close on a mech in 3025 with a mix of 15-9 range weapons. You can build a fast mech with range 3 weapons, but you will be chewed up enough closing in, without being able to return fire that it's a loser. Beaer in mind too, that a 3 range weapon at 3 hexes needs an 8 on two dice to hit. Bad.

    Finally. Every ton you spend on a weapon/ammo is a ton you don't spend on armour, and armour is important.Eight points of armour is Worth four rounds of the same tonnage of machinegun hitting you, and armour ALWAYS COUNTS. Weapons can be out of range, heat or in too close, or explode ammo etc, but armour is always valuable. I always found the best thing to buy was maximum armour.

  4. I agree that maximum armor is clearly efficient in Battletech. In a theoretical sense, when a game unit can buy linear amounts of offense and defense, it wants to spend an equal amount on offense and defense. So there is a sweet spot for how much armor you want to buy (this is made trickier to calculate by various practical concerns, but it still exists). The maximum tonnage you are allowed to spend on armor in Battletech is generally rather less than 50% of the tonnage you have available to spend, so the sweet spot is greater than the maximum armor allowed, so it is optimum to get as much armor as you can.

  5. Futher comment on armour: in the earlier editions (the ones I played), The armour is usually penetrated after 2-3 shots from a decent weapon, allowing critical hits that can take out weapons, slow down your mech, etc.

    The only really bad result of a critical hit is killing the pilot, or blowing up the ammo, either of which kills the mech.

    Less armoured locations are less likely to be hit, so roughly speaking, with optimal armour, damage tends to even out over a mech.

    There is (perhaps you could comment mathematically) some advantage in having damage go to one location rather than spreading out, but I don't think it extends to the massive cost inefficiencies of the really big weapons.

    At first, it looks like your aim is to get critical hits - leading to the thinking it *looks* like really heavy weapons that can punch through armour in one hit, and allow for follow up with "critseekers" are an advantage.

    (Critseekers are weapons that spew lots of small damage items over multiple locations, to try and hit an already penetrated location.)

    However, play experience showed me that as the armour and criticals tended to even out over two or three mechs in the course of a game. That is, critical damage was significant, but most mechs tended to die from multiple criticals in multiple locations, rather than two or three big hits in one location.

    What really mattered was being able to consistently hit mechs, turn after turn, in the medium range bracket - accumulating overall damage was more important that the game mechanics of critical hits etc, because the damage and critical hits "evened out".

    The evening out is becasue of the hit location table - Head or Centre Torso Critical hits need a 2 or 12 (1/36), whereas the Centre Torso is hit on a 7 - the most likely number, and the location that allows for the most armour. other locations have a similar probability spread.

    This can be manipulated by stuffing locations with heatsinks, where the loss of a single one is less important, or underarmouring areas with less important items.

    The exception was carrying an ammo-dependant weapon - basically a mech can lose a weapon, sensors, leg actuators for a minor penalty - the cockpit on a default mech is the only instant kill.

    An ammo hit, though, kills the mech. This means that even if your ammo weapon is slighly better (and few are) than its energy counterpart, you're probably giving yourself something like a 1% better chance per ammo block of being instantly killed.

    It's worth noting too that overheating can cause the same ammo explosion-no such penalty for energy weapons.

    In summary, in a game with a "lance" each (4 mechs per player); maximum armour and medium to long range energy weapons (with no minmum range penalty) give the best chance of success.

  6. As far as the mathematics of bigger weapons:

    Assume a mech has a perfect armor distribution (more armor over harder to hit spots and so on). If you attack that mech with a huge number weapons that do 1 point of damage each, the law of averages will tend to take over, and the damage will be evenly spread out over the mech according to relative hit probabilities of different locations. What is important is that, because the damage is spread out so evenly, you will need to destroy essentially every point of armor on the mech before you start to cause internal damage - before you actually affect its capabilities in any way.

    On the other hand, if you are using an AC/20, even your first hit might cause a critical hit that would start to affect the foe. But more importantly, long before you have hit the foe enough times to strip him of his armor, you will, by random chance, have hit the same location twice, destroying the location. And destroying any torso or leg location on a Mech is likely to have a drastic effect on his ability to fight. Sure, it might take you a while longer to actually destroy the Mech - but it is a lot easier to destroy a Mech when it can't fight back too well.

    Bigger weapons have more randomness, and tend to incapacitate or destroy the mech while leaving it with more armor left. And that means it took less damage overall, meaning the weapons are more efficient. Even if, with typical weapons, you end up de-armoring a few locations rather than just one, it is better than having to destroy all of the armor.

    I think there is no question that the AC/20 is substantially better than 4 medium lasers. The real question is whether a hit from a PPC is really substantially better than 2 medium lasers, or whether the difference is not really important. That I couldn't say without running a computer simulation (sounds like fun - but so many fun ideas, so little time).

    Anyway, my recollection of Battletech was that causing critical hits was fun, but the real money was in destroying entire hit locations, that was what really hurt the opponent.

  7. Fair points. an AC 20 is better than 4 medium lasers, but costs (with 10 shots) the same as 16 medium lasers.

    You're right that it's better to kill or cripple a mech without having to eat up all the armour, but it's worth noting that simply taking out all the armour in one location isn't the only way to do this:

    Head hits damage the pilot without having to penetrate.

    Any roll of 12 causes a centre torso critical hit.

    Overheating can cause an internal ammo explosion

    The mech can run out of ammo, effectively disabling a weapon.

    Taking more than 20 points of damage in any form means the mech might fall over and hurt itself.

    Jumping and moving through difficult terrain can make a mech fall over and hurt itself.

    The rear armour is much weaker and smaller weapons can cause criticals there.

    Anyway, as you say the fun is in causing critical hits, and they do hurt. BUT...

    Mechs with energy weapons can mitigate the effects of critical hits much more that those with big, ammo dependant weapons, and those critical hits oftem come from things other than weapons fire, or depleted armour.

    To use our example, an AC 20 might be better than 4 medium lasers, but is a destroyed AC 20 better than 3 medium lasers?

    The trend in reading recent classic battletech forums (and there is simulator software)- seems to favour "Zombie" mechs - mechs that can take the maximum punishment for the longest, with minimum degradation of firepower.

  8. The fact that Mechs can take damage from things other than weapons fire doesn't have much effect on the question of what the best weapon is. But I had forgotten that rolls of 2 and 12 cause "automatic" damage (critical or pilot damage) regardless of armor. This is another factor that weighs in favor of smaller weapons, especially super-small weapons like small lasers and machine guns. And certainly Mechs with numerous small energy weapons, and lots of heat sinks, take criticals much better (though it doesn't help them against total destruction of hit locations).

  9. You need to factor in the pilot skill roll for damage in excess of 20 points per round. That adds a slight weight to higher damage weapons, like the AC20. It's more of a gamble weapon due to it's eggs being within the single shell, but that single round can knock the enemy mech (regardless of size and armor) off it's feet, causing damage in proportion to it's weight. That also significantly hampers the target's offensive and defensive capabilities in the next round, with the potential for another fall.

    This next point is untestable, but I find that luck favors certain weapons. It completely ignores the smaller SRMs, but rewards the AC20 at significant moments.

    I wonder why some weapons (AC2/AC5/Flamers) are so worthless in most recordable respects, on top of the discrepancy between energy and ballistic weapons.

  10. The pilot skill roll is an interesting point to consider. The fact that the AC/20 does exactly the sweet spot is another factor in favor of that weapon. I had forgotten about this since I always played with a house rule that the knockdown damage was 40% of the tonnage of the target mech.

    I think this has more to do with hitting the sweet spot than the mere size of the weapon. I'd rather have 6 5-point attacks than 2 15-point attacks if I were trying to knock the target over. Also, the advantage of larger weapons in terms of knockdown disappears when you have enough firepower that your average damage is 20 points per round or greater.

    I think the main luck factor with the AC/20 is what I was mathematically describing, that you are much more likely to destroy or cripple the enemy mech "early" by a "lucky" hit to a weak point, whereas large numbers of weak weapons are more likely to have to wear down the opponent fair and square (except that really large numbers of really weak weapons can hope for hit location rolls of 2 or 12).

    It is also true that if you get on a lucky streak with an AC/20 you can be an awesome engine of destruction. But conversely, you can get on an unlucky streak and be a big zero.

  11. > Fair points. an AC 20 is better than 4 medium lasers, but costs (with 10 shots) the same as 16 medium lasers.
    I think that is not the correct comparison - you have to consider the heat sink cost. If you only ever got one shot with your mech, then heat would be irrelevant, but you hope for more, normally. Let'S say we want to be able to take at least 10 shots without getting into serious heat troubles - then we might not want to get to more than 14 heat after 10 shots. (a heat level of 14 is the first heat level where your mech may shutdown). A heat level of 14 is a bit hard would mean you wouldn't make more than 1.4 heat per turn. Let's be generous and assume you get a few firing puases in between, and let's say you don't want to gain more than 2 heat per turn.

    For the AC/20, that means you need 5 heat sinks (it produces 7 heat on its own), putting the total weight of the weapon system at 14+2+5 = 21 tons.

    For Medium Lasers, to get to 20 damage, you need 4. 4MLs produce 12 heat, so you need 10 heat sinks. That's about 14 tons for the MLs.
    5 MLs give us 25 damage and 18 tons, and 6 MLs give us 30 damage and 22 tons in MLs and heat sinks.

    So basically, the advantage of the single hit location would need to be worth either 7 tons or about 10 damage. Is it that? I am not convinced. But at least we know the ballpark region, I suppose - and if we were crazy enough to try to make Battletech more balanced, this methodology may be a starting point.

  12. I think this is a really thoughtful analysis, but there are some points missing.
    1. Machine guns and flamers are both hugely useful with the additions from the Citytech rules. A machine does 2d6 rather than 2 damage against an infantry platoon not in a building, so they'll average 7 damage per hit. In open terrain they'll take double damage, which means on average a 28 man platoon will be wiped out by 2 machine gun shots. A small laser on the other hand does 3. Also, while woods ignite fairly easy with lasers, buildings are harder and they're easier to ignite with either a flamer or an inferno round.
    As far as spreading out damage weapons like the SRM really shine when used against vehicles like tanks, hovercraft and VTOLs because it's easier to get a crit or rotor hit on them as compared to 'mechs and the critical effects are usually more devastating.
    2. For hit locations LRM's are allocated in 5 point groups, and SRM's are allocated individually, so the former are not as "spread out" as this analysis seemed to suggest. LRM's also allow for indirect fire when you don't have LOS; no energy weapon can do that trick.
    3. The utility of a weapon like a medium laser depends at least in part on the mobility of the 'mech. Assault 'mechs usually have weapons like PPC's and AC/5's because most of them have a walk/run of 3/5 or 2/3 and it would be trivial to keep your light or medium 'mechs either out of range or at long range if they had only medium lasers or an AC/20. A standard gunnery score is 4, so add 2 for running, add the targets movement modifier (lets say they moved 6, so it's a +4), and then add 4 for long range. That's a target number 14, so you automatically miss.
    4. The fact that the AC/20 takes up 10 critical slots and each ton of ammo only provides 5 shots can't be emphasized enough. It forces the designer to put the weapon in either the right or left torso, which by the way are the special critical hit on a 2 locations when being fired at from the left or right side. I've seen the gauss rifle from later editions of Battletech decide many a fight with its 15 damage and great range, but I don't think I've ever seen an AC/20 with its crummy range and 5 extra damage get that "lucky shot" and turn the tide. It doesn't help that the massive weapon only appears on designs like the Hunchback or Atlas which are some of the slowest 'mechs in 3025. Why would anyone purposefully let you close the distance if they're faster and they know you have an AC/20?

  13. 1. That is very interesting, I did not know they added rules to make MG's good against infantry. And that is a good point that against vehicles, unconcentrated damage is deadly at killing their mobility.

    2. I took that into account. An LRM/5 hits with an average of only 3 damage, which is pretty unconcentrated. An LRM/20 is a bit more concentrated since it gets into 5 points groups, but not quite as concentrated as a medium laser since a typical hit might be 5, 5, and 2, you usually get one small hit along with the 5 points hits.

    3/4. Good point that it is hard to get those Hunchbacks into range against the targets they would really like to hit, the light mechs. On the other hand, a lot of those light mechs have SRM's and medium lasers with exactly the same range characteristics as the AC/20, so if they want to attack the Hunchback and his buddies, they are going to have to come in at some point. It is very complicated and depends on the situation.

    4. It is a good point that the massive spaces of the AC/20 restrict where you can place it, but that seems like a pretty minor limitation. It is certainly nice with medium lasers that you can put 2 in the center torso and 1 in the head, but you've still got to put the rest somewhere, and the side torso is as good a place as any. They may be more vulnerable to lucky crits from side shots, but they are more survivable against regular damage (more internal structure, more max armor, and if side torso is killed arm is useless, but not vice-versa).

  14. Hi! I wouldn't have commented, but for the fact that you still seem to be interested in discussing this analysis. :D It's great overall, but I feel like you haven't fully accounted for density issues.

    To wit: All 'mechs have the same internal volume, but heavier ones are, well, heavier.

    This was emphasized in the later versions of the game with all the "new technology" equipment that was designed to eat up space, but it still bears mentioning in the 3025 context.

    For instance, machine guns are much better on a light 'mech than on an assault, not just because of speed issues, but also because light 'mechs sacrifice less by mounting a low-density weapon like a machine gun; they'll run out of tonnage long before they run out of space. By contrast, if you're building a 100-ton assault 'mech, filling every space with a machine gun and leaving a couple for ammo would result in a grand total of approximately (off the top of my head) 39 machine guns. That's less than 10 tons of armament, whereas you could probably fit 14-15 medium lasers in together with enough heat sinks to negate most of the heat buildup.

    On another note, as for those medium lasers... I have to go with David's point 3 on this, with a caveat. If you're playing on standard size double-maps like the ones that came in the box, then a single Panther (35 tons, 4/6/4 movement, one PPC) is enough to ruin your day; all else being equal, mobility and range outweigh firepower by virtue of the ability to play "keep-away". By contrast, if you're fighting in close quarters with multiple units to herd the enemy into the laser killzone, the medium lasers might be a bit more useful... again with the caveat that your opponent likely has equal-sized forces, and hence the ability to defeat your herding units in detail via force concentration. Finally, you're describing a theoretical custom 'Mech armed with many medium lasers; on that basis alone, I'm afraid I have to disagree with your assumption that your opponent is going to field only "average" light 'Mechs armed with close-range weapons when facing a custom 'Mech of any kind, regardless of whether they know what it is.

    OTOH, if the scenario is such that your opponent doesn't have keep-away as an option - for instance, if your mission is to destroy a fixed target, or if the opponent has to reach a very specific area of the map - then the utility of short-ranged weapons becomes considerably higher.

    In summary, weapon balance depends really heavily on the type of scenario being played, the force sizes on both sides, and the tonnage/mobility of both the firer and target. Even if we assume a deathmatch scenario and thus eliminate that variable, the other two are factors which really can't be avoided. But perhaps they constitute a topic all by themselves...

    1. ...oops, I forgot that the 3025 machine guns were 0.5 tons, not 0.25 tons like in later editions. Still, downwards of 20 tons isn't a lot of armament on an assault-class machine.

    2. I skipped the issue of spaces because for most mechs they aren't really an issue. Space is more of a constraint than a weapon efficiency issue. It is always desirable for a weapon to have a lower weight or to do more damage. But for a weapon to take up fewer spaces is meaningless unless your Mech would otherwise not have enough space for its weapons. A 100 ton pure machinegun/small laser mech isn't so much inefficient as it is illegal. You can build that kind of mech at 40 tons, you can't build that kind of mech at 100 tons. Space is mainly important when building custom mechs. But my article isn't about building custom Mechs, or what would happen if both sides tried to build their own custom Mechs to counter the other side's custom Mechs. I have not really interested in that topic. It is about the efficiency of the weapons on the 3025 Mechs, which weapons are more efficient than others. The 3025 Mechs do not generally use up all the spaces they have, so the space taken by the weapons isn't really important. This changes once you get to 2750/3050 and mech design becomes more complicated.

      As I mentioned, it is difficult to evaluate the benefit of range, it is affected so much by play style and scenario design. Your play style is certainly different from mine if a Panther can play keep away on a standard map. My experience was that Mechs couldn't really hit anything while running around at long range on a map full of obscuring terrain. And it wouldn't matter if the Panther couldn't be attacked back, the enemy force would just concentrate fire on one of his buddies while he was playing hide and seek.

  15. Interesting discussion. The prevailing wisdom on the official Catalyst boards are that medium and small lasers rule the game, at least in 3025. What you seem to be discounting in your analysis is the cost to putting an AC 20 or PPC on a Mech. When comparing apples to apples, a unit that favors more smaller energy weapons will always not only cause more raw damage on an average number of hits, under similar target numbers than the bigger guns, but will also give more opportunities to cause criticals, which is arguably the real point of the game. Not only is the AC20 a huge tonnage cost, but it has medium laser range, and it takes up so many critical spaces that it can be easily nullified with a single critical to the right location. Not a good plan. That same critical only eliminates a single medium laser and the Mech keeps on trucking, The same is true for the PPC. Yes, high damage can be a problem for your opponent, but these large weapons are so easily neutralized that it cannot be considered worth the expenditure of tonnage. One thing that you mentioned above that bothers me is "If you attack that mech with a huge number weapons that do 1 point of damage each, the law of averages will tend to take over, and the damage will be evenly spread out over the mech according to relative hit probabilities of different locations. What is important is that, because the damage is spread out so evenly, you will need to destroy essentially every point of armor on the mech before you start to cause internal damage - before you actually affect its capabilities in any way." On 2D6, you do not expect those hits to be spread out evenly. You expect the center torso to have the most damage at a roll of 7, followed by the right or left torso at 6 and 8, etc. I think that you accept some points on a standard and reject other points based on the same standard.

  16. Sorry for the double post, but I found another interesting read related to this discussion:

  17. I don't think you understand what I mean by "the damage will be evenly spread out according to the relative hit probabilities of different locations". This doesn't mean the damage will be spread out equally among all locations. This means that if there is a 7/36 chance of hitting the central torso, then 7/36 of the damage will end up there. If there is a 5/36 chance of hitting the right arm, then 5/36 of the damage will end up there. This means, of course, that more of the damage will end up on the center torso. That is why Mechs have more armor on the center torso. If the armor on the Mech is spread out according to the same probability distribution as the incoming damage (more armor on the locations that are hit more, less armor on the locations that are hit less), then if the damage comes in 1-point chunks, all armor locations will tend to hit 0 at the same time.

    1. Yep, I misunderstood. Probability pitfalls are so common. I should have given you the benefit of the doubt. Have your thoughts on this topic changed over the last few years?

    2. I don't think anything has changed my general opinion - medium laser, small laser, and machinegun are awfully strong. I seem to have received considerable discussion on the merits of the AC/20. But while I have defended the benefits of its high damage, I still don't think it is as good as medium lasers, I just think it compares favorably to, say, an AC/10.

      I don't really get any more experience on this question now because I play my own game based on Battletech in which I've changed the weapon and Mech designs to be much better balanced compared to each other.

  18. I think the comparison of AC/5 to PPC is a little unfair because if you account for all the heat sinks needed to fire PPC continuously, you probably end up at nearly identical tonnage load.

    1. Well I don't see any buton for edits, so here goes:

      It is true that you can fire just one in a group of energy weapons to cool off, then you're talking about averaging your damage over time, and if you have to keep firing one weapon to cool off, that average keeps getting worse. That means lost efficiency of the overloaded energy platform design relative to the small number of heat sinks.

      I remember in Mechcommander that both ammo and heat sinks were factored into the gun weight invisibly. A medium laser was actually 3 tons there, because they took the liberty of considering it a 1 ton weapon with 2 tons of heat sink.

    2. I'm not sure I'm following what your comment here is exactly. I factored the weight of the heat sinks into my analysis as explained above (each point of heat is equivalent to 0.9 tons of weapon weight).