Reworking weapon/armor/shield stats

[Vezzra]

As mentioned here, I've considered to come up with new numbers for weapon, armor and shield stats. Meanwhile I've gone ahead and given it a try. It turned out that if I don't want to do things halfway, other things need to be touched/changed too.

First of all, IMO we need to do some kind of "swap" in the tech tree: instead of having the lowest-level shield part pre-unlocked and armor further down the tech tree, we should do it just the other way round: pre-unlock the lowest-level armor part, and move shields further down the tech tree. After all (especially with the latest changes to shield mechanics), shield technology is way more powerful and sophisticated than armor.

So, let me give you the new numbers. The last column shows the ratio for part strength vs. build cost for easier comparison. As you'll see, there are some new parts:

new weapon stats

Weapon_Stats.png (81.24 KiB) Viewed 3376 times

new armor stats

Armor_Stats.png (46.11 KiB) Viewed 3376 times

new shield stats

Shield_Stats.png (29.99 KiB) Viewed 3376 times

Ok, now let me comment on what is my rationale for these numbers and changes:

All three categories:

I've defined a kind of "standard line" of parts in all three categories. These can be researched and build without additional reqirements like asteroid belts and neutron stars, and most of their stats progress throughout the entire line by clear dependencies calculated by simple formulas. Apart from them are the special variants (asteroid/rock and neutron) which require "special resources" (asteroid belts or neutron stars) and are calculated a bit differently.

Weapons:

As you can see, I've preserved what we had - almost. I did remove Death Ray 5 to match the 4 types of weapons exactly (it's just been easier with the numbers that way). And I added "neutron" weapon parts to flesh out the neutron branch of the tech tree. The numbers are calculated by the following formulas:

Damage: "Standard line": The highest refinement of a weapon type has twice the damage value than the basic level. The damage values of the next higher weapon type are three times that of the preceding weapon type, so MD1 has a damage value of 3, MD2 4, Laser1 9, Laser2 12, PC1 27, PC2 36 and so on. I've set the starting value, the damage level for MD1, to 3, because this is the lowest number that avoids fractional numbers with this formula. "Special variants": The NPB is calculated a bit different: it is even more powerful than the DR, but not much, so it doesn't continue the line as the other weapon types do.

Build Cost: "Standard line": Starting with 10 for the MD, each weapon type costs twice as much as it's predecessor. By multiplying the damage value with 3 and the build costs with 2 we get progressively better cost/damage value ratios. "Special variants": The build cost for the NPB is set to a value that makes it more cost effective than the PC, but less cost effective than the DR. So these two weapon types (NBP/DR) are matched: the one gives you higher absolute damage stats, which might give you the deciding edge over your opponents in combat, the other one is not quite as powerful, while being more cost effective.

These stats should ensure that you get a sufficiently significant gain when moving to the next weapon type, but also provide a gain that's worth the effort when researching refinements, and achieve all that by keeping the same ratios through all weapon types and their refinements.

Armor:

Here you'll note some changes. I've renamed "Lead Armor Plating" to "Standard Armor Plating" - I like it more that way, but that's just a matter of personal taste. I've added another armor part "Xentronium Armor" so I have one matching armor part for each weapon type in the "standard line". Number calculation:

Strength: "Standard line": Equal to the damage value of the highest refinement level of the corresponding weapon tech "tier", so it can withstand exactly one hit from this weapon. SAP is destroyed by one hit from MD4, ZAP by one hit from Laser4, and so on. "Special variants": RAP: Between SAP and ZAP. CAP: Between DAP and XAP. NAP: Same as standard line (NAP is destroyed by one hit from NBP4).

Build Cost: "Standard line": One fifth of the costs of the corresponding weapon parts. "Special variants": RAP/CAP: Same as SAP/DAP, making them better alternatives to their respective counterparts. NAP: Same as standard line (one fifth of NBP), so the same applies what I've said regarding the NBP: A bit more powerful than the XAP, but less cost effective.

The strength for the PAC have been adjusted in accordance to these changes (roughly 8 times the strength of NAP, as it has been before).

Bottom line: Armor is cheap.

Shields:

Here we have the most prominent changes. First of all, there are two new shield parts: one for the "standard line" (Plasma Shield) and, like with the weapon parts, one for the "neutron" tech branch (Neutron Cloud Shield). You'll notice that the "standard line" only has three parts, as opposed to weapons and armor, where the standard line has 4 corresponding types/parts. Here the tech tree placement swap comes into play: shields are significantly more powerful and advanced than armor, so their "tier one" level does not correspond to the "tier one" level of weapons and armor, but is matched against their "tier two" level, and so on. That means DG corresponds to Laser and ZAP, DS to PC and DAP, PS to DR and XAP.

Number calculation:

Strength: "Standard line": Matches the strength of the armor part one level below that of their corresponding armor part, or the damage value of the highest refinement of the weapon type one level below their corresponding weapon type. That way, only weapon parts of the corresponding level or higher can penetrate a shield. This leads to a value that absorbs two third of the damage dealt by the level one part of the corresponding weapon type. "Special variants": Same ratio to the damage value of the corresponding weapon type as in the standard line, giving strength 60 for NCS (two third of NPB's damage value of 90).

Build Cost: "Standard line": Three times the cost of the corresponding weapon type, making shields extremely expensive. "Special variants": Same, so same applies what has already been said regarding NPB and NAP.

In addition to these changes I also suggest the following: making shields internal only. They are special, powerful, and you should have to give up an internal slot if you want one. And, of course, make them non-stacking, otherwise these numbers would be insane.

Bottom line: Shields are both extremly powerful and expensive.

The MSS is even slightly better than the NCS, but has the same build costs as the PS, making it the most powerful and most cost-effective shield (it is expected to be very special after all ).

Ok, that have been my new numbers so far. As you can see, we wind up with considerably higher stats than before, but I think we can't avoid that if we want reasonable progress of values and also keep reasonable ratios between the stats of the various parts.

This of course makes an already existing problem even worse, and that's the ratio between weapon/armor stats and hull structure values. The latter are far to low. Armor should enhance hull structure, not multiply it. My suggestion is simple: just increase the hull structure values by a factor of 10. This in turn requires that also the values for "Reinforced Hull" and "System Defense Mines" are adjusted accordingly.

I'll try to come up with new numbers for that.

What else needs to be done: The planetary shield and defense techs need reworking/rebalancing too. The new parts and corresponding techs must be added, the tech tree adjusted to all these changes. We also need new icons (for NPB part and tech, XAP part and tech, PS and NCS part and tech - did I miss anything?).

I'll try to do that also, except the new icon part (because my artistic skills are close to what most people would describe as "abysmal"...). For that I'd ask our graphics team

Alright, I think that's it. Now I ask for your comments, objections, suggestions, etc.

[MiniMe]

I cant really comment on all the values, i'm not that much of a numbers guy.

Just a generell question which i have been asking myself (since i first played FO):
What is the sense in having alle these different weapons, all going from level 1-4(5)?

I have never researched Mass Drivers, i always go straight for Laser 1 -> Plasma 1 -> Death Ray 1-5.

Am i missing some mechanic i dont know about, ie. Plasma is better vs. certain ship hulls or shield types?
Or why would anyone want to research down an inferior tech path?

[Dilvish]

I have never researched Mass Drivers, i always go straight for Laser 1 -> Plasma 1 -> Death Ray 1-5.
Am i missing some mechanic i dont know about, ie. Plasma is better vs. certain ship hulls or shield types?
Or why would anyone want to research down an inferior tech path?

I think it's just a time and cost assessment. For example, if you are sure you're not going to build any more armed ships until you have researched laser 1, then maybe you want to skip over Mass Driver 2, etc. But otherwise, Mass Driver 2 can be had for just 2 RP and doubles your firepower. Plasma 4 will give you firepower almost as strong (**edit, at least in the current regime**) as Death Ray I, but at a lower part cost, perhaps it might be more effective for certain ship designs.

[Dilvish]

As for the main post, I'll have to noodle over the numbers some more, but it all seems pretty well thought out. One initial reaction is that multiplying the hull values by 10 might be a bit too much at least at the lower end of hulls -- a Mark I shouldn't really need 17 shots to take down another Mark I it seems to me.

Also, with the expanded Neutron line, that reminds me -- I kinda think there ought to be an unlockable "Artificial Neutron Star", based on a red star just like Artificial Black Hole, and perhaps unlocked by the same tech.

[eleazar]

Initial thoughts:

First of all, IMO we need to do some kind of "swap" in the tech tree: instead of having the lowest-level shield part pre-unlocked and armor further down the tech tree, we should do it just the other way round: pre-unlock the lowest-level armor part, and move shields further down the tech tree.

Yeah, i've thought about doing that anyway for flavor reason. Armor is a tech we've had for centuries-- shields are still in the future (maybe).

And I added "neutron" weapon parts to flesh out the neutron branch of the tech tree.

It strikes me as odd that you get a weapon, shield and armor from neutron stars. That seems to me a little monotone, and give a lot of significance to the single luck factor of getting a neutron star in range. I'd rather see special stuff based on different special resources.

Weapons:
.....

So these two weapon types (NBP/DR) are matched: the one gives you higher absolute damage stats, which might give you the deciding edge over your opponents in combat, the other one is not quite as powerful, while being more cost effective.

Haven't tested it yet, obviously, but i'm suspicious weather two weapons with stats so similar will offer an actually interesting choice.

Armor:

Gah! Too many abbreviations.

It's a fluff issue, but the armor names don't have a clear progression, i.e. if you gave someone all these names and asked them to rank them in order of most to least powerful, i don't think many people would be close.

Ok, that have been my new numbers so far. As you can see, we wind up with considerably higher stats than before, but I think we can't avoid that if we want reasonable progress of values and also keep reasonable ratios between the stats of the various parts.

This of course makes an already existing problem even worse, and that's the ratio between weapon/armor stats and hull structure values. The latter are far to low. Armor should enhance hull structure, not multiply it. My suggestion is simple: just increase the hull structure values by a factor of 10.

I don't think it is especially important that X is always exactly 2x Y, and Z is always exactly 3x X. You've put formulaic elegance ahead of a more readily perceived elegance-- familiar and memorable numbers, like 5, 10, 25...

Also i'm not thrilled that the damage value of a single weapon is so much higher than previously (i'd say the old way already was too high). We progress very quickly toward large number "eyes glaze over" territory with a single ship, let alone a fleet. And though you've chosen a formula that locks you into that, i don't believe it is really necessary. Doubling something is a really big increase, if you continue doing it, as the parable of the chessboard illustrates -- especially when the doubled property is multiplied by multiple slots and multiple ships.

I'll try to do that also, except the new icon part (because my artistic skills are close to what most people would describe as "abysmal"...). For that I'd ask our graphics team

Nothing motivates an artist like needing to replace abysmal programmer art.

[Vezzra]

One initial reaction is that multiplying the hull values by 10 might be a bit too much at least at the lower end of hulls -- a Mark I shouldn't really need 17 shots to take down another Mark I it seems to me.

That's reasonable. I had looked at the hull structure of the weakest hulls and come up with a quick estimation based on that, so it's not unexpected that I somewhat missed the mark. Maybe a factor of 5 would be sufficent?

[eleazar]

One initial reaction is that multiplying the hull values by 10 might be a bit too much at least at the lower end of hulls -- a Mark I shouldn't really need 17 shots to take down another Mark I it seems to me.

That's reasonable. I had looked at the hull structure of the weakest hulls and come up with a quick estimation based on that, so it's not unexpected that I somewhat missed the mark. Maybe a factor of 5 would be sufficent?

I don't think it would be a bad thing if early game battles frequently lasted for a few turns-- at a time when the player is more likely to want to closely manage his few ships, and withdraw from battles not going well.

[Vezzra]

It strikes me as odd that you get a weapon, shield and armor from neutron stars. That seems to me a little monotone, and give a lot of significance to the single luck factor of getting a neutron star in range. I'd rather see special stuff based on different special resources.

Good point. The reason I came up with that was that I always felt getting an (albeit powerful) armor ship part as the only result out of a difficult to get high end tech like neutronium extraction is a bit meager.

I think I'd drop at least the shield part. Maybe even the neutron weapon, and introduce a neutronium hull line instead? Kind of makes more sense for a material like "neutronium", the primary quality of which apparently is supposed to be extreme durability. Neutronium hulls could be the most durable.

Gah! Too many abbreviations.

Sorry... It was getting late, and I grew wary of typing the whole names "XYZ Armor Plating" over and over. Bear with me

It's a fluff issue, but the armor names don't have a clear progression, i.e. if you gave someone all these names and asked them to rank them in order of most to least powerful, i don't think many people would be close.

Yep, you're right. However, that problem already existed, and I didn't have the leisure to tackle this one too. We can always get to that later, first I want to get the numbers down.

I don't think it is especially important that X is always exactly 2x Y, and Z is always exactly 3x X. You've put formulaic elegance ahead of a more readily perceived elegance-- familiar and memorable numbers, like 5, 10, 25...

You're right on that one too. Although it's not so much because I'm that bend on absolutely no deviation from formulaic elegance - it has been just easier to put some formulas in a spreadsheet and see what numbers I get when I change the base value (damage value of MD1). As these numbers are only a first draft anyway, I didn't bother to try to make them "nicer".

Because that's not as easy as it looks. Obviously you can't do much about the numbers for MD and Laser, but even the high numbers of the DR are more difficult to make "nice" than you might think at first: If I change DR1 from 81 to 80, DR2 from 108 to 110, DR3 from 135 to 130 and DR4 from 162 to 160, I get the following progression in damage value: 80, 110, 130, 160. Although the changes have been small ones, instead of an even progression of 27 for each refinement, now I have an increase of 30 from DR1 to DR2, only 20 from DR2 to DR3, and 30 again from DR3 to DR4. The difference between 20 and 30 isn't insignificant anymore. Of course you can leave DR3 at 135, that would work, but: 80, 110, 135(?), 160 - ugh. In this case I'd prefer to do steps of 20: 80, 100, 120, 140. This would also lower the max value for the highest refinement level.

Not really that much trouble to get done, but as long as the basic numbers might get changed anyway, still wasted work

Also i'm not thrilled that the damage value of a single weapon is so much higher than previously (i'd say the old way already was too high). We progress very quickly toward large number "eyes glaze over" territory with a single ship, let alone a fleet. And though you've chosen a formula that locks you into that, i don't believe it is really necessary. Doubling something is a really big increase, if you continue doing it, as the parable of the chessboard illustrates

Well, the "high" numbers have been bothering me too, although I'm not sure if for the same reason. So, to clarify: What exactly bothers you? That the absolute numbers are too high? Because that can be solved rather easily: Just scale them down e.g. by a factor of 10. However, you'll end up with non-integer values, and I don't know if that's what you want.

Or do the numbers scale up too much for your taste? Going with my numbers, the most powerful weapon would have ~50 times the firepower of the weakest one (MD1: 3, DR4: 162). That's of course a lot, and originally I'd also wanted a far less steep curve. However, I found myself faced with a "pick x out of y" scenario. We have 4 types of weapons with 4 levels of refinement each. I think we also agree that there should be a sufficiently significant gain with each advancement that it's actually worth considering for the player.

And although the progression of stats doesn't need to follow my formulas exactly, it should do so at least roughly, especially when considering the proportional gains. As the numbers are currently in the game, you can enhance your firepower by a factor of 4 and more for comparatively few RPs at the beginning of the game (MD4 has 4 times the damage value of MD1!). At the end of the weapon tech branch, you have to invest thousands of RP, but won't even double your firepower (DR1: 31, DR5: 50). That's seriously imbalanced IMO.

Finally, we don't want the numbers not to scale up too much from MD1 (low end) to DR4 (high end).

That's not going to work. We can't insist on all these requirements. One solution would be to cut down on the numbers of weapon types and refinements. If we cut one wepon type and one refinement level, we end up with 3 weapon types with 3 refinement levels each, 9 levels in total. Keeping my formulas, we can start with a damage value of 2 for MD1, and we'll end up with these numbers:

• MD: 2, 3, 4
• Laser: 6, 9, 12
• PC: 18, 27, 36

We can moderatly increase the research costs for the weapon techs so it's not too easy to get to the ultimate high end weapon, but only so much unless they become too epxensive to be worth the investment.

Another option would be to allow the weapon types to be much closer, so level one of the next higher weapon type doesn't give you higher stats than the highest refinement level of the next lower weapon type. If I keep my formula again, but set the damage value for the first level of a weapon type to the same of the third level of the previous weapon type, we max out at 28 for DR4. Build costs would have to be adjusted accordingly of course. The downside of that is that the increase in damage rating get's very low, so we might have to decrease tech costs - I wonder if it's worth to pay 2k RP to get DR4 with damage value 28 in place of DR3 with damage value 23...

I hope I could clarify a bit the problems I've been confronted with when trying to come up with these numbers.

-- especially when the doubled property is multiplied by multiple slots and multiple ships.

Well, that we can try to counter with even higher build costs. That said, I think it's not completely unreasonable that a high end ship equipped with the most powerful guns is worth entire fleets of lower tech ships. If priced accordingly, it shouldn't be possible to build them in masses, we even might balance costs in a way that you need a quite powerful industrial capacity to be able to build one of those in reasonable time.

Nothing motivates an artist like needing to replace abysmal programmer art.

Be careful what you wish for!!! Unless you want to end up with "artwork" of mine like this (my vision of a "neutronium hull"):

Spaceship.png (392.9 KiB) Viewed 3340 times

[Vezzra]

I don't think it would be a bad thing if early game battles frequently lasted for a few turns-- at a time when the player is more likely to want to closely manage his few ships, and withdraw from battles not going well.

The problem with that is that once an invader has survived one game turn of battle, he can happily pass through your system. Apparently it's not necessary ATM to actually defeat the defenders of a system to continue on to the next systems. This will make defending choke points very difficult, a problem that the AI will definitely have more trouble to deal with than a human player.

[em3]

Another option would be to allow the weapon types to be much closer, so level one of the next higher weapon type doesn't give you higher stats than the highest refinement level of the next lower weapon type.

I was going to suggest just that - make the highest refinement of a weapon better than the next tier weapon (but worst than it's first refinement).

[eleazar]

I don't think it would be a bad thing if early game battles frequently lasted for a few turns-- at a time when the player is more likely to want to closely manage his few ships, and withdraw from battles not going well.

The problem with that is that once an invader has survived one game turn of battle, he can happily pass through your system. Apparently it's not necessary ATM to actually defeat the defenders of a system to continue on to the next systems. This will make defending choke points very difficult, a problem that the AI will definitely have more trouble to deal with than a human player.

Ah, in that case-- suggestion withdrawn.

[Dilvish]

I don't think it would be a bad thing if early game battles frequently lasted for a few turns-- at a time when the player is more likely to want to closely manage his few ships, and withdraw from battles not going well.

I think I pretty strongly agree with this, subject to resolving the point Vezzra raises.

The problem with that is that once an invader has survived one game turn of battle, he can happily pass through your system. Apparently it's not necessary ATM to actually defeat the defenders of a system to continue on to the next systems.

Certainly if combat is expected to often last more than one turn before one side is vanquished, and even as it stands now, I think this would be a good dynamic to change, & seems to me it shouldn't be too tough. The intended change from current is just that you can't break a blockade unless you beat the blockaders (via combat or they retreat).

Fleets could get a new attribute 'm_last_system' (perhaps the very similar existing 'm_prev_system' could be used with minor adjustment, but I'll talk about a new attribute). All fleets would start at creation with 'last_system' being set to their initial system. A fleet stopped in a system would only be able to enter a starlane if the fleet's 'last_system'' matched the current system or the system that the starlane leads to. Fleets attempting to pass though systems mid-flight would be stopped, or not, according to the same combat test as now. If allowed to pass through, their 'last_system' would be updated as with any other entry to a starlane-- Upon (allowed) entry to a starlane, the 'last_system' is set to the ID of the system just left. Turns therafter traversing or exiting a starlane do not affect 'last_system' (unlike current 'previous_system' which is reset immediately upon system entry to that new system). At the end of the combat resolution phase, if a fleet has no armed aggressive opponents in the system, then its 'last_system' is reset to be its current system. How does that sound?

[eleazar]

Certainly if combat is expected to often last more than one turn before one side is vanquished, and even as it stands now, I think this would be a good dynamic to change, & seems to me it shouldn't be too tough. The intended change from current is just that you can't break a blockade unless you beat the blockaders (via combat or they retreat).

...

How does that sound?

It sounds excellent. Easy to understand. I was afraid it would be thorny to implement.
Unless there is some complication, that i haven't considered, that's closer to the way i've long wanted it to work.

P.S. I considered splitting this, but it does seem relevant to the OP. If the two topic diverge to the point that one hinders the other, we can split.

[AndrewW]

How does that sound?

Sounds like a plan to me.

[Ragwortshire]

Another option would be to allow the weapon types to be much closer, so level one of the next higher weapon type doesn't give you higher stats than the highest refinement level of the next lower weapon type. If I keep my formula again, but set the damage value for the first level of a weapon type to the same of the third level of the previous weapon type, we max out at 28 for DR4. Build costs would have to be adjusted accordingly of course.

I also really like this idea. (I have played the game through and been lurking for a bit.)

The downside of that is that the increase in damage rating get's very low, so we might have to decrease tech costs - I wonder if it's worth to pay 2k RP to get DR4 with damage value 28 in place of DR3 with damage value 23...

I think this issue is ameliorated a lot by the existence of non-stacking DR-shields. If your opponent is fielding ships with 16 shields, then going from (23 - 16) = 7 damage to (28 - 16) = 12 damage is definitely a big deal!

Making the refined weapons more powerful than the first refinement of the next tier up would also, in my opinion, add more strategic choice to weapons research. Given that research into refinements is always "wasted" in the long run, there ought to be some strong benefit to getting them to compensate.

You could also have interesting interaction between the research patterns of different empires. You could threaten another empire, not neccessarily because you can definitely conquer them, but because you can force them to spend research on refined weapons. Meanwhile you research a higher-tier weapon yourself, and invade them later when the investment pays off (i.e. when you have refined that weapon).

Regarding numbers, how about:

Weapons tier 1: 2 -> 3 -> 4 -> 6 for 10 cost
Weapons tier 2: 5 -> 7 -> 9 -> 12 for 15 cost
Weapons tier 3: 10 -> 13 -> 16 -> 20 for 20 cost
Weapons tier 4: 17 -> 21 -> 25 -> 30 for 25 cost

Armor: 6 (cost 2) -> 12 (cost 3) -> 20 (cost 4) -> 30 (cost 5)
Shield: 4 (cost 20) -> 9 (cost 35) -> 16 (cost 50)

Then to get through a level 1 shield, you have the choice between refining tier 1 weapons or researching tier 2. To get through level 2, you need refined tier 2 or research tier 3, etc. Even in the end, the level 3 shield would still be very useful.