Xenonauts: Community Edition discussion

[Solver]

That would surely enable some realistic Soviet research projects, costing a lot of money and resources and granting no effects.

[TrashMan]

I disagree. '****' is better because it alerts the mod-maker that they forgot to add an entry to strings.xml, which is critical if we're going to get proper translations of the game.

Failing loudly is better than failing silently in this situation.

Which is relevant mostly for descriptions. Names don't change.

F-17 Condor is F-17 Condor in english, german and italian.

And if a modder misses that the description just reads "F-17CondorDes" instead of a wall of text, then it is his fault. He is as likely to miss that as he is "****"

[kabill]

Here's another one...

Is it possible to give researches a cost? Like how manufactures work. For example, a certain tech would consume technology or would require money to research it - researching this tech will consume 10x "item x" and cost "$x".

That would be really cool. There's a workaround for that at the moment, running it through the manufacturing system (so you research a 'prototype' that you then manufacture which then unlocks the tech properly) but having a way to do it directly through research would be great.

[Solver]

Which is relevant mostly for descriptions. Names don't change.

F-17 Condor is F-17 Condor in english, german and italian.

And if a modder misses that the description just reads "F-17CondorDes" instead of a wall of text, then it is his fault. He is as likely to miss that as he is "****"

Names change. The F-17 Condor may be that in German and Italian, but a Russian translation will likely name it F-17 "Кондор" or something like that. If you're making software that is supposed to have translations, never assume that there will be a single string that is the same in all languagse. Even the full stop at the end of a sentence is not always the same.

[TrashMan]

Regarding aircraft XP, that's something I'd like to work on down the line.

The idea would be to create separate Pilot entities that can be recruited/assigned to aircraft/promoted/killed/gain XP, etc. Basically expanding on the unimplemented stretch goal. Pilots would have stats that translate to bonuses during air combat.

For portraits, I'll just use the default portraits and photoshop aviator glasses on all of them.

It should be kept simple. At least for starting implementation.

I guess you could hire pilots from the aircraft screen (or hireing would be automatic when you get an aircraft).

What about pilot stats?

If you want to keep it really simple, you'd just track kills.. or missions (and tie them to a pilot, not an airplane.. Or why not both? Airplanes can have their own service history)

Either way, for every 5 kills the pilot could gain a level. Each level grants 5% accuracy and dodge. So an experienced level 4 pilot would have a 20% accuracy bonus (while the enemy would have a 20% penalty)

Also, chance to survive being shot down should also be implemented. Higher with higher level?

[Slayerjerman]

Here's another one...

Is it possible to give researches a cost? Like how manufactures work. For example, a certain tech would consume technology or would require money to research it - researching this tech will consume 10x "item x" and cost "$x".

love this idea

[kabill]

Cannons have accuracy last time I checked. It was 1 (or 100%), but we can change that.

Just checked the air combat weapons file and there's no accuracy values for weapons. Have you seen this somewhere else? The only other reference I've seen to accuracy/evasion in air combat is an old variable used for dodging in gameconfig.

[kabill]

Sorry for double post, but Kevin's comment on the Dynamic UFO thread reminded me of something:

From what I can tell, the alien race types that UFOs can spawn with (i.e. Caesan, Sebillian and Andron) are hardcoded. If you add in a new type, be that a completely made up type or one of the other races, a UFO can spawn but will have no crew.

E.g. I could add in a "corvette.harridan" entry in ufocontents folder, and can get that UFO to spawn in game. But the UFO itself will never contain any crew members (in spite of setting them up in the file; I've tested this by copy-pasting existing files to be sure).

Is there any way that this could be unlocked so you could add in new ufo contents types. This would solve the issue of ufo crew diversity entirely, as for example you could set up two Caesan, Sebillian and Andron types for a UFO and have the lists differ for each (e.g. changes in AI scripts, changes in alien types, etc.)

[Solver]

At least extending that to other existing alien races so that a "corvette.harridan" would work is not problematic, judging by an initial look at it.

[kabill]

Cool. While a fully unlocked system would obviously be best, even adding in the existing race types would over double the possible crew loadouts at the moment which isn't bad at all.

[agris]

- aircraft experience. Since cannons and missiles have an accuracy value, implementing increased accuracy and dodge (reducing enemy accuracy) with experience should be doable.

I couldn't agree more on the aircraft experience suggestion. This is a less complex solution to implementing pilots, like bonds0097 suggests.

Regarding aircraft XP, that's something I'd like to work on down the line.

The idea would be to create separate Pilot entities that can be recruited/assigned to aircraft/promoted/killed/gain XP, etc. Basically expanding on the unimplemented stretch goal. Pilots would have stats that translate to bonuses during air combat.

For portraits, I'll just use the default portraits and photoshop aviator glasses on all of them.

That sounds great, but involves a lot more work than making 1 plane = 1 pilot. There should be caps for the increases that experience gives you, something like the next tech class equivalent minus a certain value (OFFSET). So if you're flying a F-17 Condor and the next plane is the F-99, with a maneuverability and dodge of X and Y, the max level a F-17 Condor could achieve would limit this to X-OFFSET1 and Y-OFFSET2.

Same for the weapon systems. Planes (i.e. pilots) should have experience with a given loadout on hardpoints. So using an autocannon and sidewinder missiles, you can increase their accuracy to a certain point not to exceed an amount LESS than the next available tech. If that same plane (pilot) switches the loadout on his hardpoints to the next tech, the accuracy resets to standard and he has to learn how they work all over again.. but the accuracy is never equivalent to the next level of tech. The degree of offset between "experienced old tech" and "novice new tech" would be governed by the degree of technical advance.

This isn't true, unless you count 'homing' as accuracy.

You could possibly simulate it with damage/HP increases on planes, though.

I disagree. An experienced pilot who knows the weapons behavior of his loadout will know the limitations of his homing and kinetic kill devices. This will affect how he establishes conditions for the launching of missiles / deployment of ordnance and how he leads his targets based on target speed and flight path / trajectory. There are plenty of "roleplaying" reasons why a more experienced pilot would have more accuracy with a given weapon set. I don't think HP or DMG should ever be increased. Too hard to balance and marginalizes the affects of improved tech too much.

Edited July 10, 2014 by agris

[kabill]

I disagree. An experienced pilot who knows the weapons behavior of his loadout will know the limitations of his homing and kinetic kill devices. This will affect how he establishes conditions for the launching of missiles / deployment of ordnance and how he leads his targets based on target speed and flight path / trajectory. There are plenty of "roleplaying" reasons why a more experienced pilot would have more accuracy with a given weapon set. I don't think HP or DMG should ever be increased. Too hard to balance and marginalizes the affects of improved tech too much.

I have no doubt that experienced pilots are more accurate with their weapons systems. Wasn't at all disputing that.

My point was that, in code at the moment, there doesn't appear to be any simulation of accuracy at all. I think this was the case, once, as there's a deprecated variable in gameconfig that implies that the roll ability once increased weapon miss chance rather than literally moving the plane out of the way. But there's no trace of that system that I can find (having checked after TrashMan's comment), at least not in the air combat weapons file. Of course, it could be rebuilt (or dug out of the code if it's still hanging around there) but otherwise accuracy isn't a variable the game knows.

In terms of the suggestion of damage/hp increases, this was merely a suggesting given that accuracy/evasion systems don't exist. It's not ideal, but it would function as a simulation of greater piloting skill in the absence of variables which actually allow you to do that.

Furthermore, actually the effects are pretty equivalent. On average increasing chance to hit from say 50% to 60% is the equivalent of a 20% increase in damage output, for example. I'm not actually sure then why the accuracy increase would be less hard to balance or marginalise higher level tech more.

Indeed, if you're applying a flat bonus to accuracy that's going to be harder to balance: an extreme case, but if your base accuracy is 10% and you add 10 to that, you've doubled your chance of hitting and therefore doubled your damage. In contrast, if your chance of hitting was 70% and you add 10 you've only increased your chance to hit by ~15% and therefore are only doing ~15% more damage.

Alternatively, if you're going to increase accuracy as a proportion of the base accuracy of the weapon (say, 150%*base chance) then the effect is in fact identical with the average done by the weapon and could easily be implied as a damage increase instead.

As such, I'm not sure why adding accuracy/evasion is better than adding damage/hp. They basically do the same thing, and doing the latter would avoid having to (re)construct an entire system for air combat weapons and make the experience system compatible with the vanilla game.

[Max_Caine]

Having spent a lot of time with AC, the closest thing you'd get to accuracy in-game would be with homing weapons (such as missiles). Missiles have an independant turn speed and turn rate. As a pilot becomes more experienced, you could jack that up. However there's no way to simulate accuracy with cannons without major changes to the code. Cannons fire directly at the target. That's what the game permits - direct fire at the target and nothing else. A coder would have to figure out a way for cannons to plot ahead of the target so the cannon shot would hit the target (instead of shooting where the target was a second ago). At that point you could introduce accuracy for cannons - the accuracy of the pilot would be how good they would be at leading a target with cannons.

[agris]

My point was that, in code at the moment, there doesn't appear to be any simulation of accuracy at all.

[...]

In terms of the suggestion of damage/hp increases, this was merely a suggesting given that accuracy/evasion systems don't exist. It's not ideal, but it would function as a simulation of greater piloting skill in the absence of variables which actually allow you to do that.

Furthermore, actually the effects are pretty equivalent. On average increasing chance to hit from say 50% to 60% is the equivalent of a 20% increase in damage output, for example. I'm not actually sure then why the accuracy increase would be less hard to balance or marginalise higher level tech more.

All good points, I didn't realize there was no accuracy value for missiles. Perhaps their turn radius / speed could be adjusted as 'planes' gain experience, though I'm sure that would complicate the displaying of statistics in the Xenopedia.

I would agree with your analysis that damage / accuracy are equivalent, except for one factor. The last hit required to destroy a UFO. With a given amount of health left, a lower accuracy / higher damage attack is going to have a 'spikier' probability distribution of destroying the UFO than a more accurate / lower damage attack. Or, to view it another way, it could take 3 hits of a lower damage weapon to finish a UFO (all at higher probability of the hit striking) or 1 hit of a higher damage weapon at a lower probability of success. The absolute values of damage / probability would need to be analyzed to determine what makes more sense, but the game-play elements that you want to emphasize (a more predictable 'smooth' probabilistic distribution of damage vs. a more random 'spikier' distribution of damage) would really dictate the correct course.

I think damage is one of the big things later weapon systems have going for them (in addition to AoE), so I wouldn't like older tech used by experienced pilots to co-opt that advantage. Rather, with experience should come predictable results (i.e. smoother probability curve of dealing damage). A novice using advanced weapon tech would result in a 'spikier' or more random distribution of damage over time. Could the differences between damage and accuracy be smoothed out by adjusting the damage? Sure, but then what's the incentive to upgrade (other than AoE), especially if accuracy isn't on the table?

[bonds0097]

I was merely stating that I was probably going to work on a pilot/aircraft XP system in the future. I was not laying down my proposed design for such a system. When and if I do decide to work on such a thing, I will certainly think through the design carefully.

That said, XP increasing damage can absolutely make sense. An experienced pilot will know to target more vulnerable systems like exposed engines or weapons hardpoints or weakpoints in the hull or whatever RP reason you so choose.

Again, I'm not saying that that's a thing I'll do, simply that it's premature to start armchair designing.

[kabill]

Cannons fire directly at the target. That's what the game permits - direct fire at the target and nothing else. A coder would have to figure out a way for cannons to plot ahead of the target so the cannon shot would hit the target (instead of shooting where the target was a second ago). At that point you could introduce accuracy for cannons - the accuracy of the pilot would be how good they would be at leading a target with cannons.

Is there something distinct about cannons that means they could never have tracking/homing? They certainly have a value for it (albeit set to 1). Does that not do anything with cannon-mount weapons?

(I'm ignoring the fact that homing bullets doesn't make any sense and assuming it would be a proxy for more accuracy/whatever).

@agris:

Yes, you're right, there are some differences and your point about randomness vs. consistence in damage is fair. My main point was simply that there's not as much difference as you seemed to imply.

In terms of your last point, though, I don't see why experience increasing damage would make you not want to upgrade weapons. Because the damage bonuses will apply to all tiers of weapon.

[Kiel]

Sorry to break your air combat discussion but is it possible to add melee type enemies into the game? Like heavy armoured Sibillian armed with some kind of a horrific blades/claws, able to OHKO non-protected soldier?

I guess having some close range knife fights would be cool.

[ventuswings]

Sorry to break your air combat discussion but is it possible to add melee type enemies into the game? Like heavy armoured Sibillian armed with some kind of a horrific blades/claws, able to OHKO non-protected soldier?

I guess having some close range knife fights would be cool.

Can't other modders do it already with existing tools though, without having to edit the source code?

On the different topic, is it possible to change the mechanic behind base structures being destroyed/repaired after destructive base defense mission? Currently I don't think it plays into factor at all. At very least some financial hit according to the amount of props destroyed would be nice.

[Solver]

It's not so much that it needs changing but rather implementing It's not in there at all.

[kabill]

It's not so much that it needs changing but rather implementing It's not in there at all.

Is it actually not? Could well be tempted to put out a balance patch which stops the AI from attacking base structures, at least in the mean time, as I think it's currently implemented in the AI.

[Solver]

It's not that I have looked super closely, but I haven't seen any code doing that destruction. And nobody has ever reported having the buildings destroyed, so...

[Solver]

With credit to llunak for the super-cool mod loading system and numerous other changes, and to bonds0097 for the memorial screen:

CHANGELOG for 0.25:

- Fixed misaligned TU bars in ground combat

- Fixed inability to change weapons and do other maintenance on refueling aircraft

- Fixed inability to target event markers as movement waypoints

- Fixed potential Geoscape AI freeze if too many supply run missions spawned

- From 1.08: Fixed CTD with Harridans flying out of the map

- Adjusted placement of air combat speed buttons (thanks Skitso)

- A casualty list/memorial is now available in the Barracks screen, by bonds0097

- Air superiority missions no longer have a "main" fighter

- Air combat AI improved with some basic logic, enabled by airCombatAIThinkChance in gameconfig

- Weapon range colours from air combat moved to colours.xml

- The way buildings work has been overhauled, they are now far more customizable - guide

- Researches can now upgrade buildings arbitrarily with the UpgradeBuilding hook, see researches.xml for an example

- Researches can now use the LockResearch hook. This makes another research permanently disabled and can be used to implement a tech tree with exclusive branches.

- The effect of human and alien bases on the maximum alien wave UFO count is now in gameconfig.xml

- Alien bases can now grow independently, based on how long they have been on the map, and on supply run missions. Explanation of using this.

- UFO spawns can now be made more dynamic by providing probabilities for each particular UFO type in every ticker band.

- Temporarily disabled AI logic for firing at props in Xenonaut base, since it does nothing anyway

- Support for multiple Hidden Movement screens, also depending on the tileset

And the most significant change for modders in 0.25 is that it includes a functional version of llunak's modular loading system, for which there be separate documentation, but the current thread is here. Modders should try and make as much use of it as possible so that mods do not conflict!

Edited July 13, 2014 by Gauddlike

[kabill]

No mention of dynamic UFOs. Did that not make it in, or is it missed off the changelog?

EDIT: But in any case, some cool fixes/changes. Thank you to all involved.

EDIT 2: Also, is it built onn v1.07 or v1.08? If the former, will it work on top of the latter?

EDIT 3: Just saw the changes are in there; they need adding to the change log.

Edited July 11, 2014 by kabill

[Solver]

Dynamic UFOs are also in, and I'll fix the changelog.

It's based on top of 1.08, although the only code difference between 1.07 and 1.08 is a particular CTD fix.

[kabill]

Ta. Just making a backup copy for modding and then I'll get started.