View source for Talk:Explosions

Zombie, even gas station explosions don't hurt anybody nearby? - [[User:MikeTheRed|MikeTheRed]] 18:37, 3 August 2007 (PDT)

:I'm pretty sure I've gotten civilians killed by stray shots blowing up gas pumps before.  [[User:Arrow Quivershaft|Arrow Quivershaft]] 19:59, 3 August 2007 (PDT)

::Friendly units are not affected by any explosions caused by battlefield objects on UFO (on TFTD and some of my mod terrains this does not apply). There's some sort of quirk however concerning explosions that only generate smoke: I've seen often aliens/civilians get killed because of those explosions: I think (not confirmed) one factor might be if those units are already standing on smoke and the game wrongly allocates them damage everytime a barrel explodes. [[User:Hobbes|Hobbes]] 02:07, 4 August 2007 (PDT)

:On that note, I've most certainly killed aliens by blowing up fuel barrels in the Hangar during base defense missions.  [[User:Arrow Quivershaft|Arrow Quivershaft]] 07:48, 4 August 2007 (PDT)

:: Just to throw a spanner in the works, sometimes I've seen aliens that weren't near the bowser or fuel drum die after it explodes. For Quite the explosive odd ball out. - [[User:NKF|NKF]]

: Ok, this is what I know. MapView lists those "explosive" objects as having a damage type of smoke, not high explosive. This checks out on the battlescape as when you shoot one of these objects, the explosion doesn't damage terrain and it produces a dense smoke cloud just like a Smoke Grenade does. Moving X-COM soldiers near the fuel bowser and then shooting it doesn't do them any harm - that's basically a given. 

: However like NKF and Hobbes mentioned, Civilians and aliens can <i>sometimes</i> die after you shoot one of the objects. It doesn't happen all the time though and a quick round of testing reveals they normally don't take damage from the explosions either. This leads me to believe it is a bug of some sort. I did have a civilian die last night after shooting a fuel bowser: it was about 6 tiles away and wasn't standing in a smoke cloud. But, I did shoot a couple pumps out in the previous rounds which may be incorrectly allocating damage to nearby non-human units like Hobbes points out. Still, more testing needs to be done before we can come to a logical theory. Right now we have nothing to stand on other than our words. ;)

: A main problem with the civilians is that they never get close enough to a fuel bowser on a terror mission to test this out properly and you can't use MC on them as that changes their ownership flag. I might do a little map editing and intentionally force civilians to spawn surrounding the pumps, and also give the civilians no energy or TU to force them to stay put instead of wandering off. Then , if the pump gets shot out and they all survive we'll know that thay don't take damage <b>normally</b>. (By normally, I mean no other object explosions happened on the map in previous rounds). After this I can go nuts and shoot off a whole bunch of pumps to see what happens. If the civilians die when standing in smoke, then we can come to a conclusion. --[[User:Zombie|Zombie]] 08:06, 6 August 2007 (PDT)

: I have seen aliens killed and objects damaged when the purple tables explode. This is most visible in the command centers of bases, where the whole place goes up when a blaster bomb is fired,even if the back blast shouldn't hit the rearmost segments with enough force.--[[User:(name here)|(name here)]] 10:07, 18 November 2007 (PST)

<b>Update:</b>[[Image:Bowser.png|right]]

I managed to construct a testing scenario for civilians against a fuel bowser explosion the other day. The gas station map was edited so that the fuel bowsers were inside a large structure with some lighting sources (lamps) aroung the perimeter for giggles. Since I didn't have access to the internet and since Daishiva's MapEdit doesn't allow you to change the spawn points, I ended up decoding the RMP structure and editing the spawn points myself. Of the 16 normally occurring spawn points on the gas station map, I changed all of them to be civilian with a spawn priority of 10 (meaning a high probability of a civilian showing up there) and clustered 8 of them around 2 fuel bowsers with links which circled the pump to keep the civilians from wandering off. Then I kept going on a terror mission until the structure showed up with enough civies spawned around a pump.

First thing I did was to edit all civies to have 200 health. After this, I shot a pump with 3 civilians standing around it and checked to see if any of them were harmed. Only one was injured. The picture at the right details the scenario. G stands for the gas pump (er, fuel bowser, whatever you call it), C1, C2, and C3 are the civilians and S is the soldier. The solder shot at the pump with a Laser Rifle and the civilian injured in the blast was #3 (red square). No other civilians were ever injured. I didn't have time to check out anything else but the orientation of the shot and the location of the civilians play a role in who will get hurt. 

Just for the heck of it, I decided to see how much damage was inflicted to civilian #3. After writing down about 20 numbers I realized that this would be a perfect job for BB's logger and AutoHotKey. I wrote a script, automated the whole process and then came back an hour later. Now, the fuel bowser has a listed damage potential as 46, so I half-guessed the damage would be between 23 and 69 (1/2 and 3/2 the ave damage). Wrong. It was between 0 and 120! This is more like a normal ammunition type than an explosion. After sitting there in disbelief for a while, it hit me that the Laser Rifles damage potential was probably being wrongly allocated to the civilian. When I changed the weapon to the Heavy Laser, the civilian got between 0 and 170 points of damage - exactly what would be predicted with ammo having 85 average damage. And no, the Laser Rifle shots were not missing the target and hitting the civilian directly, I can't stress this enough.

The only conclusion I can draw so far is that the more powerful of weapon that you use to shoot the fuel bowser, the more damage will be dealt to someone standing near it. Of course, more testing is in-order here, but at least we have a workable theory on damage. --[[User:Zombie|Zombie]] 18:40, 9 August 2007 (PDT)

:You can edit the .rmp files with Map View but you seemed to have managed quite well by your own :)
:What you described is the effect when a weapon's hit completely destroys a terrain feature it transfers the excess damage to units (and objects?) closeby. Or, the shot destroys the target and carries on. At least that's what it seems to me on the rare occasions that i witness that behaviour, since it's hard to get it with either bullets against human buildings or plasmas against UFO walls. 
:Still that leaves the question about far away aliens dying from gas pump explosions. One thing I remember that can help is that the smoke is generated on the alien's square before it dies. - [[User:Hobbes|Hobbes]] 22:44, 9 August 2007 (PDT)
--------
I think he means if you destroy a terrain feature that subsequently blows up, the explosion damage to nearby units would be the calculated in the same way as if they had been shot directly.

- [[User:Bomb Bloke|Bomb Bloke]] 19:45, 10 August 2007 (PDT)

:Correct, BB. @Hobbes - I'm still a little confused on who gets hit. See, if the  "bullet" were to "kill" the bowser and then continue on, the civilian damaged should be #2 (not #3). Unless the line of fire was screwed up in the intermediate effect somehow. Also, the amount of damage to the civilian is exactly what is predicted if you shot it directly. If the bowser were to "absorb" some of the damage, I'd see that reflected in the log file. And for the time being, this is only true for civilians. I haven't got around to testing if aliens suffer the same fate. --[[User:Zombie|Zombie]] 15:35, 12 August 2007 (PDT)
::Well i was *convinced* that shots would continue after destroying its target until I ran a couple of limited tests to verify this: on the first one I tried firing at soft walls with an unarmored xcom unit behind it. The unit suffered no damage at all. But then I remembered how sometimes objects when hit aren't replaced with their wrecked version (like crates....they just disappear). The crate supposedly would be replaced with the wreck but instead it simply disappears, which would mean that the wreck that replaces it gets the excess power of the shot and is obliterated as well. I can't really explain it but I feel this might be related to the dying civvies/aliens. -- [[User:Hobbes|Hobbes]] 17:22, 12 August 2007 (PDT)
------------
If you shoot a piece of terrain, the "armor" of that tile is subtracted from the damage the shot did. If the result is still positive, the tile is destroyed, or replaced with the "damaged" version.

In that later case, the armor of the new tile is again subtracted from the remaining damage total, and so on until you run out of "damage" or you run out of tiles to replace with.

(This is how my damage testing tileset (which shows the shot power against a given floor tile, on that floor tile): By drawing a number on each, giving them an armor of 1, and setting them to change to the next numbered tile when toasted).

In this situation, you see one "damage" figure used to destroy multiple tiles (I've done no testing to see whether the damage can then be passed on to units). Each tile takes a reduced amount of damage as the ones before it suck it up.

In the case of Zombies test, a new damage figure is calculated for every unit (or one for all?) that is harmed by the explosion of the tile (assuming it can explode): and that's based on the weapon used to damage that tile in the first place, not on what the tile was. The units take full weapon damage (not explosive damage, or it wouldn't be a random value).

That said, if I ever feel the need to pass on damage after destroying terrain I find an autoshot works quite nicely. But then, I always use autoshot, so go figure. ;)

Also worth a mention (whether or not it's relevant I dunno) is the IN (or was it HE?) bug. If you shoot a unit with an IN bullet of some sort, all units standing in fire (and smoke?) will take the damage of the shot (regardless as to whether they were in range of the blast).

- [[User:Bomb Bloke|Bomb Bloke]] 06:37, 13 August 2007 (PDT)

===UFO Power Source Explosions===

Months ago, I was working on the probability that UPSs and their Elerium would survive a crash. This involved the individual chance of engine explosion upon crashing, the blast damage of an exploding engine, and the configuration of UFOs (and whether blast damage could spread to other engines). But Arg, I've looked all around and can't find that work on the wiki. Or maybe it was in a forum? Anyone recall? :P - [[User:MikeTheRed|MikeTheRed]] 09:23, 9 October 2007 (PDT)

: I think it was in a forum.  Abstract:
:* Individually, each power source has a measured probability of 70% explode, 30% intact.
:* If a power source is in the range of the explosion of another power source, it is destroyed and does not explode.
:* I don't recall the calculated damage off-hand.  It is possible for the navigation consoles on a Medium Scout to survive their power plant explosion, but far from certain.
:* Power sources are checked in coordinate lexicographical order.  E.g., for the terror ship the order of checking is N, W, E, S.  [This also entails that for a Terror Ship, all four explode or none explode.  Assuming no RNG problems, this translates to a 0.81% chance of seeing four power sources.  Battleships are notable in that power source explosions do not destroy other power sources, unlike other UFOs with multiple power sources.] - [[User:Zaimoni|Zaimoni]] 12:51 9 October 2007 (CDT)

MTR: please see [[Talk:Exploits#Repairing_damaged_UFOs]] or [http://www.xcomufo.com/forums/index.php?showtopic=242025161 this topic] at the xcomufo forums. BTW, the Medium Scout, Large Scout, Abductor and Battleship all have their PS explosion probabilities calculated separately since they either only have one PS or they are separated by hulls or walls which reduce blast propagation.

The only ships you need to worry about are the Harvester, Supply and Terror Ships. However, since the power sources on those ships are always separated from each other by 6 tiles, we can do some quick and dirty calculations. First of all, a power source has a explosive rating of 180-250 at GZ. Extrapolating to GZ+6 (the location of the other power source), the damage is 120-190. Elerium has a damage rating of 20 which guarantees immediate destruction if another power source explodes (exception is the Supply Ship). Power sources are destroyed outright if the average listed explosive strength is 100 or greater at the point of impact. Since the damage at GZ+6 is always greater than 120 points, all power sources will be destroyed from the blast. In the case of the Terror Ship, if one PS goes off, the strength of that single blast is more than enough to immediately destroy the other power sources. Thus, for all the PS to survive, the game needs to roll the 30% survival dice 4 times for an overall probability of 0.3^4=0.81%. (Supply has 3 PS for a 2.7% chance while the Harvester has 2 for a 9% chance).

The Supply Ship is a special case because the PS are arranged in a straight line each 6 tiles apart. I haven't checked whether items (such as Elerium) can survive at GZ+12 of a PS explosion (I think they can). If this is the case, then a power source which explodes at the outside of the line will only destroy the one in the middle and permit a second roll to determine the fate of the surviving PS. Of course, if the central PS explodes, any successful "saving" rolls made up to that point are ignored and all the power sources are destroyed. --[[User:Zombie|Zombie]] 14:49, 9 October 2007 (PDT)

:Now it's coming back to me... great stuff! Thanks Zai and Zombie. And thanks for the links. It'll take me a little while to refresh everything in my mind, and maybe take your test map for a drive again. Ultimately I'll want to get this info onto appropriate wiki pages. - [[User:MikeTheRed|MikeTheRed]] 15:38, 9 October 2007 (PDT)

::I can shine some light onto the GZ+12 question, because I just yesterday recovered Elerium from a Supply Ship with two destroyed power sources.  Blaster Bomb explosions propagate for 11 tiles, and have no power after that, and tests have shown that a PS explodes with the same force as a BB. As well, it should be noted that sometimes the location of the PS starts on fire and may remain so for some time.  (On one final note, it makes more sense to me that Elerium is spawned under SURVIVING power sources; so there is no chance that it is destroyed by an explosion; it spawns after the explosions occur.  Of course, I could be completely wrong, it just makes more sense to me to code it that way.)  [[User:Arrow Quivershaft|Arrow Quivershaft]] 16:16, 9 October 2007 (PDT)

:Good point Arrow, thanks for that. Zombie, I've got a handle on it again and will post something about expected number of PSs at crash sites, a.k.a. average Elerium recovered. E.g. Supply ships yield 17.7 units of Elerium, on average (calcs to be shown on wiki). One thing has me curious... in your XCOMUFO thread you say you wrote down the edge strength of PS blasts (35 to 70) but then later said there are 71 possible values. But aren't there actually only 36? And again I wonder if truncation might have something to do with it... see, depending on just how they truncated what was applied (if they did) and/or how BB's tiles show the number, could have caused a fair amount of creep one way or the other... and it looks like it crept up. (''If'' there were 256 original values somewhere/somehow which got converted to a 7th as many someway/somehow (36/256) ... shrug.) But it's really only an incredibly minor issue.  .... Anybody care to recommend where I'd post the info on Elerium salvage? [[UFO_Crash_Recovery]] makes a lot of sense conceptually, but it's not a particularly info rich page. - [[User:MikeTheRed|MikeTheRed]] 18:58, 10 October 2007 (PDT)

::Heh Mike. The 35-70 damage is the amount found at R11. The 90-125 damage (range of 36) is the average at GZ for objects. The last column is the actual explosive strength of the PS against units 180-250, range=71. Sorry, lots of numbers there to confuse.

::I'd include this info in the UFO Crash Recovery page. That seems like the best place for it, no? --[[User:Zombie|Zombie]] 19:42, 10 October 2007 (PDT)

:Z, am I losing my mind? but PS explosions are quite unusual, aren't they? 1) they have a ''range'' versus terrain; all other explosions are a simple, invariant 50% of rated damage versus terrain, and 2) while they have a range, it is not +/- 50% of rated, like it is for explosives vs. units. If we are indeed seeing 50% of rated vs. terrain (which the decrement-by-5 strongly suggests), then your min-ave-max conclusion of 180-215-250 for PSs is actually +/- 16.28% of rated. How weird is that (and where'd they get ''that'' percent)? Am I doing this right?? Maybe they special coded PSs. (There was also some "hybrid weirdness" seen with the  [[Small_Launcher#Stun_Bombs_vs._HE_Block|stun bomb]].) - [[User:MikeTheRed|MikeTheRed]] 21:37, 10 October 2007 (PDT)

::LOL, Yup Power Source explosions are in a class all of their own. When I came back to this recently I was quite perplexed at the numbers since they do not follow the anticipated 50%-150% range which normal explosives dish out. The extrema of the PS at 215 damage is +/- 35 damage points which is strange. Don't know why the programmers decided to code this explosion so differently but what's done is done. It's our job to explain it which is usually hard. --[[User:Zombie|Zombie]] 22:14, 10 October 2007 (PDT)


:Ok I have added [[UFO_Crash_Recovery#Power_Source_Explosions_and_Elerium_Recovery]] (and re-organized the page a little) and also created [[Explosions#Unusual_Explosives]] and simplified [[Small_Launcher#Stun_Bomb_Explosions]].
:*Did I get that last paragraph on [[Explosions#Stun_Bombs]] right? Also see [[Talk:Main_Page#Re-organize_Explosions]]. 
:*Zombie, is there any way you'd care to do a little more testing on crashing and PS explosions - specifically, how hard are units hit? Given how PSs break the standard explosion rules, it'd be interesting to try to modify the game so that the sectoids(?) in the savegame crash scenario, have a lot of health (and armor?), so one can cross-confirm that the damage inflicted on terrain, is what is seen versus the units. (If PSs break some rules, do they break others? Is terrain really being hit by 50%? Just one look ought to answer everything.) I'm not sure how/if aliens' stats can be modified, when moving from GEO to BATTLE... can it be?
:- [[User:MikeTheRed|MikeTheRed]] 17:54, 12 October 2007 (PDT)

::Sure, I'll fool around with this a little bit tonight. But I'm letting you know right now that setting this scenario up is gonna take some time. I'll need to edit the routes (erm, spawn points) of a craft so the aliens will start directly next to a power source. Then the aliens stats will need to be modified in the executable to have high health to survive the explosion (assuming the PS explosion does a max of 250, an alien with 255 health should survive all the time so modifying armor would not be necessary). Also, to get any idea of what is happening with the blast, the quantity of aliens will need to be modified in the executable to force more of them to show up (8 is good for a GZ+1 test and 16 for a GZ+2). Will let you know what I come up with.

::Edit: I created a scenario and ran it through its paces just now. Turns out Sectoids were dying left and right with 255 health. No good. I edited them to have 200 health and 100 overall armor and checked again. A couple times I saw corpses around the power source so the strength of the blast must be at least 310. 300 + 10 GZ+1 adjustment. As a quick calculation, if GZ is slated to have a blast strength of 250 and it follows the normal explosive theory, the max should be 250*3/2 = 375. Will need to jump up to the higher difficulties in order to bump armor up to 200 though - beginner always cuts the armor rating in the executable by half. Then we can see what will happen with a 400 health unit. Heck, maybe I better crank the total up to 500 to be safe, lol.--[[User:Zombie|Zombie]] 23:31, 14 October 2007 (PDT)

::Edit2: Well, with an armor of 250 and a health of 250, the Sectoids didn't seem to take any damage so I ended up switching everything to 200. Not sure how many trials I ran but it was on the neighborhood of 200 or so and the lowest value I reached was a health rating of 73. 400-73= 327 at GZ+1 and therefore 337 at GZ. More trials will be necessary to narrow the actual number down since the damage range is extremely high (each value in the range has a 0.3% chance of showing up once). --[[User:Zombie|Zombie]] 21:48, 15 October 2007 (PDT)

:Zombie, sorry for the delay. Wow I had thought this would only need a single test - I hope this isn't too much trouble! Are you also writing down the numerical tileset (terrain) GZ damage each test? It's definitely looking like PS explosions break yet another rule - they are not 50% of average/2 (and who knows what the relation is, unless enough data is colleced). - [[User:MikeTheRed|MikeTheRed]] 15:52, 17 October 2007 (PDT)

:Looks like before the first turn, the engine will look for every tile in the map (it scans the MAP.DAT data linearly) ; when it finds a power source (it checks if the MCD special property is set to 2), there is a 25% chance that it will leave it alone. Otherwise, it'll generate an explosion at the UPS location with a strength of 180+RND*70. Whether the UPS blows up on top of that or is just destroyed, I do not know. Can someone hack the MCD data and see if it's possible to generate an explosion on a tile that is not a UPS just by messing with the special property? PS: I am almost certain of the 75% probability of explosion vs 70% that is often stated here. [[User:Seb76|Seb76]] 09:31, 12 February 2008 (PST)

===Bodies===

I've noticed that sometimes when you use a high level explosive(Alien Grenade sometimes, but more often Blaster Bombs or XComUtil High-Explosive), aliens killed will not leave a corpse.  Has anyone figured out the rhyme or reason to this? [[User:Arrow Quivershaft|Arrow Quivershaft]] 13:33, 13 October 2007 (PDT)

----

This also happens with human grenades; a fully healthy Sectoid must be at least two squares away from the detonation center to leave a corpse on Superhuman. [[User:Zaimoni|Zaimoni]] 16:38, 13 October 2007 (CDT)

----

Stranger and stranger...I just Mind Controlled two Ethereals, stood them next to each other, and had one of them prime his grenade for immediate detonation and deposit it at his feet.  The ethereal at ground zero, right on top of the grenade, was killed and left a corpse; however, his buddy standing adjacent(and to the south) got completely fragged(no corpse).  Makes me even more curious!  [[User:Arrow Quivershaft|Arrow Quivershaft]] 15:49, 13 October 2007 (PDT)

----

Arrow, I'll try to explain this the best I can. Just for the sake of understanding let's look at a piece of terrain. Say you shoot at a tile containing an object (a brick fence for instance). If you shoot at it with a low powered weapon it probably will not do anything to the fence segment. However, if you use a higher powered weapon it will destroy the fence segment which will in turn convert it into a destroyed fence section. If you use a really powerful weapon, it will obliterate the fence segment completely. What happens in this case is that the high damage weapon will destroy the brick fence segment converting it to a damaged object. If the amount of power used up in this process is less than the power of the hit, the resulting excess is applied to the damaged object. If the excess is enough, it will destroy the damaged object. Usually the process ends here and no further objects are created if more damage is applied. Since there isn't a tertiary object pointed to, the object is obliterated. (But in the case of Bomb Blokes numerical damage testing tileset, he linked a whole bunch of tiles together, each with a damage greater than the last one to form a testing scenario).

The same thing happens to live aliens as it does with objects. When an alien is killed, it leaves a corpse "object" behind. If the damage applied is high enough, it destroys the corpse object and leaves nothing behind. The reason why you may notice some strange results is that the damage done by weapons is random (this is especially true with explosives which damage an area). So even though the damage dealt at GZ wasn't enough to destroy an alien outright, it doesn't mean that aliens further away from the blast center will take less damage than those closer to GZ (damage is calculated independently for each tile but still follows the usual damage profile for units). Overall though (like 200+ trials), aliens closer to the blast stand more of a chance to be obliterated than those standing further away. Hope this helps. --[[User:Zombie|Zombie]] 18:03, 14 October 2007 (PDT)

----

Ok, thanks!  Makes alot more sense now.  I suspected that might be the reason, but I wasn't sure about it.  Considering my primary explosive at this point is the XComUtil High Explosive, it makes sense that the corpses are obliterated.  Thanks for the explanation!  [[User:Arrow Quivershaft|Arrow Quivershaft]] 18:10, 14 October 2007 (PDT)

----

[[Image:Numerical PS Screencap.png|right|thumb|250px]]
Note that although most weapons do random damage to their targets, explosives always deal a set amount of damage to tiles (which depends on the distance from GZ, any non-ground tiles in the way and the original power of the blast), as per this image.

Explosive damage to units is random, however. Dunno if it's random against items or not (I suspect the later). Damage from non-explosive weapons is never applied to those.

- [[User:Bomb Bloke|Bomb Bloke]] 19:01, 14 October 2007 (PDT)

----

Sort of. At least this is what I think happens: blast damage from explosives is random to living units. If the random damage allocated to the unit is enough, it will kill the unit creating a corpse. If the blast has even more power, it will pass the excess on to the corpse leaving nothing. Think of the death tile for the corpse as an extra amount of health added to a live alien. If that value is surpassed, it bypasses corpse creation and creates the next tile in the sequence. In most cases, this is nothing.

Non-Explosive weapons can and will have the damage applied to an alien and then the corpse if damage is enough. However, once the corpse is formed, any damage from non-explosive weapons will not destroy the body (you can't target a corpse or items to shoot at, only the ground). --[[User:Zombie|Zombie]] 19:19, 14 October 2007 (PDT)

----

Erm, I'm not following you on that last part. You're saying that if I shoot a sectoid with, say, a heavy plasma, any excess damage will be passed onto the corpse... But the corpse will never be destroyed?

In that case, how do you know any damage was applied to the corpse at all? Items don't have an individual "health" stat that I know of, they either get destroyed or they don't get scratched.

- [[User:Bomb Bloke|Bomb Bloke]] 19:46, 14 October 2007 (PDT)

----
Let's say you shoot a Sectoid with a Laser Rifle modified to dish out 255 damage. If the shot pumps out the average (255) it will kill the Sectoid, pass the excess damage on to the corpse object and "kill" the corpse as well all in one fatal swoop. This boils down to sequencing. What probably happens is the game decides what to create by comparing the shot power to the damage ratings of all the tiles in the list sequence. If the damage is high enough, the game creates the appropriate item. It never actually creates the intermediate objects on the battlescape, otherwise a corpse would permanently end the sequence as it is an item from obdata.dat. Is that better? --[[User:Zombie|Zombie]] 20:34, 14 October 2007 (PDT)

----

No sorry, now you've lost me <b>completely</b>. What the heck do tiles have to do with this?! I suspect we're not talking about the same issue here.

Terrain when "destroyed" turns into other types of terrain, units turn into items, and items just get vaporised. Terrain never turns into items and units never turn into terrain.

My question was whether or not a non-explosive weapon could damage a corpse, as your comments seem to suggest one could (while also stating one couldn't, hence my confusion as to what you mean!!). 

To make sure I know what I'm talking about, I jacked up the laser rifle damage to maximum, and crippled the Sectoid corpse endurance down to one (defaults to 26 btw). After shooting a few "volunteers" I'll say my original statement was dead on: A non-explosive weapon CANNOT "kill", destroy, or otherwise damage an item in the battlescape, even if that item is a corpse produced by the death of the original target.

If you shoot a Sectoid, it takes the damage, and that's that. There's no more damage processing unless you're using an explosive of some sort, regardless as to whether the target dies and leaves a body. Terrain/tile elements have nothing to do with the matter.

- [[User:Bomb Bloke|Bomb Bloke]] 04:56, 15 October 2007 (PDT)

----
Sorry, I didn't mean to type "tiles" in my last post it should have been "objects". As for non-explosive weapons vaporizing a body, I have seen it. Unfortunately I can't seem to replicate the result I witnessed with pumping up the damage to 255. Perhaps it is a glitch of some sort in one of my saved games? More testing is in order. --[[User:Zombie|Zombie]] 07:39, 15 October 2007 (PDT)

----

Ah! I think I see the light now. I reckon what happened was you killed an alien when the item table was full (meaning a corpse couldn't be produced). Could that have been the case?

- [[User:Bomb Bloke|Bomb Bloke]] 15:38, 15 October 2007 (PDT)

----
I don't think so. Usually I test on Beginner skill level with a small compliment (8) of crack soldiers each with a Laser Rifle and Psi-Amp. That isn't  even close to filling the item table. And I would have noticed if I couldn't finish a mission due to a clogged item table too. I may have been fooling around with the Damage Modifiers though. I wonder if the additional damage produced by a greater susceptibility rating would have done this. Perhaps. I'll need to run a few tests though. --[[User:Zombie|Zombie]] 21:48, 15 October 2007 (PDT)

----
Zombie, I think it must've been something other than the gun itself.  The susceptibility, maybe.  Because otherwise, any alien killed by a Heavy Plasma other than a Muton would usually NOT leave a corpse, because of the inhuman damage dealt by the weapon.  115 damage on average, minus 30 for Sectoid health, leaves plenty more than the 26 needed to frag the corpse, and Sectoids blasted by Heavy Plasma always leave corpses in my experience.  Hope your tests can shed some more light on this!  [[User:Arrow Quivershaft|Arrow Quivershaft]] 17:31, 16 October 2007 (PDT)

== Direct Hits ==

Is there any evidence that a direct hit with an HE round is calculated any differently than dropping the round at GZ at the target's feet? In one sense you would expect a DH to apply to the facing armour rather than under armour (even though that would make a DH less effective than splash damage at GZ). Is there any chance that "weapon" damage mechanics rather than explosion mechanics are used for a DH? What about DH from Incendiary rounds?

I have a test rig set up to test this: under Armour maxed out,  Auto Cannon accuracy maxed out. I wanted to ask before running the tests.

[[User:Spike|Spike]] 00:55, 11 April 2008 (PDT)

:There's no evidence I'm aware of.  The only weapon type that has "impact" damage, as I recall, is Incendiary.  [[User:Arrow Quivershaft|Arrow Quivershaft]] 00:57, 11 April 2008 (PDT)

:: Well, from all the tests I ran back in the early days when we were still figuring the damage formula out, I did strike my test subjects head on, and that still dealt damage to the under armour. It's probably easier for the game to deal everything on a tile by tile bases and deal damage to under armour from GZ and GZ+1, then deal directional damage later on. It just recycles the same explosion mechanics that you get with grenades - only it happens earlier(I suppose you could say that if the bullet hits something solid within the tile, instead of dealing directional damage to the target, it tells the game to generate an explosion from that tile location). Besides, making the explosion directional at GZ would weaken it rather than strengthen it. 

:: As for incendiary: That's a weird kettle of fish. Impact damage is dealt to ALL units in flames, so I don't know if direction or location really matters.

:: No harm in running your tests though. We've learned a lot of things here by doing just that. -[[User:NKF|NKF]] 03:15, 11 April 2008 (PDT)

: I ran more tests on explosives than I'd like to admit over at the XcomUfo forums. There is absolutely no difference between a ranged explosive (those fired from a weapon) and grenades. All affect under armor when the explosive hits at GZ or GZ+1. (As a side note, the mechanics for the stun bomb function exactly like explosives). Like NKF mentioned, a direct hit from an incendiary round is a strange creature. Where the round hits makes no difference: all units which catch fire take 5-10 damage points while those standing in fire take 1-12 damage points per turn. --[[User:Zombie|Zombie]] 06:48, 11 April 2008 (PDT)

100 AC-HE rounds fired at a Sectoid in the open with 225 Under-armour, all rounds hit his GZ square, nearly all were Direct Hits on him - zero damage to Mr Sectoid. That seems pretty conclusive. There is no special Direct Hit effect from HE, only the standard GZ explosion effect. 

[[User:Spike|Spike]] 04:54, 12 April 2008 (PDT)

== Re: When Is a Wall not a Wall  ==

That is so not the definition of a "north" wall. Is he trying to say "the hull you see when the UFO is north of your unit"? --[[User:JellyfishGreen|JellyfishGreen]] 06:57, 7 May 2008 (PDT)

----

You haven't provided enough context.  Which sentence is wrong, and what should it say?--[[User:Ethereal Cereal|Ethereal Cereal]] 19:53, 8 May 2008 (PDT)

----

Take a look at the image in the section listed. There's a big circle around the southern side of the UFO labeled "north wall", and a big circle around the eastern side labeled "west wall". My bet is that's what he's referring to.

However, that article is referring to walls on a "tile by tile" basis. There are four sorts of tiles in UFO: Ground, Object, North Wall and West Wall.

That is to say, the tiles referred to as the "north wall" region actually DO have walls placed in the northern region <i>of the tiles themselves</i>.

And the same goes for the "west wall" tiles, although they are on the eastern side of the UFO the walls themselves are in the western region of the tiles they reside in.

So when is a wall not a wall? When it's set to an "object" (similar to the lumps of dirt you see in bases, and scenery such as trees, lamp posts, etc). Unlike regular walls, these tiles do not block explosions when hit DIRECTLY by a blast, allowing you to launch attacks through a UFO's hull even if your explosives don't actually have the strength to demolish the hull itself. Aside from the true walls circled in the graphic ALL walls that make up a UFO hull fall into this category.

If it helps, try loading up MapView and enabling the TopView. You'll be able to see which of the four tile types exist in any given map location.

It's quite true that the section is very vague and confusing (and at worst just plain wrong at times) so I've taken a stab at re-writing it.

- [[User:Bomb Bloke|Bomb Bloke]] 21:47, 9 May 2008 (PDT)

---

Like BB says, I took my designations from MapView. I too wondered "north from what?", etc. But somebody who knew the code a lot better than me (the MapView author) called it something specific. I'm just a data analyst -slash- XCOM lover and didn't want to bring up a big question of, what's the right orientation. I just used MapView's orientation. I, for one, never saw a clear north or south on anything, since it's an "isometric" view (viewing the edge on of a diamond point, at the bottom of your screen), so I was happy to have someone say something definitive.

If anyone would like to propose a standard way of what is north or south, especially relative to the screen, please say so. And you're also volunteering to clean up the whole wiki, plus petition the MapView guy to change his program. My same dilemma.

So I went with what MapView said. Thanks for making it sound reasonable, BB.

Thanks for helping to explain that, BB. The whole Explosions page could use a ton of simplification, and maybe even it should be separated into two pages - one a summary page, and the other, all the supporting details. This page grew from some basics to an overflow of information when I hit it ... and it's a repetitious mess in some places, because I didn't want to delete what had been there before. The "Blast Propagation", "Blaster Bomb Pattern", and "Blast Diameters" sections could all be folded into one, for example (I think Zombie posted some original data and I didn't want to delete it). Also the footnotes to "Blast Diameters" were obsoleted by my subsequent work, but I never got back to that table.

--[[User:MikeTheRed|MikeTheRed]] 00:28, 10 May 2008 (PDT)

: While orientation is generally a matter of preference, I've always used the overhead map for compass orientation. For example, soldiers exiting the Skyranger travel North, the nose of the Skyranger points South, while the aliens exiting a Large Scout would be traveling East. The isometric view just tilts everything by 45 degrees to the right, so North is the top right of the screen. - [[User:NKF|NKF]] 03:53, 10 May 2008 (PDT)

: The default orientation for overhead maps is that north is going up.  Confirming this, the UNITREF orientation byte is 0 for North, and increases going clockwise.  [E.g., East is 2.]  MapView does correctly identify north and west walls -- as <b>map components</b> whose usage was optimized for thin walls.  As outer UFO walls are thick, the game is forced by data design into using true north walls for the southern boundary of UFOs, and true west walls for the eastern boundary of UFOs. - [[User:Zaimoni|Zaimoni]] 9:28, 10 May 2008 (CDT)

: Makes sense to me. In order to help readers a little more, the notes for that one picture could re-emphasize what MapView says. Or, of course, the picture (and text) could be changed to simply say, only south and east outer UFO walls are true ones. Plus supply a footnote re: MapView's nomenclature, so folks aren't confused if they use it. I dunno... something like that, shrug. -[[User:MikeTheRed|MikeTheRed]] 11:51, 14 May 2008 (PDT)

== HE Propagation Edge ==
[[Image:AC-HE Blast.png|thumb|right|AC-HE/HC-HE blast pattern in desert.]]
A question for our Explosive experts. The tables in this article show AC-HE (44 HE) having a blast radius of 3, and an Edge strength of 14. The Discrepancy shown is "0", so this is the "expected" radius. But why is there not an effect at radius=4, with HE strength=4 (2-6 dmg)? Of course this would have no impact on terrain and only a small impact on personnel. Are we just ignoring this radius=4 effect, or is it actually "crimped" down to zero by the game engine? The issue came up because I was comparing the gross total damage of AC-HE and HC-HE. HC-HE is stated as being "crimped" to radius=3, with Edge HE=22. Whereas you would expect it to propagate out to radius=4, HE=12 and radius=5, HE=2 (dmg 1-3, almost no chance of doing any damage to a unit). 

Could someone explain this in a bit more detail? Thanks! [[User:Spike|Spike]] 09:14, 28 February 2009 (CST)

:First see picture to the right. This is the actual pattern produced by the Auto-Cannon's HE round (44HE) and the Heavy Cannon's round (HE52). (Technically, the pattern is the same for any projectile explosion between 40-59 inclusive). Any tiles which are damaged in my pic are also the tiles where units take damage too. My guess for the reason why the blast doesn't propagate further than what is expected is due to the [[Explosions#Distance_from_Ground_Zero|equation used]]. The walking TU method models how explosions are applied to units pretty close. Unfortunately, how the game models something is not always what is expected. --[[User:Zombie|Zombie]] 10:27, 28 February 2009 (CST)

Thanks, that helped a lot, especially the reference to the walking TU rule - I had gotten that wrong. In fact I have also got it wrong in my firepower models, because I thought that for 4-square units, the 4th square would be at GZ+2, but in fact all 3 extra squares are at GZ+1. 

I'm still not clear why AC-HE is crimped at radius=3, HE=14, and does not propagate out to radius=4, HE=4. Does your desert test terrain have 0 HE block, 1 strength? In other words does it show any square where a unit would take any damage? [[User:Spike|Spike]] 10:59, 28 February 2009 (CST)

:Well, I'm using unaltered desert terrain which has 0 HEB, 5 "Base" Armor and 25 "Death" Armor. Suppose I could use Bomb Bloke's numerical testing tileset for this as his armor strengths are all custom edited and have numbers to show tile damage. But, I might just use my flat terrain map pack and edit the MCD values so that everything is 0. Still, terrain shouldn't have any effect on damage to units in an explosion if the ground tiles have 0 HEB as this lets the explosion go past unhindered (except for distance of course). Armor only determines when a tile sprite should change it's appearance. Not too sure of the exact reason why the explosion doesn't go out further than what is expected though. My only guess is that integer rounding plays a pretty big role here and it may be negating the propagation when the damage gets below 10 Average (or 5 for the min?). Sorry I can't be more helpful at this point. --[[User:Zombie|Zombie]] 11:28, 28 February 2009 (CST)


OK that makes sense. With the Desert Base armour at 5, any residual HE effect below 10 would have no effect on the terrain. It does suggest that the HC-HE radius 4, HE=12 is definitely crimped off, since HE=12 would be enough to damage the tiles at radius=4. But I suppose it's ''possible'' that the AC-HE radius 4, HE=4, is still present but undetected on the desert tile map. Since it does a maximum of 6 damage, you would very rarely see damage even to a unit. But unless the AC-HE is less constrained than the HC-HE, which would make no sense at all, there is no effect at all from AC-HE at radius=4 - not even against units. [[User:Spike|Spike]] 11:37, 28 February 2009 (CST)

:: The capped explosion radii is even more apparent for Blaster Bombs. Blaster bombs have an edge strength of 90, so very often someone RIGHT on the border of the blast gets killed, while someone just outside the blast radius is totally unharmed. ; Actually, I did some experiments here, and I found that blast radius seems to be, for most weapons, ((Power/10)/2)+1 = (Power/20)+1. (rounded down) (radii  Why /10 and /2? The /10 comes in because explosive power is reduced by 10 for each square distance from the epicentre. The +1 comes in because even an explosion with power of 0 must occur in the tile it hits, giving a minimum radius of 1. The /2 comes in because terrain takes consistent 50% damage from explosives. Unfortunatly, I suspect there was some programming formula logic inconsistencies here, because the 10 point distance reduction is applied BEFORE the 50% damage reduction resulting in "edge"... otherwise, we would have a smooth terrain destruction graph. On the other hand, currently terrain damage is consistent with unit damage. Hmm... So, guys, no, basically nothing to do with integer rounding. (I know all this cause I love making the boom boom booms.) Try editting different weapon power with an editor and you will get interesting results. [[User:Jasonred|Jasonred]] 14:48, 28 February 2009 (CST)

We already did tests on this quite extensively. The equation can be found [[Explosions#Hacking_HE_versus_Blast_Diameters|here]], and integer rounding <b>does</b> play a role. --[[User:Zombie|Zombie]] 15:29, 28 February 2009 (CST)