Blondbeard
Level: Smitemaster

Rank Points: 1510
Registered: 03-31-2005
IP: Logged

Maybe this is no bug, but are you really supposed to survive when you hit a bomb? Because if you´re standing on a door you do. No matter where you stand I think that hitting a bomb with a sword should be pretty stupid.

AlefBet
Level: Smitemaster
Rank Points: 979
Registered: 07-16-2003
IP: Logged

You shouldn't survive if you hit a bomb while standing on an open door (barring force arrows), but since a bomb explosion stops at walls and closed doors, it seems reasonable to me that you wouldn't be in the blast. In fact, I know that some of the puzzles in JtRH rely on this ability.

____________________________
I was charged with conspiracy to commit jay-walking, and accessory to changing lanes without signaling after the fact .

++Adam H. Peterson

Blondbeard
Level: Smitemaster

Rank Points: 1510
Registered: 03-31-2005
IP: Logged

Bomb exposions stops at walls. That´s alright, but I still think that hitting a bomb with your sword should resukt in death. Even if you stand on a closed door. If there are puzzels in JTRH that relies on blasting bombs by hitting them with the sword rather than lighting a fuse I don´t remember them.

Stefan
Level: Smitemaster

Rank Points: 2119
Registered: 05-25-2004
IP: Logged

Backtracking through L21:2N2E relies on hitting a bomb with your sword while standing on a door.

Anyway, I think the report should be specified a little, because it was probably a cause of this discovery.

Due to the order things happen in, you will survive hitting a bomb while you stand on an open door provided that the bomb blast closes the door you're standing on*. IMO this is rather unintuitive and should perhaps be changed.

*Edit: I tested a little bit further. It requires a pretty specific configuration to work, so it's not very severe.

[Edited by Stefan at Local Time:05-24-2005 at 11:12 PM]

____________________________
0.099³

schep
Level: Smitemaster

Rank Points: 865
Registered: 03-01-2005
IP: Logged

File: Bomb Test.hold (795 bytes)
Downloaded 63 times.
License: Other
From: Unspecified

Interesting bug. Here's a room that demonstrates how unintuitive the current behavior is. A door that closes or opens during an explosion has its effects executed in the engine when some squares have already been dealt with, and it affects the squares that haven't yet been exploded.

It's not entirely clear what should happen in this situation, but I think it ought as much as possible preserve the illusions that the explosion happens in all directions at once and that each individual connected door opens and closes as a unit.

And I don't have an answer, but two ideas come to mind:
1. Implement things as though the explosion spreads more quickly than doors can begin to move and lasts longer than they take to finish moving. So when a bomb blows up, the explosion first kills and lights what it can, then any orbs activated move their doors, and if any doors open up new regions in the bomb's range, more things may get burned. This could get complicated, especially if an opening door exposes another orb. And if a door around an orb closes, then opens during the same explosion, should the orb activate twice? (Probably not, at least to avoid the halting problem.)

2. Implement things as though the explosion is exhausted before doors begin to move. So yes, Beethro can hit the bomb while standing on a closed door and survive even if a nearby orb drops him in among the hot embers. But it avoids the problems of the other idea, and doesn't sound too hard to implement (queue orb activations to be dealt with after all the explosions have been calculated, right?)

And then there's the related question of when one explosion activates two orbs which attempt to open and close the same door. You might try for sorting by distance to the bomb somewhere in there, but there will always be some reliance on direction. But this situation is one I wouldn't expect to be predictable in the first place.

AlefBet
Level: Smitemaster
Rank Points: 979
Registered: 07-16-2003
IP: Logged

Surviving a blast while standing on a door is useful for puzzles and pretty easy to understand. However, it's not very intuitive to have doors toggling in the middle of an explosion. So Mike is busy refining those semantics and the behavior will probably be more straightforward and intuitive in the next patch. (This may break some demos, but probably not that many.)

____________________________
I was charged with conspiracy to commit jay-walking, and accessory to changing lanes without signaling after the fact .

++Adam H. Peterson

VortexSurfer
Level: Delver

Rank Points: 90
Registered: 04-04-2004
IP: Logged

AlefBet wrote:
(This may break some demos, but probably not that many.)

Hey, it may even fix some, mine for example

mrimer
Level: Legendary Smitemaster

Rank Points: 5466
Registered: 02-04-2003
IP: Logged

schep wrote:
Interesting bug. Here's a room that demonstrates how unintuitive the current behavior is. A door that closes or opens during an explosion has its effects executed in the engine when some squares have already been dealt with, and it affects the squares that haven't yet been exploded.

It's not entirely clear what should happen in this situation, but I think it ought as much as possible preserve the illusions that the explosion happens in all directions at once and that each individual connected door opens and closes as a unit.

I've improved the semantic of how bomb explosions expand in 2.0.8 -- an explosion will completely expand (including chain reactions) before any doors affected by struck orbs change state. Also, orbs hit multiple times during a chain reaction will only activate once. Multiple bombs lit by fuses that detonate on the same turn will similarly act as if they were part of the same chain reaction where all explosions happen simultaneously, striking each orb at most once and activating doors only after the explosions have completed.