Celtic Minstrel

Regarding webs, I believe only NPCs of the "bug" type are immune to their effects. This immunity applies to the barrier (they simply pass through without affecting it), so if something were to directly apply the status effect, I think they would be affected. There's most likely no way to actually do that though.

I just want to note two things here: that the Roman pilum is rather different from your typical javelin, and that an Ermarian stone doesn't necessarily have to be 14 pounds.

I think the "target button" idea could be implemented with just special encounters in OpenBoE, actually. I added a way to trigger special encounters by casting a spell on a location, and while I'm not sure if that currently works for missiles, it probably wouldn't be hard to make it work. Of course, that all depends on me finding the time…

Since this thread was originally intended for me to post new releases, perhaps we can move all the recent discussion to a new (but still locked) thread and unlock this again? I'd do it myself, but I'm just not sure what I'd call that new thread… or just where the split should start… EDIT: I've gone ahead and done it.

If you wanted to build it yourself, naturally it's possible. It's not easy though. There are some basic instructions in the readme on GitHub, but I'm not sure how complete they are.

There will be, whenever I get around to making a new build. I think I have several unreleased fixes already on the GitHub, but I haven't found found the time to work on it much lately.

I think I remember doing it over and over and it was eventually cleared out, but I don't remember how many times I had to defeat them. It was a long time ago though so I could be imagining this.

I fixed a bunch of Lyceum/Alcritas-related links, but the archive of Drizzt's website is quite broken. It looks like it was downloaded through a browser using the "save whole web page" feature, because there are a bunch of "this_web_page_files" folders which oftentimes contain duplicate content. There are also a whole bunch of missing images, and the zip files seem to be absent too.

The only other place I can think of is Truesite4Blades, but I'm pretty sure the OpenBoE archive is directly derived from that, so there's probably no point checking. (It is currently archived at OpenBoE.com though, if you want to take a look. I think it's untouched from the archive Jewels sent me save for updating broken internal links, though I might've missed some.)

Yeah, I thought as much. Sorry I couldn't help, then. Presumably the official Spiderweb archive is even less up-to-date, right?

I assume the versions on OpenBoE.com aren't more up-to-date than what you have?

I think counters are transparent, so they wouldn't affect line of sight. If you had a line of pillars though, then yes. It just uses a naive line-drawing algorithm to figure out which hexes affect your vision. The algorithm described on that page only works for lines with small slope (less than 1), but Blades of Exile swaps the role of x and y if the line has a large slope (greater than 1), so it still produces a reasonable result.

Nice detective work! From this it seems fairly likely that it was indeed a mistake.

It's a degree of obscurity based on a sum of opacity over the path between the two tiles. A fully-opaque tile has opacity 5, so that's why there's a comparison to 5 in can_see. It checks a total of max(dx, dy) tiles to determine the final obscurity. Opacity is normally based on the terrain (5 if it blocks sight, 1 if it's obstructed but you can shoot through it, 0 if it blocks nothing), unless there's a fire or force barrier there, in which case it's 5. A web on the space increases opacity by 2, and a crate, barrel, or town special spot increases it by 1. When in combat, lava and pits are also forced to opacity level 5. This time I looked at OBoE rather than the original source, but I don't think there are any appreciable changes in this area (if there are, it's a bug).

That argument does make sense, but… this is the code for shockwave in Blades of Exile (original version, not the OBoE source): void do_shockwave(location target) { short i; start_missile_anim(); for (i = 0; i < 6; i++) if ((dist(target,pc_pos[i]) > 0) && (dist(target,pc_pos[i]) < 11) && (adven[i].main_status == 1)) damage_pc(i, get_ran(2 + dist(target,pc_pos[i]) / 2, 1, 6), 4,-1); for (i = 0; i < T_M; i++) if ((c_town.monst.dudes[i].active != 0) && (dist(target,c_town.monst.dudes[i].m_loc) > 0) && (dist(target,c_town.monst.dudes[i].m_loc) < 11) && (can_see(target,c_town.monst.dudes[i].m_loc,0) < 5)) damage_monst(i, current_pc, get_ran(2 + dist(target,c_town.monst.dudes[i].m_loc) / 2 , 1, 6), 0, 4); do_explosion_anim(5,0); end_missile_anim(); handle_marked_damage(); } As you can see, there's a clear call to the "can_see" function. It seems odd that he would have inserted that by mistake.

I'm pretty sure he doesn't read the forums very often, so if you want his opinion that much, you'd probably have to ask by email or something. Given that it changed between E2 and E3, and the manual never says it's supposed to pass through walls, I'm currently leaning towards it being a bug that he fixed in E3.

Yeah, the link from the main page is outdated, I'll fix it. As for ADoS's issue, it looks like you need to omit the www. I'll look into fixing that too. EDIT: I was wrong about the www, looks like the webserver had crashed and my browser was just showing the cached page. Fixed now.

Assuming you mean for a custom scenario, that's documented. Note that that's the documentation for the experimental version of the game; there are a few differences if you're using the original BoE. Most notably, the ID of an item graphic is different, instead of adding 1000 you add something else (I don't remember what, though). Also, the original BoE doesn't support custom missiles or explosions. Swapping out existing graphics totally works, but I wouldn't recommend doing it for any of the types of graphics documented on that page. It's fine to swap out user interface graphics though.

So now the question is… should Shockwave pass through walls, or is this a bug that was fixed in Exile 3?

I did implement this. I'm pretty sure it works perfectly (unlike some other parts of OpenBoE).

If I recall correctly, monsters can drop items with that treasure value or a lower treasure value. The actual place_treasure function is indeed quite complicated… it looks like treas_char holds indices into min_chart and max_mult, which appears to limit the minimum and maximum value of the dropped items. The amt variable is how much gold drops, and id_odds is indexed by your Identify skill to determine the chance of the item being identified… the level parameter of the function is actually half the level of the monster that died, and the loot parameter is the monster's "treasure" value. It also looks like the rightmost column of treas_chart and treas_odds are never used, since both loot and j are in the range 0..4.

That's not CBoE, it's OBoE. I realize this is confusing, but CBoE refers to an older version that only works on Windows. I'm mildly amused that you're explaining to me the file formats that I designed.

OpenBoE.com is back up now. I have no idea about the magnet link, though - how does that even work? I do still have it set to seeding in my torrent app. Just saying, but the preset party, and any other newly-made party, needs to be able to play through the tutorial. This sounds like you're talking about OBoE, but I think ADoS was asking about CBoE? If this is about CBoE, it probably only uses a different format for the saved games. I'm pretty sure it can run legacy custom scenarios. Scenarios created with CBoE still use the .exs extension, so I'm pretty sure the format is 90% identical and only adds some new allowed values in some places.

This probably isn't what you want to hear, but I suspect there's just no way to make this work unfortunately. The relocate_character call is documented under cutscene calls, so presumably Jeff never thought that someone might used it to implement teleportation.

Ah nice; and good to know you figured out a solution. I believe 68k code on MacOS was stored in the resource fork, right? Is that what you meant by using its own executable format, or is there something unusual about the code itself? I'm pretty sure there's actually a disassembler that runs right in ResEdit… but I could be wrong about either of those assertions.