lynx

I've looked it up and the overhead on subviews is actually 1400 commits at the moment. If anyone was wondering why the odd codename, there's a lengthier explanation here: https://sorbusaria.wordpress.com/2017/11/28/anniversary-gemrb-release/

You're right, either HPPercent* or See shouldn't match for dead actors. The loot dumping would happen in the next tick IIRC, so that death script is problematic. The second case looks like something managed to interrupt him, but it needs further inspection. Is that behaviour consistently reproducible?

Exe mod, I think. GemRB displays them as green/red.

GemRB is a portable open-source implementation of Bioware's Infinity Engine. It was written to support pseudo-3D role playing games based on the Dungeons & Dragons ruleset (Baldur's Gate and Icewind Dale series, Planescape: Torment). It has been a while since our last release, but plenty of goodies have landed since. This release brings a slew of minor features and fixes, of which a select few are mentioned in the changelog. Of those that aren't, a last minute AttackReevaluate fix stands out, since that action is used all the time in AI scripts. For the details, try GemRB yourself or check the linked changelog for some highlights.

The GemRB team is proud to announce a new release. It has been a while since our last release, but plenty of goodies have landed since. This release brings a slew of minor features and fixes, of which a select few are mentioned below. Of those that aren't, a last minute AttackReevaluate fix stands out, since that action is used all the time in AI scripts. Wait a minute, but where's the promised return of biography editing, you may exclaim. That branch is very much still alive and being worked on, with more than 1000 extra changes. The rewrite just spiralled out of control. Interested users can take a peek at the subviews branch. You can get the sources and packages from here. Full changelog digest: GemRB v0.8.5 (2017-11-26): New features: - SDL2 resolution-independent window scaling, environmental audio, IRIX compatibility - regeneration during sleep, time dilation - iwd2 ability stacking, iwd ZZ* weapon bonuses - portrait.2da handling, avatar shadows, dialogF.tlk handling Improved features: - infravision - better compatibility with bgt, 10pp, ia, kelsey, dr - sdl2 input, mouse scrolling, opengl driver - iwd2 cg&lu, item (un)usability, avatar sizes, permanent clabs - ambients, verbal constants, random walk, walk sounds - bugfixes Packagers note that a few CMake options have been tweaked and are now togglable as true CMake options.

If you have the ogg vorbis and vorbis file library, you just have to rebuild gemrb for the plugins to support these files. Then they can be used everywhere. You'd only need to modify game files that also specify the file extension — the other cases would just work if you removed the original copies (otherwise they would likely get found first). BUT, looking at our build bot logs, the mac builds do have ogg vorbis support: https://travis-ci.org/gemrb/gemrb/jobs/306231521#L3184 So you can grab one of those dev snapshots and just roll with it. Current master is very stable, we're right before a new release.

But you see the point parameter isn't set to anything specific and the two locations are way too far apart for this just to somehow be a result of searchmap jumping. Argent is spot on in the sense that the launch point is what does it. And the golem one has it elsewhere too.

Of course, but that's elsewhere.

I am rechecking oddities I found during a play through and one of them were the wolfweres in old red's dungeon, just before you get to the area with the elemental well. I remember they were at the end of the corridor leading to it. LP footage confirms: https://youtu.be/6WaaFPDT8i0?t=4m13s Looking at the area (ar1202), there are no spawn points there, but there is an undetectable trap. However, even in an unmodded game, that trap runs spwngol.bcs, which gives you golems. That's what I saw ingame too. So two things are not clear to me: 1. Why are there wolfweres there? 2. Why are they there immediately, not via the usual trap triggering mechanism? They are not embedded in the area. Neither the area script nor baldur does anything special. The rest interrupts give you undead and they'd be nearby anyway. I am running the gog version, which is different from the video, however it seems unlikely they'd go changing something like this. WTF am I missing?

Just verified it's not this fixpack's fault, so let's just forget about it.

If the bit is set, the creature name is used for the scriptname instead. That matches my experience, which is good, but doesn't explain anything — who embedded the copies with the same name? I've loaded a vanilla bg2 and there udgith03..05 are properly named, so there was no bug.

you could adapt the bam composer in iesh or it's base format handler: https://github.com/gemrb/iesh/blob/master/infinity/bam_composer.py https://github.com/gemrb/iesh/blob/master/infinity/formats/bam.py it's already using PIL (python imaging library), so it should be that hard to extend it to do arbitrary scaling.

heh, thanks!

The area has no overrides for them in this install (says DLTCEP ... wait, it only has fields for dialog/scripts, grr). I have the are in override, but weidu --change-log override/ar2400.are says no mods are affecting it, which is of course wrong, as at least fixpack does.

What do you mean by scaling?

During my playthrough (with v10) I noticed that some of Simyaz's companions aren't affected by the cutscene scripts. Looking at the detailed changelog for the fixpack, it doesn't appear to be its doing (other mods are npcs + item randomizer + dr + alternatives). They're not affected, since both have the same scripting name / death var ("githyanki" iirc). CREature-wise, there is only udgith02.cre. Extracting all the scripts and dialogs shows the problem, similarly for Kruin later: $ grep -rin gith0 bafs/ bafs/ar2100.baf:209: Dead("udgith02") // ~Githyanki~ bafs/ar2100.baf:210: Dead("udgith03") // udgith03 bafs/ar2100.baf:211: Dead("udgith04") // udgith04 bafs/ar2400.baf:125: Dead("udgith02") // ~Githyanki~ bafs/ar2400.baf:126: Dead("udgith03") // udgith03 bafs/ar2400.baf:127: Dead("udgith04") // udgith04 bafs/ar2400.baf:143: Dead("udgith02") // ~Githyanki~ bafs/ar2400.baf:144: Dead("udgith03") // udgith03 bafs/ar2400.baf:145: Dead("udgith04") // udgith04 bafs/cut44j.baf:15: ActionOverride("udgith02",JumpToPoint([2043.1864])) bafs/cut44j.baf:16: ActionOverride("udgith03",JumpToPoint([2105.1871])) bafs/cut44j.baf:17: ActionOverride("udgith04",JumpToPoint([2016.1881])) bafs/cut44j.baf:18: ActionOverride("udgith05",JumpToPoint([2016.1881])) bafs/cut44k.baf:8: ActionOverride("udgith02",JumpToPoint([1963.2057])) bafs/cut44k.baf:9: ActionOverride("udgith03",JumpToPoint([1987.2101])) bafs/cut44k.baf:10: ActionOverride("udgith04",JumpToPoint([2057.2140])) bafs/cut44k.baf:11: ActionOverride("udgith05",JumpToPoint([1872.2016])) bafs/cut44l.baf:8: ActionOverride("udgith02",JumpToPoint([2679.1339])) bafs/cut44l.baf:9: ActionOverride("udgith03",JumpToPoint([2730.1354])) bafs/cut44l.baf:10: ActionOverride("udgith04",JumpToPoint([2721.1298])) bafs/cut44l.baf:11: ActionOverride("udgith05",Kill(Myself)) bafs/udgith.baf:78: Name("udgith03",Myself) bafs/udgith.baf:93: Name("udgith04",Myself) bafs/udgith.baf:108: Name("udgith02",Myself) $ grep -rin gith0 dlgs/ dlgs/kruin.d:92:CreateCreatureOffscreen("gith02",0) dlgs/kruin.d:93:CreateCreatureOffscreen("gith02",0) dlgs/kruin.d:94:CreateCreatureOffscreen("gith03",0) dlgs/kruin.d:95:CreateCreatureOffscreen("gith04",0) dlgs/kruin.d:96:CreateCreatureOffscreen("gith04",0) dlgs/kruin.d:97:CreateCreatureOffscreen("gith05",0) dlgs/kruin.d:98:CreateCreatureOffscreen("gith06",0) dlgs/udgith02.d:2:// argument : UDGITH02.DLG dlgs/udgith02.d:8:BEGIN ~UDGITH02~ 1 // non-zero flags may indicate non-pausing dialogue Can someone verify? I don't remember having this problem before and it seems like something that would have been noticed.

if it's not in the archive subforum, my guess would be no, we don't have the original research.

Thanks, so that's a 72 radius circle.

my guess would be one of the big ones is a timing mode switch. 112 didn't change with the ees, it's even the same in all engines. So it'd be easy to check the actual behaviour rather than argue over labels in code. One thing I'm interested in is what range the Help action uses.

Thank you Roxanne, you're the reverse of a child of Bhaal.

DLTCEP

unless you want to ping pong around, I suggest you create a map anew, just of the relevant planescape area.

iesdp.gibberlings3.net and gibberlings3.net/iesdp are baked in many places, so it would still be helpful. But now that you have access to ftp, you could also just delete the old iesdp and add a custom 404 page with another redirect (if supported). Or if there's cron, even better ...

that's odd, this should work on the web root level. Anyway, at least IESDP can be fixed via html. Replace its index.htm with this one: <html> <head> <title></title> <meta http-equiv="refresh" content="0;url=https://gibberlings3.github.io/iesdp/" /> <style></style> </head> <body> </body> </html>

I don't know what the base is (RewriteBase), but something like this should do: RewriteEngine on RewriteCond %{HTTP_HOST} ^forums.gibberlings3.net RewriteCond %{REQUEST_FILENAME} !-f RewriteCond %{REQUEST_FILENAME} !-d RewriteRule ^(.*)$ http://gibberlings3.net/forums/$1 [R=301,L] test case: http://forums.gibberlings3.net/index.php?showtopic=29062 http://gibberlings3.net/forums/index.php?showtopic=29062