Version 5 Bug Reports

[CamDawg]

List 'em here.

 

I've run into my first bug. I'm unable to enter the orc cave (ar1201). Entry point is unreachable. Anyone else encounter this?

 

Edit: I must correct myself. The entrance is accessible, but I had to move into the cave opening before I could reach the trigger. Difficult entry, but not impossible.

[CamDawg]

I've run into my first bug. I'm unable to enter the orc cave (ar1201). Entry point is unreachable. Anyone else encounter this?

 

Edit: I must correct myself. The entrance is accessible, but I had to move into the cave opening before I could reach the trigger. Difficult entry, but not impossible.

 

 

Original IWD up top, IWDTutu below. Looks like the travel trigger area is the same, but you can not walk to the one in IWDTutu. The area height/shadow/etc. BMPs match as well.

[CamDawg]

Alright, had a closer look. It looks like the search map is not quite covering all the way to the entrance. You can enter the cave by going to the left and then straight up, but trying to click the travel area from anywhere else will move the character to the cave entrance and then stop short. The attached search map seems to be more robust. I'll eventually package this in to some kind of post-conversion fix pack a la Tutufix but for now just drop this in your override.

 

Let's really hope this isn't a systematic issue--guys and gals, pay close attention to your pathfinding. If you can't reliably hit a travel trigger let us know.

 

AR1200SR.BMP

[DavidW]

There are some systematic issues with entrances and exits, actually. More accurately: there were such issues, though I'd hoped they were largely solved.

 

Fairly systematically, IWD exits are untraversable if the search maps are just carried across to BG2. The engines seem to interpret them differently - basically, the point that BG2's engine wants you to move to in order to actually exit is untraversable. This has been known since my original version.

 

I tried to solve this in v1-4 by applying a heuristic algorithm to resize the exit, but it didn't work so well. In v5, I have code that makes the entire space covered by the exit party-traversable. So far (as of end chapter 1, which is as far as I've tested), that makes all the exits okay, though occasional ones are a little awkward.

 

Unless anyone has a really brilliant idea for solving it more systematically, I think the right solution is probably to make further ad hoc fixes like Cam's on exits that are a real bitch.

 

(Incidentally, this is one of an annoyingly large number of small game-engine differences. The task is/was much harder intrinsically than TUTU/BGT, because the BG1 engine is really just a subengine of the BG2 engine, whereas the IWD and BG2 engines had about a year of development differences. Of course, in another sense it was/is much easier than TUTU, simply because of how much the community learned from TUTU itself.)

[Jarno Mikkola]

Hmm, I was wondering, I don't have issues like this cause my characters(all 6) usually pack themselves tightly around each others, and "push each others" inside the cave when they try to go to the point of the transition(the green movement target) and then stop moving(called that: land), a small picture:

Now, if others don't have that or none of their characters can land on the exact spot when all the characters are in a small area(about 10x10 feet area around the point of transition) because they are all trying to get there at the same time, it can be a problem, that is easy to solve, just try again, and move the characters with the right click if needed but do not try to left click cause the characters that have landed to the area then try to move to the area again causing further problems.

 

It could be that the "push each others" is enabled when the party is not in combat, or it has option in one of the that enables it... and sometime my characters can pass each others circles too, so I really don't understand why people have issues like this.

[CamDawg]

Fairly systematically, IWD exits are untraversable if the search maps are just carried across to BG2. The engines seem to interpret them differently - basically, the point that BG2's engine wants you to move to in order to actually exit is untraversable. This has been known since my original version.

 

I tried to solve this in v1-4 by applying a heuristic algorithm to resize the exit, but it didn't work so well. In v5, I have code that makes the entire space covered by the exit party-traversable. So far (as of end chapter 1, which is as far as I've tested), that makes all the exits okay, though occasional ones are a little awkward.

Hmm. Well, in this case the original and converted search maps were identical, as were the travel triggers. Is it possible it was skipped?

[DavidW]

Fairly systematically, IWD exits are untraversable if the search maps are just carried across to BG2. The engines seem to interpret them differently - basically, the point that BG2's engine wants you to move to in order to actually exit is untraversable. This has been known since my original version.

 

I tried to solve this in v1-4 by applying a heuristic algorithm to resize the exit, but it didn't work so well. In v5, I have code that makes the entire space covered by the exit party-traversable. So far (as of end chapter 1, which is as far as I've tested), that makes all the exits okay, though occasional ones are a little awkward.

Hmm. Well, in this case the original and converted search maps were identical, as were the travel triggers. Is it possible it was skipped?

 

I'd be delighted if so... will check. (I'm at work right now, so it might take a bit of time)

[CamDawg]

No worries. Manually tweaking the search map isn't difficult, but if we have to do a few hundred... :/

[DavidW]

Aha. Yes, the orc cave (and also the Easthaven finale areas) are missed out.

 

To see if this sorts the problem (comment mostly aimed at Cam): edit iwd_in_bg2.tp2, component 155. After a couple of macros are defined, there's a COPY_EXISTING_REGEXP_GLOB block (around line 880) in which I've randomly left out ar11.* and ar12.*. Add them in, and delete all the other area references (not that it really matters if they're left in, but it'll run faster), then try installing component 155. (Copy it to a hotfix tp2 if you'd rather.)

 

(component 155, like most components, isn't run locally when you install, I just package the result.)

[CamDawg]

Aha. Yes, the orc cave (and also the Easthaven finale areas) are missed out.

 

To see if this sorts the problem (comment mostly aimed at Cam): edit iwd_in_bg2.tp2, component 155. After a couple of macros are defined, there's a COPY_EXISTING_REGEXP_GLOB block (around line 880) in which I've randomly left out ar11.* and ar12.*. Add them in, and delete all the other area references (not that it really matters if they're left in, but it'll run faster), then try installing component 155. (Copy it to a hotfix tp2 if you'd rather.)

 

(component 155, like most components, isn't run locally when you install, I just package the result.)

Thanks, I'll check when I get home (my turn to be at work). Assuming this works I'll probably throw together a fix tp2 (and just add to it as we go along) now that we have some actual code instead of a lone file.

[DavidW]

Aha. Yes, the orc cave (and also the Easthaven finale areas) are missed out.

 

To see if this sorts the problem (comment mostly aimed at Cam): edit iwd_in_bg2.tp2, component 155. After a couple of macros are defined, there's a COPY_EXISTING_REGEXP_GLOB block (around line 880) in which I've randomly left out ar11.* and ar12.*. Add them in, and delete all the other area references (not that it really matters if they're left in, but it'll run faster), then try installing component 155. (Copy it to a hotfix tp2 if you'd rather.)

 

(component 155, like most components, isn't run locally when you install, I just package the result.)

Thanks, I'll check when I get home (my turn to be at work). Assuming this works I'll probably throw together a fix tp2 (and just add to it as we go along) now that we have some actual code instead of a lone file.

 

Great. (In due course this fix, at least, can be wrapped back into the original code, but it takes about three hours to build the mod from scratch, and then it's 20MB to download, so I don't plan on doing it every time we find a problem.)

[Jarno Mikkola]

Great. (In due course this fix, at least, can be wrapped back into the original code, but it takes about three hours to build the mod from scratch, and then it's 20MB to download, so I don't plan on doing it every time we find a problem.)

Well, you could make a .tp2 file, a .bat file and package them as "iwd_in_bg2-v5a.zip" and be done with that. 20KB's at most.

[DavidW]

Great. (In due course this fix, at least, can be wrapped back into the original code, but it takes about three hours to build the mod from scratch, and then it's 20MB to download, so I don't plan on doing it every time we find a problem.)

Well, you could make a .tp2 file, a .bat file and package them as "iwd_in_bg2-v5a.zip" and be done with that. 20KB's at most.

 

That's not how the mod works. (Most of) the tp2 file doesn't get executed locally, it gets executed on my machine. This is Easytutu, not original Tutu.

[CamDawg]

Worked a treat. The post-install fixpack has been added to DavidW's main v5 beta release thread and will generally live in that thread, though we should continue to keep bug reports and fixes in this thread.

[grogerson]

I've cleared Kresselack's Tomb, level 2. Here's what I've found so far.

 

Created a druid, had 9hp v. 8hp.

 

After Arundel sent me to the Vale, I found I couldn't reach it if I left Kuldahar on the left side. When I return, I'll check if it remains unreachable from the left. I'll also rest there and see if I get the same bug I did with v4...

 

When yeti die, I get the sound of a mountain lion, and the dead body is likewise that of a mountain lion.

 

All traps work. No CTDs to report.