1155 replies to this topic

[Argent77]

Is this bit fully supported? I've tested it in both BG2 and BG2EE, and my half-orc character could equip items with enabled bit 31 without problems.
 

[-Hurricane-]

The only game I know that currently supports it is BG:EE patched to the latest version v1.3.2053. From the patch notes:

Modder Features
- Items can now be flagged as unusable by Half-Orcs

I've checked and indeed this didn't work yet in v1.2 but it does now in v1.3.

[The Imp]

I've checked and indeed this didn't work yet in v1.2 but it does now in v1.3.

Any idea how they fixed it ? So The BG2 could use that fix... say, in the (G3)BG2Fixpack.

[-Hurricane-]

Sorry, only Beamdog knows.

[Argent77]

Update: NearInfinity v1.35.0-snapshot-20141017

Changes:

• Added support for case-sensitive filesystems (as found on Linux and derivates). This feature supersedes the (already dated) CI versions of NearInfinity and removes the requirement of tolowered game installations at the same time.
• Added a CHU resource viewer
• Added a proper Modron Maze structure to PST's GAM resources
• Several minor fixes and updates

[CamDawg]

Dammit Argent, stop being awesome.

[Sam.]

• Added support for case-sensitive filesystems (as found on Linux and derivates). This feature supersedes the (already dated) CI versions of NearInfinity and removes the requirement of tolowered game installations at the same time.

1^st post has been edited to reflect this .

[-Hurricane-]

Thanks for the update!

For the Poison opcode (25), NearInfinity says the parameter #2 "Poison type" can be one of four values (0 through 3). The label for value=3 says "Amount damage per round". I think this should be "1 damage per 'Amount' seconds" because that is what I am experiencing in BG:EE

 

I'm using both IESDP and DLTCEP as sources for effect opcode descriptions which are not always in sync

Huh, I figured out why NI had the "Amount damage per round" label before. It was the correct description for vanilla BG1. The opcode parameter is different in BG1 vs. BG2. Sorry for not checking before. I've tested the behavior of Eldoth's arrows in vanilla BG1 and indeed it's 2 damage per round in that game (doled out as 1 dmg every half round), instead of 1 dmg every 2 seconds as in BG1:EE. So I guess either the old or the new label should be displayed depending on which game is loaded.

[Argent77]

Poison type 3 in BG1 is pretty weird. It's more like "1 damage per amount+1 seconds". Even BG1EE uses a strange pattern to calculate the poison damage. It's close to "1 damage per amount seconds", but not quite.

[-Hurricane-]

If you say so.

It's more like "1 damage per amount+1 seconds". Even BG1EE uses a strange pattern to calculate the poison damage.

Hm, I didn't know that. Is there a tool or code available that shows how this works? In other words, can you tell me where you know all this from, if there is any source other than in-game testing?

[Argent77]

Hm, I didn't know that. Is there a tool or code available that shows how this works? In other words, can you tell me where you know all this from, if there is any source other than in-game testing?

Only from pure observation. I'm testing with customized spells however, to have more control over the specific spell effect.

[-Hurricane-]

Thanks.

[Mad Mate]

Argent77, I have one bug and one feature request:

 

First bug: When I try to save result.txt from triggers & actions check in dialogues it always saves just first few lines, never whole list.

 

And feature request, when checker encounters CreateVisualEffect & CreateVisualEffectObject in script or dialogue it searches for VVC, and reports an error if it isn't found, but some of those actions refer to BAMs.

Is it possible to search for BAM, if VVC isn't found, and then reports an error?

From iesdp: "...the engine first looks to play a VVC file of the specified name, if no such a file exists, the engine looks to play a BAM file of the specified."

 

Thank you for your work.

[Argent77]

Argent77, I have one bug and one feature request:

 

First bug: When I try to save result.txt from triggers & actions check in dialogues it always saves just first few lines, never whole list.

It looks like only errors are saved into the text file even if you switch to the "warnings" tab. I'll fix it in the next snapshot. I'd like to keep errors and warnings separated however to avoid cluttering the output file.

 

And feature request, when checker encounters CreateVisualEffect & CreateVisualEffectObject in script or dialogue it searches for VVC, and reports an error if it isn't found, but some of those actions refer to BAMs.

Is it possible to search for BAM, if VVC isn't found, and then reports an error?

From iesdp: "...the engine first looks to play a VVC file of the specified name, if no such a file exists, the engine looks to play a BAM file of the specified."

I can't reproduce this behavior on my side. If a VVC isn't found, it will search for BAM files. Only if neither of the resource types have been found, a warning will be issued. I'll change the search order for resource types however, so it will first look for VEF, VVC and BAM, in that order.

[Mad Mate]

It looks like only errors are saved into the text file even if you switch to the "warnings" tab. I'll fix it in the next snapshot. I'd like to keep errors and warnings separated however to avoid cluttering the output file.

 Thank you!

snip

I can't reproduce this behavior on my side. If a VVC isn't found, it will search for BAM files. Only if neither of the resource types have been found, a warning will be issued. I'll change the search order for resource types however, so it will first look for VEF, VVC and BAM, in that order.
 
 

Let me give you more details, what is happening, I'll give you an example:
In vanilla BG2 in DRSHSP01.dlg in action 0 NI displays warning (Resource not found: S:DialogFile* - "SPENTAXI") related to:
CreateVisualEffectObject("SPENTAXI",Myself)
but SPENTAXI.BAM is in the game. Things like this occurs in script check, also.
In BG2:EE same files, same line and no warnings are displayed.

Edited by Mad Mate, 23 October 2014 - 07:16 AM.

[Argent77]

Ah, ok. The cause for this warning is an incorrect definition of CreateVisualEffectObject() in ACTION.IDS. The first parameter is defined as "S:DialogFile*" instead of "S:Effect*". I had tested it only in BG2EE where this issue has already been fixed. I'll add a work-around for it in the next snapshot.
 

Edited by Argent77, 23 October 2014 - 08:52 AM.

[CamDawg]

Ah, ok. The cause for this warning is an incorrect definition of CreateVisualEffectObject() in ACTION.IDS. The first parameter is defined as "S:DialogFile*" instead of "S:Effect*". I had tested it only in BG2EE where this issue has already been fixed. I'll add a work-around for it in the next snapshot.
 

In some ways I'm torn on this. On the one hand I feel like NI should accurately reflect the current IDS files, but the flipside of this is that action.ids is wrong and NI shouldn't really waste user's time with false reports.

 

Maybe we can get that Fixpack jerk to fix this for vBG2.

[CamDawg]

While I'm here, I may as well actually contribute something.

 

If you use Tools > Check > Scripts on IWD you get about 4000 warnings, mainly about script names not found, e.g.

 

File: 4001LMTR.BCS  Error: Script name not found: S:Name* - "LIZARD_KING"  Line: 20

 

By default, IWD uses the Name field of an actor in an area file as a creature's script name. LIZARD_KING is a valid script name since it's in ar4001.are as actor 31's name.

 

While that's not default behavior in BG2, the same thing can happen when the 'override script name' flag is set there. BG2 doesn't use the trick often, but one example are the githyanki in ar2400.

[Argent77]

By default, IWD uses the Name field of an actor in an area file as a creature's script name. LIZARD_KING is a valid script name since it's in ar4001.are as actor 31's name.

 

While that's not default behavior in BG2, the same thing can happen when the 'override script name' flag is set there. BG2 doesn't use the trick often, but one example are the githyanki in ar2400.

That's one source of script-related warnings (which I'll fix or work around eventually).

Another source are the checker routines themselves. The checker routines are a bit unreliable sometimes and might show many false positives on occasion. You can minimize this effect by opening any BCS file manually first before starting a check. It's most likely caused by a race condition or something like that. It's very difficult to track down these kinds of issues, which is why I haven't found a solution for it yet.
 

Edited by Argent77, 23 October 2014 - 09:50 AM.

[CamDawg]

The checker routines are a bit unreliable sometimes and might show many false positives on occasion.

Oh, I know. Between the various Fixpacks and EE development, I spend a lot of time with the mass check tools.