Skip to content

New versions of NearInfinity available

1171820222344

Comments

  • argent77argent77 Member Posts: 3,434
    edited October 2017
    kjeron said:

    The preset list of Scripting States/Stats for Opcode 282 is off.
    From 32 onward, all stats should be one less.
    32 is Chaos_Shield
    33 is NPC_Bump
    34 is Use_Any_Item
    35 is Assassinate
    etc...
    42 is StoneSkinsGolem
    43 is LevelDrain
    44 is DonotDraw

    The current stat for 32, Summon_Disable_Action, is skipped.
    The final stat, Ignoredraindeath109, cannot be set by this opcode.
    This may be EE-only (maybe a side effect of giving it a proper upper/lower bound, I don't know).

    NI reads values for opcode 282 directly from STATS.IDS. To take the arbitrary skips into account I would be forced to turn it into a static list of entries. Are scripting states 32 or higher actually set via opcode by any spells? I'm more inclined to limit range of valid scripting states from 0 to 31 instead.
  • kjeronkjeron Member Posts: 2,367
    argent77 said:

    NI reads values for opcode 282 directly from STATS.IDS. To take the arbitrary skips into account I would be forced to turn it into a static list of entries. Are scripting states 32 or higher actually set via opcode by any spells? I'm more inclined to limit range of valid scripting states from 0 to 31 instead.

    There are only a handful of places left where the game still sets the non-Scripting-State stats. While Beamdog has done a good job migrating these over to Spellstates, those I've found are all "BD____" files, from SoD.

    None of them above #9(Stat 165) should be set by this opcode though, unless your intentionally trying to override it's normal function.

    A static list would unfortunately be appropriate, as it really is one. It would also allow STATS 176-182 to be labeled correctly, as Thieving skill modifiers.
    Alternatively, just reversing the display, so it is ordered by STAT, instead of LABEL, would make it a lot easier to work with, IMO.
  • TarakTarak Member Posts: 12
    edited October 2017
    Thanks for the update, argent77!
    The color slider is much better now and copying & pasting entries also works in the correct order.
    What looks broken now in the CRE files are plenty Unknowns for Allegiance, Class, Race, etc.

    About the color slider again: how about changing the slider into a scrollable, vertical list always showing the codes and color shemes side by side. That would vastely speed-up looking for the desired colors instead of going through all numbers over and over again.
  • argent77argent77 Member Posts: 3,434
    edited October 2017
    Tarak said:

    What looks broken now in the CRE files are plenty Unknowns for Allegiance, Class, Race, etc.

    Yeah, that's quite a nasty regression from one of the changes. It affects every field that presents values from an IDS file. I'll fix it asap.

    About the color slider again: how about changing the slider into a scrollable, vertical list always showing the codes and color shemes side by side. That would vastely speed-up looking for the desired colors instead of going through all numbers over and over again.
    I had the same thought when I started working on the slider. It'll require a bit of work to implement though, so it might take a while.
  • argent77argent77 Member Posts: 3,434
    I'm currently working on a new color gradient selection control that allows you to choose a value from a list of entries. It is meant to replace the current control using a slider.

    I'm unsure how to visualize the color gradient. There are two options available:

    1. Use color gradients directly, without filtering:


    2. Interpolate colors to create smooth gradients:


    What do you prefer?
  • AndreaColomboAndreaColombo Member Posts: 5,525
    While option 2 undeniably looks prettier, I would personally prefer option 1 for clarity (it is easier for me to tie that visual back to the squares in EEKeeper, which I'm accustomed to.)
  • MirandelMirandel Member Posts: 526
    Depends how it works - smooth gradient allows infinitely :) more colors. Can the game even handle it?
  • AndreaColomboAndreaColombo Member Posts: 5,525
    I don't think so. My understanding is that you'd still have to select colors by their numerical index, and the only change would be in how each index is displayed (smooth or blocky.)
  • egbertegbert Member Posts: 99
    1
  • argent77argent77 Member Posts: 3,434
    edited October 2017

    I don't think so. My understanding is that you'd still have to select colors by their numerical index, and the only change would be in how each index is displayed (smooth or blocky.)

    Correct. The gradients are all predefined and can only be selected by an index number. Each gradient consists of twelve shades of colors.

    First option allows you to make out every individual color. Second option may give you a better feeling how it looks in-game, since most of the time you'll only see a partial or distorted gradient.
  • MirandelMirandel Member Posts: 526
    Tough call. I'd say first option gives better idea about limitations, hence, more logical. But it's me.
  • Sam_Sam_ Member Posts: 172
    I vote 1st option (non-blended).
  • argent77argent77 Member Posts: 3,434
    I'm also tending to option 1. It may take a while until the next NI version is released though.
  • TarakTarak Member Posts: 12
    #1 is better, as object parts usually don’t show the full spectrum but only some parts of the gradient. These small block-ranges can be easier compared to other codes then. It simply looks closer what we get to see in the game.
  • kjeronkjeron Member Posts: 2,367
    Issue with Spell Icons:
    The icon displayed on the "View" tab is always the first frame of the specified BAM file, rather than the first frame of the first Cycle, as is used in-game.
    I'm not certain how important it is though. I don't think this affects any non-mod content, nor do I know of any reason why you need such inconsistency in a BAM file, I only noticed it because I forgot to delete the unused frames before saving the file.
  • argent77argent77 Member Posts: 3,434
    There is a similar issue with item icons (a good example are spell scrolls in BG1EE). It should be easily fixed.
  • islandkingislandking Member Posts: 426
    edited November 2017
    Hello argent77,
    Could you change the color of travel region or maybe make it a separate option in "Regions" layer in .ARE viewer? Because as of now the the travel regions are the same color as traps (both are red), which is a bit confusing to read, especially for those where traps are near the entrance, thanks.
    Post edited by islandking on
  • argent77argent77 Member Posts: 3,434
    Well, technically there is no real difference between a trap, an info region and a travel region. All three types are defined by the same kind of structure internally. I could experiment with different color shades though. Maybe that helps a bit.
  • GwendolyneGwendolyne Member Posts: 461
    argent77 said:

    Update: NearInfinity v2.1-20171027

    Changes:
    • Lots of changes in BAM Converter:
      • Adds a list of recently used BAM sessions for quick access.
    I certainly missed something... But how to display this list?
  • argent77argent77 Member Posts: 3,434
    edited November 2017

    argent77 said:

    Update: NearInfinity v2.1-20171027

    Changes:
    • Lots of changes in BAM Converter:
      • Adds a list of recently used BAM sessions for quick access.
    I certainly missed something... But how to display this list?
    You can find a new submenu "Load recent sessions" on top of the BAM session menu (see screenshot). It's grayed out (and possibly easy to miss) if the list is empty.


  • GwendolyneGwendolyne Member Posts: 461
    Thanks.

    And it is still in memory after closing and relaunching NI: Great ! ;)
  • islandkingislandking Member Posts: 426
    argent77 said:

    Well, technically there is no real difference between a trap, an info region and a travel region. All three types are defined by the same kind of structure internally. I could experiment with different color shades though. Maybe that helps a bit.

    Thank you @argent77, for the continued support of this great program.
  • jasteyjastey Member Posts: 2,671
    edited November 2017
    Thank you for all the work that went into the BAM Converter. Using it I feel like a BAM artist now! :-)

    One question (v1.32.1. beta 24): I have a mod added .cre in the game and run a search "Find/references to this file/BCS" and it lists me more or less all bcs files of the game as a reference, whereas the cre doesn't appear in any of them. Do you have an idea what this means, I am ready to blame my mod cre-file but I wouldn't know what went wrong.
  • argent77argent77 Member Posts: 3,434
    Are you certain you want to use NI v1.32? That version is already over 10 years old.
  • jasteyjastey Member Posts: 2,671
    edited November 2017
    Nah, I'm just an idiot - I took the "from version" number instead of the big, bold one on top: 2.1-20170907

    NI crashed once (disappeared) and wrote an err.log, too - if you give me an email address I can send it to you, if you want.
  • kjeronkjeron Member Posts: 2,367
    jastey said:

    One question (v1.32.1. beta 24): I have a mod added .cre in the game and run a search "Find/references to this file/BCS" and it lists me more or less all bcs files of the game as a reference, whereas the cre doesn't appear in any of them. Do you have an idea what this means, I am ready to blame my mod cre-file but I wouldn't know what went wrong.

    Does this CRE have an empty script name / death variable, offset 0x280? That will cause the results you are seeing.
  • argent77argent77 Member Posts: 3,434
    kjeron said:

    jastey said:

    One question (v1.32.1. beta 24): I have a mod added .cre in the game and run a search "Find/references to this file/BCS" and it lists me more or less all bcs files of the game as a reference, whereas the cre doesn't appear in any of them. Do you have an idea what this means, I am ready to blame my mod cre-file but I wouldn't know what went wrong.

    Does this CRE have an empty script name / death variable, offset 0x280? That will cause the results you are seeing.
    That's probably it. Currently, only script name "None" is ignored by the reference search. I will also add empty script names to the ignore list in the next release.
  • jasteyjastey Member Posts: 2,671
    Indeed! Thank you very much, I didn't create this cre and wasn't even aware it's midding a DV.
Sign In or Register to comment.