32x32 Universal Tileset

Collapse
X
 
  • Time
  • Show
Clear All
new posts

  • nppangband
    replied
    Originally posted by buzzkill
    Moved the seeker arrow effect tiles at the tail end of row 0x81 in order to entirely vacate column 0xFF. Also created the 4 missing tiles of that type. As far as I can tell (searching all prf files in both V and NPP), these tiles weren't utilized at all. Just FYI.
    Did you move the whole set? I just started using those. If you fire an arrow, it displays the regular arrow tile (pointed to the upper right corner), regardless of direction. Now, in NPP, I use those two arrowsets. if you fire an arrow (or bolt), it will point will point in the right direction as it travels.

    It is hard-coded rather than being part of the pref files. Please let me know where you moved it to. Thanks,

    Leave a comment:


  • buzzkill
    replied
    Moved the seeker arrow effect tiles at the tail end of row 0x81 in order to entirely vacate column 0xFF. Also created the 4 missing tiles of that type. As far as I can tell (searching all prf files in both V and NPP), these tiles weren't utilized at all. Just FYI.

    Leave a comment:


  • buzzkill
    replied
    Originally posted by nppangband
    There is probably some sort of overflow issue with 0xFF, since that would be loaded into a byte or char variable. Maybe the code would be adjusted to properly read it, but is reality we would have to fill up a 127x127 grid with tiles before that becomes an issue.
    Well, now that I know about it, I'll move those tiles. Would physically shrinking the tile sheet, eliminating that last column so that this can't happen again, cause any foreseeable problems?

    Leave a comment:


  • nppangband
    replied
    Originally posted by PowerWyrm
    Unfortunately, this won't work. There seems to be a problem with displaying tiles having their char set to 0xFF, which seems to explain why there are no tiles in that column (except on the second line, which is not used).
    There is probably some sort of overflow issue with 0xFF, since that would be loaded into a byte or char variable. Maybe the code would be adjusted to properly read it, but is reality we would have to fill up a 127x127 grid with tiles before that becomes an issue.

    Leave a comment:


  • buzzkill
    replied
    Originally posted by PowerWyrm
    Why did you put a dark tile at 0x80:0x80? It's not required, as the code puts a dark background anyway when drawing tiles. Reverting to a transparent tile and leaving lurkers/trappers at 0x80:0x80 should be ok...
    Because it's referred to as "darkness" in the prf, and I thought that it might be a good idea for it to represent darkness rather than nothing. At the same time, I figured that any unused space would suffice as the lurker/trapper tile. I chose 0x80:0xFF because I thought it would remain unused long into the future. Will fix.

    Leave a comment:


  • nppangband
    replied
    Originally posted by PowerWyrm
    What would be perfect is to add a graf-dvg-base.prf file with the mappings for all common tiles, and load this file at the end of the graf-dvg.prf file for each variant. This would improve readability and ease maintainance...
    The difficulty in that is coming up with a standard where each variant can read the tiles. Many of them are assigned to different monster race numbers or slightly different names, so it would be a challenge to create a universal graf-dvg that could be loaded by all variants.

    Leave a comment:


  • buzzkill
    replied
    Originally posted by PowerWyrm
    Hmm I've just looked at the updated 32x32 graphics for NPP Angband... Many tiles have been redrawn/remapped for monsters/objects/terrain that are present in all variants, eliminating many duplicates in the process, still the mappings in graf-dvg files differ. This completely breaks the purpose of an universal 32x32 tileset...
    Yep. Vanilla Angband is still the special case. I'm very hesitant to make changes to Vanilla in deference to David Gervais, who's tileset has been the gold standard forever, but I wouldn't say that breaks the purpose of UT32. There are many variants that support tiles, 1/2 a dozen I can think of. The main point of UT32 was to have the all the tiles in one location so that variants can easily share tiles and even non-graphically inclined maintainers can do some of the (prf tile assignment) work themselves. Don't forget, UT32 is still a long way from completion.

    I wasn't my intent for UT32 to alter Vanilla in any significant way. I just wanted it to be compatible with Vanilla so that there would be only need for one 32x32 tileset.

    Leave a comment:


  • PowerWyrm
    replied
    I've just downloaded the new UT32 file and found this:

    03-07-2012 Reassigned the lurker and trapped to a transparent tile 0x80:0xFF.
    Unfortunately, this won't work. There seems to be a problem with displaying tiles having their char set to 0xFF, which seems to explain why there are no tiles in that column (except on the second line, which is not used).

    Why did you put a dark tile at 0x80:0x80? It's not required, as the code puts a dark background anyway when drawing tiles. Reverting to a transparent tile and leaving lurkers/trappers at 0x80:0x80 should be ok...

    Leave a comment:


  • PowerWyrm
    replied
    Hmm I've just looked at the updated 32x32 graphics for NPP Angband... Many tiles have been redrawn/remapped for monsters/objects/terrain that are present in all variants, eliminating many duplicates in the process, still the mappings in graf-dvg files differ. This completely breaks the purpose of an universal 32x32 tileset...

    What would be perfect is to add a graf-dvg-base.prf file with the mappings for all common tiles, and load this file at the end of the graf-dvg.prf file for each variant. This would improve readability and ease maintainance...

    Leave a comment:


  • PowerWyrm
    replied
    Review of monster tiles, part III: hounds

    This is mainly a remap to match colors in the txt file. The only problem I have with this one concerns clear hounds, which would probably need a more "transparent" tile.

    Light hound 0x90:0xC6
    Dark hound 0x90:0xC3
    Clear hound 0x90:0xBB
    Fire hound 0x90:0xC5
    Energy hound 0x90:0xC7
    Acid hound 0x90:0xC2
    Vibration hound 0x90:0xBD
    Nexus hound 0x90:0xBE
    Gravity hound 0x90:0xBA
    Inertia hound 0x90:0xC9
    Impact hound 0x90:0xC0
    Nether hound 0x90:0xC8
    Time hound 0x90:0xCB
    Plasma hound 0x90:0xCC
    Ethereal hound 0x90:0xBC
    Chaos hound 0x90:0xC1
    Aether hound 0x99:0xDB

    Leave a comment:


  • buzzkill
    replied
    Originally posted by PowerWyrm
    In fact, I fear the current tile for Morgoth is the one that was supposed to be used by Sauron. So basically the mapping for V could use the old Morgoth tile for Sauron and the new Morgoth tile for Morgoth.
    We both agree that the old Sauron tile is unsuitable. Jeff and I had a lot of trouble creating a tile for Sauron that looked good and we both liked. Sauron is staying put in NPP which means that I'm very inclined to use that same tile for Sauron in all variants including Vanilla.

    If old Morgoth goes anywhere, he'll probably go away. I not a big fan of the tile itself. The multi-colored robes don't do anything for me. I'm content to leave him as he is now in Vanilla, but other variants will get the new Morgoth.

    I'll post the latest version of everythnig later today.

    Leave a comment:


  • PowerWyrm
    replied
    Review of monster tiles, part II: undead

    As hard as I tried, I couldn't come with something completely acceptable. There are clearly tiles missing for the zombified/mummified creatures, and a lot of useless skeleton tiles. Here's how I would see the mapping for V (with the associated comments):

    Vampires (changed mostly to match color in the txt file)
    Master vampire 0x94:0xC7
    Vampire lord 0x94:0xB3
    Elder vampire 0x94:0xB4

    Wights & liches (the original mapping simply doesn't match the description of the creatures -- corpse-like undead; using same tile family, with tile matching color in txt file)
    Forest wight 0x94:0x90
    Grave wight 0x94:0x8E
    Barrow wight 0x94:0x91
    Emperor wight 0x94:0x8F
    Lich 0x94:0x99
    Master lich 0x94:0x9B
    Demilich 0x94:0xC5
    Archlich 0x94:0xCA
    Black reaver 0x94:0x93

    Wraiths (the original mapping simply doesn't match the description of the creatures -- skeletal ghosts; seems like the person who did the mapping inverted wights and wraiths; using same tile family, with tile matching color in txt file; for Nightcrawler, its description mentions a giant wormlike creature, the tile was simply inappropriate; for Nightwalker, its description mentions a giant black figure, the tile was changed accordingly)
    White wraith 0x94:0xCC
    Grey wraith 0x94:0x8C
    Black wraith 0x94:0x8D
    Nether wraith 0x94:0x89
    Nightwing 0x94:0x8A
    Nightcrawler 0x91:0x99
    Nightwalker 0x99:0xDC

    Ghosts (changed mostly to match color in the txt file)
    Poltergeist 0x94:0x94
    Green glutton ghost 0x94:0x82
    Lost soul 0x94:0x97
    Moaning spirit 0x94:0x95
    Banshee 0x94:0x80
    Ghost 0x94:0x96
    Shade 0x94:0xC3
    Spectre 0x94:0x86
    Dread 0x94:0x84
    Shadow 0x94:0xC8
    Phantom 0x94:0x98
    Dreadmaster 0x94:0x81
    Dreadlord 0x94:0xCB

    Skeletons (just changed these two tiles to keep the proportions to human/troll size)
    Skeleton human 0x94:0xAA
    Skeleton troll 0x94:0xB8

    Zombies (remapped to the "mummy"-like tiles instead of the original "skeleton"-like tiles; "zombified" and "mummified" creatures have been remapped to the same tile, since no distinct tile currently exists in the tileset; other tiles changed mostly to match color in the txt file)
    Zombified kobold 0x91:0xB1
    Zombified orc 0x91:0xB1
    Zombified human 0x91:0xB1
    Mummified orc 0x91:0xB7
    Mummified human 0x91:0xB7
    Ghoul 0x91:0xB6
    Ghast 0x91:0xB0
    Greater mummy 0x91:0xB3
    Mummified troll 0x91:0xB7

    Leave a comment:


  • PowerWyrm
    replied
    Originally posted by PowerWyrm
    Cncerning Sauron, the NPP tile (if it's the one I'm thinking of) is great. Please use that. For Morgoth, the original tile is ok.
    In fact, I fear the current tile for Morgoth is the one that was supposed to be used by Sauron. So basically the mapping for V could use the old Morgoth tile for Sauron and the new Morgoth tile for Morgoth.

    Leave a comment:


  • buzzkill
    replied
    Originally posted by PowerWyrm
    Here you should really keep the original mapping, which corresponds to the colors in the txt file:

    Elder aranea (r) 0x8F:0xB3
    Mirkwood spider (g) 0x8F:0xB8
    Aranea (R) 0x8F:0xB6
    Phase spider (B) 0x8F:0xB4
    Wood spider (U) 0x8F:0xB5
    Cave spider (D) 0x8F:0xB7
    Giant spider (V) 0x8F:0xB2
    I'm going to have to flip-flop on this. Given the tiles we have at our disposal right now, I like the current mappings. I'd rather phase was light blue than purple, but I've got no problem with the rest. AFAIK, the current mappings are the original mappings or at least that's the way they been for the past several years since I started hanging out here at oook.

    Leave a comment:


  • PowerWyrm
    replied
    Review of monster tiles, part I: spiders

    Here you should really keep the original mapping, which corresponds to the colors in the txt file:

    Elder aranea (r) 0x8F:0xB3
    Mirkwood spider (g) 0x8F:0xB8
    Aranea (R) 0x8F:0xB6
    Phase spider (B) 0x8F:0xB4
    Wood spider (U) 0x8F:0xB5
    Cave spider (D) 0x8F:0xB7
    Giant spider (V) 0x8F:0xB2

    Considering the Giant spider, I've filed a bug report IIRC about monsters using the light purple color, which is now used for the special "Display uniques in purple" option. But it shouldn't affect the tile...

    The only problem concerns the Giant tarantula (o). In UT32, there's a new tile that isn't used for anything in V and could match: 0x99:0xD3

    Leave a comment:

Working...
😀
😂
🥰
😘
🤢
😎
😞
😡
👍
👎