Bugs and complaints on current master

[Derakon]

So it sounds like the issue really is that the relative frequency of "endgame-quality" ego weapons is higher than that of artifacts. It's not like most standart players find any of Ringil/Deathweaker/Doomcaller either.

Whether the answer is to make top-tier egos more rare or to make top-tier artifacts more common is a question for the game balance folks. Keep in mind of course that you can win the game with punching if you're patient and have enough consumables (i.e. there's no objective definition of what "endgame-quality" actually is), so really the question is how long you want fights to last.

[Nick]

/* Place the stairs in the north wall */
x = rand_spread(z_info->town_wid / 2, z_info->town_wid / 12);
y = 2;
while (square_isperm(c, y, x) || square_isfiery(c, y, x)) y++;
y--;

This makes commit 9651d36 unnecessary.

Yeah, that's much nicer than my flawed attempt at a solution.

[Estie]

The unstated presumption is that an artifact must have extra attacks to be an endgame-viable weapon. You could also find an artifact like Deathwreaker, which just does monstrous damage. Anything like that in the randart file?

Yes, Deathwreakers exist, but they are very rare, and no, you dont need extra attacks to be endgame viable. But you need something good and if the pool for good mods gets smaller, you get fewer good weapons. The presumption is not only unstated, it is also nonexistant.

I have created a few (about a dozen) randart files, and only one had a weapon in the same league as a top ego. Standarts have at least 3 (Deathwreaker, Ringil, Doomcaller).

It is certainly possible to make good enough randart weapons without the extra attack mod, but why ? Btw, extra shots still exist - its what makes me think this must be an error.

[Derakon]

The unstated presumption is that an artifact must have extra attacks to be an endgame-viable weapon. You could also find an artifact like Deathwreaker, which just does monstrous damage. Anything like that in the randart file?

[Estie]

I havent been playing much, but looking at the randrat file of my latest characters, not a single randart had extra attacks. Standarts are still the same.

Basically, this means that randart weapons are even worse than before, so I would almost always end up with an ego weapon in the endgame, which is not what I want when picking randarts. I assume this is an error ?

[PowerWyrm]

The town generation lava problem may have been fixed, but apparently there's also a town wall generation problem. [ATTACH]1473[/ATTACH]

In close up... [ATTACH]1475[/ATTACH]

In PWMAngband, I just did this to avoid that problem:

/* Place the stairs in the north wall */
x = rand_spread(z_info->town_wid / 2, z_info->town_wid / 12);
y = 2;
while (square_isperm(c, y, x) || square_isfiery(c, y, x)) y++;
y--;

This makes commit 9651d36 unnecessary.

[PowerWyrm]

place_new_monster_one():

/* Add to level feeling, note uniques for cheaters */
c->mon_rating += race->level * race->level;

/* Boost rating by power per 10 levels OOD */
c->mon_rating += (race->level - c->depth) * race->level * race->level;

The "per 10 levels OOD" is gone. I suppose the second line misses a division by 10.

[Nick]

This is from git hash 24570ae7d7c32ffcebab8e80edd0747ecf47f7f2, last Wednesday.

Thanks for the report.
That looks like a bug with null objects that has since been fixed. I'm not sure about the second one - it looks like a problem with monster names, but it might be memory corruption from one of the other problems with that version.

[kandrc]

Same version:

Program received signal SIGABRT, Aborted.
0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
(gdb) bt
#0 0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
#1 0x00007ffff6d81088 in abort () from /lib64/libc.so.6
#2 0x00007ffff6d78966 in __assert_fail_base () from /lib64/libc.so.6
#3 0x00007ffff6d78a12 in __assert_fail () from /lib64/libc.so.6
#4 0x00000000004e325e in textblock_vappend_c (tb=0xcecb58, attr=1 '\001',
fmt=0x514e05 "%-*s%s\n", vp=0x7fffffffd728) at z-textblock.c:115
#5 0x00000000004e3584 in textblock_append_c (tb=0xcecb58, attr=1 '\001',
fmt=0x514e05 "%-*s%s\n") at z-textblock.c:186
#6 0x00000000004bc7db in monster_list_format_section (list=0xb765e8,
tb=0xcecb58, section=MONSTER_LIST_SECTION_ESP, lines_to_display=45,
max_width=52, prefix=0x514e6f "You are aware of", show_others=true,
max_width_result=0x7fffffffd9f8) at ui-mon-list.c:160
#7 0x00000000004bcb36 in monster_list_format_textblock (list=0xb765e8,
tb=0xcecb58, max_lines=49, max_width=52, max_height_result=0x0,
max_width_result=0x0) at ui-mon-list.c:302
#8 0x00000000004bce03 in monster_list_show_interactive (height=88, width=214)
at ui-mon-list.c:422
#9 0x00000000004b82ad in do_cmd_monlist () at ui-knowledge.c:3387
#10 0x00000000004ad719 in textui_process_command () at ui-game.c:312
#11 0x00000000004ad765 in textui_get_cmd (context=CMD_GAME) at ui-game.c:324
#12 0x00000000004ada3e in play_game (new_game=false) at ui-game.c:432
#13 0x00000000004e5c5d in main (argc=1, argv=0x7fffffffdc48) at main.c:524

[kandrc]

Program received signal SIGABRT, Aborted.
0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
(gdb) bt
#0 0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
#1 0x00007ffff6d81088 in abort () from /lib64/libc.so.6
#2 0x00007ffff6d78966 in __assert_fail_base () from /lib64/libc.so.6
#3 0x00007ffff6d78a12 in __assert_fail () from /lib64/libc.so.6
#4 0x00000000004e325e in textblock_vappend_c (tb=0xc19808, attr=1 '\001',
fmt=0x4fed93 "Dropped by %s %s", vp=0x7fffffffd758) at z-textblock.c:115
#5 0x00000000004e34d7 in textblock_append (tb=0xc19808,
fmt=0x4fed93 "Dropped by %s %s") at z-textblock.c:173
#6 0x000000000045feb0 in describe_origin (tb=0xc19808, obj=0xa09548,
terse=false) at obj-info.c:1676
#7 0x0000000000460276 in object_info_out (obj=0xa09548, mode=2)
at obj-info.c:1771
#8 0x0000000000460599 in object_info (obj=0xa09548, mode=OINFO_SUBJ)
at obj-info.c:1826
#9 0x00000000004c1724 in textui_obj_examine () at ui-object.c:1504
#10 0x00000000004ad719 in textui_process_command () at ui-game.c:312
#11 0x00000000004ad765 in textui_get_cmd (context=CMD_GAME) at ui-game.c:324
#12 0x00000000004ada3e in play_game (new_game=false) at ui-game.c:432
#13 0x00000000004e5c5d in main (argc=1, argv=0x7fffffffdc48) at main.c:524

This is from git hash 24570ae7d7c32ffcebab8e80edd0747ecf47f7f2, last Wednesday.

[Nick]

OK.

The latest build has trap, door and stair detection completely removed, and mapping a bit enhanced to compensate. I'm interested to hear opinions, but preferably after a reasonably thorough test...

[Nick]

The town generation lava problem may have been fixed, but apparently there's also a town wall generation problem. [ATTACH]1473[/ATTACH]

In close up... [ATTACH]1475[/ATTACH]

*facepalm*

[Ingwe Ingweron]

The town generation lava problem may have been fixed, but apparently there's also a town wall generation problem.

In close up...

[Nick]

New build up with a fix to the level generation crash and some tweaks to randarts (which will probably cause weirdness in existing randart games from the previous build). Template room changes not done yet, maybe tomorrow.

[Nick]

This newest version is much more stable. Made it to DL 98 before it crashed, immediately upon level arrival down the stairs. Haven't been able to duplicate it, so I don't know what the crash was from, but it must have had trouble generating something.

Edit: Yep, it's a crash on level generation, so no savefile will do much to troubleshoot it. However, if you go to DL 98/99 and stair scum, it shouldn't take too long to generate a crash.

Found it, I think. It's been a while since I've had an infinite recursion bug