4.0.2 Bugs

[fph]

Yes, I dimly remember reading a post by Nick where he said it got fixed. Unfortunately I cant find the post anymore.

I think it got fixed in 4.0.2. The release notes on http://angband.oook.cz/forum/showthread.php?t=7462 have "Fix lack of artifacts" in them. There was a longer discussion here http://angband.oook.cz/forum/showthr...?t=7436&page=3 ending in Nick writing "Bingo. Check was there for deleting individual monsters (eg banishment), but not when the whole monster list was wiped on leaving the level."
This happened the day before 4.0.2 was released, and it looks like a fix which is game-changing enough to release a new minor release especially for it.

[Timo Pietilä]

I think it got fixed in 4.0.2.

Just bad luck then. OTOH I too find very few artifacts at early levels. There just aren't many that can be generated in early game.

[Estie]

Definitely not "bad luck". With my latest character, I have cleared 5 checkerboard vaults now (I think thats the right name ? the ones with single cells and a snakelike line of granite winding through), not a *single* artifact in all 5. Clearing one of these would normally provide a dozen or so artifacts and be enough to set me up for the endfights.

[Timo Pietilä]

Definitely not "bad luck". With my latest character, I have cleared 5 checkerboard vaults now (I think thats the right name ? the ones with single cells and a snakelike line of granite winding through), not a *single* artifact in all 5. Clearing one of these would normally provide a dozen or so artifacts and be enough to set me up for the endfights.

That definitely sounds like a bug. Maybe there are two bugs that cause artifacts not be available. One got fixed but other remains.

[Nick]

Definitely not "bad luck". With my latest character, I have cleared 5 checkerboard vaults now (I think thats the right name ? the ones with single cells and a snakelike line of granite winding through), not a *single* artifact in all 5. Clearing one of these would normally provide a dozen or so artifacts and be enough to set me up for the endfights.

Is this randarts, and is it played over a savefile from before the bug that was fixed was fixed? I don't know if that's relevant, but they seem like things that might be. Any other info you think might be relevant is worth mentioning, too - in fact, if you have a savefile, that could be useful.

[Estie]

Savefile:

http://www85.zippyshare.com/v/cp9QOn8M/file.html

Its a fresh download + install after I read about the artifact rarity fix. Yes, it is randarts.

[Ingwe Ingweron]

Was the lack of protection against "...commands you to go far away" ever fixed in development? @ has full protection but that @#$% Sauron just forced him through the ceiling. With only one * left and @ ready to deliver the killing blow! ARGH!!!!

I confess though, I'm playing the branch 4.0.1 version because I wanted the "cone breath". Has that been put into 4.0.2?

Now I have to find some stairs and track him down all over again. That bastard is going to be instantly at full health.

EDIT: Update. @#$%! The bastard did it again! @ tracked him down, killing a few uniques along the way. Had him down to one *, and @ is teleported away through the ceiling.

FURTHER EDIT: Third time is the charm. That bastard Sauron commanded @ to "go away" through the ceiling TWICE! Both times when he was down to only one *. Finally, on the third attempt @ was able to take him out, but only because he got lucky and was able to resist a teleport away command. Since @ has all the resistances necessary to block such a command, this BUG is quite frustrating. Ate up too much !Healing and !*Healing*, since had to basically fight and kill Sauron three times over. At least @ killed several uniques in his quest to reacquire Sauron as a target. http://angband.oook.cz/ladder-show.php?id=18155

[kandrc]

The artifact bug is not fixed in 4.0.2. Nick did post that he fixed it, but the context of that post leaves in unclear whether it's fixed in 4.0.2 or in devel. If he believes it's fixed in 4.0.2, he's mistaken. I've done the same analysis I did in the older thread and confirmed that the issue persists.

[Nick]

The artifact bug is not fixed in 4.0.2. Nick did post that he fixed it, but the context of that post leaves in unclear whether it's fixed in 4.0.2 or in devel. If he believes it's fixed in 4.0.2, he's mistaken. I've done the same analysis I did in the older thread and confirmed that the issue persists.

Well, that's disappointing. I wonder what I did wrong. I'll get back to you on this.

[Nick]

I confess though, I'm playing the branch 4.0.1 version because I wanted the "cone breath". Has that been put into 4.0.2?

The cone breath branch has everything from 4.0.2 except the actual version update. Sadly, the RNexus preventing teleport level fix has only gone in post 4.0.2.

[Ingwe Ingweron]

The cone breath branch has everything from 4.0.2 except the actual version update. Sadly, the RNexus preventing teleport level fix has only gone in post 4.0.2.

Is the RNexus fix in the nightlies? But not the cone breath, is that what you're saying?

On another note, one minor bug I've noticed that was introduced during the 4.0beta fix for id by use problems. =Open Wounds identifies itself on activation, even if @ is at full hp. I think it is only supposed to ID when @ has some healing to do, much like !CLW, etc. At least, that's how it worked prior to 4.0.

[kandrc]

"The Berserker tramples over Lorgan, Chief of the Easterlings."

Not sure if this is a bug or not. Are uniques supposed to be able to get trampled? And should ordinary monsters be able to trample uniques?

[kandrc]

I loaded up my Sonic Screwdriver:

name:139:of the Doctor
base-object:sword:Sonic Screwdriver
graphics:|:B
info:1:10:50000000
alloc:99:1 to 127
power:0:17d61:71:71:400
flags:PROT_FEAR | PROT_BLIND | PROT_CONF | PROT_STUN | SLOW_DIGEST | FEATHER
flags:REGEN | TELEPATHY | SEE_INVIS | FREE_ACT | HOLD_LIFE | BLESSED | NO_FUEL
flags:SUST_STR | SUST_INT | SUST_WIS | SUST_DEX | SUST_CON
act:ENLIGHTENMENT
time:0+1d1
values:STR[125] | INT[125] | WIS[125] | DEX[125] | CON[125]
values:STEALTH[125] | INFRA[125] | SEARCH[125]
values:TUNNEL[125] | SPEED[70] | BLOWS[70] | SHOTS[70] | MIGHT[70] | LIGHT[70]
values:BRAND_ACID[5] | BRAND_ELEC[5] | BRAND_FIRE[5] | BRAND_COLD[5]
values:BRAND_POIS[5] | SLAY_ORC[5] | SLAY_TROLL[5] | SLAY_GIANT[5]
values:SLAY_DRAGON[5] | SLAY_DEMON[5] | SLAY_UNDEAD[5] | SLAY_EVIL[5]
values:SLAY_ANIMAL[5]
values:RES_ACID[3] | RES_ELEC[3] | RES_FIRE[3] | RES_COLD[3] | RES_POIS[3]
values:RES_LIGHT[3] | RES_DARK[3] | RES_SOUND[3] | RES_SHARD[3]
values:RES_NEXUS[3] | RES_NETHER[3] | RES_CHAOS[3] | RES_DISEN[3]
desc:It's full of wibbly-wobbly, timey-wimey stuff, but it's not a weapon.
desc: Never a weapon.

scummed dl1 until I found it, then dove. Fastest win ever. Anyway, after killing M, I tunnelled out the entirety (literally) of DL100, then started farming summoners. I was having a fantastic time when, suddenly:

Program received signal SIGSEGV, Segmentation fault.
0x00000000004b6075 in object_kind_attr (kind=0x7ffff710c7b8 <main_arena+88>)
    at ui-object.c:99
99              return use_flavor_glyph(kind) ? flavor_x_attr[kind->flavor->fidx] :
(gdb) print kind
$11 = (const struct object_kind *) 0x7ffff710c7b8 <main_arena+88>
(gdb) print *kind
$12 = {name = 0xc3cfd0 "", text = 0x8085d0 "", base = 0x80c2a0, 
  next = 0x8085d0, kidx = 8298864, tval = 0, sval = 8311968, pval = {base = 0, 
    dice = -149895208, sides = 32767, m_bonus = -149895208}, to_h = {
    base = 32767, dice = -149895192, sides = 32767, m_bonus = -149895192}, 
  to_d = {base = 32767, dice = -149895176, sides = 32767, 
    m_bonus = -149895176}, to_a = {base = 32767, dice = -149895160, 
    sides = 32767, m_bonus = -149895160}, ac = 32767, dd = -149895144, 
  ds = 32767, weight = -149895144, cost = 32767, 
  flags = "(\310", <incomplete sequence \367>, kind_flags = "\377\177", 
  modifiers = {{base = -149895128, dice = 32767, sides = -149895112, 
      m_bonus = 32767}, {base = -149895112, dice = 32767, sides = -149895096, 
  <SNIP>
(gdb) print *kind->flavor
$14 = {text = 0x7ffff710c9d8 <main_arena+632> "\310\311\020\367\377\177", 
  next = 0x7ffff710c9d8 <main_arena+632>, fidx = 4145072616, 
  tval = 255 '\377', sval = 127 '\177', d_attr = 0 '\000', 
  d_char = -149894680 L'\xf710c9e8'}
(gdb) up
#1  0x00000000004b0a46 in grid_data_as_text (g=0x7fffffffb670, 
    ap=0x7fffffffb6a4, cp=0x7fffffffb69c L"Ă\001", tap=0x7fffffffb6a0, 
    tcp=0x7fffffffb698 L"\037Ă\001") at ui-map.c:240
240                             a = object_kind_attr(g->first_kind);
(gdb) print *g
$15 = {m_idx = 0, f_idx = 1, first_kind = 0x7ffff710c7b8 <main_arena+88>, 
  trap = 0x0, multiple_objects = false, unseen_object = false, 
  unseen_money = false, lighting = LIGHTING_LOS, in_view = true, 
  is_player = false, hallucinate = false, trapborder = false}

On a side note, it's good there are no Wood Golems. The Sonic is rubbish on wood.

[Carnivean]

"The Berserker tramples over Lorgan, Chief of the Easterlings."

Not sure if this is a bug or not.

It's not new to this version. The last time I saw it posted the Berserker trampled it's summoner.

Are uniques supposed to be able to get trampled? And should ordinary monsters be able to trample uniques?

I'd say that they shouldn't. In fact I'd say that summoners should always be stronger than the summoned as well.

[Nick]

I was having a fantastic time when, suddenly:

Program received signal SIGSEGV, Segmentation fault.
0x00000000004b6075 in object_kind_attr (kind=0x7ffff710c7b8 <main_arena+88>)
    at ui-object.c:99
99              return use_flavor_glyph(kind) ? flavor_x_attr[kind->flavor->fidx] :
(gdb) print kind
$11 = (const struct object_kind *) 0x7ffff710c7b8 <main_arena+88>
(gdb) print *kind
$12 = {name = 0xc3cfd0 "", text = 0x8085d0 "", base = 0x80c2a0, 
  next = 0x8085d0, kidx = 8298864, tval = 0, sval = 8311968, pval = {base = 0, 
    dice = -149895208, sides = 32767, m_bonus = -149895208}, to_h = {
    base = 32767, dice = -149895192, sides = 32767, m_bonus = -149895192}, 
  to_d = {base = 32767, dice = -149895176, sides = 32767, 
    m_bonus = -149895176}, to_a = {base = 32767, dice = -149895160, 
    sides = 32767, m_bonus = -149895160}, ac = 32767, dd = -149895144, 
  ds = 32767, weight = -149895144, cost = 32767, 
  flags = "(\310", <incomplete sequence \367>, kind_flags = "\377\177", 
  modifiers = {{base = -149895128, dice = 32767, sides = -149895112, 
      m_bonus = 32767}, {base = -149895112, dice = 32767, sides = -149895096, 
  <SNIP>
(gdb) print *kind->flavor
$14 = {text = 0x7ffff710c9d8 <main_arena+632> "\310\311\020\367\377\177", 
  next = 0x7ffff710c9d8 <main_arena+632>, fidx = 4145072616, 
  tval = 255 '\377', sval = 127 '\177', d_attr = 0 '\000', 
  d_char = -149894680 L'\xf710c9e8'}
(gdb) up
#1  0x00000000004b0a46 in grid_data_as_text (g=0x7fffffffb670, 
    ap=0x7fffffffb6a4, cp=0x7fffffffb69c L"Ă\001", tap=0x7fffffffb6a0, 
    tcp=0x7fffffffb698 L"\037Ă\001") at ui-map.c:240
240                             a = object_kind_attr(g->first_kind);
(gdb) print *g
$15 = {m_idx = 0, f_idx = 1, first_kind = 0x7ffff710c7b8 <main_arena+88>, 
  trap = 0x0, multiple_objects = false, unseen_object = false, 
  unseen_money = false, lighting = LIGHTING_LOS, in_view = true, 
  is_player = false, hallucinate = false, trapborder = false}

Dodgy object of some kind - hard to track down short of live debugging, or without at least a stack trace.

On a side note, it's good there are no Wood Golems. The Sonic is rubbish on wood.

You just have to trick them into shooting themselves.