FAangband 0.3.0

Collapse
X
 
  • Time
  • Show
Clear All
new posts

  • Arralen
    replied
    Anyone found a ring of XY that actually does give a resistance bonus, not some vulnerability? Even the black market only offers me rings with negative modifiers ?! Or do I have to wait until past Clvl 25 or so to get a basic "ring of fire protection"?

    Leave a comment:


  • Nick
    replied
    Originally posted by takkaria
    That was why 3.0.9 was going to be 3.0.8fix. Luckily I was steered away from that madness by ajps.
    Right. I'll let my brother know he dropped the ball on this one...

    Leave a comment:


  • takkaria
    replied
    Originally posted by Nick
    (b) it would have made it harder to pretend that nothing had gone wrong.
    That was why 3.0.9 was going to be 3.0.8fix. Luckily I was steered away from that madness by ajps.

    Leave a comment:


  • Nick
    replied
    Originally posted by Arralen
    Racial ability "shadow" still list "vulnerability to bright light" on the specialities screen for Dark Elves.
    Thanks.

    Dunno if it has been reported for that screen already or if it was some other:
    Standard arrow keys do not, numpad arrows keys do work there.
    It was the specialty choosing screen it was reported (and I fixed) for; this one still actually needs numbers to scroll.

    Leave a comment:


  • Arralen
    replied
    Racial ability "shadow" still list "vulnerability to bright light" on the specialities screen for Dark Elves.

    Dunno if it has been reported for that screen already or if it was some other:
    Standard arrow keys do not, numpad arrows keys do work there.

    Leave a comment:


  • Nick
    replied
    Originally posted by petebull
    Don't be so stingy: Version numbers are for free. A bugfix to 0.3.0 should be labeled 0.3.1, even when the first released 0.3.0 should have been rc1 or so.
    True, and I thought of doing that. Things that stopped me included (a) the changes of version number would have required more editing than the actual bugfixes and (b) it would have made it harder to pretend that nothing had gone wrong.

    Leave a comment:


  • petebull
    replied
    Don't be so stingy: Version numbers are for free. A bugfix to 0.3.0 should be labeled 0.3.1, even when the first released 0.3.0 should have been rc1 or so.

    Thanks for your work!

    Leave a comment:


  • Nick
    replied
    OS X port available now.

    Leave a comment:


  • Nick
    replied
    Bugfixed versions of FAangband 0.3.0 (source and windows port) are now available here. OS X port hopefully available soon, WinCE port available when I can work out what's wrong with it as it is.

    Bugs fixed include the notes bug which was causing crashes and other weird behaviour, arrow scrolling in the specialty menu, initialisation bug which was making vaults generate incorrectly (if you're just copying the new executable over the old one, removing lib/data/vault.raw will make sure this is fixed), monsters speaking in tongues.

    Bugs not fixed include dodgy bonus updates (ctrl-R to refresh the screen if you think your speed may be displayed wrong), occasional placing of the player in a wall when entering a river level, arrow keys giving unexpected results for firing and throwing in Win XP (this has me baffled; currently I'm blaming a pref file issue, as usual).

    Probably worth mentioning too that the file lib/pref/font-win.prf has been renamed to font-win.prfx because loading it in Vista makes walls display as blanks; if you're using an earlier version of windows (a) congratulations and (b) remove the x from the end of the name and you'll get block walls back again instead of #s.

    Current plan is that the next thing I release will be 0.3.1, which will have all the menus converted properly, low level code base (I hope) more uniform and stable, better terrain handling, and maybe some overhauling of the monster list (notably dragons). But if there are any more showstopper bugs, let me know and I'll do another bugfix.

    Leave a comment:


  • Nick
    replied
    Originally posted by pav
    Run native Linux app on Linux, and fix Windows-specific bugs. That's the beauty of it.
    That's what I usually do - the problem this time is that there is no hint of the bug in the linux native app.

    [Goes away and checks this is the case, finds that it isn't and locates bug]

    OK, thank you everyone. I will fix this (and some of the other bugs I've found while I'm at it) and get the fixed version out in the next couple of days. The bug was not windows related - I was simply trying to look at entry -1 of the notes array. This also explains the latest occurrence of Nowhere Town...

    Moral: Just because you remember doing something doesn't mean you've actually done it.

    Leave a comment:


  • pav
    replied
    Run native Linux app on Linux, and fix Windows-specific bugs. That's the beauty of it. Also make sure you have debug symbols in - the backtrace you describe looks like it was made on a binary that's stripped.

    Leave a comment:


  • Nick
    replied
    Originally posted by takkaria
    I recommend running the game under valgrind and do the same things. Even though the problem doesn't manifest on Linux, it's still there, and valgrind will find it if it's memory-related. I've found a fair few Windows-only bugs by doing that.
    Yes, this was something I tried, and I just have again. If I run the SDL port, all I get is a few entries inside libraries, and a few from initialising SDL. Once the game is actually running, nothing. This is part of what is leading me to blame windows. For my windows compiles, I have msys and mingw set up in Vista; if I attempt to use gdb in my msys window, the game crashes nicely but gives me ~6000 lines of backtrace, most of them empty and the rest (that are left in the window) uninformative; the windows executable runs fine using wine in linux, but valgrind on that gives nothing, I suspect because it's actually memchecking wine.

    That's what you get for trying to help

    Leave a comment:


  • takkaria
    replied
    Originally posted by Nick
    OK, I've tested in XP and am getting exactly the same problems - firing weirdness, craching at the same places. Tried recompiling it in XP - still the same. None of these happen in Vista (although it does crash on exiting). I'll try and work out what's gone wrong and do a new compile...

    Anyone else having windows issues?
    I recommend running the game under valgrind and do the same things. Even though the problem doesn't manifest on Linux, it's still there, and valgrind will find it if it's memory-related. I've found a fair few Windows-only bugs by doing that.

    Leave a comment:


  • Nick
    replied
    OK, I've tested in XP and am getting exactly the same problems - firing weirdness, craching at the same places. Tried recompiling it in XP - still the same. None of these happen in Vista (although it does crash on exiting). I'll try and work out what's gone wrong and do a new compile...

    Anyone else having windows issues?

    Leave a comment:


  • Nick
    replied
    OK, the morning brings fresh perspective.
    • For the firing thing, try activating the lib/pref/font-win.prf file - I disabled it by renaming it font-win.prfx so as to avoid the invisible walls problem in Vista, but forgot to tell anyone...
    • The crash bug still has me beaten. It appears to be a memory handling issue with the windows port; I have a couple of ideas that I'll check out later today.
    • While you're in testing mode, does (a) picking up a bunch of extra items or (b) setting a subscreen to display the extra player info cause a crash?

    Leave a comment:

Working...
😀
😂
🥰
😘
🤢
😎
😞
😡
👍
👎