Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - ant

Pages: [1] 2
1
Classic Roguelikes / Re: Rogue - let's beat it
« on: July 25, 2017, 11:56:08 PM »
elwin's site is down.  I've tried to message him on here to no avail.

Anyone know how to reach him?  I love his site for playing Rogue.  Any recommendations for playing Rogue elsewhere?

aw.  that's too bad.  sometimes it gets to be too much work to keep
up a website when life gets busy.

i've always compiled rogue to run on my own machine...  what are you
running for an OS?  perhaps your distribution has it packaged...

2
now at:

   www.anthive.com/rog/rog.html and
   www.anthive.com/rog/other/things.html


  as i've been updating my website the page has a
new location at:

  http://www.anthive.com/project/rogue/

  the other/things are now on the same page as the
rogue and rogomatic items.

  not much has changed with the code for some time
(i'm distracted by other/things...  :) )  i'm hoping to
pick up a new machine soon so i can code and test
on 64 bit versions, but i don't expect to start getting
back to the debugging and coding cycle until the winter.

luckily the fate of the world does not depend upon this...

3
Classic Roguelikes / Re: Rogue - let's beat it
« on: December 05, 2016, 03:21:43 PM »
i'm not sure which version you normally play, but the
code i am looking at it seems like about a 15% chance of
a weapon being +'ed (1-3) and 10% chance of being
cursed (-1 - -3).

4
Classic Roguelikes / Re: For those who have played unix Rogue a lot
« on: December 05, 2016, 03:09:54 PM »
how has it gone?

as to your original question, other than in treasure rooms it is
pretty rare to see a large number of figures on the screen at
one time.  so my guess is that three of each would work.  3x26
is a large set!  me being rather frugal would use old 1x1 tiles
with something painted on.

if you are having multiple players though, it would seem that
you would add more monsters to balance.

5
Classic Roguelikes / Re: Rogue - let's beat it
« on: August 03, 2015, 04:04:58 PM »
...
The funniest part might be that I died with zero gold. Leprechauns robbed my ass four times.

*laughing*  that's a rotten streak of bad luck...

6
Classic Roguelikes / Re: Rogue - let's beat it
« on: July 02, 2015, 11:08:48 PM »

@ant: I came across your work on Rog-O-Matic a few months ago and was intrigued. Right now I'm busy with new stuff for the Roguelike Gallery, but in a few weeks I should have the time to take a close look. Then maybe we can collaborate on squashing bugs.

thanks!  that would be fun.  :)

7
Classic Roguelikes / Re: Rogue - let's beat it
« on: June 19, 2015, 12:25:31 AM »
Elwin, it's great to see someone who's actually debugging code of the rogues.  i've done a fair bit of that too but sometimes i'm not sure i've helped.  :)

i have been working more on rogomatic so that really play tests the version of rogue i've been debugging too (a version of 5.4.4).  if you want to talk code/bugs perhaps i've found and fixed some you've not hit yet?  anyways the source code for both rogomatic and rogue that i've been messing with are at www.anthive.com/rog.html.

both rogomatic and rogue version i've been using are set up to use environment variables so that i can have replicable runs when debugging.   and i have debugging scripts (mostly bash) that i use to check my progress on finding/fixing bugs.

at the moment i'm stuck on fixing a bug in rogue (the hero gets into a strange place in the maze).  some of my previous changes may have brought this on.  hard for me to be sure as i'm getting so far now that it takes many tries to find a new bug.  and i find the code a bitch to read/understand at times.

some other things i've done with rogue is to make sure it generates more distinct messages because at times they are ambiguous or non-existent.  this helps rogomatic know what is going on.

as this is gardening season i'm not spending much time debugging or such, but if you want to send me a note at the e-mail (ant at anthive dot com) i'm always willing to chat about rogue and rogomatic.

8
Classic Roguelikes / Re: Rogue - let's beat it
« on: June 19, 2015, 12:08:21 AM »
the scare monster scroll won't turn to dust if you move on it without picking it up.

the drain life wand is very powerful but it also takes some of your own hit points.

in later levels the polymorph can save you in some tough spots.  also the teleports can at least buy you some time.

9
Classic Roguelikes / Re: Rogue - let's beat it
« on: April 23, 2015, 02:54:45 PM »
yes, there are still bugs in various versions.  i've tried to debug some of the things found in mine (5.4.4), but the code is not easy to read or trace at times...  i've run into some bugs recently that i've given up on finding for now (too little time until the gardening season is over).

i am working on rogomatic code too so that gives me insight into some things when i read one and then the other...

as for cursed items and equipping before reading uncurse scroll, you can only use one weapon and one armor and two rings so that limits your uncurse to four items.  when you uncurse the item does not go to zero, it's still bad armor, but the uncurse lets you take it off to get into something better.  the whole curse thing is that you get stuck with something that is glued to you and you are forced to cope with it.

my armor strategy is to wear the best i have until i get up to where the aquators are at, and then switch to the second best armor or leather armor (which is better than nothing) then when i get a chance to check scrolls i put the best armor back on and read them hoping to find a protect armor or enchant armor.

the mysterious trap is purely annoyance IMO, i don't see it doing much of anything for damage.  i think it was put in to annoy the rogomatic coders as some other changes seem to be geared towards that too (make it harder to know what is going on because there may not be a message in reaction to reading a scroll or zapping).  my own answer to this was to go and add more messages for those cases in rogue so that rogomatic could know what was going on.  source code...  :) :) :)

strange mist is just a sleep trap, it holds you in place for a bit.  hope you aren't being chased by a difficult monster at that time and really low on HP.

the various messages (cry of despair) usually mean something, but in some versions they can mean more than one thing, which is frustrating.  i've changed the version i mess with to make all messages unique, so that if you see it you know what it means.  otherwise it's just too hard to keep track of all the things going on in rogomatic if the messages might mean one thing or another and you have to wait to be sure (or get an identify scroll)...

potion of healing should cure blindness and potion of extra healing.  some versions will give you a one point HP bump if you quaff an extra healing potion when at max HP already.

as for your glitch, i'm wondering if you hit the Q instead of the q (Q is quit and some versions won't ask you if you want to quit for sure)?  well anyways, yes, crashes and strange stuff do happen.  hard to replicate some bugs.

10
Classic Roguelikes / Re: Rogue - let's beat it
« on: December 22, 2014, 05:26:39 AM »
it's not a graphical real time shooter with explosions and other excitement.  times have changed, games have gotten really good.

i was at an arcade the other day and was rather amazed by the visual quality of some of the games (i'd not been in one for many years).

rogue 5.4.4 that i have been debugging is very tough to beat.  and for most rogues in general there's none of this multiple lives stuff.  when you're dead, you're dead...

at the lower levels some important things that may save you are teleportation or polymorph staffs or wands, extra healing potions and knowing when to just run and dive for a trap.

my focus has shifted quite a bit since i started working on the rogomatic code.  rogue is still fun, but now it is more of a part of the whole puzzle that i work on to debug for rogomatic.  once in a while i do find a bug in rogue so any fixes i find might be useful to others.

a line count of about 18,000 lines of C for each of rogue and rogomatic makes it a tiny program compared to some other games.

anyways, i'm pretty sure i severely limit the audience of the code i distribute simply because i don't do any version for MS nor do i distribute binaries of any kind.  yet the rogue part should be fairly easy.

11
now at:

   www.anthive.com/rog/rog.html and
   www.anthive.com/rog/other/things.html


version 1.1.1 for rogue

version 2.0.1 for rogomatic

version 3.0.0 for debugging scripts

the fixes and workarounds in the rogomatic code are to get past a few bugs that have been hanging around quite a while, but are rarely encountered.

one bug you will see is that when scrolling happens it corrupts some underlying information used by searching.  to work around this i've created a slightly modified vt100 to put in the users .terminfo/v directory and that will keep the program from scrolling.  it is a work around, but it is good enough for me until someday when i can figure out the underlying bug.  the users environment variable TERM should be set to vt100.

the other bug is a followmap direction invalid error.  instead of going through all the searching code again (i looked at it several years ago but have not gotten back to it) and tracking down why the value is being returned out of range i just catch it when it comes back out of range and adjust it then.  another work around, but it is a rare bug and this is the least intrusive fix for now.

the other changes to rogue, rogomatic, debugging scripts are for improving the speed and reliability because under the old method i was getting hung up during testing when the processes were finishing too quickly for the scripts to find the PIDs needed to know what was running.  now both rogue and rogomatic will tell me their PIDS when i ask them (using environment variables GETROGOMATICPID and GETROGUEPID).  a normal user will not see these new files appear as they would have no reason to set these variables.

Testing, bug fixes, etc. continues when i have time.

More details than you probably want to know are included in the NEWs, ChangeLog, TODO, NOTES, etc files...

have fun!

ant

12
Classic Roguelikes / Re: Rogue - let's beat it
« on: December 01, 2014, 03:35:56 AM »
there are many versions so that would be an important factor to know.

i play the versions i have installed here from time to time for testing them out and to make sure they work as expected for the version of rogomatic that i am working on.

some versions are much easier than others.  :)

13
Traditional Roguelikes (Turn Based) / Re: rogue and rog-o-matic
« on: December 01, 2014, 03:15:52 AM »
The fun continues:  :)

My versions adapted from others:

(use these together for best results)

rogomatic is now at version 2.0.0
rogue is now at version 1.1.0

www.anthive.com/rog/rog.html

  I did redo most of the database code for keeping track of interesting items while working on fixing the invalid inferences problems.  The identifying and reading scrolls of identify has also been worked on.

  I also fixed some of the older bugs that I'd found before.

  More remain, like the "Followmap: direction invalid" error and some searching issues, and then there are general gaps too in that certain scrolls or potions provide information that is ignored.  So there is a lot of room yet for development.  I'm not sure how much of it I will actually do until I get to it.

  Right now I have a few weeks where I'll be doing other things and I have to decide if I want to work on the searching related stuff or to fine tune what I've got already...  Hmmm...

14
Traditional Roguelikes (Turn Based) / still going - rogue and rog-o-matic
« on: November 16, 2014, 12:34:13 AM »
updates ongoing.  at version r1.0.9 for both rog-o-matic and rogue (just by accident the versions do match at the moment).  debugging scripts at r2.0.5

current debugging in rog-o-matic to fix database code.  i'm leaning towards redoing completely.  would be faster to do something a bit better than keep trying to patch this current setup which lacks in some important ways.  this would also let me revert a few cosmetic changes i made to rogue to work around the bugs in the rog-o-matic code.

bugs in rogue, on the other hand, are much easier to work around or fix.  :)

http://www.anthive.com/rog/rog.html

15
Programming / Re: Slow progress is slow.
« on: September 08, 2014, 03:06:21 AM »
i know how it can go at times...

what is your platform coding language?  is it text file descriptions that get turned into some other language and then compiled or???

ant

Pages: [1] 2