Announce DaJAngband 1.2.0 released

[will_asher]

okay I added spells3.c to Makefile.src for the next release. Do I need to do that for those other makefile files too? I've never messed with makefiles.

[zaimoni]

The point of Makefile.src is that you should only have to specify the *.c files once. You should be fine.

[Nick]

Originally Posted by zaimoni

The point of Makefile.src is that you should only have to specify the *.c files once. You should be fine.

Should - but I know in FA at least there are some ports which still have the file list included. I tend to search through all the Makefiles for one filename (spells2, for example, is unlikely to crop up in a Makefile for any other reason) to make sure.

[SaThaRiel]

Well, Makefile.src is included by Makefile.inc which is included by Makefile.std and Makefile.osx. I think this are the only 2 you really should support at the moment. .win is done by you and there was a discussion about dropping Risc-OS (.ros) in Vanilla already. So here you are "clean" when you just add the new "to-build" object-files in Makefile.src.

Another idea might be to rewrite the build process to use a configure script which makes it a lot more flexible and easier to handle for the one who compiles it. But i havent really looked into the generation of configure scripts so i dont know what this will mean for you. But i may look around and gather some useful informations. The point is that, IMO, you will mainly find ppl using Linux or BSD-based systems playing *band-variants. Windows has "better" games (actually many ppl just use Windows to play games that arent supported on another platform and use another OS to "work").

Edi: This looks like a nice tutorial/intro: http://inti.sourceforge.net/tutorial...lsproject.html

[andrewdoull]

Originally Posted by will_asher

okay I added spells3.c to Makefile.src for the next release. Do I need to do that for those other makefile files too? I've never messed with makefiles.

In my experience, I've found avoiding adding additional files the best way to go. It minimizes the amount of messing with the make files and on modern processors you don't really notice the increase in compilation times.

Andrew

[will_asher]

The download for DaJAngband 1.2.0 has been updated with a few minor changes:

fixed bug: disguised monsters (mimmics) are a little too easily detected.
fixed bug: NONMONSTERs (trees & statues) can cause a disturbance when running.
removed stealth from description of trees & statues.
score is now shown on (c)haracter screen and character dump (in place of Maximize Y/N)
fixed kindof a bug: grenades have no graphic tile
fixed bug: can't enchant grenades
power sprite player race can fly over rubble effortlessly

I also slightly tweaked the rarity of the slaying and pseudo-randart egos.
The download link on the announcement post has been updated.

This update is mainly to prepare for a DaJAngband competition: I wanted the scores shown on character dumps since I redid the scoring.
This is how the redone scoring works:
- score for each monster you kill:
Morgoth: 1000pts
Saruon: 500pts
other uniques: score = monster level (minimum 5)
other monsters: score = monster level / 5 (rounded up)
there is a slight bonus for the 1st (nonunique) monster of each race you kill which is based on the monster's rarity.
if you kill a whole lot of the same monster race, the score for additional kills of that monster race get reduced.
*Monsters which come back from the dead are worth half the points for each kill because it is assumed you will have to kill each monster at least twice.
- score for each artifact you find (and recognise as an artifact)
- score for experience:
score for XP works by finding the 'simple' character level of the character based on a 100% experience factor (not modified by race or class).
your 'simple' character level x 10 is added to your score
- score for max depth = max depth x 4
- score is reduced 1 point for every 10000 game turns (that about 2 hours and 46 minutes of in-game time)

[Derakon]

Your scoring system sounds flawed, since it encourages farming monsters for points. How about this alternative? Give points for the first kill of every monster type, based on the difference between your level and the monster's level (more unfavorable comparisons == more points), and with a bonus for unique monsters. So for example, your score for the first kill of a monster could be given by max(1, mlvl - clvl) * 2, multiplied by 5 for uniques, 10 for Sauron, and 20 for Morgoth.

This would encourage cherry-picking your fights and going for low-level games, though...but it does strictly cap your potential score, which is a good thing in my opinion. You can toss in score decay over time as well if you like.

[SaThaRiel]

Just one little point: If you bring new code/changes to a release it would be good to also change the version number Also if its a minor change.

[Nick]

Originally Posted by Derakon

Give points for the first kill of every monster type, based on the difference between your level and the monster's level (more unfavorable comparisons == more points), and with a bonus for unique monsters. So for example, your score for the first kill of a monster could be given by max(1, mlvl - clvl) * 2, multiplied by 5 for uniques, 10 for Sauron, and 20 for Morgoth.

The FA 1.1 score is close to this - you just get the experience for each first kill added to your score. This automatically rewards killing hard monsters.

I got this idea from Eddie's suggestion that you should only get experience for first kills...

[will_asher]

Originally Posted by Derakon

Your scoring system sounds flawed, since it encourages farming monsters for points. How about this alternative? Give points for the first kill of every monster type, based on the difference between your level and the monster's level (more unfavorable comparisons == more points), and with a bonus for unique monsters. So for example, your score for the first kill of a monster could be given by max(1, mlvl - clvl) * 2, multiplied by 5 for uniques, 10 for Sauron, and 20 for Morgoth.

I guess it's probably a good idea to only score for first kills. I'll probably change it. I wanted to see the scores for the competition so I can see how it turns out with my original method. I don't want to use the scores to decide who wins the competition yet though, because I haven't finalized my scoring system.