Angband.oook.cz
Angband.oook.cz
AboutVariantsLadderForumCompetitionComicScreenshotsFunniesLinks

Go Back   Angband Forums > Angband > Vanilla

Reply
 
Thread Tools Display Modes
Old September 21, 2010, 18:55   #91
Magnate
Angband Devteam member
 
Join Date: May 2007
Location: London, UK
Posts: 5,057
Magnate is on a distinguished road
Send a message via MSN to Magnate Send a message via Yahoo to Magnate
Quote:
Originally Posted by Derakon View Post
Talking about artifact frequency, here's a random thought: what if you could tune how mean/nice the RNG was for drops and monster placements? E.g. have two game options that allow you to assign weights from -10 to 10 for monster OOD frequency and item drop quality. A setting for 0 would use the current behavior; setting for -10 would make monsters frequently be OOD or make items almost never be of high quality (effectively reducing ego-item / artifact drop rate); setting for 10 would make monsters never be OOD even in vaults and would significantly increase how often good items drop.

Basically they're difficulty tuners. The game would only be intended to be balanced when they're set to 0, but if you want a greater or lesser challenge, they give you ways to achieve it that don't rely on powerdiving, save scumming, etc.
Hugely radical, very interesting idea. One thought is that it makes it harder to filter subsequent observations, as you'd need to know what level each poster set the tuners at and whether they had departed from their normal setup for this game (to account for perception bias). You could print the tuner levels in dumps, of course.

It's actually surprisingly easy to code such an option. But I'm conscious that we're trying to beat a path back to the days of 3.0.x or even 2.9.x when wins were rare and didn't happen for new players for a while. At least, that's my understanding of the response to the "game is too easy" threads, but I may be over-reacting.
Magnate is offline   Reply With Quote
Old September 21, 2010, 19:53   #92
Derakon
Prophet
 
Derakon's Avatar
 
Join Date: Dec 2009
Posts: 8,504
Derakon is on a distinguished road
That point on measuring peoples' observations is a good one. It's hard enough right now to tell what "too many artifacts" means and if things have changed from previous versions. I don't have a good answer for that.

As for moving back towards a harder base game, I wouldn't see these options as preventing us from changing the reference frame. A "balanced at 0" game might be easy or hard; the options would just be there to make it easier or harder, depending on taste and aptitude.
Derakon is online now   Reply With Quote
Old September 21, 2010, 21:20   #93
Magnate
Angband Devteam member
 
Join Date: May 2007
Location: London, UK
Posts: 5,057
Magnate is on a distinguished road
Send a message via MSN to Magnate Send a message via Yahoo to Magnate
Quote:
Originally Posted by Derakon View Post
That point on measuring peoples' observations is a good one. It's hard enough right now to tell what "too many artifacts" means and if things have changed from previous versions. I don't have a good answer for that.

As for moving back towards a harder base game, I wouldn't see these options as preventing us from changing the reference frame. A "balanced at 0" game might be easy or hard; the options would just be there to make it easier or harder, depending on taste and aptitude.
Agreed on the latter point. I do actually have an answer to the observation issue, which is to gather hard data. Takkaria pointed me to wiz-stats.c, which I plan to expand into a proper monte carlo test suite for item and monster generation. That way I can tell it to generate a million monsters (or items) and get output which tells me all about which artifacts drop at which depths, whether stat potions are too common or too rare etc. etc. Then each set of changes can be re-tested to see how far they've changed those results. I think it's high time we had a rigorous test suite for both monsters and items, but particularly for items.

The thing is, this is like number 31 on my ever-increasing to-do list, and I spend far more time adding to it than reducing it. That's partly RL - work has been crazy this summer and I'm hoping it will get a bit easier this winter and allow me to have some free evenings. But also because I'm getting less and less focused in what I want to do. I think this one might actually be pretty important.
Magnate is offline   Reply With Quote
Old September 21, 2010, 23:17   #94
Tiburon Silverflame
Swordsman
 
Join Date: Feb 2010
Posts: 405
Tiburon Silverflame is on a distinguished road
Oh yes, standalone testbeds are incredibly happy.

Down the line, an even more complete approach...one that incorporates some of the more complex variables involved...would be to have the capability to generate a full dungeon level, of whatever depth, with monsters. Then there's perhaps 2 options:

--Dive strategy: @ would probably kill 25% of the monsters and find 25% of the items on the floor
--Clear strategy: @ would probably kill 90% of the monsters and pick up 90% of the items on the floor

Kill the appropriate percentage of the monsters, and add their drops to the floor loot gathered.

Next point: create every 5th level twice; these are the levels you recall from, then back to. This holds except for stat gain territory, up until probably around 90th level, or whichever level the diving pros would say you park to get your final kit built up. Run the level pair (say, level 96 and 97) 10 times each...?

Toss in...oh, let's say 5 scrolls of acquirement, and 3 *acquirement*, at some plausible depth.

One trick is that artifact generation will have to go through a side loop in the test code here, because this is where the artifacts already found, would be kept. Another issue is what to do about pits (we love clearing troll pits, but a big-time undead pit is likely gonna be avoided) and vaults.
Tiburon Silverflame is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
very low priority feature request - metric system option bebo Vanilla 20 April 28, 2012 10:08
Missing the top line in character file dumps. Kyle Development 2 July 25, 2009 20:57
feature request - spellbook number OPTION PowerDiver Vanilla 18 July 18, 2009 19:16
[Feature Request]: 'g' gets top item from stack awldune Vanilla 0 December 14, 2007 22:31
RGRA portal stealth post feature zaimoni Oook! 13 December 13, 2007 18:53


All times are GMT +1. The time now is 01:01.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.