Magnate's new egos

[sethos]

I do hope that my midnight mace of majesty Didn't really make anyone upset. I haven't been playing Angband for more than a year, and I'll admit that the flavor of it is charming in it's simplicity. I do think that a bit more variance is a good, no, great thing. However, there will be the issue of BALANCE, and as Magnate has said - that will be addressed. I do know that the changes have been really really fast, and that the "NEW" vanilla will not be the same as the "OLD" vanilla. Perhaps it will one day split off and become a variant - I don't see the problem with that. Then again - I WAS about 5 years old when "Vanilla" came out - shouldn't there be room for growth and change over DECADES? the fact that it's happening all at once? not a bad thing. once all the new flavor and variance is added, who's to say that Angband cannot again become as difficult as it once was... or even more so. the variance in the items will HELP that, I believe.

Example - how long would YOU spend in Diablo 2 looking for that "perfect" set of gear? it could now be made very tough to get exactly what you want in angband - resulting in more variance in the endgame, and just maybe, a new thrill with each character that you might not be able to achieve with what we had in the old angband.
Anyways, Anyone with the time and skill required to update and modify a game - and keep it interesting to at least some people, ought to be lauded, and not criminalized.

Good work on the items magnate - I'll be thrilled to see the new form of item generation, and I'll do my best to help with balancing - You might not want that "midnight" theme to show up, though. might be a bit too much - as I believe I stated at the start. Just got excited, y'know?

Sorry for the rant - It's just not cool to give lashes to people who work at making something - if you don't like it, don't play it. If you want to play, try to be constructive - or join the DEVTeam.

[Antoine]

If you want to play, try to be constructive - or join the DEVTeam.

In a spirit of being constructive, can I say that multiple pvals are pretty cool.

A.

[sethos]

In a spirit of being constructive, can I say that multiple pvals are pretty cool.

A.

AMEN to that. NPP needs that... think I'll go post @ tonnppaf.

[nppangband]

AMEN to that. NPP needs that... think I'll go post @ tonnppaf.

No need. I am here too. It is already on my to-do list.

[fizzix]

Well, I'm on record as suggesting you stick with the status quo on all counts for 3.4.

For about 3.8, I'd be happy with a "yes" to questions 1-5. For question 6, I think it should be sufficient for some mobs to drop more items, with a higher probability of egos/artifacts. It shouldn't be necessary to veto those mobs from producing low-grade egos. Those can be squelched or sold.

A.

Ok, can I ask about this in a bit more detail? Assuming 3.8 means 5 releases from now, there must be some sort of path forward to get from here to there. Maybe one of those changes for each release? I don't know what the ideal path is. However, if you move to affixes and don't implement even one of the above changes, then there's really no point, and I think this is one of Magnate's issues with not adding anything new in the next release. So, how do we get from here to a Vanilla 3.8 that has changes 1-5 in it?

On a side note, let's say we take the incremental approach and do 1 change for each version between 3.4 and 3.8. Each of those changes will require a significant amount of playtesting and rebalancing. This is very time consuming and difficult. Even for someone like me who loves running simulations and analyzing statistical variations, I don't think I'd be up for doing a rebalancing effort 5 times in the next 2.5 years. This is probably the main motivations for doing a large chunk of potentially unbalancing changes at once.

[Magnate]

Just to head off too much more wrangling: following this debate with Antoine (and numerous identical previous debates with Timo and others), I've closed the pull request and decided not to push for the incorporation of affixes into 3.4. I strongly suspect I have merely pre-empted the decision takkaria would have made, but I'm pleased to have made the decision myself.

Fizzix and I are doing some preliminary balancing work on affixes. After that, they may or may not be made available in a playable version somewhere, but wherever they end up, it will have nothing to do with V.

[Nomad]

That's a real shame. Well, I for one would definitely be interested in seeing a Windows playable version of AffixBand if it's made available.

One aspect I wonder if we could still incorporate into V is the ability to assign things different depths at which they rank as good/great/uber, which seemed like an awesome idea to me. Could we potentially replace the current flat "min depth to max depth" allocation on egos and objects with something like that? So instead of an ego of Flames, min depth 10, max depth 100, you might define an ego of Flames, uber at 10, great at 25, good past 50.

It seems like it would allow much finer control over drops than the current system where high-end egos are just as likely to show up at their min depth as later in the game. It could potentially be used to refine pseudo-ID and quality squelch as well.

[bio_hazard]

I'll just go on record saying I'm a little sad about that as well. I'm not ready for ToME4-like explosion of egos, but a little more flavor would have been nice.

Is there any argument for adding in the affix system in a flavor-only way now? So you examine your short sword and see a "well-made short sword of orcish construction" or "heirloom quality short-sword of hobbit-make" or something similar, that at this point would be randomly generated and have no effect on the bonuses or attributes of the item?

[Antoine]

Ok, can I ask about this in a bit more detail? Assuming 3.8 means 5 releases from now, there must be some sort of path forward to get from here to there. Maybe one of those changes for each release? I don't know what the ideal path is.

Don't ask me! I'm a variant developer, I just change whatever I like whenever I feel like it. You guys have a much harder job

A.

[Antoine]

Fizzix and I are doing some preliminary balancing work on affixes. After that, they may or may not be made available in a playable version somewhere

I'm sure they'll be great in that context, and look forward to them migrating back to V from that variant in a judicious manner

A.

[EDIT: Should we get these changes into V now?
* "Special" artifact generation is gone -- the game either generates an artifact from the list of available artifacts, or it generates an item. Hooray!
* Allocation probabilities are on a 1000-point scale instead of a 100-point scale, so we can make Ringil, Feanor, etc. 10 times more rare. But artifact rarity is now independent of base item rarity, so all rarities will need to be rescaled.

And should we be looking at incremental fixes to these problems? http://trac.rephial.org/wiki/ObjectDistribution
?]

[ogozi4]

I'll Try

Looks nice, i'll definitely try it.

[EpicMan]

I concur that adding in the new affix-based item generation should conincide with new egos, (+0, +0) egos, etc. If you completely and drastically change how items are created, you almost certainly are going to alter the balance from what you had before. Therefore adding new stuff that needs to be balanced should be thrown in at the same time, because then you can get it all balanced out at once.

Furthermore, there are issues right now with item generation/distribution and game balance, so altering/breaking balance is not a valid reason to not make this change because balance is messed up right now.

I am sad to see these changes won't be in 3.4; I hope they will go in 3.5.

[Derakon]

Furthermore, there are issues right now with item generation/distribution and game balance, so altering/breaking balance is not a valid reason to not make this change because balance is messed up right now.

I don't really understand this logic. If things are broken now, then we can either fix what we have now and then replace it with something which will be "broken" until fixed again, or we can just skip the first "fix it" step and replace one broken thing with another broken thing.

About the only reason I could see for first fixing the current broken system (beyond things like concerns about the pace of new development) is to establish a baseline for what we want object allocation to be like. But we have Angband 2.9.x / 3.0.x for that.

[Iapetus]

I am sad to see these changes won't be in 3.4; I hope they will go in 3.5.

Me too. I don't know about the development stuff but as a player more variation in egos would be good thing and personally I don't see any reason not to change them. Granted balance will be an issue but this change provides more tools/options to do the balancing.

Anyway, thanks to everyone involved in developing Angband and it's variants I've wasted a lot of time playing them.

[EpicMan]

I don't really understand this logic. If things are broken now, then we can either fix what we have now and then replace it with something which will be "broken" until fixed again, or we can just skip the first "fix it" step and replace one broken thing with another broken thing.

About the only reason I could see for first fixing the current broken system (beyond things like concerns about the pace of new development) is to establish a baseline for what we want object allocation to be like. But we have Angband 2.9.x / 3.0.x for that.

I was a bit unclear. I was actually trying to say what you just said but apparently failed badly