"Nick is going to butcher the game"

[Philip]

It's interesting to me that what the official version is is so important to people who will swear that there is one true perfect Angband. It feels like unintentional validation of Nick's theory that the game has to be maintained in order to survive.

Again, all of you underestimate the extent to which the game has changed in the past. The classes are, perhaps, more of a legacy than most aspects, but even then, did you know Mages used to be a heavily melee class in the late game (before Globe of Invulnerability was removed)? Later still, they were effectively a more utility-focused ranger (they then lost most of Tenser's which made this strategy a lot weaker).

Anyway, I think it's funny that dos350 is back talking about how all of these changes are ruining Angband because a) that's what tangar reminded me of this whole time and b) dos350 would always inevitably switch from saying that 3.1.2v2 was the one true angband and the 3.2.0 devbranch was an abomination to saying that 3.2.0 was the one true angband and that the 3.2.1 devbranch was an abomination shortly after 3.2.0 officially came out. Let's see if that pattern holds up.

[dos350]

thanks for reply nick, u really are doing a good job talking to the community about this, but theres more to it than talking,


alot of users do not follow the development or use the forum, but they download the latest versions in hopes of enjoying the game they love, the game they know, i dare to say most of this category hope for general improvements, and that does not necessarily mean changes to the lore or class/item/combat systems, or additions to the game OR things being removed entirely (besides options)

i think very small incremental changes to these things are better than an all at once overhaul,

not only will the final product be more refined but it will not run a risk of shocking and terrifying returning players who do not follow the development

things like "fixing randarts" is an important thing to do, but personally i think that having wide range is fun , they are rand after all,

things liek removing search though, well thats a slippery slope
,,, i grew to accept it but i question its value and in the context of legacy i think maybe its the wrong move, same with rune id

even considering removing hunger however, i believe is a huge mistake

people in favor of massive changes please understand that once classes and monsters that have been there for 20years+ get changed in a big way, or removed or replaced, angband is not far from becoming a variant in the eyes of traditionalist players, and although that may not matter to the developers, it does really mean alot to these users

play an older version is not an answer, wats important is the live version because thats the default for all but the most hardcore users

i know its not possible to please everyone, but to me angbands history and living legacy is more important than a direction for development

[Nick]

the people opposed to the sweeping changes are clearly being dehumanized in this thread, treated as a joke or brushed off, or spammed into oblivion~

i havnt seen a single person abusing the users who are for the changes, just trying to show another perspective in a polite manner,

I think dehumanized is a bit OTT, dos, but I take your point.

As maintainer I'm in the odd position of advocating for the changes I'm making - because if I wasn't in favour, I wouldn't be making them (apart from rune ID, but that's another story) - but also needing to be a bit even-handed. IMHO I haven't always got the balance right, hence the apology upthread.

I don't think there's a good general answer to how change like this should work. I will try to list my reasons for change, though, so at least everyone has a shot at understanding why I'm doing stuff.

1. People are complaining about it. A good example here is the current situation where randarts are overpowered - it's not working as intended, everyone can see that, and it just needs fixing.
2. It has been discussed for a long time and there is a general acceptance of a needed change. Rune ID is the classic here. Alternatives to the old ID scheme had been discussed over a long period, and rune ID had emerged as the favourite contender (I initially preferred an alternative scheme that I had used in FA, but I now think I was wrong).
3. There is a game mechanic which has become pointless, usually because of other changes in the game. Traps and curses (and bad objects generally) are a good example of this, and possibly hunger falls here too. In this case either the mechanic can be removed altogether (potions of Death), or if it is seen as too integral a part of the game it has to be re-made in some way that fits with the current state of the game.
4. There is something which seems like it was a good idea in theory, but doesn't work that way in practice. This is how I see the point that the old classes had got to as of 4.1. For example, mages had 57 spells in 9 books, but one of the books only had 2 spells, and many of the spells were in practice never or rarely used. So it seemed like redoing classes to have less spells, but actually useful ones, would be a good idea. There was a big thread about it. The problem here is that redoing a class, or making a new class, involves a lot of creative work, and everyone is going to have their own opinion about the outcome. So in this case "Can't we just go back to the way it was?" has a certain amount of appeal.
5. Something has been a particular way for a long time, and it's not exactly bad, but it could be better. This is probably the most difficult one, because a lot of opinion comes into it, and it's how I see most of the monster changes. I'm relying on my own instincts a lot, but also on other people's opinions; note that I have made a lot of the changes suggested by people, particularly those who have playtested the changes. This sort of change is probably the most fun to do, but also the most difficult because of everyone being different and having different personal histories with the existing monsters. The arguments I have tended to push back against the most are the "you shouldn't change it because I don't want it to change" variety, which is understandable but (as I've explained a number of times in different ways) in my opinion doesn't get us anywhere.

So, there it is. Wall o' text I'm afraid, but whatever

[Youssarian]

dehumanized?!?

Perhaps degnomed?

[takkaria]

the people opposed to the sweeping changes are clearly being dehumanized in this thread

dehumanized?!?

[dos350]

but you don't have the right to engage in verbal diatribe against the efforts of the maintainer. That's just a dick move. Seriously, I don't understand why people have to be told these kinds of things.

the people opposed to the sweeping changes are clearly being dehumanized in this thread, treated as a joke or brushed off, or spammed into oblivion~

i havnt seen a single person abusing the users who are for the changes, just trying to show another perspective in a polite manner,

angband is important and needs to remain angband at all costs

[Carnivean]

In the 5 years I've been playing Angband I don't remember changes to the game as drastic as the ones that are occurring now. It might not be a bad idea for Nick to slow down a bit; pick one aspect of the game to overhaul and get it right before going to the next.

I can think of 2 main reasons why an overhaul should be done holistically rather than by honing a single piece at a time:

1) Nick has finite time on his hands to maintain and update the code. Following the 80/20 rule we'll get significantly more complete per unit of Nick's work than if we get him to spend more time polishing (in the 20%). Crowd-sourcing the balance is going to help and while that happens Nick can go on ruining other parts of the game. Your thoughts on shield bashing might have come to Nick eventually, but over how long a time?

2) Parts of the game don't exist in isolation. They exist in balance with and in tension with each other. Honing a single change and bedding it in is a waste of time if, when the next area is changed, Nick needs to go back and redo the balancing against the new changes. Rather he should introduce as many changes as he has in mind to functionality and then iteratively hone them to a balance. Yes the development versions might have bugs and weird situations and even parts that feel wrong, but they are development versions. Come a full release I'd expect a cohesive, "balanced" game to have emerged.

Finally, I think I've got a bit too snippy a couple of times in this thread, sorry about that. Maybe it's the title

With your patience you might want to consider becoming a monk.

[Derakon]

Done.

[Hounded]

I confess I am weary of seeing this title at the top of the forum every day (There I go, adding to the popularity by responding to it). Perhaps I am too polite for Tangars taste but the title feels combative.

Is there any way the genuine feedback discussion(s) can be pulled into the development discussions thread rather than fueling this one?

[Nick]

In the 5 years I've been playing Angband...

(Shortened )

Thanks for a very considered post, Grotug.

I won't answer everything, but I'll say a few things that come to mind immediately:

1. The suggestion that I shouldn't have done two major changes in one release is quite reasonable. Possibly that's a mistake, which I'm still currently making. The reason I'm inclined to persist with it is that the new classes needed balancing anyway, and it seemed like double-handling to balance them for the existing monster list, and then need to reconsider. The other reason is that this is taking a long time anyway, and I'm impatient
2. Your reasoning about shield bashes is excellent and helpful. I think having the chance and quality of bashing depend more on the weight of the shield and less on other things (currently DEX, to-hit, player weight and gear weight) is a good idea, with probably no bashes from wicker shields, and progressively more as you get heavier. I like the idea of factoring in the monster properties more thoroughly, too.
3. Yes, the randart situation is terrible, but I was trying to get the big changes done first to give more time for them to settle (see point 1). It is high on my priority list.

On how to handle having the old classes available - I need to think about that a bit. I have used the "copy this file here" method rather than in-game options for a few things (eg full monster lore, using old randart sets), but it's worth having a reconsider. Reloading files based on a birth option is already done - if randarts are chosen as a birth option, the standard artifacts are reloaded so that the new randart set is based on them, rather than on some other randart set. I note Mars and fph's words of caution, too.

I should say, too, that I'm going to start specific, targeted (well, hopefully) discussions too on the new classes and on the new monsters and possibly on artifacts, randarts and egos (more scope creep...) in order to work out in detail what should happen (for example the new classes have many fewer books - is it too few now?).

Finally, I think I've got a bit too snippy a couple of times in this thread, sorry about that. Maybe it's the title

EDIT: Also meant to say something about my approach to monster changes. One of the reasons for the whole thing is that the game has changed considerably around the monster list, and more possibilities are available now, so I was trying to re-imagine the monster list in that light. For example, monsters used to be technically limited to four blows; they mostly still are for balance and predictability reasons, but it enabled an obvious variation for hydras.

[wobbly]

shield bash I thinks needs looking into, as it's something that works well in O's different style of combat & doesn't necessarily translate so well in V. Not saying it's bad or good, it's just, I don't think it's exactly playing right.

In regards to the latest changes being dramatic, I think maybe load up an older version of V or old-style variant & take a look? TO change is more dramatic then anything that changed in monster list or class balance. Hound & pit quanity another. A bunch of other changes. Older variants definitely have a very different feel in how they play.

[Grotug]

In the 5 years I've been playing Angband I don't remember changes to the game as drastic as the ones that are occurring now. It might not be a bad idea for Nick to slow down a bit; pick one aspect of the game to overhaul and get it right before going to the next.

While I am by nature more prone to align with the "preserve sacred game" mentality of Tiberius and Tangar, I'm also open to hear the arguments for why that sort of mentality is perhaps not very wise at all and may even be misguided. And I don't get the impression Tangar and Tiberius have really heard those arguments. I actually found Tangar's post very compelling and very similar to the type of sentiment I might have about something dear to me when I read it. But I also found the counter arguments very compelling, too, and the sheer number of them to be quite persuasive. I also don't feel I am very wise in terms of knowing what is best for Angband in terms of its lore or themes, given my limited knowledge of Tolkien and D&D. So... weighing that all in my mind, I don't really feel I'm in a position to say that some kind of butchering is going on; but I do relate to the tendency to feel that way when I see drastic changes happening: as such, I avoided the new branch for a long time because I felt the class changes were way too drastic, which I guess is why I think maybe not change two major aspects of the game simultaneously.

I'm not sure I agree with the idea that the game always needs to change or it will die. I think the game should be changed for the purpose of perfecting it, not out of fear it will stop being interesting. The game is well interesting enough as it is. It's the balance that could always be improved, imo. The games imbalance or bugs should be fixed before any overhauling should take place. The most glaring bug in my mind is the randart weapon power bug. Seems to me it should have immediate priority over everything else to be fixed.

With regards to the need to change something to keep it fresh, I'll refer to car design as an example that affects me personally. Car manufacturers feel they need to update how their cars look each model update, changes that are purely aesthetic (and to some extent to increase fuel economy). Engineering updates make sense; technology is always improving what cars can do and their safety. But for a long time now cars have not been getting any better looking (imo). They peaked in the 90s. The Mclaren F1 is still the best looking supercar of all time, yet new mclarens didn't bother keeping that perfect look. The "purist" in me complains about this. The purist in me also complains about new doom games (the original doom is still the best in my mind). Fortunately Nick isn't removing any monsters from Angband that are iconic (imagine if some maintainer of Doom decided demons, imps and cacodemons didn't fit in Doom!) Good thing the equivalent of that isn't happening to Angband under Nick's watch. I guess my point is, the attitude that Angband should keep being changed to keep it from dying seems too close to the idiotic car manufacturing logic of: thing should be updated for the sake of being updated. Car manufacturers update the look of their cars when the look of their cars was just fine and didn't need an update. Angband should keep being changed because something about it can genuinely be improved by being changed, not out of a fear of it becoming stale. That all said, let's imagine a scenario where Angband, by the power of some super-genius maintainer achieved perfection in the eyes of most players. This does not mean the game will remain unchanged for eternity and become stale, even in this unlikely scenario, what will be perceived as perfect will change in time (by virtue of tastes changing and technological advancements), and so valid impetuses to change/improve Angband would still come about.

I am not so married to the Lore of the game to be overly bothered by the lore changes, and the lore changes seem to be by and large reasonable: few monsters I'm attached to seem to be getting the axe (I'm still hoping Osse survives). This is a good opportunity to again stress that maybe too much change too quickly to a beloved game is not a good idea. Giving the monster lore more cohesion seems a good idea; going full Tolkien on the lore and removing D&D elements seems maybe too much change too soon; better to make small changes, release to public, see how they like, first. I don't even have much experience with D&D, yet I would like to see most of the D&D stuff remain, because I feel like it has a long-standing heritage and I like the idea of conserving that heritage. Of course, if Nick is more wise than Tiberius, Tangar and me, then with the passage of time, it will be revealed that the lore changes do in fact make Angband a better game.

I've never been crazy about the classes overhaul, but since I mostly play Warrior it doesn't affect me all that much. That said, I am not terribly happy with the current implementation of shield bash. I think its implementation is a bit crude at the moment: it's both too powerful and at times annoying.

1. Too powerful: even with a dinky wicker shield (which can be bought for 2 gold pieces) you can stun and confuse monsters, almost for free during melee.

2. Annoying: When @ learns a monster cannot be stunned or confused he still bashes it with his shield.

What I propose: the size of the shield should affect the efficacy of the shield bash. How much damage @ does with his weapon should affect how often he attempts a shield bash. If you are doing 800 damage per round, shield bashes are sorta pointless. Overall, right now shield bashes are a bit too frequent.

If shield bashes cannot be fine tuned, I'd like them to removed, or have their frequency greatly reduced.

I think a simple solution if Nick (or anyone else) doesn't want to spend a lot of time fine tuning shield bash, is to have only large and mithril shields give shield bashes, and reduce how often shield bashes happen. Or, only artifact shields should have the possibility of giving shield bashes.

Fine tunings I'd propose:

If @ can kill monster in less than 4 rounds, do not attempt shield bash.
If more than 3 rounds but less than 7 rounds is required for @ to kill monster, attempt shield bash early in the fight, but don't bother once monster is below 50% health.
If @ requires more than 7 rounds to kill monster, attempt shield bashes regularly.
If @ knows a monster cannot be stunned or confused, never attempt shield bash.

I don't know what percentage shield bashes get attempted currently, but whatever that number is, that probably should be the max @ attempts them, the criterion for that being he has a large or mithril shield and/or an artifact shield with a shield bash buff and the monster he is fighting will take more than seven turns to defeat.

I guess in summary I should say that since Nick is attempting a lot of big changes simultaneously, he take all the necessary time to ensure that they are super fine tuned and balanced before calling them good enough. And let's not forget to fix the current game imbalances, too.

It's actually kinda interesting people get more up in arms about the big changes to familiarity but they seem less concerned about changes that affect gameplay balance (I seem to be the only one bothered by 8d5 branded/evil weapons being generated on DL36 with a 6 treasure feeling). Two changes that are affecting my enjoyment of new angband the most are relatively very small: extra shots nerf (boring!) and shield bash (crudely implemented). I think in time I'll probably (hopefully) grow to see the new classes and lore changes as improvements to Angband, but I'm less convinced I wouldn't be much happier to see Ranger lose extra shots in its current nerfed form and have it replaced with extra might.

[Derakon]

If you're going to have an officially maintained and distributed alternative configuration file--which is what the old_class.txt provided with the 4.2 development builds actually is--then there is no reason not to make it a birth option instead. The code complexity would hardly be increased: the only changes would be the addition of a global variable to tell the program whether to read from class.txt or old_class.txt whenever it needs to reference class configurations, and a birth option to set that variable.

Unfortunately it's not quite that simple: unless I miss my guess, the data files are loaded prior to character creation/loading, so either you need a pre-loading options file, or you need to delay options loading until you've processed the "use old classes or new classes" question during chargen. Either is likely to be nontrivial.

Think of the old vs. new classes as like being a game mod, because that's effectively how they're implemented. Most games either just blindly load all installed mods as part of startup, or have a prompt screen that asks you which mods to load before you get into the game proper. Angband happens to do the former, and you're basically asking for it to switch to the latter.

[luneya]

The other danger is having too many things to keep tested, bug-free and balanced. Same reason why many uncommon game options got removed from the = menu.

If you're going to have an officially maintained and distributed alternative configuration file--which is what the old_class.txt provided with the 4.2 development builds actually is--then there is no reason not to make it a birth option instead. The code complexity would hardly be increased: the only changes would be the addition of a global variable to tell the program whether to read from class.txt or old_class.txt whenever it needs to reference class configurations, and a birth option to set that variable. Whether you make these changes or not, as long as class.txt and old_class.txt are both part of the official distribution, it is the maintainer's responsibility to ensure that they are at least tested and bug-free. Balance is less of an issue in such cases because these are just options; if you don't think they're balanced correctly, either modify the configuration file yourself or just play with the default settings, for which the maintainer does have an obligation to ensure balance.

A case can be made for eliminating the old classes entirely, as it does take some work to ensure that changes in the rest of the code don't break old_class.txt. But if Nick thinks it's worth the trouble to maintain the legacy classes as an official configuration file, then there is absolutely no case for not going further and including them as a proper birth option.

[Youssarian]

The danger of that approach is feature bloat, which is detrimental for both maintainer and player ("gee, which of these gazillion things to choose from. Meh, never mind.").

Yes, but I'm not talking about leaving everything in the game that was ever there and just adding to it. Classes are a base function and why even bother to include the option to switch to them if its too hard to balance and maintain them all? By that reasoning there should never be a way to regress the game.

From my perspective, the new jobs change the game which is fine. But the core classes are well known commodities by players who have enjoyed this game for many years. I am merely stating my preference to keep them and the others. Or give it as a base option just like no gold, or randarts, or no selling. Put the option there for the player to opt for original classes; new classes or both.

If it is too much work, then I understand as I cannot code so can't answer as to that.