Developer release of knightmare.g

Developer release of knightmare.g

[Lincoln Peters]

I've finally gotten knightmare.g to a point where I think it's ready for
release, at least in beta form.  You can download it at:

http://homepage.mac.com/lmpeters/knightmare.g

I hope to add even more features prior to releasing v1.0, but I know
that at least one person on this list has been eagerly awaiting a
developer release.

Some changes that have occurred since this game was last discussed on
the list:

* There are no longer multiple age categories of each type of dragon.
This caused two major problems: I couldn't figure out a reasonable (and
playable) way to make them grow more powerful, and I started getting
weird logic errors when I tried to implement the most powerful dragons
(which exceeded 20th level).  I may try again later.  Fortunately, with
a total of 274 units, I doubt anyone will miss them too much.

* Orc and kobolds no longer exist, as they proved to be either redundant
or pretty much useless (and I had to do a bit of trimming at one point
when I had to do an almost complete re-write of the GDL code).  Maybe
I'll re-introduce them before v1.0.

* I re-engineered the process by which a side acquires monstrous units.
Dragons are now produced by temples corresponding to their alignment
(lawful, chaotic, good, evil), and require extra advanced to be build
(the "Dragonkind" advance and at least one form of elemental magic).
Undead are now produced by "death temples", and there are now "life
temples" that can produce many natural (?) monsters that would otherwise
only be available by capturing or conjuring a lair specific to that
monster.


A few Xconq bugs that I uncovered while developing this module (all of
them have been posted at SourceForge):

* When a city or temple upgrades by changing type, the new type has a
longer reach than the old type.  However, the reach of the unit is
unchanged.  This drastically slows their growth.

* When a change-type occurs, the affected unit seems to lose materials
in the process.  Although with the supply and economy systems in place,
this bug may be hard to pin down.


I would encourage people to test this game and send me bug reports,
feature requests, comments, or anything else that would help me to
improve the game.

---
Lincoln Peters
<sampln@sbcglobal.net>

By protracting life, we do not deduct one jot from the duration of death.
		-- Titus Lucretius Carus

Re: Developer release of knightmare.g

[Elijah Meeks]

Hey Lincoln,

I opened up knightmare.g and received an error
reporting that 'The unit type form has no property
named size-min' which seems to relate to the various
cities.  Was that meant to be a size-max?

Otherwise it seems to run just fine and, wow, that's a
lot of units.  This'll take a while to do some serious
play-testing, but I've noticed some issues:

Category 1 - Relatively Serious
1.  The AI builds ships in land-locked cities.

Category 2 - Relatively Frivolous
1.  There's a trident.gif wagon that looks better than
the b/w one.
2.  The image for the Roc is Godzilla.  My Monster
Manual says its a giant bird, not a radioactive,
fire-breathing lizard.
3.  There's a thing called a 'thorp'.  What the heck
is a 'thorp'??

As you can tell, not too many serious issues yet. 
I'll keep hammering it.  Is there any particular
aspect that you'd want testing on?

Elijah


		
__________________________________ 
Do you Yahoo!? 
The all-new My Yahoo! - Get yours free! 
http://my.yahoo.com 
 

Re: Developer release of knightmare.g

[Lincoln Peters]

On Mon, 2005-01-03 at 14:48 -0800, Elijah Meeks wrote:
> I opened up knightmare.g and received an error
> reporting that 'The unit type form has no property
> named size-min' which seems to relate to the various
> cities.  Was that meant to be a size-max?

No.  "size-min" is the minimum size that an advanced unit is allowed to
be (e.g. a metropolis of size 1 would be a bit too weird).

It didn't give me any errors.  When did you last update your copy of
Xconq?

> 
> Category 1 - Relatively Serious
> 1.  The AI builds ships in land-locked cities.

Sounds like a bug in the "don't build ships without a shore" code,
although I can't really be sure without studying that part of the AI.

> 
> Category 2 - Relatively Frivolous
> 1.  There's a trident.gif wagon that looks better than
> the b/w one.

So there is.  I'll change it.

> 2.  The image for the Roc is Godzilla.  My Monster
> Manual says its a giant bird, not a radioactive,
> fire-breathing lizard.

It is supposed to be a giant bird (and is implemented as such), but
there aren't very many images of giant birds in the image library.

At least for now, I'm more inclined to use unique images for each unit
if possible, even if it doesn't look right, rather than assign the same
image to two or more unrelated units.


(Of course, if you want to find a monster that looks like nothing in the
image library, try looking up the otyugh in your Monster Manual!)

> 3.  There's a thing called a 'thorp'.  What the heck
> is a 'thorp'??

It's a populated areas that is even smaller than a hamlet, usually with
a population less than 100.

I suppose I should document the game more thoroughly before I upgraded
its status to anything higher than "developer release".

> 
> As you can tell, not too many serious issues yet. 
> I'll keep hammering it.  Is there any particular
> aspect that you'd want testing on?

Nothing particular, but I wouldn't be surprised if one or two game
balance issues were to come up (e.g. maybe high-level units can be built
too rapidly, or not rapidly enough).

And, of course, I'm hoping that you and a few other game developers can
have some fun while helping me to test this thing more rapidly...

---
Lincoln Peters
<sampln@sbcglobal.net>

The way to make a small fortune in the commodities market is to start
with a large fortune.

Re: Developer release of knightmare.g

[Elijah Meeks]

> No.  "size-min" is the minimum size that an advanced
> unit is allowed to
> be (e.g. a metropolis of size 1 would be a bit too
> weird).
> It didn't give me any errors.  When did you last
> update your copy of
> Xconq?

That is strange...  I'm running 20041227, so I don't
know why I'd catch errors...  Still, if there are no
other reports, let's chalk it up to something uniquely
strange to my system, since the game runs fine.

> Sounds like a bug in the "don't build ships without
> a shore" code,
> although I can't really be sure without studying
> that part of the AI.

You can set your naval units to be (naval true) as a
unit property.  I threw in (add naval naval true) and
it didn't seem to have any effect, which is strange
because I've noticed the AI is pretty good about not
producing naval units when land-locked.  I'll see if I
can find out why.

> It is supposed to be a giant bird (and is
> implemented as such), but
> there aren't very many images of giant birds in the
> image library.

Well, there are still another five or six files worth
of Angband gifs that I have left to format for XConq,
so hopefully one of them has a great, big falcon. 
(Three of them are nothing but armor, potions and
weapons...  You know, if Eric wants to throw in some
support for items, hint hint)

> (Of course, if you want to find a monster that looks
> like nothing in the
> image library, try looking up the otyugh in your
> Monster Manual!)

And, if you do, try to differentiate it from a
neo-otyugh, at 32x32 pixel...  

As a side note, I think your Tarrasque can be
implemented by using a seed-unit (I use these in
Opal-rules) with a low-possibility starvation and an
initial stock of material that allows it to last as
many turns as you want.  After that, since the thing
is nigh-invulnerable, I'd say you could give it
another low-possibility attrit and wreck-type it back
into the seed-unit (That's it 'going back to sleep').

> It's a populated areas that is even smaller than a
> hamlet, usually with
> a population less than 100.

I figured.  It's apparant by the image used and its
functions, I'd just never encountered the word before.

Back to building my Behir army...



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Find what you need with new enhanced search.
http://info.mail.yahoo.com/mail_250

Re: Developer release of knightmare.g

[Eric McDonald]

Lincoln Peters wrote:
> On Mon, 2005-01-03 at 14:48 -0800, Elijah Meeks wrote:
> 
>>I opened up knightmare.g and received an error
>>reporting that 'The unit type form has no property
>>named size-min' which seems to relate to the various
>>cities.  Was that meant to be a size-max?
> 
> No.  "size-min" is the minimum size that an advanced unit is allowed to
> be (e.g. a metropolis of size 1 would be a bit too weird).
> 
> It didn't give me any errors.  When did you last update your copy of
> Xconq?

An error would cause an abort. It was probably a warning.
And you may not have seen it if you started Knightmare from the command 
line:
	./xconq -g knightmare
as opposed to starting it from the GUI list of games. Add 'knightmare.g' 
to your 'game.dir' file in the 'lib' directory and then start it from 
the Tkconq games list, and see what happens.

>>Category 1 - Relatively Serious
>>1.  The AI builds ships in land-locked cities.
> 
> 
> Sounds like a bug in the "don't build ships without a shore" code,
> although I can't really be sure without studying that part of the AI.

You guys should look at 'minimal-sea-for-docks' which Hans added a year 
or more ago, I think. I doubt it ever got documented, since external 
documentation was not one of his strong points apparently. Also, it 
should have been named 'ai-minimal-sea-for-docks', but wasn't. Anyway:
(unit-type foo
   (minimal-sea-for-docks 30)
)

Actually, I thought 30 liquid cells was the default, so I am surprised 
that a naval unit is being built in a land-locked city. Maybe a new bug 
has arisen.

> Nothing particular, but I wouldn't be surprised if one or two game
> balance issues were to come up (e.g. maybe high-level units can be built
> too rapidly, or not rapidly enough).
> 
> And, of course, I'm hoping that you and a few other game developers can
> have some fun while helping me to test this thing more rapidly...

Yeah, I think it would be good if you and Elijah tested each other's 
stuff. You are both experienced Xconq game developers, and could provide 
useful feedback to each other. That way I could focus on bugfixes and 
the growing pile of feature requests. (Not that I mind testing your 
games; both of you have now made games that could keep me from touching 
Xconq code for weeks, while I, um, "playtested" them.)

Eric

Re: Developer release of knightmare.g

[Eric McDonald]

Lincoln Peters wrote:

> http://homepage.mac.com/lmpeters/knightmare.g

I noticed a couple of typos when I was reading through it. I believe 
"benevolent" was spelled as "beneavolent" in one place. I have not 
forgotten the other typo.

But, wow. It is much bigger than last time I looked at it. I am glad to 
see that the 20 levels of knights are still intact though, since the 
progression through combat experience is part of what made the original 
'knights.g' appealing to me.

Hopefully, I will get a chance to playtest it soon.

> A few Xconq bugs that I uncovered while developing this module (all of
> them have been posted at SourceForge):
> 
> * When a city or temple upgrades by changing type, the new type has a
> longer reach than the old type.  However, the reach of the unit is
> unchanged.  This drastically slows their growth.
> 
> * When a change-type occurs, the affected unit seems to lose materials
> in the process.  Although with the supply and economy systems in place,
> this bug may be hard to pin down.

Both of these are likely easy to fix. However, it would be quite helpful 
to have a saved game demonstrating them.

> By protracting life, we do not deduct one jot from the duration of death.
> 		-- Titus Lucretius Carus

I see the Romans thought about mathematical cardinality.

Eric

Re: Developer release of knightmare.g

[Lincoln Peters]

On Mon, 2005-01-03 at 19:00 -0700, Eric McDonald wrote: 
> I noticed a couple of typos when I was reading through it. I believe 
> "benevolent" was spelled as "beneavolent" in one place. I have not 
> forgotten the other typo.

I guess I should have run "spell check" before the developer release.

And as I think about it, I found a few typos in the IMF files while
working on this project:

* In fantasy.img, "heroes-yellow-shieldman" is misspelled "heroes-yellow-shiledman"
* In angband.imf, "scorpion" is misspelled "scorpian" in every case.

I'm not entirely sure how to fix this without also fixing it in every
game that uses those images (not that there are very many of them).

> 
> But, wow. It is much bigger than last time I looked at it. I am glad to 
> see that the 20 levels of knights are still intact though, since the 
> progression through combat experience is part of what made the original 
> 'knights.g' appealing to me.

The combat experience mechanism (which only became possible with the
implementation of auto-upgrade a month or two before I started
developing it) was the initial reason I wrote it, both as a (hopefully)
enjoyable game and as a demonstration of how the change-type mechanism
can be used.

I had hoped to increase the number of levels to 27 (when I added the
ancient, wyrm, and great wyrm dragons, which were eventually removed),
but when I tried to do so, the entire hit-chance table mysteriously
broke down.  I found it was easier to re-write it from the original
knights.g than to try to fix it.  Maybe I will try again later.

> 
> Hopefully, I will get a chance to playtest it soon.

Enjoy!

> 
> > A few Xconq bugs that I uncovered while developing this module (all of
> > them have been posted at SourceForge):
> > 
> > * When a city or temple upgrades by changing type, the new type has a
> > longer reach than the old type.  However, the reach of the unit is
> > unchanged.  This drastically slows their growth.
> > 
> > * When a change-type occurs, the affected unit seems to lose materials
> > in the process.  Although with the supply and economy systems in place,
> > this bug may be hard to pin down.
> 
> Both of these are likely easy to fix. However, it would be quite helpful 
> to have a saved game demonstrating them.

I'll see what I can do.  I'll e-mail you the saved games directly, since
they'll probably be pretty big.

---
Lincoln Peters
<sampln@sbcglobal.net>

Epperson's law:
	When a man says it's a silly, childish game, it's probably
	something his wife can beat him at.

Re: Developer release of knightmare.g

[Elijah Meeks]

> * In fantasy.img, "heroes-yellow-shieldman" is
> misspelled "heroes-yellow-shiledman"
> * In angband.imf, "scorpion" is misspelled
> "scorpian" in every case.


Are you sure it's not spelled with an 'a'?  Isn't
language normative, anyway?  If enough of us spelled
scorpion with an 'a', we'd be in much better shape,
don't ya think?

The only game I've got that uses a scorpion image is
opal-rules.g and it calls the imfs itself, rather than
from angband.imf, so it should be unaffected when
changing the a's to o's.  Everybody told me to mind my
p's and q's, not my a's and o's.

Elijah



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250

Re: Developer release of knightmare.g

[Lincoln Peters]

On Tue, 2005-01-04 at 12:24 -0800, Elijah Meeks wrote:
> Are you sure it's not spelled with an 'a'?  Isn't
> language normative, anyway?  If enough of us spelled
> scorpion with an 'a', we'd be in much better shape,
> don't ya think?

Not according to "spell check"...

But seriously, this is the first time I've seen it spelled "scorpian",
and I'm not inclined to rock the language boat, at least not without
more of a precedent.

Of course "shiledman" is not nearly as debatable for being a typo (and,
yes, I know that that typo was created by someone else).

---
Lincoln Peters
<sampln@sbcglobal.net>

It's easy to get on the internet and forget you have a life
	-- Topic on #LinuxGER

Re: Developer release of knightmare.g

[Eric McDonald]

Lincoln Peters wrote:

> Of course "shiledman" is not nearly as debatable for being a typo (and,
> yes, I know that that typo was created by someone else).

None of these should be a big deal to fix.

> It's easy to get on the internet and forget you have a life

That's odd. I think some people's life is on the Internet.

Eric