Hello,

I'm currently working on a heavily changed atlantis 4.0.10 code with
the intention to have a 4.1.0 soon. I'd like to setup a testgame,
however up to now I haven't found enough players who are interested in
testing.

New features are:

1) reworked gamedata: new (balanced) weapontable, armortable,
races,terrains.
2) Possibility to train monsters for combat
3) reduced strength of magic
4) AMTS now gives +3 to OBSE, as RING gives +3 to stealth
5) reworked skills: horsetraining and cameltraining were replaced by
animaltraining, weaponsmith and armorer are replaced by smithing,
weaponcraft and armorcraft were added for producing advanced weapons
and armor, animaltaming and monstertaming were added to train
monsters, carpenter now is toolmaking, longbow now is archery. Several
magic skills were changed, too.
6) non-magic skills that rely on eachother, e.g. monstertaming needs
animaltraining
7) non-leaders can study two skills
8) lots of new items
9) for each race, there is a racial leader and a racial hero. Normal
leaders are disabled.
10) much more

I will start a testgame if I can get at least 20 playtesters.

Regards

Georg (mailto: jegron @ hotmail.com)

> 1) reworked gamedata: new (balanced) weapontable, armortable,
> races,terrains.
> 2) Possibility to train monsters for combat

Magical or usual? Like it was in Ceran (pretty limited)?

> 3) reduced strength of magic

A good thing to correct.

> 4) AMTS now gives +3 to OBSE, as RING gives +3 to stealth

This one i would like to see tested extensively...

> 5) reworked skills: horsetraining and cameltraining were replaced by
> animaltraining, weaponsmith and armorer are replaced by smithing,
> weaponcraft and armorcraft were added for producing advanced weapons
> and armor, animaltaming and monstertaming were added to train
> monsters, carpenter now is toolmaking, longbow now is archery. Several
> magic skills were changed, too.
> 6) non-magic skills that rely on eachother, e.g. monstertaming needs
> animaltraining
> 7) non-leaders can study two skills
> 8) lots of new items
> 9) for each race, there is a racial leader and a racial hero. Normal
> leaders are disabled.
> 10) much more

--
-------------------------------------------------------------------
        Aivars Liepa           Dept. of New Inform. Technologies
  Latvian Academic Library     mailto:aliepa@...
    10, Rupniecibas str.       http://www.acadlib.lv/~aivars/e/
   Riga, LV-1235, LATVIA       AOL IM: aivars9  Phone: (371)-7106231
-------------------------------------------------------------------

> 1) reworked gamedata: new (balanced) weapontable, armortable,
> races,terrains.

Will you publish the data for inspection?

> 2) Possibility to train monsters for combat
> 3) reduced strength of magic
> 4) AMTS now gives +3 to OBSE, as RING gives +3 to stealth

That will effectivly make assassins and scouts useless. Why not keep
it as is, but tweak the game data to make assassination more
dangerous for the assassin (i.e. limit the number of free attacks an
assassin can get, rule out the more powerful weapons from
assassination).

> 5) reworked skills: horsetraining and cameltraining were replaced by
> animaltraining, weaponsmith and armorer are replaced by smithing,
> weaponcraft and armorcraft were added for producing advanced weapons
> and armor, animaltaming and monstertaming were added to train
> monsters, carpenter now is toolmaking, longbow now is archery.
> Several magic skills were changed, too.

Good idea to merge skills like horse/camel training to simplify the
system. I'd also remove tools, thus no need for toolmaking.

> 6) non-magic skills that rely on eachother, e.g. monstertaming needs
> animaltraining
> 7) non-leaders can study two skills

This could be a lot of fun. So now you can have non-leader knights
(c3 r2), stealth troops, or versitile producers/crafters.

> 8) lots of new items

hmmm.... sounds like clutter to me

> 9) for each race, there is a racial leader and a racial hero. Normal
> leaders are disabled.

How is a hero different from a leader?

> 10) much more

More==more fun or more==more clutter?

Cheers,
Tom

> > 2) Possibility to train monsters for combat
>
> Magical or usual? Like it was in Ceran (pretty limited)?
>

Both ways were improved. You can controll more animals by magic, and
there are several different monsters you can train just with a skill.
Unlike Ceran II, monstertaming isn't forbidden to leaders, so everyone
can get it, though the maximun level is limited for some races.

> > 3) reduced strength of magic
>
> A good thing to correct.
>

It was necessary to correct this, however I'm not sure whether my
changes are still balanced here. This is what I need testing for,
most.

If you (or anyone else) wants to join the testgame, you should sign up
for the ceran3dev group, there isn't much info there yet, but I will
add the info as I get them ready.

Georg

--- In atlantisdev@yahoogroups.com, "tzarg" <tzarg@y...> wrote:
> > 1) reworked gamedata: new (balanced) weapontable, armortable,
> > races,terrains.
>
> Will you publish the data for inspection?

I'll publish it once it is finished. In the meantime, you can check
out http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/tarmellion/source/
though the gamedata there is for Henriks Tarmellion game, my items are
already in there (or at least most of them). However the Ceran
directory at sourceforge is not up to date.

> > 4) AMTS now gives +3 to OBSE, as RING gives +3 to stealth
>
> That will effectivly make assassins and scouts useless. Why not keep
> it as is, but tweak the game data to make assassination more
> dangerous for the assassin (i.e. limit the number of free attacks an
> assassin can get, rule out the more powerful weapons from
> assassination).
>

Maybe I'll change this later, however I had to do something against
all those ringed assassins (at the end of Ceran II, there were more
than 500 assassins with rings and longbows (my weaker version of
doublebows))

> > 5) reworked skills: horsetraining and cameltraining were replaced
by
> > animaltraining, weaponsmith and armorer are replaced by smithing,
> > weaponcraft and armorcraft were added for producing advanced
weapons
> > and armor, animaltaming and monstertaming were added to train
> > monsters, carpenter now is toolmaking, longbow now is archery.
> > Several magic skills were changed, too.
>
> Good idea to merge skills like horse/camel training to simplify the
> system. I'd also remove tools, thus no need for toolmaking.
>

Henrik and me even added more tools, most item production is now
boosted by tools and tools also aid in building. In Ceran II they were
used a lot.

> > 7) non-leaders can study two skills
>
> This could be a lot of fun. So now you can have non-leader knights
> (c3 r2), stealth troops, or versitile producers/crafters.
>
> > 8) lots of new items
>
> hmmm.... sounds like clutter to me
>

Well, there are no longer tradegoods, but you'll have to produce them
instead. We tried to make it more interesting to play a trade faction,
and make it possible to play without without having the final war in
mind.

> > 9) for each race, there is a racial leader and a racial hero.
Normal
> > leaders are disabled.
>
> How is a hero different from a leader?
>

Every race is expert in up to four skills and good in four more
skills. All other skills are at default.

Racial leaders know the racial skills at higher level and have some
limited magical skills. They can be bought instead of the normal
leaders.

racial heros have all skills racial leaders have, plus a skill default
of 5, thus they no longer have the limit in magic. Every player starts
with one hero, the leader of his faction, and cannot get another one
unless his hero dies. The race of the hero also is the race of the
faction, so e.g. wood elves will never follow an orcish hero.

> > 10) much more
>
> More==more fun or more==more clutter?

just give it a try.

>
> Cheers,
> Tom

Georg

> Maybe I'll change this later, however I had to do something against
> all those ringed assassins (at the end of Ceran II, there were more
> than 500 assassins with rings and longbows (my weaker version of
> doublebows))

Simple: If a weapon is too powerful, set the flag so it can't be used
in assassinations. It's perfectly realistic to outlaw longbows from
assassinations. Maybe rule out all ranged weapons in assassinations,
and then you can protect key leaders by giving them good combat skill
and weapons (so they have a good chance to kill the assassin).

> Henrik and me even added more tools, most item production is now
> boosted by tools and tools also aid in building. In Ceran II they
> were used a lot.

If they're there, people will use them. But it's extra player work to
manage all the tools. Anything which can be done to reduce the amount
of time a player has to put into atlantis to be competitive, should
be done, I reckon.


> > > 8) lots of new items
> >
> > hmmm.... sounds like clutter to me
>
> Well, there are no longer tradegoods, but you'll have to produce
> them instead. We tried to make it more interesting to play a trade
> faction, and make it possible to play without without having the
> final war in mind.

So there are lots of different, but very rare items which can be
produced? That would be fine.

> Every race is expert in up to four skills and good in four more
> skills. All other skills are at default.
>
> Racial leaders know the racial skills at higher level and have some
> limited magical skills. They can be bought instead of the normal
> leaders.
>
> racial heros have all skills racial leaders have, plus a skill
> default of 5, thus they no longer have the limit in magic. Every
> player starts with one hero, the leader of his faction, and cannot
> get another one unless his hero dies. The race of the hero also is
> the race of the faction, so e.g. wood elves will never follow an
> orcish hero.

Sounds good.

> > > 10) much more
> >
> > More==more fun or more==more clutter?
>
> just give it a try.

Can't argue with that ;-)

Cheers,
Tom

--- In atlantisdev@yahoogroups.com, "tzarg" <tzarg@y...> wrote:
> > Maybe I'll change this later, however I had to do something
against
> > all those ringed assassins (at the end of Ceran II, there were
more
> > than 500 assassins with rings and longbows (my weaker version of
> > doublebows))
>
> Simple: If a weapon is too powerful, set the flag so it can't be
used
> in assassinations. It's perfectly realistic to outlaw longbows from
> assassinations. Maybe rule out all ranged weapons in assassinations,
> and then you can protect key leaders by giving them good combat
skill
> and weapons (so they have a good chance to kill the assassin).
>

Ok, current status: only knives, daggers and crossbows can be used in
assassinations. Let's see how this will do during the testgame.

> > Henrik and me even added more tools, most item production is now
> > boosted by tools and tools also aid in building. In Ceran II they
> > were used a lot.
>
> If they're there, people will use them. But it's extra player work
to
> manage all the tools. Anything which can be done to reduce the
amount
> of time a player has to put into atlantis to be competitive, should
> be done, I reckon.
>

I think the main goal should be to give players much choices what they
can do. It isn't actually necessary to use tools, but you will produce
a bit cheaper if you do. it's up to you.

> > > > 8) lots of new items
> > >
> > > hmmm.... sounds like clutter to me
> >
> > Well, there are no longer tradegoods, but you'll have to produce
> > them instead. We tried to make it more interesting to play a trade
> > faction, and make it possible to play without without having the
> > final war in mind.
>
> So there are lots of different, but very rare items which can be
> produced? That would be fine.
>

Yes, but there are also higher amounts of normal ressources. There are
a few very rare ressources, and if you can get control of a region
which has one of them, you will have a good trade base. However I
think those regions might be faught over.

> > > > 10) much more
> > >
> > > More==more fun or more==more clutter?
> >
> > just give it a try.
>
> Can't argue with that ;-)
>

I saw you just signed up ;) Let's see when there are enough testers.

> Cheers,
> Tom

Georg

> Ok, current status: only knives, daggers and crossbows can be used
> in assassinations. Let's see how this will do during the testgame.

Sounds good to me. Does that mean the target can also only use
knives, daggers and crossbows?

I think there might be a case for splitting
CAN_BE_USED_IN_ASSASSINATION or whatever the flag is called into two
separate flags - one for assassins (CAN_BE_USED_BY_ASSASSINS) and one
for targets (CAN_DEFEND_IN_ASSASSINATIONS) - that way your leader
with a longbow (or DBOW or SWOR or whatever) gets to defend with his
primary weapon, and if he lives, use it for counter attack. That
would make assassinations a lot more dangerous for the assassin.

I'm also in favour of limiting the number of assassination rounds
(say, to 4 rounds), and possibly including a new order (e.g. AMBUSH),
which would allow hit-and-run attacks by units or armies, as well as
assassins.

AMBUSH could work as follows:
- If the unit issuing the AMBUSH order has STEA>OBSE of the target
faction, the ambush succeeds and the attacker gets a free round.
- If OBSE==STEA, the ambush succeeds, but there are no free rounds.
- If OBSE>STEA, the ambush fails, and a normal battle ensues.
- Perhaps we could also include a modifier for TACT - the side with
greater TACT skill gets +1 on the OBSE/STEA comparison.
- AMTS or TRUE will not be included in the OBSE/STEA comparison,
except as below.
- If the unit issuing the AMBUSH order is a 1-man unit, the battle
will be between 2 individual soldiers, as in an assassination attempt.
- If the unit issuing the AMBUSH order has more than 1 man (or there
are multiple units AMBUSHing the same target, then a normal battle
will ensue, with a few limitations, as below.
- Defenders in an AMBUSH will be chosen randomly from the target
faction's units which would normally muster in the same hex, and only
2-3x the number of attacking men will be included as defenders.
- AMTS will only be included in the OBSE/STEA comparison if one of
the chosen defenders is carrying an AMTS and his total OBSE is
greater than the faction's OBSE without AMTS.
- An AMBUSH will last only one round (in addition to any free rounds).

This needs some more thought, but it could be a good replacement for
the ASSASSINATE order, and allow for more tactical variation.

On Tools:
> I think the main goal should be to give players much choices what
> they can do. It isn't actually necessary to use tools, but you will
> produce a bit cheaper if you do. it's up to you.

True, but they are extra items to clutter inventories. And all
traders will use tools to optimise their productivity.

But my main objection to tools is I feel they are outside of the
spirit of the game, which is to simplify interactions, and only
report on things which have major in-game effects. For example, the
rules state re combat: "Troops which are fighting hand-to-hand
without specific weapons are assumed to be irregularly armed with
makeshift weapons such as clubs, pitchforks, torches, etc.".

Therefore, I assume that trade units with lumberjack, mining or other
skills are assumed to have axes, picks, hammers and anvils and other
tools of the trade, or they wouldn't be able to produce anything.

How can a miner without a pick, shovel, wheelbarrow or mining cart,
etc. produce any iron? So why choose pick, and not shovel? And for
smiths, why hammer and not anvil?

Basically, I think tools are silly timewasters for players.

> > So there are lots of different, but very rare items which can be
> > produced? That would be fine.
>
> Yes, but there are also higher amounts of normal resources. There
> are a few very rare resources, and if you can get control of a
> region which has one of them, you will have a good trade base.
> However I think those regions might be fought over.

Having things to fight over is good for the game, I think.

> I saw you just signed up ;) Let's see when there are enough testers.

Yep. Does that mean we should move this discussion to the ceran3dev
group?

Cheers,
Tom

> Maybe I'll change this later, however I had to do something against
> all those ringed assassins (at the end of Ceran II, there were more
> than 500 assassins with rings and longbows (my weaker version of
> doublebows))

>Simple: If a weapon is too powerful, set the flag so it can't be used
>in assassinations. It's perfectly realistic to outlaw longbows from
>assassinations. Maybe rule out all ranged weapons in assassinations,
>and then you can protect key leaders by giving them good combat skill
>and weapons (so they have a good chance to kill the assassin).

How about making it so that assassinating with ranged weapons is not forbidden,
but it will mean the unit will be detected and
attacked immediately after the assassination (if it's successful of course).
This means if you want a successful assassination, you
must most likely sacrifice your assassin.

I'm also thinking perhaps a change to truesight would be in order, so that it's
more efficient against INVI. Perhaps give OBSE+1 for
each TRUE level simply. (So a true4 could not only detect, but also identify the
assassin, though true 5 would be pointless but for
creating more amulets.) Also, perhaps grant TRUE automatic OBSE5 to begin with,
so you don't have to waste eight turns studying OBSE
with your mage.

/Tomaz

Hello Tomaz,

> How about making it so that assassinating with ranged weapons is not
forbidden, but it will mean the unit will be detected and
> attacked immediately after the assassination (if it's successful of
course). This means if you want a successful assassination, you
> must most likely sacrifice your assassin.
>

Assassination with ranged weapons should, if possible at all, be the
safest way for an assassin. He doesn't have to approach the target,
but could just sit on the roof of some building and wait, while an
assassin with a dagger has to sneak up his victim. However the used
weapon shouldn't require much movement. An assassin with a bow will
definitely be seen and wouldn't have a second shot.

> I'm also thinking perhaps a change to truesight would be in order,
so that it's more efficient against INVI. Perhaps give OBSE+1 for
> each TRUE level simply. (So a true4 could not only detect, but also
identify the assassin, though true 5 would be pointless but for
> creating more amulets.) Also, perhaps grant TRUE automatic OBSE5 to
begin with, so you don't have to waste eight turns studying OBSE
> with your mage.
>

I don't want to make magic faction even more powerful. It should be
expensive to get a trueseeing mage, and it should happen in aboout the
same time another faction can get assassins and invisibility-mages
trained. The additional OBSE-bonus is already included as a gamedef
option.

> /Tomaz

Georg

> Assassination with ranged weapons should, if possible at all, be the
> safest way for an assassin. He doesn't have to approach the target,
> but could just sit on the roof of some building and wait, while an
> assassin with a dagger has to sneak up his victim. However the used
> weapon shouldn't require much movement. An assassin with a bow will
> definitely be seen and wouldn't have a second shot.

I agree. No code changes needed to implement this, just config.

> > I'm also thinking perhaps a change to truesight would be in order,
> so that it's more efficient against INVI. Perhaps give OBSE+1 for
> > each TRUE level simply. (So a true4 could not only detect, but
also
> identify the assassin, though true 5 would be pointless but for
> > creating more amulets.) Also, perhaps grant TRUE automatic OBSE5
to
> begin with, so you don't have to waste eight turns studying OBSE
> > with your mage.
> >
>
> I don't want to make magic faction even more powerful. It should be
> expensive to get a trueseeing mage, and it should happen in aboout
> the
> same time another faction can get assassins and invisibility-mages
> trained. The additional OBSE-bonus is already included as a gamedef
> option.

Yes, magic is too powerful. However, it takes a lot longer to get to
TRUE 5 than it does to get to INVI 2 or 3, which is all you need for
a group of invisible untouchable assassins. However, if assassination
is risky for the assassin, this isn't such a bad thing. You'll get
more people using the STEAL order instead.

Just to let you know that a new project is underway developing Atlantibots.

Below is the 'announcement'  and below that the relevant links :

Well the Atlantibot project is now open.
We have even been joined by Pete Christie who wrote the first
Atlantibots I ever came across in Pascal.

Most of us are working in Python although we have one C++ writer and
another coder who aims to work in Java and Python.... so it should prove
fun.

The eventual aim of the project is to produce several fully featured
'Atlantibots' to battle it out on an Atlantis world. These are scripts
that read Atlantis Turn reports and generate orders....... These
'automated factions' would have to cope with exploring their world and
dealing with other factions as they discover them.

Along the way we will produce various tools of use to the general
Atlantis player though - for example an automated sentry monitoring
script, automated mapping scouts (including generating movement orders
and building a map), faction summary and more.....

There is already some basic code being bandied around........ as soon as
we have anything firm - alpha releases, protocols or file format
conventions etc... they'll be released here. Any input or suggestions
welcomed.

Fuzzyman

Project homepage :        http://atlantibot.sourceforge.net/
Sourceforge Page :        http://sourceforge.net/projects/atlantibot/
The Discussion Group :    http://groups.yahoo.com/group/atlantis_talk/
The Atlantibot Atawiki entry (old) ! :
http://www.atlantis-pbem.org/ataWiki/index.php/AtlantiBots


--
Michael Foord
Sales Daventry Depot
TBS Builders Merchants Ltd
Hackwood Road
Daventry
Northants
NN11 4ES
Tel : 01327 877378, Fax : 01327 300598, Mobile : 07740 774204
http://www.TBSMerchants.co.uk
http://www.voidspace.org.uk or http://www.fuchsiashockz.co.uk
http://groups.yahoo.com/group/void-shockz

---

Everyone has talent. What is rare is the courage to follow talent
  to the dark place where it leads. -Erica Jong
Ambition is a poor excuse for not having sense enough to be lazy.
          -Milan Kundera

-----

This email is confidential & intended solely for the use of the individual to
whom it is addressed.  Any views presented are solely those of the author & do
not necessarily represent those of Towcester Building Supplies.
If you are not the intended recipient, be advised that you have received this
email in error & that any use, dissemination, forwarding, printing, or copying
of this email is strictly prohibited.
Any files attached to this email will have been checked with virus software
before transmission. However, you should carry out your own virus check before
opening any attachment.
Towcester Building Supplies accepts no liability for any loss or damage which
may be caused by software viruses.

At 10:26 AM +0100 7/9/03, Michael Foord wrote:
>There is already some basic code being bandied around........ as soon as
>we have anything firm - alpha releases, protocols or file format
>conventions etc... they'll be released here. Any input or suggestions
>welcomed.

You're welcome to use the smoketest scripts (in the CVS, under
smoketest) - they might give you a head start. The basic idea was to
have bots that run around, testing various features to make sure that
they don't crash. I'd almost finished the report-reading-in part of
it last time I worked on it.

Oh yeah, and it's in python too.

Hope that helps,

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

> At 10:26 AM +0100 7/9/03, Michael Foord wrote:
> >There is already some basic code being bandied around........ as soon as
> >we have anything firm - alpha releases, protocols or file format
> >conventions etc... they'll be released here. Any input or suggestions
> >welcomed.
>
> You're welcome to use the smoketest scripts (in the CVS, under
> smoketest) - they might give you a head start. The basic idea was to
> have bots that run around, testing various features to make sure that
> they don't crash. I'd almost finished the report-reading-in part of
> it last time I worked on it.
>
> Oh yeah, and it's in python too.
>
> Hope that helps,
>
> Ant


Thanks Ant,

Pete Christie pointed out your code to us.
We're taking a couple of different 'paths'. Rob is working cutrrebntly
on a 'report parser' which reads in the report and builds up a faction
sumamry - probably similar to smoketest. That code will probably be
re-used in most of what we do - but we may all develop seperate versions
as its not terribly complicated.

I'm starting with a sentry-monitoring script - it does  read through the
report but is looking for specific events.

Next I'll need to build up a map from a 'meta-report' - we'll be saving
them in the format outputted by Atlantis - so our 'meta-reports' of map
data can be read by any Atlantis Client........   I'd like to use this
to write an automatic mapper - a script that takes a few units and sends
units out to map. In order to do this I'll need to store the map data in
memory in a logical way -

e.g. an array where the computer knows that if it  is at location 4,3 it
  moves NE to location 4,4 for example.

As an array is square and the map is hex I'll have to sit down with
pencil and paper and work out the ruleset for describing movement -
unless someone can describe it simply for me !! Obviously if I create a
square array there will be some empty holes - but that doesn't matter.
The good thing about python is arrays are dynamic and you can change the
size as you go - a great thing for a unit that doesn't know the size of
the map he is in !!

Fuzzyman

--
Michael Foord
Sales Daventry Depot
TBS Builders Merchants Ltd
Hackwood Road
Daventry
Northants
NN11 4ES
Tel : 01327 877378, Fax : 01327 300598, Mobile : 07740 774204
http://www.TBSMerchants.co.uk
http://www.voidspace.org.uk or http://www.fuchsiashockz.co.uk
http://groups.yahoo.com/group/void-shockz

---

Everyone has talent. What is rare is the courage to follow talent
  to the dark place where it leads. -Erica Jong
Ambition is a poor excuse for not having sense enough to be lazy.
          -Milan Kundera

-----

This email is confidential & intended solely for the use of the individual to
whom it is addressed.  Any views presented are solely those of the author & do
not necessarily represent those of Towcester Building Supplies.
If you are not the intended recipient, be advised that you have received this
email in error & that any use, dissemination, forwarding, printing, or copying
of this email is strictly prohibited.
Any files attached to this email will have been checked with virus software
before transmission. However, you should carry out your own virus check before
opening any attachment.
Towcester Building Supplies accepts no liability for any loss or damage which
may be caused by software viruses.

At 11:52 AM +0100 7/9/03, Michael Foord wrote:
>Pete Christie pointed out your code to us.
>We're taking a couple of different 'paths'. Rob is working cutrrebntly
>on a 'report parser' which reads in the report and builds up a faction
>sumamry - probably similar to smoketest. That code will probably be
>re-used in most of what we do - but we may all develop seperate versions
>as its not terribly complicated.

Sure. Just trying to save you some of the boring work.

>Next I'll need to build up a map from a 'meta-report' - we'll be saving
>them in the format outputted by Atlantis - so our 'meta-reports' of map
>data can be read by any Atlantis Client........   I'd like to use this
>to write an automatic mapper - a script that takes a few units and sends
>units out to map. In order to do this I'll need to store the map data in
>memory in a logical way -

I believe that smoketest already stores map information in the way
that you're after, but you'll have to have a look at it to be sure.
Outputting a meta-report is a different thing, but it shouldn't be
too hard to do once you have the information in a list.

>As an array is square and the map is hex I'll have to sit down with
>pencil and paper and work out the ruleset for describing movement -
>unless someone can describe it simply for me !! Obviously if I create a
>square array there will be some empty holes - but that doesn't matter.

The square array is pretty much the easiest way to go, and the
easiest to think about when coding. I remember trying to fit it all
into a square grid with no gaps when I first started writing
smoketest, and I seem to remember it being possible, but it's fairly
brain hurty and nonintuitive (nw, ne, sw and se all go in different
directions depending on whether you're on an odd or even column).

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

> As an array is square and the map is hex I'll have to sit down with
> pencil and paper and work out the ruleset for describing movement -
> unless someone can describe it simply for me !! Obviously if I create a
> square array there will be some empty holes - but that doesn't matter.
> The good thing about python is arrays are dynamic and you can change the
> size as you go - a great thing for a unit that doesn't know the size of
> the map he is in !!

Well, the obvious place to find some freely available code that handles
hex arrays with variable connectivity is the atlantis source!  It may
not be in your target language, but it shouldn't be too hard to translate.

You shouldn't have to work out the relationship yourself anyway, just
derive it from the report - after all, there is no guarantee that 0,0
connects to 1,1 to the southeast (underworld, for example), so you
should really just read the hex reports and construct your map from
that!  The atlantis region classes are made to handle this kind of
geometry, so adapting them could be a good place to start.

Hope this helps,

Steve.

Hi All,

I've figured out a way to run a game using only a dial up account and
GetMail, a program to get mail via a POP account. I've written up a
how-to, which I've pasted on to the end of this email - hopefully
it'll lead to a few more 'boutique' games ;)

It *might* even let you set up a game running under Windows using
cygwin, but I haven't tried this (too lazy). If you do manage to get
a game going using this under Windows, let me know how you did it,
and I'll put it into the how-to (and credit you for it). You may have
to tweak the ~ (user home directory) parts out, but other than that
it should be cross platform.

I'm open to any other feedback you want to send my way, including
whether it works for you, and what problems you run into. Even if
it's just a typo.

Anthony

---

Running a game of Atlantis with a dial-up account
-------------------------------------------------

The maintainer of this how-to is Anthony Briggs,
<abriggs@...>. Email him with any suggestions, changes or
criticisms. This how-to is currently in the beta stage, since I've
only just set this up, and haven't tested it thoroughly yet. It works
in principle though.


Blurb:

Want to run a game of Atlantis for your friends? Except that you're
not 1337 3nUf to have a UNIX box connected to the internet 24/7? Or
perhaps you're a skinflint like me, and don't want to pay for a
permanent broadband connection? Read on...

The steps:

0. What you will need
1. Get Atlantis and your GM scripts up and running
2. Install python
3. Download and install GetMail
4. Configure GetMail
5. Configure your GM scripts
6. Decide on settings/policy
7. Testing
8. Advertise your game


0. What you will need

To set this up you'll need a basic knowledge of how various network
things work, such as POP, SMTP and email in general, plus be
reasonably confident in setting up scripts and running them.

To put this how-to into practice literally, you'll also need some
sort of UNIX-like computer. Since GetMail is written in Python
though, you may be able to apply what you learn here to other
situations, eg. Windows running cygwin.

As if you haven't guessed from the title, you'll also need a dialup
account, which you can obtain from your nearest friendly ISP. Oh
yeah, and a modem to use with it ;)


1. Get Atlantis and your GM scripts up and running

This is the first step - I'll assume that you've already compiled
Atlantis, set up some sort of GM script such as Thera or JT's
scripts, and run a few test games using local email addresses to make
sure everything's working ok.


2. Install python

If you haven't installed python yet, now's the time to do it,
although you'll probably need it for your scripts anyway. Most
UNIX-like distributions will have v1.5 installed - upgrading to v2 is
usually a good idea, since it makes your scripts run much faster.


3. Download and install GetMail

GetMail is a script similar in concept to FetchMail, but written in
python. It has the advantage that it can feed mail directly to
scripts - FetchMail doesn't appear to be able to do this from what
I've read of the documentation.

GetMail can be put pretty much anywhere you want it - there's a
wrapper script included that sets the appropriate variables and runs
the right python program. But for now you can unzip it into your home
directory.

The homepage for GetMail is at
<http://www.qcc.ca/~charlesc/software/getmail-3.0/getmail.html>.


4. Configure GetMail

Here's where the fun starts. There are two places that GetMail needs
tweaking - in the getmailrc file (usually within ~/.getmail) and in
the getmail wrapper script (I put mine in ~/bin). The getmailrc file
contains all sorts of options to do with getting your email from a
POP server, and the getmail wrapper script is run to actually
download the mail.

The wrapper script is fairly simple, and just needs to know where the
rest of the GetMail scripts are so that it can feed them the email.
Here's what mine looks like:

      #!/bin/sh

      /usr/local/bin/python /home/fracas/getmail/getmail.py $@

Pretty straightforward, huh? The one included with GetMail has all
sorts of fanciness for figuring out where python is, and setting the
path to the rest of the scripts, but I have to cut all of that out
when I got the following error from cron:

      exec: : permission denied

If anyone can shed any light on what's going on, I'd appreciate it.
It'd be nice to have GetMail work 'out of the box'. I suspect that
the line 'exec "$PYTHON" ${GETMAILPATH}/getmail.py $@' is the
culprit, but I'm not sure how to fix it.

To configure ~/.getmail/getmailrc, just copy across the template from
the GetMail directory, and edit it to your taste. Here's what mine
looks like:

      [default]

      verbose = 1                        # Be verbose by default
      readall = 0                        # Retrieve only new messages
      delete = 0                         # Delete mail after retrieval
      message_log = /var/log/getmail.log # Log message retrieval and delivery
      timeout = 240                      # TCP timeout value
      max_message_size = 0               # Don't retrieve messages over
this size.
                                         # 0 means no limit.

      [iinet]
      server = mail.iinet.net.au
      port = 110
      username = fracas
      password = "SekRitPassWeRd"

      postmaster = "|/home/fracas/atlantis/thera/atlantis.py"
      message_log = "/home/fracas/getmail.log"

      no_delivered_to = 1       # Don't add a Delivered-To: header
      no_received = 1           # Don't add a Received: header

The important parts for you to change are the server, username and
password settings, which should match your ISP's settings. The
postmaster and message_log settings, which should match your wrapper
script and the entry point for your script (in this case I'm using
Thera, hence the atlantis.py).

Once you've got to this stage, you should be able to run the wrapper
script, and have it download messages (GetMail will give you a nice
series of status lines as it works) and feed them into your script.
Depending on how your GM scripts work, you should be able to see the
emails appear in the logs.


5. Configure your GM scripts

Right, you've got emails being delivered to your scripts, all you
have to do now is make your scripts treat them properly! I'll show
you how to set up Thera to respond properly - hopefully that'll give
you the idea for other scripts. Or some nice people will email me and
tell me how to do it.

Firstly, you need to set the smtp server and the return email address
for the script, so that the scripts can send emails out, and players
can reply to them and have their email go right to the server. In
Thera, you do this by settings in the config.py file. Here's what the
server part of my config.py looks like:

      server = { 'smtpserver' : 'mail.m.iinet.net.au',
                 'adminemail' : 'abriggs@...',
                 'basedir'    : '/home/fracas/atlantis/',
                 'baseemail'  : 'fracas@...',
                 'emailsep'   : '-',
                 'logfile'    : '/home/fracas/atlantis/log.txt',
                 'scriptdir'  : '/home/fracas/atlantis/thera/',
                 'tar'    : '/usr/bin/tar',
                 'zip'    : '/usr/bin/bzip2',
                 'unzip'  : '/usr/bin/bunzip2',
                 'zipsuffix'  : '.bz2',
                 }

mail.m.iinet.net.au is my ISP's SMTP server, and the script's email
address is fracas@.... The admin email is still the same; so
are the rest of the settings.

At this point, you should be able to send a test message (eg. #help
or #create) to your script, and get a response back. You're almost
there!

The final step for me was to lobotomise Thera somewhat, since my ISP
doesn't do hyphenated names in emails (eg.
abriggs-fracas@...). I settled for a secondary email
account, which I can get for free. I commented out the part that gets
the game name from the email address, and made it 'fracas' all the
time:

      # Try and get the game name
      #gamenamesearch = '\w+\\'+server['emailsep']+'(\w+)@'
      #gamenametemp = re.search(gamenamesearch, mail['me'])
      #if gamenametemp == None:
      #    mail['gamename'] = None
      #    failnicely(mail['them'], 're:'+mail['subject'],
      #               'Couldn\'t find the game name in the address: '+mail['me'],
      #               None, 'Unknown')
      #else:

      mail['gamename'] = 'fracas'
      thisgame = game[mail['gamename']]


6. Automation

Of course, you don't want to be manually running GetMail every once
in a while to download your player's emails. The easier way to do it
is to set up a cron job so that your UNIX box will do it for you. I
use the following crontab:

      # My crontab

      # Run getmail every 10 minutes to pull in new mail
      0,10,20,30,40,50 * * * * /home/fracas/bin/getmail > /dev/null 2>&1

Now cron will fire up GetMail every 10 minutes and download email to
my GM script - easy, and I don't have to worry about remembering to
type '~/bin/getmail' every 10 minutes.

That little "> /dev/null 2>&1" on the end sends all of GetMail's
output to /dev/null, which means you won't get lots of annoying cron
output emails. You might want to leave it off if you're having
trouble, that way any error messages will appear in your email. Of
course, there's also your GM script and the GetMail logs to help you
figure things out.


7. Testing

Now comes the dull bit. You need to run a couple of turns, and make
sure that your game works as you'd planned it, emails don't go
astray, etc. No glitches? Good. Nothing annoys your players more than
obvious bugs that you should have caught.


8. Decide on settings/policy

Now here comes the tricky part - making sure that your players are
kept in the loop. You need to decide on a policy: how often the turns
are run, and when, and then stick to it. Your players won't always be
able to contact the server - after all, it's on an unreliable dialup
account. You need to make sure that they understand this, and know
that they need to get preliminary turns in early, and what will
happen if their turns aren't in on time.

One option is to try and be online at least once a day, and make sure
that your players know roughly when you'll be available (eg. For at
least an hour between 6pm and 10pm) so that they know when they can
contact you and have a reasonable response time.

Also, games with shorter turn times will be more stressful for your
players, since they might only get a couple of goes to get their
turns right, and might miss out if they get their last update in a
bit late.


9. Advertise your game

All done? You're ready to advertise, and open the game to the world! Have fun!


10. Other bits

Here are a couple of ideas that might come in handy:

If you set you email client and GetMail to save email on the server
(eg. for 2-3 days), you can use your regular account for both normal
email and game email :) Setting up a filter in your client to
automatically delete orders might be good for your sanity, though.

You could probably also use something along the lines of procmail to
filter incoming email. If you make your players put a specific word
in the subject, and/or search for words in the body (#atlantis
springs to mind) you can send those emails off before they reach your
inbox.

--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

Hello all,

I was upgrading my current code with the 4.0.10i patch (yep I'm
late :) ) and I found some little issues in the modified market code.
Nothing important, but just discrepancies into the process of market
creation. I don't know how the 5.0 version fix them and I don't
locate any comment from the group on it, so I put them here.

In the aregion.cpp code,
@@ -498,8 +516,13 @@

+// finished goods have a higher offset than raw goods
  Market * m = new Market (M_BUY, i, price, 0,
- 1000+population, 4000+population, 0, amt);
+ population+cap+offset, population+citymax+offset, 0, amt);
+// Offset added to maxamt too to reflect lesser production capacity
+// the more items are on sale.
  markets.Add(m);

1- The comment is irrelevent cause the first
population+citymax+offset value is minpop and the second one is
maxpop, maxamt is last value (amt).

2- I'm not sure about the desired effect but the maxpop should be
keep to population+citymax.


@@ -537,8 +565,11 @@
 	 price = ItemDefs[ i ].baseprice;
  }

-Market * m = new Market (M_SELL, i, price, 0, 2000+population,
-  4000+population, 0, amt);
+cap = (citymax *3/4) - 1000;
+if(cap < citymax/2) cap = citymax / 2;
+offset = (citymax/20) + ((citymax/5) * (magic++ +1));
+Market * m = new Market (M_SELL, i, price, amt/6, population+cap,
+  population+citymax, 0, amt);

3- The minpop should be population+cap+offset.


Jean-Michel

> In the aregion.cpp code,
> @@ -498,8 +516,13 @@
>
> +// finished goods have a higher offset than raw goods
>  Market * m = new Market (M_BUY, i, price, 0,
> - 1000+population, 4000+population, 0, amt);
> + population+cap+offset, population+citymax+offset, 0, amt);
> +// Offset added to maxamt too to reflect lesser production capacity
> +// the more items are on sale.
>  markets.Add(m);
>
> 1- The comment is irrelevent cause the first
> population+citymax+offset value is minpop and the second one is
> maxpop, maxamt is last value (amt).

Typo. I meant maxpop (but really this affects the maximum amount of goods
that will be on sale in the city).

> 2- I'm not sure about the desired effect but the maxpop should be
> keep to population+citymax.

No. The desired effect was to make the reduce the overall amount of goods on
sale for each item beyond the first (and the offset was a simple but
indirect way to do this).

> @@ -537,8 +565,11 @@
>   price = ItemDefs[ i ].baseprice;
>  }
>
> -Market * m = new Market (M_SELL, i, price, 0, 2000+population,
> - 4000+population, 0, amt);
> +cap = (citymax *3/4) - 1000;
> +if(cap < citymax/2) cap = citymax / 2;
> +offset = (citymax/20) + ((citymax/5) * (magic++ +1));
> +Market * m = new Market (M_SELL, i, price, amt/6, population+cap,
> + population+citymax, 0, amt);
>
> 3- The minpop should be population+cap+offset.

You're right. I was sloppy about the magic items! Tsk, tsk. Not that I know
any game where magic items are enabled on the markets... :)

Thanks Jean-Michel!

    Jan

Hi All,

There seems to be a crashing bug in the latest version of v5 - It's
coredumping when adding to guardsmen units (I think). I've confirmed
the bug for Wyreth and Fracas, but interestingly it doesn't seem to
happen for the standard game.

Here's the coredump:

#0  0x809f729 in Game::AdjustCityMon (this=0xbfbffb18, r=0x8212100,
u=0x8210600) at game.cpp:2230
2230            if(FindSkill(wp->baseSkill) == FindSkill("XBOW"))
skill = S_CROSSBOW;
(gdb) where
#0  0x809f729 in Game::AdjustCityMon (this=0xbfbffb18, r=0x8212100,
u=0x8210600) at game.cpp:2230
#1  0x809f4f5 in Game::AdjustCityMons (this=0xbfbffb18, r=0x8212100)
at game.cpp:2186
#2  0x817b9e5 in Game::PostProcessTurn (this=0xbfbffb18) at runorders.cpp:1112
#3  0x81732d9 in Game::RunOrders (this=0xbfbffb18) at runorders.cpp:105
#4  0x8097307 in Game::RunGame (this=0xbfbffb18) at game.cpp:1033
#5  0x81412a4 in main (argc=2, argv=0xbfbffbdc) at main.cpp:105
(gdb) print wp
$1 = (WeaponType *) 0x0

So it's crashing because wp is null (I think)

(gdb) print ItemDefs[weapon].abr
$2 = 0x0
(gdb) print ItemDefs[weapon]
$3 = {name = 0x0, names = 0x0, abr = 0x0, flags = 134687484, pSkill =
0x0, pLevel = 134658740, pMonths = 0, pOut = 0,
    pInput = {{item = 0, amt = 134688016}, {item = 0, amt = 134687980},
{item = 0, amt = 0}, {item = 136051562,
        amt = 136051570}}, mSkill = 0x81bfb7d "Neutral", mLevel =
136051589, mOut = 136051598, mInput = {{item = 136276908,
        amt = 136051603}, {item = 136051607, amt = 136051613}, {item =
136276932, amt = 136051619}, {item = 136051623,
        amt = 136051629}}, weight = 136051634, type = 136276948,
baseprice = 0, combat = 134766744, walk = 0, ride = 134724968,
    fly = 0, swim = 0, hitchItem = 0, hitchwalk = 134766684, mult_item
= 0, mult_val = 134717264, max_inventory = 0, escape = 0,
    esc_skill = 0x0, esc_val = 0}
(gdb)

Interestingly, it doesn't look like some of the variables are being
initialised properly:

(gdb) print weapon
$1 = -1
(gdb) print u
$2 = (Unit *) 0x8210600
(gdb) print armor
$3 = -1
(gdb) print i
No symbol "i" in current context.
(gdb) print mantype
$4 = 0
(gdb)

Also, I don't seem to be able to call items.GetNum, although that may
be normal, I'm not sure:

(gdb) print u->items.GetNum(1)
Cannot resolve method ItemList::GetNum to any overloaded instance
(gdb) print u->items.GetNum(10)
Cannot resolve method ItemList::GetNum to any overloaded instance
(gdb) print u->items.GetNum(0)
Cannot resolve method ItemList::GetNum to any overloaded instance
(gdb)

(gdb) call u->items.GetNum(0)
Cannot resolve method ItemList::GetNum to any overloaded instance
(gdb) call u->items.GetNum(5)
Cannot resolve method ItemList::GetNum to any overloaded instance
(gdb)

Can anyone shed any light on what's happening? I assume from the
description that the function is intended to update existing monster
units (f1 & f2) but I can't pick where exactly it's breaking down.

Thanks,

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

Hi Anthony,

>#0  0x809f729 in Game::AdjustCityMon (this=0xbfbffb18, r=0x8212100,
>u=0x8210600) at game.cpp:2230
>2230            if(FindSkill(wp->baseSkill) == FindSkill("XBOW"))
>skill = S_CROSSBOW;
>(gdb) where
>#0  0x809f729 in Game::AdjustCityMon (this=0xbfbffb18, r=0x8212100,
>u=0x8210600) at game.cpp:2230
>#1  0x809f4f5 in Game::AdjustCityMons (this=0xbfbffb18, r=0x8212100)
>at game.cpp:2186
>#2  0x817b9e5 in Game::PostProcessTurn (this=0xbfbffb18) at runorders.cpp:1112
>#3  0x81732d9 in Game::RunOrders (this=0xbfbffb18) at runorders.cpp:105
>#4  0x8097307 in Game::RunGame (this=0xbfbffb18) at game.cpp:1033
>#5  0x81412a4 in main (argc=2, argv=0xbfbffbdc) at main.cpp:105
>(gdb) print wp
>$1 = (WeaponType *) 0x0
>
>So it's crashing because wp is null (I think)

  The problem is here:

(2229)  WeaponType *wp = FindWeapon(ItemDefs[weapon].abr);

  weapon is initialised to -1 further up, and is only changed if the unit has an
item of type IT_WEAPON. So for whatever reason, the unit in question does not
have a weapon, and since ItemDefs[-1] is undefined, FindWeapon sets wp to NULL.

  Don't know why the guard unit would have no weapons, but to fix the coredump,
just add an if (weapon != -1) {....


Ben




>(gdb) print ItemDefs[weapon].abr
>$2 = 0x0
>(gdb) print ItemDefs[weapon]
>$3 = {name = 0x0, names = 0x0, abr = 0x0, flags = 134687484, pSkill =
>0x0, pLevel = 134658740, pMonths = 0, pOut = 0,
>   pInput = {{item = 0, amt = 134688016}, {item = 0, amt = 134687980},
>{item = 0, amt = 0}, {item = 136051562,
>       amt = 136051570}}, mSkill = 0x81bfb7d "Neutral", mLevel =
>136051589, mOut = 136051598, mInput = {{item = 136276908,
>       amt = 136051603}, {item = 136051607, amt = 136051613}, {item =
>136276932, amt = 136051619}, {item = 136051623,
>       amt = 136051629}}, weight = 136051634, type = 136276948,
>baseprice = 0, combat = 134766744, walk = 0, ride = 134724968,
>   fly = 0, swim = 0, hitchItem = 0, hitchwalk = 134766684, mult_item
>= 0, mult_val = 134717264, max_inventory = 0, escape = 0,
>   esc_skill = 0x0, esc_val = 0}
>(gdb)
>
>Interestingly, it doesn't look like some of the variables are being
>initialised properly:
>
>(gdb) print weapon
>$1 = -1
>(gdb) print u
>$2 = (Unit *) 0x8210600
>(gdb) print armor
>$3 = -1
>(gdb) print i
>No symbol "i" in current context.
>(gdb) print mantype
>$4 = 0
>(gdb)
>
>Also, I don't seem to be able to call items.GetNum, although that may
>be normal, I'm not sure:
>
>(gdb) print u->items.GetNum(1)
>Cannot resolve method ItemList::GetNum to any overloaded instance
>(gdb) print u->items.GetNum(10)
>Cannot resolve method ItemList::GetNum to any overloaded instance
>(gdb) print u->items.GetNum(0)
>Cannot resolve method ItemList::GetNum to any overloaded instance
>(gdb)
>
>(gdb) call u->items.GetNum(0)
>Cannot resolve method ItemList::GetNum to any overloaded instance
>(gdb) call u->items.GetNum(5)
>Cannot resolve method ItemList::GetNum to any overloaded instance
>(gdb)
>
>Can anyone shed any light on what's happening? I assume from the
>description that the function is intended to update existing monster
>units (f1 & f2) but I can't pick where exactly it's breaking down.
>
>Thanks,
>
>Ant
>--
>----------------------------------------------------
>HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
>aBRiGgS@...
>----------------------------------------------------
>
>
>To Post a message, send it to:   atlantisdev@eGroups.com
>To Unsubscribe, send a blank message to: atlantisdev-unsubscribe@eGroups.com
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

At 7:14 PM +1200 7/11/03, Ben Lloyd wrote:
>Hi Anthony,
>
>  >$1 = (WeaponType *) 0x0
>>
>  >So it's crashing because wp is null (I think)
>
>The problem is here:
>
>(2229)  WeaponType *wp = FindWeapon(ItemDefs[weapon].abr);
>
>  weapon is initialised to -1 further up, and is only changed if the
>unit has an item of type IT_WEAPON. So for whatever reason, the unit
>in question does not have a weapon, and since ItemDefs[-1] is
>undefined, FindWeapon sets wp to NULL.
>
>Don't know why the guard unit would have no weapons, but to fix the
>coredump, just add an if (weapon != -1) {....

Yeah, I understood all of that. What I meant was why is weapon not
set to anything? I suspect it's because the correct items (men,
swords, etc.) aren't in the unit.

I could add a line as you suggest, but I don't think it would fix the
root cause of the problem (which may well lie elsewhere in the code).

I've tracked it down to something which was added between 5 and 8
weeks ago, ie. cvs -z3 update -D "8 weeks ago" works fine, but cvs
-z3 update -D "5 weeks ago" fails. I'll try to narrow it down
further, and post an update when I have more news.

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

> >Don't know why the guard unit would have no weapons, but to fix the
> >coredump, just add an if (weapon != -1) {....
>
> Yeah, I understood all of that. What I meant was why is weapon not
> set to anything? I suspect it's because the correct items (men,
> swords, etc.) aren't in the unit.

Hmmm... I didn't expect guards without weapons (but really I sloppily left
out a safety check!). Ben's suggestion would fix it all.

> I could add a line as you suggest, but I don't think it would fix the
> root cause of the problem (which may well lie elsewhere in the code).

No, the root cause is the modified code for city guards I added recently.
I'll update the code if you don't beat me to it, Ant!

    Jan

At 1:35 PM +0200 7/11/03, Jan Rietema wrote:
>  > >Don't know why the guard unit would have no weapons, but to fix the
>>  >coredump, just add an if (weapon != -1) {....
>>
>>  Yeah, I understood all of that. What I meant was why is weapon not
>>  set to anything? I suspect it's because the correct items (men,
>>  swords, etc.) aren't in the unit.
>
>Hmmm... I didn't expect guards without weapons (but really I sloppily left
>out a safety check!). Ben's suggestion would fix it all.

Ah, like mages, you mean? Now I understand :)

>  > I could add a line as you suggest, but I don't think it would fix the
>>  root cause of the problem (which may well lie elsewhere in the code).
>
>No, the root cause is the modified code for city guards I added recently.
>I'll update the code if you don't beat me to it, Ant!

I'll see what I can do...

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

At 9:32 PM +0800 7/11/03, Anthony Briggs wrote:
>At 1:35 PM +0200 7/11/03, Jan Rietema wrote:
>  >Hmmm... I didn't expect guards without weapons (but really I sloppily left
>>out a safety check!). Ben's suggestion would fix it all.
>
>Ah, like mages, you mean? Now I understand :)
>
>>   > I could add a line as you suggest, but I don't think it would fix the
>>>   root cause of the problem (which may well lie elsewhere in the code).
>>
>>No, the root cause is the modified code for city guards I added recently.
>>I'll update the code if you don't beat me to it, Ant!
>
>I'll see what I can do...

Alrighty, that seemed to fix it. Thanks all.

Ant
--
----------------------------------------------------
HyPEraCtiVE? HeY, WhO aRE YoU cALliNg HypERaCtIve?!
aBRiGgS@...
----------------------------------------------------

Hi everybody,

Just a small question : How work the Elixir of life ? It's wrote how
to make it but not how to use it. I would like to turn it on for my
players.

If somebody knows and could tell me ...

Hi all,

can anyone tell me where I can find the code for V5? As I read the
messages online, and the original message contained a "@", the line
was blanked out by yahoo. Can someone please repost and put a blank
before and after the "@"?

thanks

Georg

Georg,

         Well first off there is technicaly no true v5 yet.  But if you
want the most recent unstable code there are instructions in the GM's
guide on how to get it from CVS.  You can find the unblanked out version
of the Guild on the AtaWiki or by going to this link:
http://www.atlantis-pbem.org/modules.php?op=modload&name=Sections&file=index&req
=viewarticle&artid=1&page=1

Hope that helps,
Mal


On Mon, 14 Jul 2003, Georg Gottschalk wrote:

> Hi all,
>
> can anyone tell me where I can find the code for V5? As I read the
> messages online, and the original message contained a "@", the line
> was blanked out by yahoo. Can someone please repost and put a blank
> before and after the "@"?
>
> thanks
>
> Georg
>
>
>
> To Post a message, send it to:   atlantisdev@eGroups.com
> To Unsubscribe, send a blank message to: atlantisdev-unsubscribe@eGroups.com
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

Hi guys,

--- In atlantisdev@yahoogroups.com, Michael Foord <michael.foord@t...>
wrote:

> We're taking a couple of different 'paths'. Rob is working cutrrebntly
> on a 'report parser' which reads in the report and builds up a faction
> sumamry - probably similar to smoketest. That code will probably be
> re-used in most of what we do - but we may all develop seperate
versions
> as its not terribly complicated.
>
> I'm starting with a sentry-monitoring script - it does  read through
the
> report but is looking for specific events.
>
> Next I'll need to build up a map from a 'meta-report' - we'll be saving
> them in the format outputted by Atlantis - so our 'meta-reports' of map
> data can be read by any Atlantis Client........   I'd like to use this
> to write an automatic mapper - a script that takes a few units and
sends
> units out to map. In order to do this I'll need to store the map
data in
> memory in a logical way -
>
> e.g. an array where the computer knows that if it  is at location
4,3 it
>  moves NE to location 4,4 for example.
>
> As an array is square and the map is hex I'll have to sit down with
> pencil and paper and work out the ruleset for describing movement -
> unless someone can describe it simply for me !! Obviously if I create a
> square array there will be some empty holes - but that doesn't matter.
> The good thing about python is arrays are dynamic and you can change
the
> size as you go - a great thing for a unit that doesn't know the size of
> the map he is in !!
>

I would like to point out that parsing atlantis reports is not as easy
as it may seem from the beginning :)
In order to avoid this and the array problems as described above, I
suggest using Atlantis Helper's parser for parsing reports, storing
parsed data and providing access to it.
The parser can be separated from the GUI, it was designed with such
possibility in mind. And Python allows you to use C++ code, right?

I meant to add scripting support to ALH from the very beginning, so
sharing parser, scripting object model and hopefully script modules
could benefit both projects.
And I can help with adding functionality to the parser if you are
interested.

Max.

Hi all,

I'm a player in Piotr's Arcadia, and am cooperating with Piotr on (and
will be running) Arcadia II.

I'm trying to experiment with the source code at the moment, but
unfortunately, although I've now been through much of the code, my
programming experience is nil and I have no experience with
writing/compiling (though modifying pre-written stuff is easy).

I've downloaded the source files for 4.0.10i/k and tried compiling
them with Dev-C++, a free compiler which seems simple enough for me to
run. However, when I try to compile it produces a large number of
errors for the file i_rand.c, ranging from parsing errors to "ctx was
not declared in this scope", etc.
  Leaving i_rand.c out of the compilation produces errors from
gameio.cpp - which needs 'isaac' and 'randinit'.

Can anyone tell me what I am doing wrong?

Bradley