Exult

Is it possible! When yea'll get the hang of it! That
all the people in serpent isle can be resurrected?

...and will there be in the future! The debug mode? 

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail - only $35 
a year!  http://personal.mail.yahoo.com/

In the future of Exult! Will it be possible to get
other NPC to join? And will they have status screens?

__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - Buy the things you want at great prices.
http://auctions.yahoo.com/

On Tuesday 13 March 2001 20:05, you wrote:
> In the future of Exult! Will it be possible to get
> other NPC to join? And will they have status screens?

Maybe the distant future:-)  Currently, we're concentrating on a 
single-player engine, but there's nothing stopping someone from taking the 
code and adding multi-user support.  It sounds like a fun project.

I just wanted to know if yea'll plan to program
serpent Isle where everone of serpent isle can be resurrected?

__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/

I know this is a sick question. But i wanted ask 
anyway.
In the black gate, will the wrenches be program to 
have sex with the Avatar like the wrench Lucilla and
Frigidazzi from Serpent Isle. If possible with the
Girl at the shrine of compassion? I think this is the
best question yet.

__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/

On Tuesday 29 May 2001 10:37, you wrote:
>
> You ever get ahold of Lady MOI again, out of curiosity? 

The last time was just as she was leaving Origin.

>Or have you ever
> tried to talk with any of the original developers?

Ken Demarest (lead programmer for BG) wrote to this list about a month ago, 
but then stopped answering emails.  In the past, I've exchanged emails with 
Jeff Wilson (U7 audio and Voodoo memory manager), Mike Schafly (sp?), DrCat 
(from U6), and way, way back, a couple others (possibly W. Spector, but I 
can't remember).  But I never got any information from anyone.:-(

> Naw, it's gonna be depressing either way. Some of the deaths /were/
> intentional and planned, I suspect. And besides, you have to have something
> dark leading into Pagan. =>

Okay, then, the new ending will have the Avatar click his heals, say "There's 
no place like home", and then it will turn out that the whole thing was just 
a dream.  Well, maybe not.

>
> Although, it'd be real nice if we could just rewrite Pagan entirely.. And
> U9 is almost a lost cause.
> I understand there's a project out there to create the 'original' U9
> storyline in a tilebased engine?
> You know anything about that?

Heard of it.  There's also a 'dialog patch' for U9 that's supposed to make 
the story more consistent with the earlier ultimas.

I'd only want to do a major rewrite of an Ultima if I had written permission 
from EA.  And I'd probably want to get paid something, too.  Otherwise, why 
not write a completely new game?

On Tue, May 29, 2001 at 11:00:35AM -0700, Jeff Freedman wrote:
> The last time was just as she was leaving Origin.

> Ken Demarest (lead programmer for BG) wrote to this list about a month ago, 
> but then stopped answering emails.  In the past, I've exchanged emails with 
> Jeff Wilson (U7 audio and Voodoo memory manager), Mike Schafly (sp?), DrCat 
> (from U6), and way, way back, a couple others (possibly W. Spector, but I 
> can't remember).  But I never got any information from anyone.:-(

Yuck. Wanna try again? =>


> Okay, then, the new ending will have the Avatar click his heals, say "There's 
> no place like home", and then it will turn out that the whole thing was just 
> a dream.  Well, maybe not.

... Um.

No.

> Heard of it.  There's also a 'dialog patch' for U9 that's supposed to make 
> the story more consistent with the earlier ultimas.

... u9, as shipped, has a story?
I played about halfway through it before I got tired of playing Avatar the
ErrandBoy. It felt too much like a chunky console adventure with poor play
control.

I honestly think the team was trying to be like Zelda, because it had the
same 'feel' as the N64 Zelda games, just.. slower, and harder to control.
Oh, and I think the N64 Zelda games had a deeper plot. =p That's pretty sad.
I used to think U8 was as low as the series could go, but U9, while
extraordinarily pretty, was not very fun.
 
> I'd only want to do a major rewrite of an Ultima if I had written permission 
> from EA.  And I'd probably want to get paid something, too.  Otherwise, why 
> not write a completely new game?

Yeah, but I can at least entertain the thought, ne? =>

And I think Garriot at one point expressed an interest in having a U4
rewrite made.. Mayhaps someday he'll make the arrangements for it to happen.
Maybe. Issa hope.

On Tuesday 29 May 2001 11:29, you wrote:

> Yuck. Wanna try again? =>

Might be worth another try, since we have a lot more to show.  Anyone know 
Warren Spector's email?

> ... u9, as shipped, has a story?
> I played about halfway through it before I got tired of playing Avatar the
> ErrandBoy. It felt too much like a chunky console adventure with poor play
> control.

I assume you're not a fan of adventure games (esp. since you haven't played 
Monkey Island).  You're right, U9 was more of an adventure than an RPG.  
Ironically, King's Quest 8 was more an RPG than an adventure (but I gave up 
on that after about an hour).

> rewrite made.. Mayhaps someday he'll make the arrangements for it to
> happen. Maybe. Issa hope.

That would be a reasonable thing for us to do with Exult, since we could 
start with the world that's already there.  But I'd hate to spend 100 hours 
on it, and then get a nasty letter from EA.  And they're the ones who own the 
rights to Ultima.

Meanwhile, RG only seems to be interested in multiplayer games.

On Tue, May 29, 2001 at 11:56:12AM -0700, Jeff Freedman wrote:
> I assume you're not a fan of adventure games (esp. since you haven't played 
> Monkey Island).  You're right, U9 was more of an adventure than an RPG.  
> Ironically, King's Quest 8 was more an RPG than an adventure (but I gave up 
> on that after about an hour).

Untrue. I like adventure games. I'm a fan of the Zelda series, and many
other console adventure games have entertained me over the years. However,
PC adventures never interested me much (Kings Quest bothered me, although
Space Quest was pretty darn fun; played the first few of those on my Apple
II)..


> That would be a reasonable thing for us to do with Exult, since we could 
> start with the world that's already there.  But I'd hate to spend 100 hours 
> on it, and then get a nasty letter from EA.  And they're the ones who own the 
> rights to Ultima.
> 
> Meanwhile, RG only seems to be interested in multiplayer games.

Hmm. I had forgotten about that.

Memory block, if you will.

On Thursday 31 May 2001 00:14, you wrote:

> Just another bug that needs to be fixed from the original: In the
> Britannian mint (near Sleeping Bull) the Chaos Hierophant begins to
> speak to you. 

Does it happen in Exult?  I played through that part, but can't remember.

Thursday, May 31, 2001, 5:17:04 PM,=20
Jeff Freedman <jsf@...> wrote:

>> Just another bug that needs to be fixed from the original: In the
>> Britannian mint (near Sleeping Bull) the Chaos Hierophant begins to
>> speak to you.=20

> Does it happen in Exult?  I played through that part, but can't remembe=
r.

Yeah, when you pass the mint or you go inside it ahppens a lot. Don=B4t
know if it still happens when you click through the whole conversation
but ending it the quick way (two times saying bye and clicking yes)
makes him pseak to the avatar at least four times there (annoying).
Dom

Dancer Vesperman <dancer@...> writes:

> If we change the way those plots work, then we mess with EA's
> copyright. Not a good idea, IMO.

(IANAL) It could be argued, that the current usage of exult together
with U7 already makes it a derived work under copyright law. Whether
usecode is modified or not may not make a difference. This
interpretation is also supported by the (to be mended) fact that
currently exult is pretty useless without U7. That's in contrast with,
say, a zcode interpreter that can /also/ be used to play Infocom
games.

Does somebody have a license of one of the U7 handy to check whether
modifications are strictly forbidden?

Whether the U7+exult combination is a derived work is also an issue
with the GPL.

I agree that there is a fine line somewhere, but I'm not at all sure
where it lies.

-- 
Robbe

Robert Bihlmeyer wrote:
> 
> Dancer Vesperman <dancer@...> writes:
> 
> > If we change the way those plots work, then we mess with EA's
> > copyright. Not a good idea, IMO.
> 
> (IANAL) It could be argued, that the current usage of exult together
> with U7 already makes it a derived work under copyright law. Whether
> usecode is modified or not may not make a difference. This
> interpretation is also supported by the (to be mended) fact that
> currently exult is pretty useless without U7. That's in contrast with,
> say, a zcode interpreter that can /also/ be used to play Infocom
> games.
> 
> Does somebody have a license of one of the U7 handy to check whether
> modifications are strictly forbidden?

Just took a quick peek at the 'Complete U7' license. Nothing in there
about modifications or derived works. Just a standard "reproduction of
this work in any form is prohibited"

-Willem Jan

Robert Bihlmeyer wrote:

>Dancer Vesperman <dancer@...> writes:
>
>>If we change the way those plots work, then we mess with EA's
>>copyright. Not a good idea, IMO.
>>
>
>(IANAL) It could be argued, that the current usage of exult together
>with U7 already makes it a derived work under copyright law. Whether
>usecode is modified or not may not make a difference. This
>interpretation is also supported by the (to be mended) fact that
>currently exult is pretty useless without U7. That's in contrast with,
>say, a zcode interpreter that can /also/ be used to play Infocom
>games.
>
>Does somebody have a license of one of the U7 handy to check whether
>modifications are strictly forbidden?
>
>Whether the U7+exult combination is a derived work is also an issue
>with the GPL.
>
>I agree that there is a fine line somewhere, but I'm not at all sure
>where it lies.
>
It's something that could concievably come to argument but it's 
basically: Origin/EA engendered the separation between the data files 
and the engine. They are not coupled as a single unit, and there is no 
attempt to confuse copyright, trademarks or ownership in the eyes of the 
consumer or violate the copyright-holder's rights (reserved or 
otherwise) (Admittedly that's under international copyright law, not 
under the DMCA). We cause no material harm, and have increased sales of 
the product and gained more fans for some of EA's franchises. You can 
argue it back the other way, but that's the meat that both arguments 
have to dress up.

D

On Sun, Jun 03, 2001 at 06:32:20PM +0200, Robert Bihlmeyer wrote:

> (IANAL) It could be argued, that the current usage of exult together
> with U7 already makes it a derived work under copyright law. Whether
> usecode is modified or not may not make a difference. This
> interpretation is also supported by the (to be mended) fact that
> currently exult is pretty useless without U7. That's in contrast with,
> say, a zcode interpreter that can /also/ be used to play Infocom
> games.

I see no difference between Exult and a zcode interpreter. One just happens
to have a larger available pool of data than the other.

> Does somebody have a license of one of the U7 handy to check whether
> modifications are strictly forbidden?

Well, we haven't modified U7 yet. Won't until/unless we get permission.

> Whether the U7+exult combination is a derived work is also an issue
> with the GPL.

It's not a derived work. Exult simply is an interpreter, that happens to
only have one valid data set to date. We could disprove the concept that
Exult requires U7/U7SI by simply putting together a dataset that starts up,
lets a character move around, have a conversation, and pick up an item or
three.

> I agree that there is a fine line somewhere, but I'm not at all sure
> where it lies.

The License Agreement in full from my Complete Ultima VII manual (by 
quoting it I've also broken it)

THIS MANUAL AND THE SOFTWARE DESCRIBED IN THIS MANUAL ARE COPYRIGHTED. ALL 
RIGHTS ARE RESERVED. NO PART OF THIS MANUAL OR THE DESCRIBED SOFTWARE MAY 
BE COPIED, REPRODUCED, TRANSLATED, OR REDUCED TO ANY ELECTRONIC MEDIUM OR 
MACHINE-READABLE FORM WITHOUT THE PRIOR CONSENT OF ELECTRONIC ARTS, P.O. 
BOX 7578, SAN MATEO, CALIFORNIA 94403-7578, ATTN: CUSTOMER SUPPORT.

Pretty obviously we have already broken the license agreement. We've done 
this in a few ways.
1) Exult is a reproduction of the Ultima 7 engine.
2) The Font used in the EXULT Menu is a font from BG. We are distributing 
this copyrighted data.
3) Paperdolls in BG require us distributing data, some of which is based on 
copyrighted art.
4) The new in game menus require us to distribute data that is based upon 
copyrighted art.
5) The SFX waves that we are distributing are a reproduction of the sounds 
from the game.

Anything about us changing the game is IMO fairly insignificant. It could 
be argued that we already have. Here is a list of things different in Exult 
to the originals:
1) Paperdolls in BG
2) Wave based SFX
3) In Game Menu
4) Party Flocking
5) Combat

Where is the line drawn?

On Sunday 03 June 2001 20:28, you wrote:
>
> I see no difference between Exult and a zcode interpreter. One just happens
> to have a larger available pool of data than the other.

Exactly.

On Sunday 03 June 2001 17:18, you wrote:

> under the DMCA). We cause no material harm, and have increased sales of
> the product and gained more fans for some of EA's franchises. 

I wonder if sales of "Ultima Collection" have increased lately?  Of course, 
U7run could be at least as responsible.

On Monday 04 June 2001 08:19, you wrote:

> Pretty obviously we have already broken the license agreement. We've done
> this in a few ways.
> 1) Exult is a reproduction of the Ultima 7 engine.

I'm not sure if a reverse-engineered engine is a 'reproduction'.

> 2) The Font used in the EXULT Menu is a font from BG. We are distributing
> this copyrighted data.
> 3) Paperdolls in BG require us distributing data, some of which is based on
> copyrighted art.
> 4) The new in game menus require us to distribute data that is based upon
> copyrighted art.
> 5) The SFX waves that we are distributing are a reproduction of the sounds
> from the game.

If challenged, we could easily drop 2) through 5).  Also I thought we were 
just displaying data from the user's copy of U7, not distributing it 
ourselves (except for the SFX's).

> Where is the line drawn?

I don't know.  Maybe this is a good topic for Advogato or Slashdot.

Ryan Nunn wrote:

> The License Agreement in full from my Complete Ultima VII manual (by 
> quoting it I've also broken it)
>
> THIS MANUAL AND THE SOFTWARE DESCRIBED IN THIS MANUAL ARE COPYRIGHTED. 
> ALL RIGHTS ARE RESERVED. NO PART OF THIS MANUAL OR THE DESCRIBED 
> SOFTWARE MAY BE COPIED, REPRODUCED, TRANSLATED, OR REDUCED TO ANY 
> ELECTRONIC MEDIUM OR MACHINE-READABLE FORM WITHOUT THE PRIOR CONSENT 
> OF ELECTRONIC ARTS, P.O. BOX 7578, SAN MATEO, CALIFORNIA 94403-7578, 
> ATTN: CUSTOMER SUPPORT.
>
> Pretty obviously we have already broken the license agreement. We've 
> done this in a few ways.
> 1) Exult is a reproduction of the Ultima 7 engine. 

Actually, no, it isn't.

>
> 2) The Font used in the EXULT Menu is a font from BG. We are 
> distributing this copyrighted data.
> 3) Paperdolls in BG require us distributing data, some of which is 
> based on copyrighted art.
> 4) The new in game menus require us to distribute data that is based 
> upon copyrighted art. 

For these three, what are we transmitting as files (or parts of files) 
that originally came on the CD? I don't believe that using the font to 
display our own words counts.

>
> 5) The SFX waves that we are distributing are a reproduction of the 
> sounds from the game. 

The SFX actually are a problem.

>
>
> Anything about us changing the game is IMO fairly insignificant. It 
> could be argued that we already have. Here is a list of things 
> different in Exult to the originals:
> 1) Paperdolls in BG
> 2) Wave based SFX
> 3) In Game Menu
> 4) Party Flocking
> 5) Combat
>
> Where is the line drawn?
>
> _______________________________________________
> Exult-general mailing list
> Exult-general@...
> http://lists.sourceforge.net/lists/listinfo/exult-general

On Mon, Jun 04, 2001 at 10:41:34AM -0700, Jeff Freedman wrote:
> > I see no difference between Exult and a zcode interpreter. One just happens
> > to have a larger available pool of data than the other.
> 
> Exactly.

IANAL and I mostly agree to this, but on the other hand,
zcode-interpreters don't tend to fix bugs in the original data files...

Michael

Jeff Freedman wrote:

>On Monday 04 June 2001 08:19, you wrote:
>
>>Pretty obviously we have already broken the license agreement. We've done
>>this in a few ways.
>>1) Exult is a reproduction of the Ultima 7 engine.
>>
>
>I'm not sure if a reverse-engineered engine is a 'reproduction'.
>
International copyright law protects the rights of a third-party to 
reverse-engineer. The DMCA does not, but has no force outside the USA. A 
reproduction however, is in this sense (for example) a photocopy or a 
recording of the original. If I go 'cp u7.exe reproduction.exe' then 
I've reproduced it....

>>2) The Font used in the EXULT Menu is a font from BG. We are distributing
>>this copyrighted data.
>>3) Paperdolls in BG require us distributing data, some of which is based on
>>copyrighted art.
>>4) The new in game menus require us to distribute data that is based upon
>>copyrighted art.
>>5) The SFX waves that we are distributing are a reproduction of the sounds
>>from the game.
>>
>
>If challenged, we could easily drop 2) through 5).  Also I thought we were 
>just displaying data from the user's copy of U7, not distributing it 
>ourselves (except for the SFX's).
>
Same here.

>
>
>>Where is the line drawn?
>>
>
>I don't know.  Maybe this is a good topic for Advogato or Slashdot.
>
>_______________________________________________
>Exult-general mailing list
>Exult-general@...
>http://lists.sourceforge.net/lists/listinfo/exult-general
>

On Tue, Jun 05, 2001 at 12:49:48AM +0930, Ryan Nunn wrote:
> THIS MANUAL AND THE SOFTWARE DESCRIBED IN THIS MANUAL ARE COPYRIGHTED. ALL 
> RIGHTS ARE RESERVED. NO PART OF THIS MANUAL OR THE DESCRIBED SOFTWARE MAY 
> BE COPIED, REPRODUCED, TRANSLATED, OR REDUCED TO ANY ELECTRONIC MEDIUM OR 
> MACHINE-READABLE FORM WITHOUT THE PRIOR CONSENT OF ELECTRONIC ARTS, P.O. 
> BOX 7578, SAN MATEO, CALIFORNIA 94403-7578, ATTN: CUSTOMER SUPPORT.

> Pretty obviously we have already broken the license agreement. We've done 
> this in a few ways.
> 1) Exult is a reproduction of the Ultima 7 engine.

Hmm. Tricky, and theoretically debatable.

> 2) The Font used in the EXULT Menu is a font from BG. We are distributing 
> this copyrighted data.

Easily solved. We could have a new font fairly quickly.

> 3) Paperdolls in BG require us distributing data, some of which is based on 
> copyrighted art.

It does? I thought it mostly required U7SI?

> 4) The new in game menus require us to distribute data that is based upon 
> copyrighted art.

That should also be easy to solve.

> 5) The SFX waves that we are distributing are a reproduction of the sounds 
> from the game.

Granted. But there's a catch, see below.

> 1) Paperdolls in BG
> 2) Wave based SFX
> 3) In Game Menu
> 4) Party Flocking
> 5) Combat

Those are things that are different in Exult from the original engine. They
do not entail modifications to the original software.
At no time do we ever modify the data that ships with the game.

I still say we're mostly safe, legally. There are a few points, but I think
all of them are resolvable.

Out of curiosity, the wav-based SFX.. That's from a wavetable MIDI device,
right? Who made those patches?
Those would belong to the patch author, not necessarily Origin.

On Monday 04 June 2001 13:47, you wrote:

> IANAL and I mostly agree to this, but on the other hand,
> zcode-interpreters don't tend to fix bugs in the original data files...
>

We haven't done that yet:-)  If we do, it will probably involve a patch to 
the 'usecode' file.

Michael Banck wrote:

>On Mon, Jun 04, 2001 at 10:41:34AM -0700, Jeff Freedman wrote:
>
>>>I see no difference between Exult and a zcode interpreter. One just happens
>>>to have a larger available pool of data than the other.
>>>
>>Exactly.
>>
>
>IANAL and I mostly agree to this, but on the other hand,
>zcode-interpreters don't tend to fix bugs in the original data files...
>
Nor do we patch the usecode in any way.
Most zcode-interpreters _do_ however contain assorted workarounds for 
bugs in game-data-files, however (and occasional misuse and abuse of 
zcode opcodes). I should know...I've written zcode (both raw, and with 
inform) and zcode interpreters in the past.

D

At 09:01 AM 5/06/2001 +1000, you wrote:
>Jeff Freedman wrote:
>>On Monday 04 June 2001 08:19, you wrote:
>>>Pretty obviously we have already broken the license agreement. We've done
>>>this in a few ways.
>>>1) Exult is a reproduction of the Ultima 7 engine.
>>>I'm not sure if a reverse-engineered engine is a 'reproduction'.
>International copyright law protects the rights of a third-party to 
>reverse-engineer. The DMCA does not, but has no force outside the USA. A 
>reproduction however, is in this sense (for example) a photocopy or a 
>recording of the original. If I go 'cp u7.exe reproduction.exe' then I've 
>reproduced it....

<delurk>

Umm... IANAL, but in .au the reverse engineering laws are a pain in the 
posterior. From memory, until about 12 months ago it was illegal to RE 
software. Now you CAN RE software, but it's illegal to distribute the 
information you gather. I can provide more information if required, but 
most of my references are from talking to a couple of lecturers at 
university. :(

But to my knowledge REing is not a form of 'reproduction'.

Have fun,

Patrick


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

Patrick Burke wrote:

> At 09:01 AM 5/06/2001 +1000, you wrote:
>
>> Jeff Freedman wrote:
>>
>>> On Monday 04 June 2001 08:19, you wrote:
>>>
>>>> Pretty obviously we have already broken the license agreement. 
>>>> We've done
>>>> this in a few ways.
>>>> 1) Exult is a reproduction of the Ultima 7 engine.
>>>> I'm not sure if a reverse-engineered engine is a 'reproduction'.
>>>
>> International copyright law protects the rights of a third-party to  
>> reverse-engineer. The DMCA does not, but has no force outside the 
>> USA. A  reproduction however, is in this sense (for example) a 
>> photocopy or a  recording of the original. If I go 'cp u7.exe 
>> reproduction.exe' then I've  reproduced it....
>
>
> <delurk>
>
> Umm... IANAL, but in .au the reverse engineering laws are a pain in 
> the  posterior. From memory, until about 12 months ago it was illegal 
> to RE  software. Now you CAN RE software, but it's illegal to 
> distribute the  information you gather. I can provide more information 
> if required, but  most of my references are from talking to a couple 
> of lecturers at  university. :(
>
> But to my knowledge REing is not a form of 'reproduction'.
>
> Have fun,
>
>
For .au, http://www.copyright.org.au has a set of fact sheets in PDF.

D

Hi Dude,I just visit you site and I like it.
I want to offer you to link exchange.

My link exchange is eazy.
Just submit your site to my partners center,and put link back to my
site.its all.
your site will be show on all my pages.
please go to
http://www.hotfreesite.com/cgi-bin/partners/accounts.cgi
and submit your site.

In the future will exult make it possible to make an 
option where you choice how the party follows you!
The original or exult way?

Also will there be an update on the mouse. I'm hoping
that exult gets it to the original.

PS. Keep up the good work.



__________________________________________________
Do You Yahoo!?
Send instant messages & get email alerts with Yahoo! Messenger.
http://im.yahoo.com/

On Sun, Aug 12, 2001 at 08:29:47AM -0700, Parker Charles wrote:
> In the future will exult make it possible to make an 
> option where you choice how the party follows you!
> The original or exult way?

None of the official developers seem to be interested in cloning the party
formation code of the original, so it would have to be contributed by
someone who was interested.  You seem to be interested...

> Also will there be an update on the mouse. I'm hoping
> that exult gets it to the original.

What does this mean?

-- 
	Paul

On Sunday 12 August 2001 08:29, Parker Charles wrote:
> In the future will exult make it possible to make an
> option where you choice how the party follows you!
> The original or exult way?

Yes, if someone writes (and supports) the code to have them follow the old 
way.

Someone on our forum has claimed that savegames no longer work with the 
latest snapshot.  I haven't tried Exult since Saturday, when I did lots of 
code-churning (so maybe that was the cause...).  Anyone else seen the problem?

At 9:51 Uhr -0700 13.08.2001, Jeff Freedman wrote:
>Someone on our forum has claimed that savegames no longer work with the
>latest snapshot.  I haven't tried Exult since Saturday, when I did lots of
>code-churning (so maybe that was the cause...).  Anyone else seen the problem?

No problem here with save games from early july. not sure about older 
stuff, but that shouldn't matter in this case.



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

On Tue, 14 Aug 2001 03:03, Max Horn wrote:
> At 9:51 Uhr -0700 13.08.2001, Jeff Freedman wrote:
> >Someone on our forum has claimed that savegames no longer work with the
> >latest snapshot.  I haven't tried Exult since Saturday, when I did lots of
> >code-churning (so maybe that was the cause...).  Anyone else seen the
> > problem?
>
> No problem here with save games from early july. not sure about older
> stuff, but that shouldn't matter in this case.

No problem with any save files created "recently" (withing the last 6-8 
weeks) either, in both BG and SI, load/save/load/save-in-another-slot/load, 
works as expected, AFAICT. "journey onword" (and the save/load from it) also 
works, as does starting a new game (and save/loading it).

Patrick

On Monday 13 August 2001 10:40, Patrick Burke wrote:

> No problem with any save files created "recently" (withing the last 6-8

Thanks.  I'll be testing tonight.  Maybe the user has corrupted savegames.

Max:  I was looking at the #Exult logs and saw that you were experience 
slowness in some spots.  Usually when this happens, it isn't because the 
pathfinder is slow, but because it's being called too often.  

I usually find these bugs by running in a Window and looking at the console 
output.  Generally there's someone 'looking for a foe' who never gives up, 
which is a combat bug.  (The tricky thing is that I can't turn off 'looking 
for a foe' completely, because you might open a locked door that was 
separating you from the monster.  Instead, the goal is to make the checks 
happen less frequently when they've failed enough times.)

At 15:31 Uhr -0700 13.08.2001, Jeff Freedman wrote:
>Max:  I was looking at the #Exult logs and saw that you were experience
>slowness in some spots.  Usually when this happens, it isn't because the
>pathfinder is slow, but because it's being called too often. 
>
>I usually find these bugs by running in a Window and looking at the console
>output.  Generally there's someone 'looking for a foe' who never gives up,
>which is a combat bug.  (The tricky thing is that I can't turn off 'looking
>for a foe' completely, because you might open a locked door that was
>separating you from the monster.  Instead, the goal is to make the checks
>happen less frequently when they've failed enough times.)

I'll check my stdout for this. Maybe I need to add additional cout 
calls, though, the problem occurs very strongly visible in the 
britain actor's hall, during a play ,)



On a different note, I just see that speech is a memory hog. For the 
intro, the 512 K VOC data get expanded to 8 MB, ok, fine - but then 
the PCB allocates 16 MB  (!) simply because it wants to be prepared 
to grow later - which it won't of course. Sure I don't care that much 
on my 448 MB system + VM, but I think wasting 8 MB like this is a bad 
thing in any case ;)


Maybe for our PCBs, doubling the size is not the best solution... we 
should make some performance testing and see what common sizes they 
have, and perhaps use a different scheme for reserver().



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

There are two bugs i like to report! U7part2 serpent
isle

In Monitor when you pass the knights test Lucilla
will only take on of the meats saying that she needs
the rest! In order for me to get it to work. 
I used the pick pocket mode and equip Lucilla with a
backpack before giving her the meat.

another is when the banquit in moonshade is over.
Magelord disappear into infinity.

And the map layers need work.




__________________________________________________
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
http://phonecard.yahoo.com/

There's something screwy with our exception-catching.  When I provided a
bad savegame file, and another time when I wasn't reading one of the
files properly, Exult exited at the top level without any hint of what
the error or file was. I think it was better before when we printed the
filename, and it would even be better just to crash where the error
occurred.  I'm not very experienced with C++ exceptions (only Java's)...

Here's a stack trace:

#1  0x4008b3f7 in pthread_kill (thread=1024, signo=6) at signals.c:63
signals.c:63: No such file or directory.
(gdb) up
#2  0x4008b7c0 in raise (sig=6) at signals.c:189
signals.c:189: No such file or directory.
(gdb) up
#3  0x4011aef3 in abort () at ../sysdeps/generic/abort.c:88
../sysdeps/generic/abort.c:88: No such file or directory.
(gdb) up
#4  0x400c8f75 in __default_terminate ()
(gdb) up
#5  0x400c8f9a in __terminate ()
(gdb) up
#6  0x400c99dc in __throw ()
(gdb) up
#7  0x8081c75 in main (argc=1, argv=0xbffff5e4) at exult.cc:238
238             return result;

I've looked at the 3 savegames included in the two conversation-crash
bug reports, but haven't seen any crashes.  WJP:  Perhaps your recent
changes fixed them.

Jeff Freedman wrote:
> 
> I've looked at the 3 savegames included in the two conversation-crash
> bug reports, but haven't seen any crashes.  WJP:  Perhaps your recent
> changes fixed them.

The Tseramed one was reported from the German version. Maybe that had
something to do with it?

At 0:03 Uhr -0700 21.08.2001, Jeff Freedman wrote:
>There's something screwy with our exception-catching.  When I provided a
>bad savegame file, and another time when I wasn't reading one of the
>files properly, Exult exited at the top level without any hint of what
>the error or file was.

That's not so good of course. In theory, all exceptions should 
generate some meaningful error message, unless it was forgotten to 
throw it right (i.e. with a message). Also, filenames should be 
printed out... hmm.

Isn't there a feature in gdb to break on exceptions being thrown? In 
metrowerks codewarrior there is, so it never occured to me that this 
could be a problem.



>I think it was better before when we printed the
>filename, and it would even be better just to crash where the error
>occurred.

I strongly disagree that it would be better to crash where it occured 
- this may be OK on a unix system, but on a MacOS classic, it could 
tear the whole system to the abyss...

What might be a good idea is to replace usages of throw with a macro; 
then we could make that macro do a fprintf(stderr); exit(1) for DEBUG 
buiilds, and a throw for non-DEBUG builds.

Something like this.


>I'm not very experienced with C++ exceptions (only Java's)...
>
>Here's a stack trace:
>
>#1  0x4008b3f7 in pthread_kill (thread=1024, signo=6) at signals.c:63
>signals.c:63: No such file or directory.
>(gdb) up
>#2  0x4008b7c0 in raise (sig=6) at signals.c:189
>signals.c:189: No such file or directory.
>(gdb) up
>#3  0x4011aef3 in abort () at ../sysdeps/generic/abort.c:88
>../sysdeps/generic/abort.c:88: No such file or directory.
>(gdb) up
>#4  0x400c8f75 in __default_terminate ()
>(gdb) up
>#5  0x400c8f9a in __terminate ()
>(gdb) up
>#6  0x400c99dc in __throw ()
>(gdb) up
>#7  0x8081c75 in main (argc=1, argv=0xbffff5e4) at exult.cc:238
>238             return result;

That is of course pretty unusable to track down where the problem occured.



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

At 10:59 Uhr +0200 21.08.2001, Willem Jan Palenstijn wrote:
>Jeff Freedman wrote:
>>
>>  I've looked at the 3 savegames included in the two conversation-crash
>>  bug reports, but haven't seen any crashes.  WJP:  Perhaps your recent
>>  changes fixed them.
>
>The Tseramed one was reported from the German version. Maybe that had
>something to do with it?

We should contact the bug originator and make it test again with his 
german version & a current build.


Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

>>I think it was better before when we printed the
>>filename, and it would even be better just to crash where the error
>>occurred.
>
>I strongly disagree that it would be better to crash where it occured - 
>this may be OK on a unix system, but on a MacOS classic, it could tear the 
>whole system to the abyss...
>
>What might be a good idea is to replace usages of throw with a macro; then 
>we could make that macro do a fprintf(stderr); exit(1) for DEBUG buiilds, 
>and a throw for non-DEBUG builds.
>
>Something like this.

We could also expand exult_exception to also take a file and line parameter 
and use a macro to throw. For instance #define THROW(exp,msg) throw (exp 
(msg, __FILE__, __LINE__));

Another alternative would be a macro and use assertions in debug and throws 
in release. Of course I'm don't know exactly how assertions work on all 
platforms.

On Tuesday 21 August 2001 04:11, Max Horn wrote:

> We should contact the bug originator and make it test again with his
> german version & a current build.

True.  Is there a way to get a stack trace on Windows?

On Tuesday 21 August 2001 04:10, Max Horn wrote:

> That's not so good of course. In theory, all exceptions should
> generate some meaningful error message, unless it was forgotten to
> throw it right (i.e. with a message). Also, filenames should be
> printed out... hmm.

Maybe we shouldn't have the 'catch' at the top level, since it has the 
side-effect of catching every possible file error lower down.  (In Java, that 
wouldn't happen:-)).

At 23:04 Uhr +0930 21.08.2001, Ryan Nunn wrote:
>>>I think it was better before when we printed the
>>>filename, and it would even be better just to crash where the error
>>>occurred.
>>
>>I strongly disagree that it would be better to crash where it 
>>occured - this may be OK on a unix system, but on a MacOS classic, 
>>it could tear the whole system to the abyss...
>>
>>What might be a good idea is to replace usages of throw with a 
>>macro; then we could make that macro do a fprintf(stderr); exit(1) 
>>for DEBUG buiilds, and a throw for non-DEBUG builds.
>>
>>Something like this.
>
>We could also expand exult_exception to also take a file and line 
>parameter and use a macro to throw. For instance #define 
>THROW(exp,msg) throw (exp (msg, __FILE__, __LINE__));

Yeah, very good idea!


>Another alternative would be a macro and use assertions in debug and 
>throws in release. Of course I'm don't know exactly how assertions 
>work on all platforms.

That's what I basically meant.



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

On Tuesday 21 August 2001 06:34, Ryan Nunn wrote:

> >What might be a good idea is to replace usages of throw with a macro; then
> >we could make that macro do a fprintf(stderr); exit(1) for DEBUG buiilds,
> >and a throw for non-DEBUG builds.
> >
> >Something like this.
>
> We could also expand exult_exception to also take a file and line parameter
> and use a macro to throw. For instance #define THROW(exp,msg) throw (exp
> (msg, __FILE__, __LINE__));

Yes, that sounds reasonable.  One thing I'm wondering:  Will U7open() still 
get a chance to print its 'file not found' error, or will the 'catch' happen 
first?

Okay, maybe we're not quite ready today.  But it feels like we're getting 
closer, as the rate of bug submissions seems to have slowed down.  The only 
major bug I want to fix is the Lydia-tatoo one that Dominik just submitted, 
as it probably has to do with the rewrite of 'path_run_usecode()' not being 
quite right yet.

So maybe, if we don't make too many major code changes, we can release by 
Friday.  There are certainly enough new features, especially the status 
displays and the plasma.

At 09:16 21/08/2001 -0700, you wrote:
>On Tuesday 21 August 2001 04:11, Max Horn wrote:
>
> > We should contact the bug originator and make it test again with his
> > german version & a current build.
>
>True.  Is there a way to get a stack trace on Windows?

Yes and no. By default Win9x doesn't give stack traces, while Win2k does. 
Dr Watson on both will give you a stack trace when an application crashes. 
However on Win9x it doesn't show the names of the functions. On Win2k it 
only shows the names from files that have debugging info in the Microsoft 
format.

On Tue, 21 Aug 2001 21:10, Max Horn wrote:
> At 0:03 Uhr -0700 21.08.2001, Jeff Freedman wrote:
>
> >I'm not very experienced with C++ exceptions (only Java's)...
> >
> >Here's a stack trace:
> >
> >#1  0x4008b3f7 in pthread_kill (thread=3D1024, signo=3D6) at signals.c=
:63
> >signals.c:63: No such file or directory.
> >(gdb) up
> >#2  0x4008b7c0 in raise (sig=3D6) at signals.c:189
> >signals.c:189: No such file or directory.
> >(gdb) up
> >#3  0x4011aef3 in abort () at ../sysdeps/generic/abort.c:88
> >../sysdeps/generic/abort.c:88: No such file or directory.
> >(gdb) up
> >#4  0x400c8f75 in __default_terminate ()
> >(gdb) up
> >#5  0x400c8f9a in __terminate ()
> >(gdb) up
> >#6  0x400c99dc in __throw ()
> >(gdb) up
> >#7  0x8081c75 in main (argc=3D1, argv=3D0xbffff5e4) at exult.cc:238
> >238             return result;
>
> That is of course pretty unusable to track down where the problem occur=
ed.

It probably crashed in a Destructor. For example:

// this class with throw an exception when at is called with _any_ value
class Foo
{
	public:
		Foo() {};
		int dref(const unsigned int &i) { _v.at(i); };

	private:
		vector<int> _v;
};

class Bar
{
	public:
		Bar()
		{
			Foo foo;
			foo.at(2); // this will throw an exception
			s =3D new string();
		};
		~Bar()
		{
			delete s;
		};

	private:
		string *s;
};

The order of creation of an object is malloc() the space, then execute th=
e=20
constuctor on the malloced space.

The order of destruction of an object is to execute the destructor, then=20
free() the space.

Destructors are ALWAYS called on static objects when exceptions are throw=
n.

When you constuct a static object of Bar, an exception will be thrown at =
the=20
point in the code calling foo.at(), because there is no try...catch brace=
s=20
following it. Not only will *s never be allocated, but you're calling a=20
destructor on a partially constucted object. Which, unsurprisingly is bad.

The re-written function follows.

class Bar
{
	public:
		Bar()
		{
			try
			{
				s =3D new string();
				Foo foo;
				foo.at(2); // this will throw an exception
			} catch(...) {
				// insert error handling here
			}
		};
		~Bar()
		{
			delete s;
		};

	private:
		string *s;
};

A more general exception, is that you should never call anything in a=20
constructor that may cause an exception... this is of course almost=20
impossible, <grin> thus the try...catch paranoia.

In this case it's looking like something is throwing a SIGABRT... from "m=
an 3=20
malloc" / "man 3 free":

      "Recent  versions of Linux libc (later than 5.4.23) and GNU
       libc (2.x) include a malloc implementation which  is  tun=AD
       able  via  environment  variables.   When MALLOC_CHECK_ is
       set, a special (less  efficient)  implementation  is  used
       which  is  designed  to be tolerant against simple errors,
       such as double calls of free() with the same argument,  or
       overruns of a single byte (off-by-one bugs).  Not all such
       errors can be proteced against, however, and memory  leaks
       can  result.   If  MALLOC_CHECK_ is set to 0, any detected
       heap corruption is silently ignored; if set to 1, a  diag=AD
       nostic  is  printed  on  stderr;  if  set to 2, abort() is
       called immediately.  This can be useful because  otherwise
       a  crash may happen much later, and the true cause for the
       problem is then very hard to track down."

A posibility perhaps.... <grin>

Apologies if this isn't particularly coherent, I'm about to fall asleep a=
t=20
the keyboard, after a long night of kernal compiling. <grin>

Have fun,

Patrick

On Tuesday 21 August 2001 10:01, Patrick Burke wrote:

>
> A more general exception, is that you should never call anything in a
> constructor that may cause an exception... this is of course almost
> impossible, <grin> thus the try...catch paranoia.

Guilty!  One of these days, I've got to rewrite the constructors in 
actorio.cc, replacing them with 'read()' methods.  But I don't think that was 
the cause of last night's crashes (where I added a savegame that I forgot to 
gunzip).

At 9:45 Uhr -0700 21.08.2001, Jeff Freedman wrote:
>Okay, maybe we're not quite ready today.  But it feels like we're getting
>closer, as the rate of bug submissions seems to have slowed down.  The only
>major bug I want to fix is the Lydia-tatoo one that Dominik just submitted,
>as it probably has to do with the rewrite of 'path_run_usecode()' not being
>quite right yet.
>
>So maybe, if we don't make too many major code changes, we can release by
>Friday.  There are certainly enough new features, especially the status
>displays and the plasma.
>

Friday would be great! I hoped we could release tomorrow, but you 
probably right in saying friday is more realistic. In any case, I can 
take the time to fine tune the OS X version a bit more (that mainly 
means writting code for SDL on OS X ;)

No major changes except bug fixes I'd say till friday.



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

>
> No major changes except bug fixes I'd say till friday.

I'll try not to get 'inspired' to do any major changes before then:-)

BTW, this is the first fully-working OS X release, isn't it?  Could Ultima 7 
possibly be the first commercial game for it?

At 10:18 Uhr -0700 21.08.2001, Jeff Freedman wrote:
>  >
>>  No major changes except bug fixes I'd say till friday.
>
>I'll try not to get 'inspired' to do any major changes before then:-)
>
>BTW, this is the first fully-working OS X release, isn't it?  Could Ultima 7
>possibly be the first commercial game for it?

No way you are too late ,)

Oni is running on OS X since quite some time.

And of course I can play many classic MacOS games on it, e.g. Diablo 
II runs just fine (albeit a bit slower than on classic).



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

Tuesday, August 21, 2001, 7:18:51 PM, 
Jeff Freedman <jsf@...> wrote:

>> No major changes except bug fixes I'd say till friday.

> I'll try not to get 'inspired' to do any major changes before then:-)

> BTW, this is the first fully-working OS X release, isn't it?  Could Ultima 7 
> possibly be the first commercial game for it?

release, release, release!!!!

Maybe contact Mac and let them praise Exult for this?
:-)
For FAQ updates I´m still unavaillable until the middle of next week!
Dom

I closed a bug or two, including the old 3-faces-in-a-conversation bug,
but then got caught by the CVS outage before I could commit my last
changes.  I'll try again in the morning.

>Okay, maybe we're not quite ready today.  But it feels like we're getting
>closer, as the rate of bug submissions seems to have slowed down.


Sorry, didn't plan to ruin your release date. Maybe you already fixed my
submitted bugs and I didn't notice :)

On Tuesday 21 August 2001 10:30 am, Tristan de Ines wrote:
> >Okay, maybe we're not quite ready today.  But it feels like we're getting
> >closer, as the rate of bug submissions seems to have slowed down.
>
> Sorry, didn't plan to ruin your release date. Maybe you already fixed my
> submitted bugs and I didn't notice :)

I recall fixing one of them, at least.

I just released the source and Linux x86 binaries for Alpha5 and tagged
CVS.  Now it's time to bug out:-)

From: "Jeff Freedman" <freedman@...>
> I just released the source and Linux x86 binaries for Alpha5 and tagged
> CVS.  Now it's time to bug out:-)

One small point, the date and version in news file were not updated for
Alpha5.

The Exult 0.94 alpha5 win32 binary is currently online at
http://members.optushome.com.au/veemon/exult-0.94alpha5-win32.exe
Could someone please add it to the Exult sourceforge site since I don't have
the access to do it myself.
And for any other users reading this message and wanting latest version
please use the current snapshot at
http://exult.sourceforge.net/snapshots/Exultwin32.exe instead since that
other site is very limited in bandwidth per month.

At 11:23 Uhr +1000 26.08.2001, Travis Howell wrote:
>The Exult 0.94 alpha5 win32 binary is currently online at
>http://members.optushome.com.au/veemon/exult-0.94alpha5-win32.exe
>Could someone please add it to the Exult sourceforge site since I don't have
>the access to do it myself.
>And for any other users reading this message and wanting latest version
>please use the current snapshot at
>http://exult.sourceforge.net/snapshots/Exultwin32.exe instead since that
>other site is very limited in bandwidth per month.


I will upload it to the SF page ASAP


Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

At 3:34 Uhr +0200 26.08.2001, Max Horn wrote:
>At 11:23 Uhr +1000 26.08.2001, Travis Howell wrote:
>>The Exult 0.94 alpha5 win32 binary is currently online at
>>http://members.optushome.com.au/veemon/exult-0.94alpha5-win32.exe
>>Could someone please add it to the Exult sourceforge site since I don't have
>>the access to do it myself.
>>And for any other users reading this message and wanting latest version
>>please use the current snapshot at
>>http://exult.sourceforge.net/snapshots/Exultwin32.exe instead since that
>>other site is very limited in bandwidth per month.
>
>
>I will upload it to the SF page ASAP

OK, it is up at 
http://prdownloads.sourceforge.net/exult/exult-0.94alpha5-win32.exe

Webpage is also updated.
Missing as of now are *BSD, OS X, Alpha and maybe a .deb package for 
linux. But Windows, Linux and classic MacOS are up. Front page has 
been updated, too.



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

Max Horn wrote:
> 
> At 3:34 Uhr +0200 26.08.2001, Max Horn wrote:
> >At 11:23 Uhr +1000 26.08.2001, Travis Howell wrote:
> >>The Exult 0.94 alpha5 win32 binary is currently online at
> >>http://members.optushome.com.au/veemon/exult-0.94alpha5-win32.exe

> OK, it is up at
> http://prdownloads.sourceforge.net/exult/exult-0.94alpha5-win32.exe
> 
> Webpage is also updated.
> Missing as of now are *BSD, OS X, Alpha and maybe a .deb package for
> linux. But Windows, Linux and classic MacOS are up. Front page has
> been updated, too.

Thanks!

On Sun, 26 Aug 2001 04:18:47 +0200
Max Horn <max@...> wrote:

> 
> Webpage is also updated.
> Missing as of now are *BSD, OS X, Alpha and maybe a .deb package for 
> linux. But Windows, Linux and classic MacOS are up. Front page has 
> been updated, too.
> 

Okay, time to see what this puppy can do :-)

> 
> 
> Max
> -- 
> -----------------------------------------------
> Max Horn
> Software Developer
> 
> email: <mailto:max@...>
> phone: (+49) 6151-494890
> 
> _______________________________________________
> Exult-general mailing list
> Exult-general@...
> http://lists.sourceforge.net/lists/listinfo/exult-general
> 


-- 
JP Morris - aka DOUG the Eagle (Dragon) -=UDIC=-  doug@...
Fun things to do with the Ultima games            http://www.it-he.org
Developing a U6/U7 clone                          http://ire.it-he.org
d+++ e+ N+ T++ Om U1234!56!7'!S'!8!9!KA u++ uC+++ uF+++ uG---- uLB----
uA--- nC+ nR---- nH+++ nP++ nI nPT nS nT wM- wC- y a(YEAR - 1976)

Max Horn <max@...> writes:

> Webpage is also updated.
> Missing as of now are *BSD, OS X, Alpha and maybe a .deb package for
> linux.

<URL:http://incoming.debian.org/> has a debian source package and a
i386 deb of exult 0.94alpha5. Barring mishaps, those should enter
Debian unstable later today ... binaries for other architectures will
follow over the week.

Feel free to snarf the package for the SF site. Alternatively, you may
want to just point users to Debian
(<URL:http://packages.debian.org/exult> is a fine target). In any
case, you should probably remove the old 0.93 .deb from the download
page.

FWIW, if you're not that interested in building your own debs, may I
ask that debian/* be removed from the snapshots/CVS? It makes my life
just a bit more complicated.

-- 
Robbe

http://www.stampede.org/~drax/exult-0.94alpha5.openbsd-2.9.tar.gz

Please upload this to the site.


On Fri, 24 Aug 2001, Jeff Freedman wrote:

> I just released the source and Linux x86 binaries for Alpha5 and tagged
> CVS.  Now it's time to bug out:-)
> 
> _______________________________________________
> Exult-general mailing list
> Exult-general@...
> http://lists.sourceforge.net/lists/listinfo/exult-general
>

At 17:44 Uhr -0700 26.08.2001, DraX wrote:
>http://www.stampede.org/~drax/exult-0.94alpha5.openbsd-2.9.tar.gz
>
>Please upload this to the site.

Done.


Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

What happened?  When I run autogen.sh, I get:
 
aclocal: configure.in: 277: macro `AM_PATH_SDL' not found in library

Configure reports:

./configure: line 642: syntax error near unexpected token
`AM_INIT_AUTOMAKE($PACKAGE,$VERSION,no-define)'
./configure: line 642: `AM_INIT_AUTOMAKE($PACKAGE,$VERSION,no-define)'

I've updated automake to that in SuSE 7.2 to no avail.  Next I'll try
downloading the latest source from GNU.

Kind of frustrating.

Whoops.  Problem solved.  Turns out aclocal wasn't looking in
/usr/local/share/aclocal, which is where the sdl.m4 file was stored.  I
added a "-I /usr/local/share/aclocal" to the autogen.sh script.  Will
that cause any trouble?

Looks like I also finally have to upgrade to the new SDL...

I looked at the crash in SI when you tell a companion to leave and ask
for his stuff.  The problem is in Container_game_object::remove_this(): 
It calls extract_contents(), which adds the contents of the container
back to the owner. This results in an infinite loop (not sure how
exactly yet). 

Perhaps extract_contents() shouldn't be called when the container has an
owner?

At 21:11 Uhr -0700 04.09.2001, Jeff Freedman wrote:
>Whoops.  Problem solved.  Turns out aclocal wasn't looking in
>/usr/local/share/aclocal, which is where the sdl.m4 file was stored.  I
>added a "-I /usr/local/share/aclocal" to the autogen.sh script.  Will
>that cause any trouble?

It might be for me since I have noe /ust/local/share/aclocal; for me 
that is /sw/share/aclocal.

Maybe we can put in tome if [-f ...]; statments in there.

I though about putting the sdl.m4 stuff into acincludes.m4, but that 
would bad; it would either break it for me, or for everybody else; 
sincce I am using latest CVS of SDL to be able to compile on MacOS X 
(yeah! my changes are *finally* in SDL CVS  since yesterdy, Sam 
Lantinga had time, he is very busy being one of the developers on 
Worlds of Warcraft, the new Blizzard thingy).



>Looks like I also finally have to upgrade to the new SDL...

=)



Max
-- 
-----------------------------------------------
Max Horn
Software Developer

email: <mailto:max@...>
phone: (+49) 6151-494890

Jeff Freedman wrote:
> 
> I looked at the crash in SI when you tell a companion to leave and ask
> for his stuff.  The problem is in Container_game_object::remove_this():
> It calls extract_contents(), which adds the contents of the container
> back to the owner. This results in an infinite loop (not sure how
> exactly yet).
> 
> Perhaps extract_contents() shouldn't be called when the container has an
> owner?
> 

Looking at the usecode function that transfers contents (09b3) that
isn't the case. This function doesn't actually move the items, but it
creates a new one (of the same shape, frame, quality, quantity) in your
inventory, and then deletes the old object.

The problem is that extract_contents causes an infinite loop when called
on a container that is owned by an actor. What happens is that extract
contents adds the object to the actor. Actor::add tries to find a free
equipment spot, fails, and then tries to put the object in a bag... the
bag it finds is the backpack... oops :-)

Maybe we should avoid putting items in a sub-container when dont_check
is set?

-Willem Jan

Willem Jan Palenstijn wrote:
> 
> Jeff Freedman wrote:
> >
> > I looked at the crash in SI when you tell a companion to leave and ask
> > for his stuff.  The problem is in Container_game_object::remove_this():
> > It calls extract_contents(), which adds the contents of the container
> > back to the owner. This results in an infinite loop (not sure how
> > exactly yet).
> >
> > Perhaps extract_contents() shouldn't be called when the container has an
> > owner?
> >
> 
> Looking at the usecode function that transfers contents (09b3) that
> isn't the case. This function doesn't actually move the items, but it
> creates a new one (of the same shape, frame, quality, quantity) in your
> inventory, and then deletes the old object.
> 
> The problem is that extract_contents causes an infinite loop when called
> on a container that is owned by an actor. What happens is that extract
> contents adds the object to the actor. Actor::add tries to find a free
> equipment spot, fails, and then tries to put the object in a bag... the
> bag it finds is the backpack... oops :-)
> 
> Maybe we should avoid putting items in a sub-container when dont_check
> is set?
> 

To follow up on this:

doing that fixes the crash, although I have no idea if it has any
unwanted side effects (I committed it anyway, though ;-) ).

It did leave the avatar carrying 54/36 weight. Are we missing a
weight-check somewhere?

-Willem Jan

Willem Jan Palenstijn <palenstijn@...> wrote:
> It did leave the avatar carrying 54/36 weight. Are we missing a
> weight-check somewhere?

I think this is how it worked in the original as well. Or did they say
something like "no more space in backpack - I´ll leave it on the
ground"?

On Wed, 5 Sep 2001, Dominik R. wrote:

>
> Willem Jan Palenstijn <palenstijn@...> wrote:
> > It did leave the avatar carrying 54/36 weight. Are we missing a
> > weight-check somewhere?
>
> I think this is how it worked in the original as well. Or did they say
> something like "no more space in backpack - I=B4ll leave it on the
> ground"?
>

The relevant usecode functions have some strings in them like "You are to=
o
encumbered, I'll place the remaining items on the ground"

Willem Jan Palenstijn <wpalenst@...> wrote:
>> Willem Jan Palenstijn <palenstijn@...> wrote:
>> > It did leave the avatar carrying 54/36 weight. Are we missing a
>> > weight-check somewhere?
>>
>> I think this is how it worked in the original as well. Or did they say
>> something like "no more space in backpack - I´ll leave it on the
>> ground"?
>>

> The relevant usecode functions have some strings in them like "You are too
> encumbered, I'll place the remaining items on the ground"

And a big bug when the companions you want to leave carry stacks of
items (arrows, bolts, coins). See for yourself (especially with stacks
of 100)- great way to make some money.
Dom

On Wed, 5 Sep 2001, Dominik R. wrote:

>
> Willem Jan Palenstijn <wpalenst@...> wrote:
> >> Willem Jan Palenstijn <palenstijn@...> wrote:
> >> > It did leave the avatar carrying 54/36 weight. Are we missing a
> >> > weight-check somewhere?
> >>
> >> I think this is how it worked in the original as well. Or did they s=
ay
> >> something like "no more space in backpack - I=B4ll leave it on the
> >> ground"?
> >>
>
> > The relevant usecode functions have some strings in them like "You ar=
e too
> > encumbered, I'll place the remaining items on the ground"
>
> And a big bug when the companions you want to leave carry stacks of
> items (arrows, bolts, coins). See for yourself (especially with stacks
> of 100)- great way to make some money.
> Dom
>

Not sure if this is the cause, but there is a bug in
Container_game_object::create_quantity when called from the item transfer
function. Both quality and quantity are passed. First one object is
created, and the quality is set to the passed quality. This, however,
causes the quantity to be set too, since quality and quantity are stored
in the same field. When create_quantity then tries to add
the remaining N-1 items, the quantity is actually set to 2N-1.

On Wed, 5 Sep 2001, Jeff Freedman wrote:

> Willem Jan Palenstijn wrote:
>
> >
> > Not sure if this is the cause, but there is a bug in
> > Container_game_object::create_quantity when called from the item transfer
> > function. Both quality and quantity are passed. First one object is
> > created, and the quality is set to the passed quality. This, however,
> > causes the quantity to be set too, since quality and quantity are stored
> > in the same field. When create_quantity then tries to add
> > the remaining N-1 items, the quantity is actually set to 2N-1.
>
> That does seem flawed.  But if we store quality and quantity as separate
> fields, I'm not sure how we'll store them.  It should be called with one
> or the other, but not both.  Or is it being called with the quality set
> to represent the quantity?

 One way to fix this would be to make the set_quality() function check if
the object does indeed have a quality. (using the object class)

We do have to check for not-proper uses of set_quality(), though. (Where
it actually tries to set something else than the quality)

-Willem Jan

Willem Jan Palenstijn wrote:

> 
> Not sure if this is the cause, but there is a bug in
> Container_game_object::create_quantity when called from the item transfer
> function. Both quality and quantity are passed. First one object is
> created, and the quality is set to the passed quality. This, however,
> causes the quantity to be set too, since quality and quantity are stored
> in the same field. When create_quantity then tries to add
> the remaining N-1 items, the quantity is actually set to 2N-1.

That does seem flawed.  But if we store quality and quantity as separate
fields, I'm not sure how we'll store them.  It should be called with one
or the other, but not both.  Or is it being called with the quality set
to represent the quantity?

Fixed.  I just made sure the packpack was removed first.

As for the create_quantity() bug, I think the problem is that my version
of add_cont_items() isn't returning the right value.

Willem Jan Palenstijn wrote:

> > And a big bug when the companions you want to leave carry stacks of
> > items (arrows, bolts, coins). See for yourself (especially with stacks
> > of 100)- great way to make some money.

> 
> Not sure if this is the cause, but there is a bug in
> Container_game_object::create_quantity when called from the item transfer
> function. Both quality and quantity are passed. 

Okay, after a bit of hacking, this part of the bug is fixed.  And weight
is now checked as well.

But it's still not perfect, since items seem to disappear when there's
no capacity and no one else to carry them.  I think our interpretation
of the 'add_party_items()' intrinsic is wrong for SI, so the Usecode
doesn't know that it failed.  It will take some detailed tracing of the
Usecode (only two places call it).

Purely a guess:  It's the tailor who shows up repeatedly in these bad
savegames.  And it's the 'sew schedule' where we've had several
memory-corruption bugs.  Could those be the cause?

Jeff Freedman wrote:

>On Tuesday 21 August 2001 10:01, Patrick Burke wrote:
>
>>A more general exception, is that you should never call anything in a
>>constructor that may cause an exception... this is of course almost
>>impossible, <grin> thus the try...catch paranoia.
>>
>
>Guilty!  One of these days, I've got to rewrite the constructors in 
>actorio.cc, replacing them with 'read()' methods.  But I don't think that was 
>the cause of last night's crashes (where I added a savegame that I forgot to 
>gunzip).
>
>
>_______________________________________________
>Exult-general mailing list
>Exult-general@...
>http://lists.sourceforge.net/lists/listinfo/exult-general
>
Actually, NEVER EVER throw in a DEstructor. You _can_ throw in a 
CONstructor, but you have to remember that the object that does it won't 
have it's destructor called since only 'fully constructed' objects are 
subject to automatic destruction when an exception is thrown. Partially 
constructed objects (ie: those which have not had their constructors 
complete cleanly will not be destructed.

D

Hi Dude,I just visit you site and I like it.
I want to offer you to link exchange.

My link exchange is eazy.
Just submit your site to my partners center,and put link back to my
site.its all.
your site will be show on all my pages.
please go to
http://www.hotfreesite.com/cgi-bin/partners/add.cgi
for more info.

best regards.

In the future! Is is possible to make Vasculio in
Serpent
Isle a playable character and also will it be possible
to merge Ultima and Serpent Isle into one Big Game?

__________________________________________________
Do You Yahoo!?
Find a job, post your resume.
http://careers.yahoo.com

Not really, no. That would require violating the copyright of
Origin Systems and EA, which would be wrong.

D


Parker Charles wrote:

>In the future! Is is possible to make Vasculio in
>Serpent
>Isle a playable character and also will it be possible
>to merge Ultima and Serpent Isle into one Big Game?
>
>__________________________________________________
>Do You Yahoo!?
>Find a job, post your resume.
>http://careers.yahoo.com
>
>_______________________________________________
>Exult-general mailing list
>Exult-general@...
>https://lists.sourceforge.net/lists/listinfo/exult-general
>

What its someone lacks the ability to add the patch
for the program. Is it possible that the exult team 
can patch the Exult Studio in the next version of the
beta?

I tried patching it. But it either crash or doesn't
see it.

__________________________________________________
Do You Yahoo!?
Send FREE video emails in Yahoo! Mail!
http://promo.yahoo.com/videomail/