Not totally sure what you mean by having functions allow
copy-by-reference. Something else/in addition to ref()?
Basically, ref() does it all. I meant like... how in c++:
void doX(int x) {}
...passes in a reference to x, without forcing the programmer to learn anything
about pointers, and
An alternative to a copy() operator could be a ref()
operator.
Personally, I think a ref() operator is somewhat inconsistent. This is how I
see it:
1) When I call copy(), some work is done, and a copy of my string is MADE --it
didn't exist before.
2) When I call ref(), what happens? Is a
Nice! However, the JAR file doesn't work for me:
Weird. Try from the command line/shell:
java -jar Desktop_HVM_0.1.jar
Failing that, there's a slim chance that your JVM just hates you, and isn't
including the JAR itself in the classpath. Then try:
java -cp Desktop_HVM_0.1.jar
That is cool! I am glad to see this coming along so nicely.
Three cheers
for Seth! ... and 1,1,2,3,5,8... cheers for Fibonacci!
Thanks! Also, I fixed that bug; now all the examples from the wiki work. Error
reporting's so-so, but I'm more than happy to help you guys debug HF scripts
Hey all,
The latest screenie of the HVM in action is here:
http://gilgamesh.hamsterrepublic.com/wiki/ohrrpgce/index.php/User:Sorlok_reaves
It can now run everything that a Pistachio HVM is required to handle. The
screenshot shows the Fibonacci sequence being calculated correctly for
variable (a, b)
a := A
b := a
b += B
For the record, here's what Java does:
1) a is A
2) b is A (same reference)
3) b is AB, and a is still A
That's because Java enforces two properties about strings, at all times:
A) Two strings with the same value MUST have the same reference.
B) Any
The desktop VM looks nice. Gives me the urge - but not the
time - to play with it.
I'll mail the list when the desktop VM is ready for release. Hope to get some
good sample code going, too.
The VM documentation is really sprawling. Since it's so
huge I've been
avoiding reading it until
Hey y'all. Just thought I'd drop a line to let everyone know the progress being
made on the OHRRPGCE FMF. Currently, development is focusing on the Henceforth
Virtual Machine, which is described here:
http://gilgamesh.hamsterrepublic.com/wiki/ohrrpgce/index.php/Henceforth_VM
As you can tell by
I like this approach to slices a lot! In a few releases, I'll incorporate this
approach into the OHRRPGCE FMF --it makes a lot more sense.
Here's an idea: make a slice insurmountable. If you have:
parent_slice
child_slice_1
child_slice_2
child_slice_3, (insurmountable = true)
Given my limited knowledge of how the Standard OHR actually works, this still
seems to handle things well. My main question is: what takes up slots 32 to
9? I don't like the idea of excessive wasted space, unless it's
unavoidable.
--Seth
--- On Wed, 1/14/09, James Paige
at 09:38:35PM -0800, S'orlok
Reaves wrote:
Given my limited knowledge of how the Standard OHR
actually works, this still seems to handle things well. My
main question is: what takes up slots 32 to 9? I
don't like the idea of excessive wasted space, unless
it's unavoidable.
--Seth
I've played through WH several times before
Werewaffle, and I remember:
1) The cactus fleeing when you're behind the castle.
2) The cactus jumping off the mountain (didn't notice
the spinning, but I guess it's there).
--Seth
--- James Paige [EMAIL PROTECTED] wrote:
On Mon, Oct 06, 2008 at
for example, you want a sparkle sprite to
be positioned to the leader but parented to map
layer 2
Just a quick nudge: an excellent demonstration of
Slices would be a minimap, in the lower-left corner of
the screen, that updates your position on said map in
real-time. I'm thinking
Actually, now that I type this, why should the slice
drawing behaviour be modal?
Haha, I started typing my last email before I saw this
post; seems we've come to many of the same conclusions
in our own different ways.
I guess since I have the podium, I'll just say that
I'm really proud some
Not to bust in on your debate, but I think the idea is
that pumping max stats with spells might be useful for
some avant-garde games. Consider:
1) A game with fixed max MP (of, say, 100) and no way
to restore it inside dungeons. The first room of a
dungeon resets your stats, and the max-stat
except that attempts to do indenting
Haha, yeah, indenting sucks. I want to add the option
to strip indents at some point
if the value of a ${} code changes when the box
is being displayed, the box changes?
Oh, no, that's not what I meant. That would, indeed,
cause errors as you
Hmmm... what about the case where somebody writes a
line with a ${} code
in it that expans and makes the line too long. Is
there any reasonable
way to handle that case at run-time rather than
edit-time?
The OHRRPGCE FMF handles this by dyanmically expanding
text boxes at run time, and
I can't wait for OHR ports to handheld
devices.
It's getting there; I've just been a bit busy these
days. :)
--Seth
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
I am really curious as to whether or not a
frequently updated file would
be a performance hit or not.
Knowing the way Windows locks files (seemingly
randomly) you might *cough* (crash your system)
*cough*. Maybe there's a better way? Like CUSTOM
(in Windows) is just based off of Win32
Hmm just thought of this:
For the current process ID, you could create a file
called, say, 3455.pid in the temp directory. Make it 1
byte, and RE-create it every 10 seconds. I don't think
this is too intensive... and it gives you PID
identification accurate to 10 seconds.
...Or is the
Hey all,
I bet you were all just thinking, right now, Gee,
I wish I could browse the subversion history of the
OHRRPGCE FMF! I know I was! And now you can. We now
have a google code page:
http://code.google.com/p/ohrrpgce-fmf/
...which is mainly for source code. All bugs will
still be
Sorry for the late reply; your message was filtered as
spam. Looking closely, MY OWN MESSAGE from my yahoo
account to the list to the SAME yahoo account was also
filtered as spam. Way to go, Yahoo.
I'm working on a GUI-from-scratch as
well..., taking an identical approach, so will be
to the wiki; I expect that
some people might want to extend the FMF's main menu
for their own games, in the far future.
Cheers,
--Seth
--- James Paige [EMAIL PROTECTED] wrote:
What, praytell, is a slice?
---
James
On Mon, Mar 31, 2008 at 11:20:17AM -0700, S'orlok
Reaves wrote:
I've
I've been refactoring the FMF's menu system to be more
integrated... it now works off of something called a
slice which is faster and more powerful. So I didn't
really notice the lack of mail...
PyWeek looks cool. Best of luck!
--Seth
--- Simon Bradley [EMAIL PROTECTED] wrote:
On Mon, Mar
Ah, good, Yahoo's back.
The best way to organize this is to have your
higher-level language spec, and then translate
programs into syntax trees using a parser-generator
(lex/yacc, antlr, javacc are all good).
You can optimize this tree independently, and then use
it to output a bytecode of your
to confirm this, but I have spent
tens of hours of my
life trying to get something going with those types
of programs, and I
have failed every time.
I'm not looking forward to trying again, but maybe I
will someday.
On Thu, Mar 13, 2008 at 10:13 PM, S'orlok Reaves
[EMAIL PROTECTED] wrote
()). So, when I've gotten my nerves
back in order,
I'll see what I can do with that.
On Thu, Mar 13, 2008 at 10:37 PM, S'orlok Reaves
[EMAIL PROTECTED] wrote:
Ah, ok. I _do_ have successful projects with all
of
these (I'm not just repeating the popular
options) but
I can see
Hey James,
Sorry to take so long, but I finally got time to
look at the code. The first tutorial works for me.
I'll be trying the rest out later this week.
So far, good job. I like the fact the game engine
can be started in so little code. Keep up the good
work!
--Seth
--- James Paige
Other people might be better informed on this one, but
I'd expect that the graphics card doesn't have a lot
to say about the script interpreter.
--Seth
How much of that is this option, and how much of
that is the fact that
the OHR is not letting the CPU offload any help to
the graphics
What's wrong with NPCTAG?
James walks too fast.
I'm serious! I manually compiled the castle script, so
that's the only one you can test. To get past the plot
boulders, you need to talk to James. But, currently,
he walks too fast; it's hard to catch him. I didn't
want people to get frustrated
The latest fortnightly of the FMF is up:
http://fmf.ohrdev.com/files/fortnightly/webdemo.jnlp
Be warned that it's rigged:
- You can only play WH
- You start at the castle
The idea is to show you all how plotscripting is
coming along. Walk up into the blue X's and see a
guard come out and
Hey all,
If you have time, could you please help me answer
this question? I don't mind going through the source
code, but this problem is a bit big-picture, as it
involves threads (which haven't been implemented in
source yet).
Consider a hero on tile (X,Y). To the right, on (X+1,
Y), is NPC
Thanks, Mike. That sounds good to me; I think I can
implement the current behavior leave room for the
future one with no problem. Mutexes are a good simple
way of managing concurrency; at least, the FMF will
have no problem implementing them.
Two more questions, just to clear things up:
1)
I also tried checking current map throughout all
instances of write map block in the loop, and still
nothing changed.)
This doesn't seem right... shouldn't current map
return the current map's ID? I.e., shouldn't current
map be updated AFTER switching maps but BEFORE
updating the current
that? Then ditch that reference when the
enemy has fully dissolved.
(- SR
--- Mike Caron [EMAIL PROTECTED] wrote:
On Nov 22, 2007 3:39 PM, James Paige
[EMAIL PROTECTED] wrote:
On Thu, Nov 22, 2007 at 11:32:03AM -0800, S'orlok
Reaves wrote:
I've changed line 2003 to read:
#define
I've changed line 2003 to read:
#define DEFAULT_TRANSITION 3
Then I run make. game.exe is created (checked the
datestamp) and when I load a game (I tried both wander
and vikings) its lump data is updated. All is going
well...
However, the old death animation persists... even if I
load a new
Sorry, just noticed the related bug whereby
transitions get reset to 1:
For completeness, I also reverted (updated) to this
exact revision (1634) and the problem remains. I
expected it to: I'm not seeing Mike's new default,
I'm seeing the old one.
Any ideas?
(- SR
--- [EMAIL PROTECTED]
Mostly for Kizul, I revved up the FMF's speed. It
should run noticeably quicker (Large Screen should
still be left un-ticked.)
(- SR
PS: Thanks to Adam for testing my Ant script.
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam
:
On Thu, Oct 04, 2007 at 10:01:03PM -0700, S'orlok
Reaves wrote:
Hello all!
This paragraph's mostly for Adam: I replaced the
Applet with a proper Swing App. It works just fine
for
me (except that Gisli's menu crashes.) Note to
applet
users: the applet WILL be back up, as soon as Adam
Hello all!
This paragraph's mostly for Adam: I replaced the
Applet with a proper Swing App. It works just fine for
me (except that Gisli's menu crashes.) Note to applet
users: the applet WILL be back up, as soon as Adam
confirms if this solution works for him or not. As
always, experiment via:
Okay! I've abused my .htaccess powers again to create
a Java Web Start demo; this is leaps and bounds above
the applet ---even though it's STILL an applet. Figure
that one out. Or, click on:
http://fmf.ohrdev.com/files/webdemo.jnlp
(Note that the link from the main page now points
here.)
Adam,
Ah! Thanks for the comments. Let's see:
1) Keys on the game selection meny are flakey.
This is because I do a lousy job simulating the DOS
timer. I'll probably end up buffering key input, so
if you hit Enter in the period that the DOS timer
wouldn't tick, it'll stay stuck until it does tick.
Hey all,
In honor of the release of Ubersetzung, and because
it's sooo difficult to play the Midlet otherwise, I've
made the OHRPGCEFMF web-enabled! Here is a link to the
applet:
http://fmf.ohrdev.com/files/webdemo.html
If it runs too slowly for you, de-select the Large
Screen box. I've had
Which version do you have? I used 1.5 for convenience
(generics and for each loops) but I am willing to
backport it to 1.4 if you are using this. If it is
earlier than 1.4 I am uneasy
--Seth
--- Adam Perry [EMAIL PROTECTED] wrote:
On 10/1/07, James Paige [EMAIL PROTECTED]
wrote:
That
ASAP.
--Seth
--- Kizul Emeraldfire [EMAIL PROTECTED]
wrote:
On 10/1/07, S'orlok Reaves [EMAIL PROTECTED]
wrote:
Hey all,
In honor of the release of Ubersetzung, and
because
it's sooo difficult to play the Midlet otherwise,
I've
made the OHRPGCEFMF web-enabled! Here is a link
Okay, I've updated the page:
http://fmf.ohrdev.com/files/webdemo.html
To Adam: I set the compliance level to 1.4. I see no
reason why this won't work, but please send the
console output if it fails and I'll work on fixing it.
To Kiz: At the bottom of the webdemo page are two
links, and
Ah! I use SciTE for Ruby. Fantastic program!
Hmm... James might know this one best: Are there any
editors that support plugins of the kind that link
words with the documentation.
So, e.g., if you open a script in this editor and
hover over suspend player, then it will show a small
caption
Then just wait until I implement plotscripts in the
FMF. I can't guarentee what language .HSX will be
converted to, but I CAN guarentee that it'll be
stack-based. ;)
(- sR
--- James Paige [EMAIL PROTECTED] wrote:
On Sat, Sep 01, 2007 at 04:50:10AM +1200, Ralph
Versteegen wrote:
On 9/1/07,
Indeed... I think it was also part of an integral
chapter in the Wiki -yep, here:
http://gilgamesh.hamsterrepublic.com/wiki/ohrrpgce/index.php/Battles
Er... I'd change it to a new example myself, but this
would just feel dirty.
(- sR
--- Kizul Emeraldfire [EMAIL PROTECTED]
wrote:
*sniffs*
James,
I've been toying with the floppy release (playing
sounds in game, editing in custom, etc.) and so far
it's fine. (I'll install the windows one later when I
have time.)
But... what exactly should we be testing? I'll go
through all the things in whatsnew.txt, but is there
anything else
Perhaps you'd all benefit from pausing for a minute to
consider not what reflect should do
(programmatically), but rather what it should DO
(e.g., why we have it.)
I've put some (boring) examples below; my point is
that you should seperate reflect into its most BASIC
components. For instance:
Hello,
GAME and CUSTOM build correctly; thanks.
From the title screen, starting a new game in Wander
crashes (your standard Windows XP game.exe has
encountered a problem and needs to close.) I deleted
wander.sav, and it still crashes.
Any idea what's up?
(- SR
--- [EMAIL PROTECTED]
Hello all,
I know it's (probably) abundantly clear by now, but
this fix also fixed the problem I was having with
WANDER.
Cheers!
(- SR
--- [EMAIL PROTECTED] wrote:
http://gilgamesh.hamsterrepublic.com/cgi-bin/bugzilla/show_bug.cgi?id=435
[EMAIL PROTECTED] changed:
Hooray! However
mapsubs.bas(1487) : error 59: Illegal inside a SUB or
FUNCTION, before: ')'
DIM destdoor() as door
^
...occurs when I try to compile CUSTOM, after a full
update. (GAME compiles fine)
(- SR
--- [EMAIL PROTECTED] wrote:
james
2007-08-18 22:13:33 -0700
I'm not filing this as a glitch, because I'm guessing
it's just a bug in WANDER:
The beds in the Monestary are covered by top-level
rug tiles -you can only see the pillows.
(- SR
Looking for a deal?
--- James Paige [EMAIL PROTECTED] wrote:
truncated
My vote is for refusing to import WAV if oggenc
isn't available.
I concur. If we assume oggenc is easy to install, then
we need only allow WAV if it presents some benefit.
(WAV's higher quality is not a good argument; the
difference only
Good afternoon (midnight) all:
It is my pleasure to announce, in anticipation of
Ubersetzung, the latest alpha release of the OHRRPGCE
FMF: Prima Facie. For posterity's sake, I'll send an
email to the list when a new version is released from
now on.
This version features a nascent menu
I just verified that the latest source stores 00 00
for SFXDATA.BIN's size in BINSIZE.BIN for new games as
well.
--Seth
--- David Gowers [EMAIL PROTECTED] wrote:
As I was reading The Art of Unix Programming
(http://www.faqs.org/docs/artu/) I was struck with
the idea to
implement a
--- Comment #9 from [EMAIL PROTECTED]
[paraphrase] What all should the delete tile key
delete? ... I always hated having to cycle all the
way back to zero...
Perhaps a local delete key (say, Del) that deletes
the data you're currently editing (foemap) and then a
global delete key
--- Mike Caron [EMAIL PROTECTED] wrote:
On 4/29/07, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
teeemcee
2007-04-29 18:07:09 +0800 (Sun, 29 Apr 2007)
33
Shortcut evaluation for || and
---
U wip/docs/plotdict.xml
U wip/game.bas
U wip/moresubs.bas
U wip/plotscr.hsd
--- Mike Caron [EMAIL PROTECTED] wrote:
Well, the rule of thumb is: If there's a way,
someone has done it
I can't think of any particular reason why you would
want it to *not*
short circuit, but then, I'm not everyone in the
universe.
The only one I can think of is side-effects.
Hello,
In yetmore.bas:loadsay(), saytag(2) is checked, and
the appropriate tags are set to on/off. Then, the text
box is displayed. Finally, in game.bas, when the box
closes, all effects of the box are computed (inns are
shown, etc.)
For the OHRRPGCEFMF, can I safely move the
tag-setting
.
Then the user will see text box 15 (not 16), and will
gain $50.?
If that is the case, then I understand. Thanks. :)
(- SR
--- Bob the Hamster [EMAIL PROTECTED] wrote:
On Mon, Apr 16, 2007 at 01:46:02AM -0700, S'orlok
Reaves wrote:
Hello,
In yetmore.bas:loadsay(), saytag(2) is checked
Indeed, this is the most important. As for the
black-pixel dilemma, don't lose heart. Remember that
CUSTOM could, when a layer is enabled, prompt the user
and replace all black in a tileset with 0x63 0x63 0x62
(nearly black). Or, there can be a means of setting
the transparent color.
What I
Here's my experience, using Edgy to run Trailblazers.
(I promise I'll try it on Vista soon! Just need to
steal my roommate's computer.)
1) Everything runs fine, and at the proper speed.
2) Music doesn't work.
The second one is (possibly) due to me not installing
Timidity correctly. I'll tinker
] wrote:
On Fri, Apr 13, 2007 at 12:17:53PM -0700, S'orlok
Reaves wrote:
Here's my experience, using Edgy to run
Trailblazers.
(I promise I'll try it on Vista soon! Just need to
steal my roommate's computer.)
1) Everything runs fine, and at the proper speed.
2) Music doesn't work
Keith,
Could you please download the latest OHRRPGCE FMF
(fort)nightly from:
http://fmf.ohrdev.com/files/fortnightly/GAME_FMF.jad
http://fmf.ohrdev.com/files/fortnightly/GAME_FMF.jar
and run WANDER (no need to run the other two). The
error messages have been updated; please re-type (or
just
How can we have our cake and eat it too?
Make all chairs NPCs, and change the draw order to
draw the highest NPCs first? This is how RPG maker
does it, if I recall. However, it DEFINITELY breaks
existing games, so you'd need a new bit: DRAW_FROM_TOP_DOWN.
That would not support heroes, so isn't really a
better solution.
Ooh, snap. I thought it was clear that when I said
NPCs I meant heroes too. Certainly, the blunder
was on my part. So you'd have three choices:
1) Draw heroes first.
2) Draw NPCs first.
3) Draw ALL heroes/NPCs ordered by
Remember that you're working with an (unstable)
nightly.
Speaking of which... since uber isn't out yet, you
could simply re-specify the map format: if layers are
ON, then a tile's Overhead settings are IGNORED.
I know, this sounds crazy, but consider the arguments
thus far:
Camp 1) We must keep
The biggest problem here is not the specific
behaviour of overhead tiles, but
the fact that they are not drawn the same way in
custom as they are in game. If
they were drawn consistently, then then it would be
possible to spot the
problem immediately when editing your map, rather
than
PROTECTED]
wrote:
On Mon, Mar 05, 2007 at 12:16:12AM -0600, Keith
Gable wrote:
On 3/5/07, Bob the Hamster
[EMAIL PROTECTED] wrote:
On Sun, Mar 04, 2007 at 09:37:10PM -0800,
S'orlok Reaves wrote:
Hey all,
For the past few days, Gilgamesh has been
acting
up
Hey all,
For the past few days, Gilgamesh has been acting
up. Is this a common problem for everyone?
Also, the next version of the OHRRPGCE FMF is out
(Amphitheatre); here you go:
http://www.comp.nus.edu.sg/~sethhetu/segments/alpha_amphitheatre.zip
This one has a lot more content: message
--- Mike Caron [EMAIL PROTECTED] wrote:
On 2/15/07, S'orlok Reaves [EMAIL PROTECTED]
wrote:
Technically, this makes these features
available
to plotscripting strings, but to use them, you'd
need to embed those characters directly into the
plotscript file, which while doable
Gentlemen, please. If you must quip, I'd ask you not
to Reply-All.
About the issue of keys, I feel we've reached an
impasse. I like the Alt+1/2/3 idea. Perhaps Alt++
can turn the layers above the current one off.
Still, I have a question:
Do we _need_ shortcuts to make the top two layers
Mike,
Thanks for the detailed description; I feel more
confident with the format now. I know what my problem
was: I forgot that importing a font changes the font
for CUSTOM *as well as* GAME and the font I
imported was offset by a few characters.
Question: Is this appropriate? Should
76 matches
Mail list logo
