my sprite demo has now been released.
its not perfect but it does work.

<a
href="http://www.jumpstation.co.uk/arrwdemo/">http://www.jumpstation.c
o.uk/arrwdemo/</a>

Enter your vote today!  Check out the new poll for the vmu-dev
group:


Spam/Advert OT postings.  Should the poster be banned?  I'm not naming
any names, but anyone who's been on this list will know who this is
directed to (posts that are just automated website plugs)

I'd like to keep this democratic, so what do you think?

(Oh yeah, and I'm keeping this poll anonymous, to prevent lashbacks!)

   o Don't ban - No harm's been done
   o Ban if OT posts continue
   o Ban straight away
   o Moderate posts from this person


To vote, please visit the following web page:

http://www.egroups.com/polls/vmu-dev

Note: Please do not reply to this message. Poll votes are
not collected via email. To vote, you must go to the eGroups
web site listed above.

Thanks!

The Game Room plays a psychopath with an axe in Nightmare Creatures
2, which isn't much of stretch from reality for us...
http://www.projectgames.com

What happens when you combine axe chopping bloody mayhem, loud Rob
Zombie music and 1984 graphics on the Dreamcast? You get Nightmare
Creatures 2! This is not a bad game by any means, but is the
dictionary definition of "shovel-ware," where a Playstation game was
just shoveled over to the Dreamcast without any graphics
enhancements. Now, any game with mass beheadings and Zombie's music
can't be all bad, and it isn't. It's a fun game. But shame on the
programmers for not making it look better.

can any one tell me where i can get the jap godzilla vmu game
i accdently deleated it of my memory card

email me at sonic2000@...
thanks

Is it my port? try "vmuasm -V"

I've got version 1.8 on my website and I've compiled a version 1.9 this
morning and hope to test it and post it tonight.

-- john maushammer

www.maushammer.com/vmu.html

rednuht wrote:

> the dos port i have does not even tell you which version it is.
>
>
> --- mezmo@... wrote: > > > Am I using an old version of
> vmuasm?
> (aslc8kc v1.8 by Marcus)
> > >
> > > Yes you are.  Get v1.9.
> >
> > Is there a precompiled version for dos/win available??
> >
> > -TL
> >
> >
>
>
> =====
> Later ...
>
> rednuht@...
> -+-..=|<[{(_"!"_)}]>|=..-+-
>

the dos port i have does not even tell you which version it is.


--- mezmo@... wrote: > > > Am I using an old version of vmuasm?
(aslc8kc v1.8 by Marcus)
> >
> > Yes you are.  Get v1.9.
>
> Is there a precompiled version for dos/win available??
>
> -TL
>
>


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail - Free email you can access from anywhere!
http://mail.yahoo.com/

I try and avoid it :)

im only using 1.7 so it proberly that

--- Marcus Comstedt <marcus@...> wrote: > --- In
vmu-dev@egroups.com, rednuht <rednuht@r...> wrote:
> > If you type in a lot of code in one go and make a simple mistake
> (i.e.
> > missing an attribute for a mnuemmonic DEC with no operand)
> > then it seems to accept the next instruction as the value that
> should
> > have been there and so on cascading down the code, very anoying, i
> > ended up commenting out all my new code and then uncommenting
> > assembling uncommenting assembling etc.
>
> Hm?  Can you show me an example of this?  If I assemble for example
>
>  .org 0
>
>  nop
>  dec
>  mov #0,$100
>  ret
>
> I only get one error, a parse error on the line immediately after the
> erroneous "dec".  How do you trigger the "cascading" thing?
>
>
>   // Marcus
>
>
>
>
>


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail - Free email you can access from anywhere!
http://mail.yahoo.com/

> > Am I using an old version of vmuasm? (aslc8kc v1.8 by Marcus)
>
> Yes you are.  Get v1.9.

Is there a precompiled version for dos/win available??

-TL

--- In vmu-dev@egroups.com, rednuht <rednuht@r...> wrote:
> If you type in a lot of code in one go and make a simple mistake
(i.e.
> missing an attribute for a mnuemmonic DEC with no operand)
> then it seems to accept the next instruction as the value that
should
> have been there and so on cascading down the code, very anoying, i
> ended up commenting out all my new code and then uncommenting
> assembling uncommenting assembling etc.

Hm?  Can you show me an example of this?  If I assemble for example

	 .org 0

	 nop
	 dec
	 mov #0,$100
	 ret

I only get one error, a parse error on the line immediately after the
erroneous "dec".  How do you trigger the "cascading" thing?


   // Marcus

--- In vmu-dev@egroups.com, mezmo@h... wrote:
> Am I using an old version of vmuasm? (aslc8kc v1.8 by Marcus)

Yes you are.  Get v1.9.


   // Marcus

If you type in a lot of code in one go and make a simple mistake (i.e.
missing an attribute for a mnuemmonic DEC with no operand)
then it seems to accept the next instruction as the value that should
have been there and so on cascading down the code, very anoying, i
ended up commenting out all my new code and then uncommenting
assembling uncommenting assembling etc.


--- mezmo@... wrote: > Hi.
>
> Problem:
> As my assembler code grows bigger, my 8-bit relative branches grow
> out of range. The problem is that the assembler lists the last line
> of the code as the line containing the error, thus leaving me no
> means whatsoever to find out where the actual error was generated.
>
> Is there something that can be done about this? Except, of course, to
>
> code smarter. (But that is really out of the question at this
> time.  :-)
>
> Am I using an old version of vmuasm? (aslc8kc v1.8 by Marcus)
>
> regards -TL
>
>
>


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/

Hi.

Problem:
As my assembler code grows bigger, my 8-bit relative branches grow
out of range. The problem is that the assembler lists the last line
of the code as the line containing the error, thus leaving me no
means whatsoever to find out where the actual error was generated.

Is there something that can be done about this? Except, of course, to
code smarter. (But that is really out of the question at this
time.  :-)

Am I using an old version of vmuasm? (aslc8kc v1.8 by Marcus)

regards -TL

--- In vmu-dev@egroups.com, biggs_avalanche@h... wrote:
> post by torbjorn
> > Primarily an option to monitor the registers (and variables, or
any
> > address in memory for that matter). And that cpu cycle count that
> > john mentions would really come in handy too.
> >
> > Now all we need is someone to code this!!  ;-D
> > (Add debugging to SoftVMS 2.0? YesPlease!!!)
> >
> > -Torbjorn
>
> Seems like a great idea, but I was wondering though, by "any
address
> in memory" should there by just a display of all the RAM at once?
or
> does anybody else have a better idea? It's a good idea to have the
> ACC, B, and C registers always displayed, right? But for other RAM
> areas, maybe just 16 - 64? entries in the RAM at a time? or all at
> once?

I'm imagining it like this:

The debugger has a console window. You have some variables that you'd
like to check the value of. You know that var1 is at 30h and var2 is
at 31h and so on. You could maybe do something like "? $30" to
display contents at 30hex!?

Hi guys,

My friend had written a complete package - a VMU Debugger, Emulator,
Assembler, Disassembler (VMU DEAD, as he called it) for DOS.  It has
all sorts of cool features - breakpoints, conditional breakpoints,
watches, profiling instructions, battery life measurement - basically
every feature you've named so far.  He did it independently of
everyone else's efforts about 6 months ago, so the syntax of his
assembler is slightly different.  It also generates a symbol table
during assembly, so label, constant, and variable names remain intact
upon emulation and disassembling.

Anyhow, I've finally convinced him to release it - he's got a little
more polishing to do.  The only drawback is the fact that it's not
even remotely portable, since it's written in Borland Pascal 7, but
for anyone using DOS it's really a fantastic tool.

I'll keep you guys posted as to when/where to get it.

Wayne Bartnick : VMU Animator
http://www.vmudev.org (still down, looking for new host)

--- In vmu-dev@egroups.com, john maushammer <john@m...> wrote:
> Two observations-
> 1. A profiler would be nice. Since the CPU's speed is limited,
> optimizing is a good thing. This would tell us which code is the
most
> heavily used so that optimization efforts can be focused there.
> 2. There is a gnu command-line debugger called "gdb". Perhaps we can
> model the command line syntax after that? I've got a copy, but I
don't
> really use it, so I'm not that familiar with it.
>
> Two features that would be nice that I haven't seen on other
debuggers-
> 1. A loop counter that tell's the main process's speed (just like
what I
> implemented above)
> 2. A current measurement. Battery life is an issue, too, so we can
> measure the time spent at 32kHz and 600kHz and give this as
feedback.
>
> -- john

I have just post a Java application in the files/programs section.
It takes an image file (preferably .GIF) and outputs a text file with
the assembly code.
Its not pretty, but worked for me.

hope somebody can use it :)

details on use in ZIP file.

source code and bad comments included.

=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/

i think this is common question (not asked).

this is what i my sprite routine does (trys).

1: take Y coord for sprite and find correct offset to 1st byte of that
row in LCD memory. $80 + offset
2: take the X coord and by dividing by 8 find how many bytes should be
added to the offset and the subtracting that value from the original X
coord get how mang Bits need to be shifted.
3: shift the byte the correct number bits and as it is shifted take the
shifted bits into another location.
4: output shifted data to the LCD memory.

i.e.

if X = 20 and Y = 22
1: $80 + offset($04) = $84 (its only four because it is in the second
LCD bank)
2: 22/8 = 2
offset ($84) + $02 = $86
2*8 = 16
22-16 = 6
shift by 6 bits
output byte one of pixel data to $86 and second byte (if there is one)
to $87.

I'm trying to get the correct values from my bit shifter routine at the
mo, debugger anyone? :)


--- Isaac Rounds <isaac@...> wrote: > Okay, okay, I am
embarrassed to ask this, but I think it seems kind
> of
> important. I know how to show a full screen image on the VMU's LCD,
> but how
> would I show a smaller (7x5 or 8x8 or something) image to a certain
> portion
> of the screen? I'm sure it cannot be THAT difficult, or else it would
> be
> pointless to even try to program much of anything for the VMU. :)
>
> Isaac Rounds
>
>
>


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/

Hi,

  Do you know where can I download for free the latest version of
netzip?

  Thank you in advance for your response.

Okay, okay, I am embarrassed to ask this, but I think it seems kind of
important. I know how to show a full screen image on the VMU's LCD, but how
would I show a smaller (7x5 or 8x8 or something) image to a certain portion
of the screen? I'm sure it cannot be THAT difficult, or else it would be
pointless to even try to program much of anything for the VMU. :)

Isaac Rounds

post by torbjorn
> Primarily an option to monitor the registers (and variables, or any
> address in memory for that matter). And that cpu cycle count that
> john mentions would really come in handy too.
>
> Now all we need is someone to code this!!  ;-D
> (Add debugging to SoftVMS 2.0? YesPlease!!!)
>
> -Torbjorn

Seems like a great idea, but I was wondering though, by "any address
in memory" should there by just a display of all the RAM at once? or
does anybody else have a better idea? It's a good idea to have the
ACC, B, and C registers always displayed, right? But for other RAM
areas, maybe just 16 - 64? entries in the RAM at a time? or all at
once?

> That would save me so much time, i posted a message to the VMU-CODERS
> aksing a simular question but i think i have been banned (anyone else
> still get regular updates from the list?)

The last message I saw was "[vmu-coders] DevKit / VMU Roleplaying Game"
dated 13 Jul 2000 04:17:00 +0100 (i.e. well over a week ago)  I know it's a
human-moderated list, so maybe all the moderators are on holiday, or
something ;-)

--
    _  Richard Munn  RAMTronics Software
   //  www   : http://come.to/ramtronics
\X/   email : richard.munn@...


__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Sorry for answering my own post :-) , just wanted to tell my most
wanted debug features.

Primarily an option to monitor the registers (and variables, or any
address in memory for that matter). And that cpu cycle count that
john mentions would really come in handy too.

Now all we need is someone to code this!!  ;-D
(Add debugging to SoftVMS 2.0? YesPlease!!!)

-Torbjorn

I remember John saying that he'd modified his version of the emulator to
support a few debug commands, but I think we need to work out something
 

I don't think my technique will work for a general-purpose solution. My "debug instruction" technique is commonly used on normal debuggers -- they replace breakpointed instructions with code to generate a special interrupt. This allow the code under test to run at full speed until it reaches the breakpoint. However, we're luckier - since we've got an emulator we can set breakpoints by keeping a breakpoint table that is checked before execution of an op-code is performed. There would be no need to change the user's code.

Two observations-
1. A profiler would be nice. Since the CPU's speed is limited, optimizing is a good thing. This would tell us which code is the most heavily used so that optimization efforts can be focused there.
2. There is a gnu command-line debugger called "gdb". Perhaps we can model the command line syntax after that? I've got a copy, but I don't really use it, so I'm not that familiar with it.

Two features that would be nice that I haven't seen on other debuggers-
1. A loop counter that tell's the main process's speed (just like what I implemented above)
2. A current measurement. Battery life is an issue, too, so we can measure the time spent at 32kHz and 600kHz and give this as feedback.

-- john
 
 
 

proper.

What I think we need is this:

1) Add support into Aslc to make it output some form of annotated source
listing

2) Make the emulator display this listing in a console window whilst it runs
(switchable possibly, since not everyone wants debug options)

3) Add special compiler directives for breakpoints, etc

4) Add keys that let you step though instructions, run till the next
breakpoint, etc. (or make it that when the debugger is running you have to
type commands at the console)
 

I think this could all be implemented without any major modifications, since
we wouldn't need to create any kind of debug gui as we'd just write text
(possibly with ansi formatting) to a console window (which would work fine
for the unix, windows and amiga versions of softvms, and could probably be
tweaked into the dos version somehow)

What do other people think? (oh and can anyone come up with a good list of
features that the debugger should have)

--
   _  Richard Munn  RAMTronics Software
  //  www   : http://come.to/ramtronics
\X/   email : richard.munn@...
 
 

--- In vmu-dev@egroups.com, Omar Cornut <cornut@c...> wrote:
> A question I am asking myself, because I saw that kind of code in
> a lot of source, and including the official VM samples:
>
>  MOV  #$1,ACC
>  ST   XBNK
>
> Why by the hell don't you write that? :)
>
>  MOV #$1, XBNK

One reason for the first version is if you really need ACC to be 1 for
some later operation.

   MOV #1,ACC
   ST  XBNK

is shorter than

   MOV #1,XBNK
   MOV #1,ACC

Another is if you need to set at least three more registers to 1 as
well:

   MOV #1,ACC
   ST  XBNK
   ST  FOO
   ST  BAR
   ST  FUM

is shorter than

   MOV #1,XBNK
   MOV #1,FOO
   MOV #1,BAR
   MOV #1,FUM

That's the reasons I can think of.  :-)


   // Marcus

That would save me so much time, i posted a message to the VMU-CODERS
aksing a simular question but i think i have been banned (anyone else
still get regular updates from the list?)

I would be so damm cool to set a break point and see the contents of
the Registers (plus memory would be usefull).

Everything else Richard mentions would be nice but for speed of
something to use today we should keep it simple.


--- Richard Munn <richard.munn@...> wrote: >
> > Is there any way to debug the assembled stuff on a VMU?
> > Are any of you emulator guys working ona debugging option for the
> VMU
> > emulators??
>
> I remember John saying that he'd modified his version of the emulator
> to
> support a few debug commands, but I think we need to work out
> something
> proper.
>
> What I think we need is this:
>
> 1) Add support into Aslc to make it output some form of annotated
> source
> listing
>
> 2) Make the emulator display this listing in a console window whilst
> it runs
> (switchable possibly, since not everyone wants debug options)
>
> 3) Add special compiler directives for breakpoints, etc
>
> 4) Add keys that let you step though instructions, run till the next
> breakpoint, etc. (or make it that when the debugger is running you
> have to
> type commands at the console)
>
>
> I think this could all be implemented without any major
> modifications, since
> we wouldn't need to create any kind of debug gui as we'd just write
> text
> (possibly with ansi formatting) to a console window (which would work
> fine
> for the unix, windows and amiga versions of softvms, and could
> probably be
> tweaked into the dos version somehow)
>
> What do other people think? (oh and can anyone come up with a good
> list of
> features that the debugger should have)
>
> --
>    _  Richard Munn  RAMTronics Software
>   //  www   : http://come.to/ramtronics
> \X/   email : richard.munn@...
>
>
> __________________________________________________
> Do You Yahoo!?
> Talk to your friends online with Yahoo! Messenger.
> http://im.yahoo.com
>


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/

> Is there any way to debug the assembled stuff on a VMU?
> Are any of you emulator guys working ona debugging option for the VMU
> emulators??

I remember John saying that he'd modified his version of the emulator to
support a few debug commands, but I think we need to work out something
proper.

What I think we need is this:

1) Add support into Aslc to make it output some form of annotated source
listing

2) Make the emulator display this listing in a console window whilst it runs
(switchable possibly, since not everyone wants debug options)

3) Add special compiler directives for breakpoints, etc

4) Add keys that let you step though instructions, run till the next
breakpoint, etc. (or make it that when the debugger is running you have to
type commands at the console)


I think this could all be implemented without any major modifications, since
we wouldn't need to create any kind of debug gui as we'd just write text
(possibly with ansi formatting) to a console window (which would work fine
for the unix, windows and amiga versions of softvms, and could probably be
tweaked into the dos version somehow)

What do other people think? (oh and can anyone come up with a good list of
features that the debugger should have)

--
    _  Richard Munn  RAMTronics Software
   //  www   : http://come.to/ramtronics
\X/   email : richard.munn@...


__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Hi guys.

Is there any way to debug the assembled stuff on a VMU?
Are any of you emulator guys working ona debugging option for the VMU
emulators??

regards Torbjorn

Hello.

Just wanted to let everyone know that my second self-made VMU game is now
finished.  It's a simple puzzle / board game called "Vision", based on the
old classic.  This one is not as advanced as "Pacman" (because I wrote it
in only three days), but I think it's quite fun to play nonetheless. :)

It can be downloaded at

         http://www.franken.de/users/deco/myfiles/vision.html

And _yes_, the sound routine is the same one that I used in "Pacman". :)

Bye


Alessandro
---
You get what anyone gets. You get a lifetime.

A question I am asking myself, because I saw that kind of code in
a lot of source, and including the official VM samples:

	 MOV  #$1,ACC
	 ST   XBNK

Why by the hell don't you write that? :)

	 MOV #$1, XBNK

Do you want to hear something scarry?

A function i wrote works !!

AGGHHhhhhh.......... .   .  !

Thanks for all the support guys (and girls?)

; Input ACC as y coordinate
; Output ACC as pointer to requested LCD line
; Output XBNK = correct bank for this coordinate
; Preserves B,C

findline:
	 PUSH B
	 PUSH C
	 ST   B 	 ; back up the ACC before we trash it.
	 AND  #%11110000  ; mask all the lower bits
	 BZ   .bank0 	 ; if its zero no high bits were set so number was less
than 16 (not equal) [0-F]
	 MOV  #$1,ACC  ; prepare
	 ST   XBNK 	 ; the xbank is set for 1 [16-32-(255)]
	 LD   B 	 ; restore
	 SUB  #$10 	 ; correct it
	 ST   B 	 ; we need this later.
	 BR   .bank 	 ; skip to next bit(we are ready!)
.bank0:
	 ST   XBNK 	 ; the xbank is set for 0 [0-15]
	 LD   B 	 ; Restore the input.
; at this point we have ACC equal to the line number (minus 16 if it
was 16 or higher)
.bank:
	 CLR1 PSW,CY 	 ; clear CY
	 ROLC 		 ; times by 2
	 ST   C
	 ADD  C 	 ; equiv of times by 4 (cos of above instruction)
	 ADD  C 	 ; equiv of times by 6 (cos of above instructions)
; now ACC is input * 6 number of bytes per line.
; we now need to add spare bytes cos of the way the LCD memory is
arranged
	 XCH  B 	 ; backup the number and restore the original to ACC
	 CLR1 PSW,CY 	 ; clear the CY flag
	 RORC 		 ; divide by 2
	 BNZ  .mbyte 	 ; need to loop it
	 LD   B 	 ; retrive the completed value we backed up
	 BR   .fin 	 ; skip the loop
.mbyte:
	 XCH  B 	 ; B is now the loop counter.
	 ADD  #$04 	 ; and 4 (the missing bytes)
	 DEC  B 	 ; decrement the loop counter
	 XCH  B 	 ; swap B for ACC
	 BNZ  .mbyte 	 ; check the loop counter
	 XCH  B 	 ; now we have finished with the loop
.fin:
	 ADD  #$80
	 POP  C
	 POP  B
	 RET 		 ; return to caller.


=====
Later ...

rednuht@...
-+-..=|<[{(_"!"_)}]>|=..-+-


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/

Omar,

    AAAARRRRRRRRRGGGGGGGGHHH!
    DOH! DOH! DOH! DOH! DOH!
    Thank you for pointing out my STUPID error with the AND
statement.  Somehow I knew it would turn out to be something like that ;)