Jazz Jackrabbit
Mattbrad3
one thing has not been brought up yet (atleast I think it hasn't)... is
the fact that Jazz seems to use a parallax scrolling background; Using the
standard way of writing tile-based platform type games in mode-x, more
than 1 layer would not be possible (or so it seems). Could someone please
expand more on how this is done in mode-x?
Ian Ward
Multiple layers in mode-x is quite possible.. You just have to choose
one plane to scroll around (preferably the one that covers most of the
screen) and update all the others each frame. Jazz uses a simple
trick for it's background parallaxing -- the backgrounds are limited
to thick horizontal bars of different colors, so that it only has to
update the edges of the bars that change, and not the whole thing.
[ Ian Ward | iw...@chat.carleton.ca ] - all the usual disclaimers apply.
Bryan Wilkins
iw...@chat.carleton.ca (Ian Ward) wrote:
>
>Multiple layers in mode-x is quite possible.. You just have to choose
>one plane to scroll around (preferably the one that covers most of the
>screen) and update all the others each frame. Jazz uses a simple
>trick for it's background parallaxing -- the backgrounds are limited
>to thick horizontal bars of different colors, so that it only has to
>update the edges of the bars that change, and not the whole thing.
>
>[ Ian Ward | iw...@chat.carleton.ca ] - all the usual disclaimers apply.
Jazz uses a simple trick but I don't think that it updates any edges...
It is simply a tile like all of the others and when the player walks, the
palette is scrolled. Thus, producing a very good parallax scrolling LIKE
effect. I don't think that jaz really is PARALLAX even though it IS possible
in mode x.
-Bryan
bwil...@swcp.com
Darren Reid
matt...@aol.com (Mattbrad3) wrote:
>In article <47coo4$k...@ulke.hiMolde.no>, lon...@hiMolde.no (John C.
>Lonningdal) writes:
>
>>Isn't it just to draw the background first and then the platform tiles
>>on top... Seems pretty obvious to me.. Although I think this will be
>>better to do on an offscreen buffer and blit it to the screen afterwards.
>>
>>John
>>
>
>If it were that easy my friend... but the secret to mode-x's smooth
>scrolling is syncing to the vertical retrace. Having to deal with more
>than 1 layer (ie. blitting more than 1 layer to the VGA buffer in 1
>frame), takes a taxing amount of time... you can barely get away with
>copying the 1 layer and all the objects in that 1/70th of a sec you have
>to spare before the v-retrace catches up to you. Could someone with
>knowledge of MODE-X please respond to this VERY disturbing question.
I missed the first part of this, but one thing to keep in mind is that almost
NO games run their framerate at 70/sec. Most games wait 2-4 v-retraces
inbetween page flips. Twenty-odd frames per sec gives you very smooth
animation. Sixty/sec is a waste, IMHO.
John Paul D'India
: Yeah, I figured that, but what about the way some objects can pass in
: front of AND behind tiles... this would indicate a fringe layer in their
: somewhere. Thus... it has to use more than 1 layer of tiles to accomplish
: this... you can't do THAT with palette cycling.
Go and find an old 386sx16 with an ancient 8bit card and play Jazz (it
will be slow enough for you to see flicker). By simple deduction you'll
figure out _exactly_ what he's doing. The bars are palette cycled though
(and that's what you asked). Also, just because some things go behind
others doesn't mean much, because there are two simple methods around
that (1) update any area where a sprite can go behind after drawing the
sprite (2) divide up the palette into foreground and background colors
and don't overwrite foreground colors. I would probably go with method
#2, since it's less work. Keep in mind, there aren't that many overlap
situations in Jazz (at least not on the first level [the only level I
played]).
--
====================================
John Paul D'India - D'India Software
project: 4gen (space shooter)
web page: http://www.ax.com/dindia
====================================
Mark Rein (Epic)
>simple colour cycling, beloved of Demo coders for years.
You're right about this. HOWEVER, Jazz 2 which is now in development
uses 8 levels of true parallax scrolling and runs at 70 fps under
Windows 95. For you Jazz freaks out there the new Jazz character has
over 1100 frames of animation compared to the original 40 frame
character. Plus there is a second playable character (Spaz, Jazz's
goofy brother), tons of really cool special visual effects and all
kinds of gameplay improvements. Check out this months UK PC Gamer
magazine for a sneak preview screen shot. Jazz 2 won't be out until
some time next year but the early demos are very, very impressive and
by the time it's finished this could be the best platform game ever
done for the PC. We'll probably start showing screen shots and
progress reports of Jazz 2 on our Web site in January 96.
Mark Rein,
Epic MegaGames.
Check out the Epic Web Site at http://www.epicgames.com
Download our latest shareware games from ftp.uml.edu in /msdos/games/epic
Vito Miliano
>You're right about this. HOWEVER, Jazz 2 which is now in development
>uses 8 levels of true parallax scrolling and runs at 70 fps under
>Windows 95. For you Jazz freaks out there the new Jazz character has
>over 1100 frames of animation compared to the original 40 frame
>character.
70fps under Win95? Does that mean you're running in a DOS shell from Win95,
or is it native? And 1100 frames? Really? Wow... Will he be a larger
character? I always thought Jazz was pretty small on the screen.
--Vito Miliano
---The Perilith Project
\ / = Vito Miliano <peri...@aol.com>
\ / _____ ___ = Head Developer for The Perilith Project (tm)
\ / | | | | =
\/ | | |___| = PERILITH: A Fantasy Spanning the World
------------------------= A Battle Lasting a Thousand Years
Darren Reid
din...@netcom.com (John Paul D'India) wrote:
>(and that's what you asked). Also, just because some things go behind
>others doesn't mean much, because there are two simple methods around
>that (1) update any area where a sprite can go behind after drawing the
>sprite (2) divide up the palette into foreground and background colors
>and don't overwrite foreground colors. I would probably go with method
>#2, since it's less work. Keep in mind, there aren't that many overlap
>situations in Jazz (at least not on the first level [the only level I
>played]).
Ha! You learn something new every day...I never thought of masking out certain
palette registers as opaque for foreground objects...hmm...sounds slow.
Neat idea, anyway.
sl...@cc.usu.edu
> In article <47lib3$k...@sam.inforamp.net> mr...@inforamp.net (Mark Rein (Epic)) writes:
>
>>You're right about this. HOWEVER, Jazz 2 which is now in development
>>uses 8 levels of true parallax scrolling and runs at 70 fps under
>>Windows 95. For you Jazz freaks out there the new Jazz character has
>>over 1100 frames of animation compared to the original 40 frame
>>character.
> 70fps under Win95? Does that mean you're running in a DOS shell from Win95,
> or is it native? And 1100 frames? Really? Wow... Will he be a larger
> character? I always thought Jazz was pretty small on the screen.
Having seen the demo when I visited Arjan in the Netherlands... I must say it
is most impressive. however the characters are still about the same size, and
this was running in fullscreen (or should I say fullscream) mode... but I think
that in a window, although it would be dang small, it would run close to
that on highend pcs.
wReam...
Stuart Mackereth
Organization: Internet Africa public service
Distribution:
Jason Doucette (019...@axe.acadiau.ca) wrote:
: vmil...@aol.com (Vito Miliano) wrote:
: >In article <47lib3$k...@sam.inforamp.net> (Mark Rein (Epic)) writes:
: >
: >>You're right about this. HOWEVER, Jazz 2 which is now in development
: >>uses 8 levels of true parallax scrolling and runs at 70 fps under
: >>Windows 95. For you Jazz freaks out there the new Jazz character has
: >>over 1100 frames of animation compared to the original 40 frame
: >>character.
Sounds like fun, but it will only be out in June/July next year, or
so I have heard.
As an alternative, look out for CYRIL CYBERPUNK which boasts extra
features than Jazz. Jazz uses palette cyriling for most animation effects
which is a cheap cheat really. Also the enemy types are sparse, and
pretty darn stupid, and I don't remember seeing a Jazz enemy actually
shooting a projectile...
CYRIL CYBERPUNK uses really animation frames for waterfalls, doors,
flickering torches on walls, teleporters etc etc etc, and has over
30 detailed enemies, all with individual personalities and ai. Some
dro pslime which chases your around, other fire rockets, other try
bounce onto you, yet others still rush at you shooting.
The artwork is far superior to Jazz, and the engine seems to run smoother
and faster, even on a 386dx-40 - although VL-BUS is really a must!
But then, the game is about 40MB, much much bigger than Jazz, so it is not
really fair to compare.
Look out for it, the shareware version is probably on the net by now.
Stuart Mackereth
Senior Programmer
Reality Interactive
Kiwifruit
>> meathod would result in a disgusting mess! The parallax effect is
>> simple colour cycling, beloved of Demo coders for years.
>background trick? Noone in their right mind would do such a thing.
Oh dear. Didn't you notice that the colour bar pattern *repeated*
after a few lines? No need for 180 palette colours there! In fact, it
would be most logical for it to be the same number of colours as the
height of a foreground tile (Is that 32 pixels high? I didn't count).
Carl Muller, Probe Software: S5g+l+y-z--n+ox-a+u(-)v-j++ f++s+e+
A.R.Campbell
>
>...I really don't think this is Mark Rein writing this, unless you
>got your facts mixed up? but I don't think so, it's not a very good
>marketing scheme. Maybe he's just having fun with all the novice
>programmers out here...what do you think?
Yes, that really was Mark Rein, and, hard though it may be to believe,
everything he said about Jazz II is true. Just wait and see [grin].
Regards,
Ammon R. Campbell
Epic MegaGames, Inc.
---
E-Mail:
am...@epicgames.com
am...@teleport.com
71441...@compuserve.com
Check out Epic's latest and greatest news and shareware games at:
http://www.epicgames.com/
ftp.epicgames.com
sl...@cc.usu.edu
> As an alternative, look out for CYRIL CYBERPUNK which boasts extra
> features than Jazz. Jazz uses palette cyriling for most animation effects
Yes, but is Cyril cute? And does it have cool 3d Bonus stages? ;)
wReam...
P.S. Why do I get the impression this thread is now nothing but a commercial?
;)
David Wareing
<019...@axe.acadiau.ca> wrote:
>vmil...@aol.com (Vito Miliano) wrote:
>>In article <47lib3$k...@sam.inforamp.net> mr...@inforamp.net (Mark Rein
(Epic)) writes:
>>
>>>You're right about this. HOWEVER, Jazz 2 which is now in development
>>>uses 8 levels of true parallax scrolling and runs at 70 fps under
>>>Windows 95. For you Jazz freaks out there the new Jazz character has
>>>over 1100 frames of animation compared to the original 40 frame
>>>character.
>
>Now THIS will be something to see... even the hottest Pentiums can't
>handle 8 levels of true (he did say true) parallax scrolling.
That's a pretty meaningless assertion. It would depend entirely upon what
is being scrolled, how much is visible, etc etc.
>I can maybe
>see using 2 levels, with the 2nd level split up, as in Sonic...
>And under windows 95 too...of course this could mean pressing F8 and
>choosing Dos 7 instead of win95...hey, Microsoft misleads, why can't game
>publishers?
Get up on the wrong side this morning?
>1100 frames is a bit ridiculous, I must say. I really don't think this is
>Mark Rein writing this, unless you got your facts mixed up? but I don't
>think so, it's not a very good marketing scheme. Maybe he's just having
>fun with all the novice programmers out here...what do you think?
According to the May issue of 'Edge', the hero in 'Heart of Darkness'
has access to 1600 frames. Why can't Jazz 2 have something approaching
that level?
--
David Wareing dwar...@adelaide.on.net
Belair, South Australia http://www.AmbrosiaSW.com/~dwareing/
Macintosh Games & Multimedia Programming
Robin Ward
I talked to Mark Rein on the phone about 6 months ago and he was talking about
it, so it's obviously true. Check out this excerpt from the latest Epic
Meganews:
==================
JAZZ JACKRABBIT II
==================
Jazz is back and better than ever in the fantastic sequel to last
year's best selling Jazz Jackrabbit. Featuring 8-level parallax
scrolling, a new digital sound system, new characters and over
1200 frames of animation for each main character. Jazz Jackrabbit
II will run in Windows 95 at 70 frames per second and features
multi-player game play modes. Watch Epic's web page for more
information on JAZZ II in early 1996.
Jason Doucette
>In article <8159832...@reachout-1.acadiau.ca>, Jason Doucette
><019...@axe.acadiau.ca> wrote:
>
>>vmil...@aol.com (Vito Miliano) wrote:
>>>In article <47lib3$k...@sam.inforamp.net> mr...@inforamp.net (Mark Rein
>(Epic)) writes:
>>>
>>>>You're right about this. HOWEVER, Jazz 2 which is now in development
>>>>uses 8 levels of true parallax scrolling and runs at 70 fps under
>>>>Windows 95. For you Jazz freaks out there the new Jazz character has
>>>>over 1100 frames of animation compared to the original 40 frame
>>>>character.
>>
>>Now THIS will be something to see... even the hottest Pentiums can't
>>handle 8 levels of true (he did say true) parallax scrolling.
>
>That's a pretty meaningless assertion. It would depend entirely upon what
>is being scrolled, how much is visible, etc etc.
>
regardless of what is seen. This is obviously impossible on any machine
with only the power of a Pentium, and probably useless since 8 full
screens would have too much going on to see each individual layer. If the
game does have 8 levels of parallax, it is obviously optimized not to
draw what will not be seen...but I can't see this happening at 70fps with
complicated layers such that most of them need to be fully drawn. 2
layers of *true* parallax is hard enough on a PC.
>>I can maybe
>>see using 2 levels, with the 2nd level split up, as in Sonic...
>>And under windows 95 too...of course this could mean pressing F8 and
>>choosing Dos 7 instead of win95...hey, Microsoft misleads, why can't game
>>publishers?
>
>Get up on the wrong side this morning?
>
I get up on the wrong side of the bed every morning. What I meant by the
above paragraph was perhaps Jazz II uses 2 layers of true parallax
scrolling (as Sonic does), but split the 2nd layer into many horizontal
planes (as Sonic does in it's first level) to make it appear as though
there are more than 2 layers. Adding sprites to the horizontal layers
(Sonic didn't do this) may make it appear as though some of the
horizontal layers are in front of another. The Genesis can only handle 2
true parallax layers, but programmers used tricks like these to get
around that. Since the PC has no hardware for this, everything is done in
the software, and using bitmap graphics modes (unlike the Genesis),
anything is really possible, it's only a question of speed. And I truely
do not think 8 layers parallax is possible at 70fps. Of course the
machine on which this is running isn't mentioned, but I assume a fast
Pentium, which will be widespread by the time the game is released.
>>1100 frames is a bit ridiculous, I must say. I really don't think this is
>>Mark Rein writing this, unless you got your facts mixed up? but I don't
>>think so, it's not a very good marketing scheme. Maybe he's just having
>>fun with all the novice programmers out here...what do you think?
>
>According to the May issue of 'Edge', the hero in 'Heart of Darkness'
>has access to 1600 frames. Why can't Jazz 2 have something approaching
>that level?
For ONE character? I really can't argue this since it is possible, but I
still think it's rediculous. I guess memory wasn't a concern here.
BTW, I just recieved Epic's MegaNews, and the about facts (8 layers, 1200
frames) are mentioned there, so I apologize for suggestion that perhaps
it was a false posting. And it's not like Epic to mislead their
customers, so I really can't see them lying. I guess I'll have to wait
and find out.
Jason Allen Doucette
Saw Tooth Distortion
--------------------
019...@axe.acadiau.ca
Christopher G Busch
Lars Troen
Christopher G Busch (cbu...@ub.d.umn.edu) wrote:
: Is Jazz Jackrabbit shareware? Is there a demo out there??
Yes it is. The demo is on nic.funet.fi:/pub/msdos/games/epic/1jazz.zip
--
(***************************************************************************)
(* Lars Trøen på Høgskolen i Sør-Trøndelag *)
(* "The key, the whole key, and nothing but the key. So help me Codd." *)
(* Who is General Failure and why is he reading my C: drive? *)
(***************************************************************************)
Neil Graham
>It is possible.. I've done a game with 6 true levels of parallax in 256
>colors which runs at 60 frames on a 486 33 Mhz with isa bus.
I'm intrigued, My routines allow for a single plane ok and I could do small
windows of parallax but for any large change it would be too slow.
Approximately how much video memory do you modify per frame? For what I would
consider true parallax the amunt of video memory to change would be more than an
isa bus could take.
>If you port it to protected mode (mine is in realmode) you get a speed
>boost of about 25%..
Why? What can you make faster that requires protected mode?
Nick Waanders
>>colors which runs at 60 frames on a 486 33 Mhz with isa bus.
my system sometimes does 35 fps, due to very full screens.
>I'm intrigued, My routines allow for a single plane ok and I could do small
>windows of parallax but for any large change it would be too slow.
>Approximately how much video memory do you modify per frame? For what I would
>consider true parallax the amunt of video memory to change would be more than an
>isa bus could take.
I moved a whole screen to mem, so 64000 bytes. 16000 dwords.
the blit itself takes most time.. it takes about 2/3 of my vsync.
*** Mail sending interrupted by the sender ***
Chris Gouldie
Don't have to load any segment registers, also, I believe that accessing 32
bit registers takes one less clock in protected mode than it does in real
mode. Plus, and this is the biggie, if you use ems or xms, there is a switch
from real to protected mode and back, which is very slow, also the memory you
want to access has to be _copied_ to lower memory, not an instant process. So
yes, you can get quite a speed boost (in writing your code as well as running
it :) when you use protected mode.
Nick Waanders
Later, Nick
--
Nick Waanders, n.b.wa...@student.utwente.nl
Programmer, 3D-Graphics-Artist and Gamedesigner for Youngblood Games.
(and for the rest of the time a student... ;)
Neil Graham
>
>In article <81794881...@aurora.co.nz>,
> Le...@aurora.co.nz (Neil Graham) wrote:
>>Why? What can you make faster that requires protected mode?
>
>
> Don't have to load any segment registers, also, I believe that accessing 32
>bit registers takes one less clock in protected mode than it does in real
>mode.
I would imagine that any speed change from not having to load them elsewhere
would be close to undetectable. I have no information on the impact of 32 bit
acesses.
>Plus, and this is the biggie, if you use ems or xms, there is a switch
>from real to protected mode and back, which is very slow, also the memory you
>want to access has to be _copied_ to lower memory, not an instant process. So
>yes, you can get quite a speed boost (in writing your code as well as running
>it :) when you use protected mode.
ah, I was kinda assuming that the real mode program was sitting in the first
meg and not talking above, but you're right, pmode is the way to go for
lots-o-mem.
+---------------------------------------------------------------------------+
| "Glook" Le...@aurora.co.nz "Stop _poking_ me!" |
+---------------------------------------------------------------------------+
J Morris
> In article <49upkt$8f4...@ots.utexas.edu> cgou...@mail.utexas.edu writes:
> >
> >In article <81794881...@aurora.co.nz>,
> > Le...@aurora.co.nz (Neil Graham) wrote:
> >>Why? What can you make faster that requires protected mode?
> >
> >
> > Don't have to load any segment registers, also, I believe that accessing 32
> >bit registers takes one less clock in protected mode than it does in real
> >mode.
> If the inner loops are done properly than they shouldn't twiddle segments so
> I would imagine that any speed change from not having to load them elsewhere
> would be close to undetectable. I have no information on the impact of 32 bit
> acesses.
>
Yes, but Jazz JAckrabbit uses 16-bit protected mode, not 32.
> >Plus, and this is the biggie, if you use ems or xms, there is a switch
> >from real to protected mode and back, which is very slow, also the memory you
> >want to access has to be _copied_ to lower memory, not an instant process. So
> >yes, you can get quite a speed boost (in writing your code as well as running
> >it :) when you use protected mode.
No, I have seen the source for HIMEM.SYS that does the XMS handling.
It does _not_ use protected mode. Groovy or what :)
> ah, I was kinda assuming that the real mode program was sitting in the first
> meg and not talking above, but you're right, pmode is the way to go for
> lots-o-mem.
>
And 16-bit Pmode is easier for porting 16-bit DOS programs.
>
> +---------------------------------------------------------------------------+
> | "Glook" Le...@aurora.co.nz "Stop _poking_ me!" |
> +---------------------------------------------------------------------------+
--
*****************************************************************************
*J.P Morris, IT-HE Software * jp-m...@csm.uwe.ac.uk * NOT a UWE spokeman *
*****************************************************************************