Reality-Tech-Demos-1994 - 1995 (Amiga A1200 AGA)

Started by kevin, October 06, 2020, 10:36:59 AM

Previous topic - Next topic

kevin

 


Reality Tech Demo's (for Amiga AGA)



 
What Is it?


       Reality was the name of our Wolf 3d styled game engine, and became an umbrella term for anything related to that pursuit.  This archive contains the last known remaining compiled exe's of my various attempts to get it up to a reasonable performance on a near stock Amiga 1200 (020+FastRam) back in the mid 90's.  

   The Reality project was originally most active between 1994 -> 1995 and apart from few screen shots and close friends was never released to the public.   It's important to note that the attached downloads are ALPHA software and while operational should not be considered stable.  They do take over the system in as an OS friendly manner as I knew back at that time, but modern systems might not like them.




 
Can I use the demo's under AMIGA emulation ?


     Yes,  that's i'd recommend running these days.



 
Home Page?


    Reality Tech Demos Home Page


 
HashTags


   #amiga #aga #doom #wolf #wolf3d #bladestone #port #aga #gloom #dread #bloom #indiedev #gamedev #retro #2d




Videos


Reality Tech Demos (Amiga AGA)  1994-1995

In this video i'm looking back at my #Wolf3D / #BladeStone styled texture mapping engines that were written between 1994 / 95.   The videos looks at three different builds, the first is the initial ray cast version, the next is first polygon edition and third includes many of the long term gol features, from polygon 3d objects to 2d planes/sprites with depth queues for lighting.  

All programming by Kevin Picone / https://underwaredesign.com
   




    Script / Captions


hi guys we're gonna take a quick look

back at some early demos of my reality

project on the Amiga this was these are

taken from about 94 and 95 and we'll

look at what I think are the only demos

of it that still exist let's take a look

okay

this first one yeah this is actually the

only one that's raycast and the way it

works is you've got a picture and we're

scanning across the the floor across

that picture to look for values that are

nonzero and any value that's not zero is

a wall fragment so it's like a cache of

wall strips and they're just drawers

it's true it's very fisheye very

accurate as he can tell whether the

bottom bottom of the walls but this was

very early probably she probably ninety

for some time we're starting to work on

Chucky bits line up so the ambition was

pretty high at the time and I think this

is a mule I had to be running on an i-20

I'm a guy that's already making 1,200 so

the performance isn't it isn't that bad

actually

as you can see there it's dipping

outside of the

the chunk of memory that defines the map

he doesn't look like like Wolfenstein

we've got like a agreed and that green

the tiles have a certain size and states

through the eyes that's pretty cool to

say that I thought I'd lost all these

dealings by the way

okay the next one is it's not ray cast

this is

yes actually further along I thought it

was

[Music]

the environments of polygon environment

that we've got limited rotation so he

just wrote area the vertical access and

we have not read straight walls we've

got also normal diagonal we can just

connect anymore but you'll notice that

it's not prospectively incorrect and

some of these same inaccuracies that can

be seen that the towards the bottoms of

the wall fragrance when scaling

yeah actually a little bit of inaccuracy

there between the walls too

I don't remember when this one was this

is probably about 95 when this is what

was mine

amongst about 10 of the things we're

trying to do at the same time I'm pretty

sure it's 64 colors as well and this the

the mountains behind it there they would

be surprised it's trying to get some

benefit out of using cowboy sprites and

those days it's not depth acute or

anything like that you've got no floor

at all and all the say the pee retains

her they're optimized for our 20s

I think that these two demos here

I think these ones are very similar

[Music]

okay there's a second take second

version of that demo that that I think

I'm testing different scaling techniques

to do full screen walls as you can see

they're over later cover each other that

he tried to do layers of these

transparent fragments the way it works

is you have a solid wall pass and then

you have a transparent all the parts and

the transparent walls are clipped to the

solid walls so anything that's being

drawn behind you because the scene is

really just a list of these wall

fragments these polygons fragments we're

just working out which which vertical

scan line will row of that that needs to

be drawn for each each section of the

screen when this multiple wires we need

to have I think has a buffer of maybe

eight why is that could appear either

top of each other and it doesn't

insertion

insert the new any fragment works I

confront all the other ones or behind

them

for every vehicle strip yeah but there's

a look there's lots of theories about

well maybe I can do different types of

rain as the sidewalk

try to optimize the code and read them

to us a 1 so 3 3 moles at once that

caused them

I think these are all the same what

happens in a lot of resolutions over

running Probie

yeah it's at least two pixels high so

that's probably I think the frame buffer

is 256 by 128 I don't know

and it would be running on an i-20 with

fast memory so

with my own cussing they see the pay and

at the time it was pretty pretty ugly

code let's check the last one there see

if it works

[Music]

some aggression he hasn't probably to my

to here

[Music]

I mean this

at this resolution it's pretty pliable

there's a lot of that fun in daytime

[Music]

that people would have been expecting on

what I thought

[Music]

you know but I don't know what what in

particular I was testing with these

demos I just found these all executables

and these are what they are so that's

all I can say about them right now such

a long time ago 25 years ago pretty much

all right next one

alright this one we started having a an

idea about what we're trying to do and

all the textures just saying Ephraim

heretic by the way no I've got one by

one

got some objects in there we have got

depth cueing as well

terrible clipping and still need to are

it's a bit larger

so the goal was to try and get this

thing to you know I try and be as

pliable as it possibly could at one by

one on the i-20 with fast memory

the theory was a bit Scala pretty well

but attend that it didn't they

when we tested on - he's in the high 40s

at the time he just chucked and

absolutely choked

[Music]

watch your wine obviously someone today

I could beat you a whole lot on that

now you probably say this one before

this generation of the demo was used

we were trying to get funding to join

just trying to stay afloat pretty much

and we change the resolution here

probably two by two

as you can say I mean for 1200 without

20 in it that's pretty dicey primarily I

think what really killed these projects

was that this by the time we were

getting I've been resolved getting out

techniques up to scratch pretty much a

lot of people have written been at once

you know I'm in gloom kind of a during

the same time and they was other day I

was out there uncle poon was he that the

same demo at the time I don't know he

write that one but it was beautiful

all right this has me to lighter

I think in this version too I'm using a

combination of our processor and flitter

assisted Chucky the planer so probably

using I had a an approach where I was

he's like a pre-calculated see the pain

the first pass would would copy down the

chip mirroring and the multiple passes

to to unscramble what was left

you're talking back soon

and it's driving the quality of the

takes you laughing

yeah the perspectives pretty awful about

you know you get that

[Music]

with these demos I'm not quite you can

say so here is there

[Music]

maybe that's TMI too

yes the total for floor thing

a Connecticut befool is quite what to

work as I imagine that sure there's one

of those things where I just thought

well I'll implement it and then I'll

worry about getting it accurately like

nice things memorize them because I

wanted you to I wanted to know whether

it was just how fast you can get it or I

could get it - that's pretty sure that

other people that they could do a better

job but I thought if I can get it to a

point where Oh about the attorney

objects in its vision this is not power

to say it's an over life pretty much

it's written to the junkie buffer and

process down the

that's not clear in this version I spend

a bunch of time and write a vision of

the polyfill on to do vehicle strips

rather than horizontal spans the idea

was to use the horizontal clips against

the solid walls to count or ticks out

there from Holland things but it never

got that far

you know just time is slipping away and

my version slipping lying of course that

was moving on

Sol's but here today I'm on in about

timing it's 96

suddenly here it's dry or anyone

everyone's been I don't remember it

running but we have running on the

emulator so I have a Tweetup a little

bit higher I want you there I think

that's it I think that's it

rot it's obviously this version I found

this with a mile to a friend of on some

got here about three special guy said

they were retains ivory optimize the

Chuck liners but you acted m/s which

patent fool was a kind an errant display

for the fries a second

[Music]

as I leave a demo game

well I think I'm away with their wolf

say enough of that

Oh guys think they have a patent for my

games the solid colorful

and the patent and the solar camera

the chalkida Pliny was doing

comparison to the previous Buffett to

see if there was any need to convert

that chunk if they were the same we'll

just skip past it since then I realized

that unit you don't even need to do that

or nothing what you do if you ran

through the whole thing

generated you vehicle's statements or a

list of all the segments they can work

out what columns need to be converted

parts don't need to be taken just how I

bleep feel for the background or

something probably anyway that much

achieved a lot that of course when

you're up near a wall like that you

wouldn't but when you're in an old

saying where you got probably a 50-50

distribution of

II meant your injury like like therefore

examples

it's a small mayor injury

couldn't work - right probably done the

death money I have no idea

[Music]

this I I had a fascination I thought all

I wanted figured make it make a shoot

him out gun and do something homes on

scrolling with death

[Music]

looks like been running a local

enthusiasts wall for always on the

products were getting in the middle of

it

reality was game and they love those Pro

X's I then it's had I've been in my

painting we were working on there was a

straight father to me were doing some

work on what we proved the idea is but

we didn't really have anything else we

had a few characters that were drawn

like static pauses and so we've taken

the backdrops from from all the games

and so a lot of those things we just

concepts you know saying they were gang

or squares just it's over the top okay

I think it's not for me thanks for

watching that I hope you learned

something about the history of HD - and

I'll put them up on the website and you

can come and have a look at them anytime

you want they're not terribly reliable

and they out

thanks for watching you know I'll see

you next time







Project Reality Amiga Tech Demo   Underware Design Promo VHS RIP 1995

  Welcome...   In this video we look at a raw VHS capture of our reality tech demo   that was  developed for the Amiga home computers, namely the A1200/020 with  fast memory.  I'm not too sure when this video was made now, but my best guess  would be late 1994 / early 1995.

     This was one of our various tech demos promotion videos, trying to get funding  locally.   Back in those days, if you wanted to make games in Australia, you pretty  much needed to leave.. In retrospect we should have..  But it is, what is it..      We did send out various beta's back in the day, so if you have one feel free to upload the demo to Aminet and don't forget to send us an email to  UnderwareDesign.com

   




(Broken) Reality 1994/1995 Amiga Tech Demo

   The original goal of the Reality project  was to  create a Wolf 3D / Blade Stone styled 2.5D game engine for Amiga A1200 / 020 with  Fastmem.  It was pretty ambitious at the time, and looking back I wouldn't say that we ever really met our objective, but certainly gave it a good shot.   The only down side was that almost nobody got to see it, so here's a warts and all version!

  The version shown here is about the only code that still assembles.  There's newer versions of the source code, but they're missing lots of external files.    None the less, this supports textured walls / floors with flat shaded polygons.  

  For polygons,  the  original idea was to draw  triangles as vertical spans, so they could be clipped against the walls 1D z buffer.   But sadly that's not in this version.   The walls are connected together line segments, to draw the view, we get a cone of verts and draw the walls attached (to the verts) into the depth buffer.    The floors clip to the smallest wall segment in the buffer, so sadly there's a lot of overdraw, not to mention  the funky texture slipping....

  The demo in the video, is running through the WinUAE emulator with an image of my A1200 hard drive.  Most stuff seems to run ok, except the (AsmOne) assemblers ?, not sure why,  but the combination seems rather unstable.  So when you factor in the obvious bugs in my code ! :),   this is best I can get it running in the short time available :( ... Hence the  "Broken" Reality title....  Shame,  I was enjoying the nostalgia.

Code By:  Http://www.underwaredesign.com
   Art By: Raven (Textures are from Heretic)
 Music By: Ardvarc