Jump to content
Sign in to follow this  
raaaid

i suggest to evolve the game from an euclidean space to a parmenides one

Recommended Posts

in an eucliden space to define an object you need 9 coordenates:

 

size on the xyz axe, position on the xyz axe and rotation on the xyz axe

 

instead if you use  a parmenides geomtry and describe just appearance your switched to just care on what appears on your screen

 

 

and what appears on your screen, any object is defined by seven courdinates size on the x and size on the y axe and position on the x and y axe and rotation on the x and rotation on the y ( and rotation in the z)

 

so on this way the pc instead of having to operate with nine coordenates they just opearte with 7 coordenates

 

i veen seen this kind of flat 3d rendition in zx spectrum 3d games where they had to maximize resources

 

serve as an example this animation i just made in which i move a 3D dimensional head around 2 axes of  3d rotation actually using a flat model in which i just move along two axes x and y:

 

 

to be finalized it would just need to be defined the edge lines equations as it moves along x or y that limit next cape

Edited by raaaid

Share this post


Link to post
Share on other sites

here another example on how to render a 3d earth globe from a flat surface

 

a substitution to the mercator map, a T shape map which measure precise distances on any axe and can be transformed into a cube in order to renderize

 

 

edit:

 

notice that with the T shape projection you can render with a flat algorythm any 3d landscape and appear to rotate the camera around it just by moving the flat sking accordingly just as you move the mouse on the x and y moving the camera around the T skin on x and y axe

 

il be checking old spectrum games to see which include somekind of this type of 3d from flat algorythm ☺

 

edit:

 

i found this reference to cube mapping:

 

1200px-Panorama_cube_map.png

https://en.wikipedia.org/wiki/Cube_mapping

Edited by raaaid

Share this post


Link to post
Share on other sites

well so my just discovered cube mapping was found out in 86, and no i wont go into tt here

 

i bet bos even uses cube mapping for the sky and stars

 

but what i propose that i bet is new is use cube mapping for objects as weel

 

here an eexample of model for earth which edge equation to limit with next cape would be a circle:

 

world_cube_net.png

qsc_concept.jpg

Edited by raaaid

Share this post


Link to post
Share on other sites

Congratulations - you've demonstrated beginning, remedial, first day in computer 3D graphics class, introduction to UV mapping workflow to a group of accomplished game developers.

  • Upvote 4

Share this post


Link to post
Share on other sites

I realised the other day that in my whole life, I've never actually owned a jacket with a hood.  Never ever.  Can you believe that???

Share this post


Link to post
Share on other sites

do you need a hood in Girt?

 

Or would you want one that tucks away when you don't need it in order to double you performance?

 

I think it would be nice if hoods were a little more flexible

 

Be nice to chat without having to have my hood up that's for sure

Share this post


Link to post
Share on other sites

I realised the other day that in my whole life, I've never actually owned a jacket with a hood.  Never ever.  Can you believe that???

 

Damn..... I never thought  about it . I never owned .. naah forget it All my hunting cloth , winter hike cloth got hoods . But apart from that I never really liked walking around with hood . It might be the fact I do not have a neck. My head starts with the shoulders almost I think

Share this post


Link to post
Share on other sites

the current generation og Hoods give me too much of a letterbox effect.

 

I sweat a lot in the top and its jolly inconvenient having to take a USB fan everywhere I think I'm likely to use it 

 

Before i next dip my toes into the world of hooded clothing, im gonna wait to see if the FOV improves...

Share this post


Link to post
Share on other sites

as you know ive done some 3d modelling and what i propose is a new approach

 

as you have seen in my wiki link cube mapping is much much faster but currently its only used for distant landscapes

 

i propose to use it for everything

 

if you were to model earth as seen from space from a rocket you would take that cube map i posted and define 360 edge lines equations for the profile and light interactions

 

those lines equations would serve to know when to render earth skin and when to render the stars, they would be the profile lines

 

those profile lines would serve as well to interact with the sun and create lightning

 

then if you were with your rocketship to get closer or farther to earth all you had to do would be increase the size or decrease of that object

 

current uv mapping is done in volumes i propose do it in planes and add the profile lines equations which would limit to next cape and shape shadows

 

in a parmenides space there are two rules:

 

motion is illusory nothing moves

 

volume is illusory everything is flat

 

and if you think about bos when you see a plane moving thats a decievemnt of your mind whats really going on its pixels changing colour and isnt your screen flat though you get a volume percpetion illusion?

 

so you see again the x and y rotation of an object would be as my first video animation to move the camera over the skin on the x and y axe, a z rotation would be rotating the skin

 

translation on the z would be increasing or decreasing the size of the object and translation on the xy would be rotation on the xy of the plane the object is on

 

as i said in my original post on this way you can fully describe a space with JUST 7 COORDENATES WHILE EUCLIDEAN SPACE NEEDS 9 


edit:

 

whats hell its translating from an euclidean space the set of newtonian equations of motion to a parmenides space set of equations where theres just changing of size on 2 axes of space plus rotation

 

edit:

 

i correct myself in a parmenides space theres no translation, a plane that appears to translate on the x its just because the x axe of the PLANE he is on is rotating

 

so i think now in a parmenides space there are five coordenates, rotation on the xyz and change of size on the xy

 

with those 5 components you can describe a situation for which an euclidean approach would need 9

 

guess what our matrix uses

Edited by raaaid

Share this post


Link to post
Share on other sites

this is pretty crude but shows what i mean

 

i render fully a 3d setup with a minimum of coordenates:

 

Edited by raaaid

Share this post


Link to post
Share on other sites

as you know ive done some 3d modelling and what i propose is a new approach

 

As someone who models, UV maps and textures at a professional level, I can say with confidence (especially having seen what you see fit to post and call 3D modeling) that

that you should probably contain yourself to the Fee Subject area.

 

Cubic mapping is fine for cubes and spheres, not complex objects that need to be textured and seen up close, with panel lines and other markings no less.

That you're taking this simplistic workflow, again the subject of a many a beginning tutorial on how to UV map a cube, the most basic of objects, and proposing it

here to the developers is strange even for you. That you think that you're educating them on something here is ludicrous beyond measure.

 

There are instances where planer mapping is appropriate, and many where it is not.

OrLok...maybe I weird, but I tend to enjoy UV mapping.

Rigging...not so much.

post-23599-0-61255800-1506280873_thumb.jpg

post-23599-0-73178300-1506280904_thumb.jpg

Edited by Gambit21
  • Upvote 1

Share this post


Link to post
Share on other sites

mybe like i known less than you i followed a different approach than stablished

 

using profile lines both in the x and y axe allow for life like illumination

 

i rediscovered independently cube mapping at least which has some merit but i gave it a different approach than it was followed

 

also you might be a 3d dev but im a nautic engineer and i know what it means for ship maps

 

we currently use mercator which doesnt show right distances the farther from the equator you move

 

cube mapping on earth on the other hand not only allow for exact courses posible to calculate without trig but also exact distances all over the globe

 

of course this was developed in 86 so my teachers didnt know about it and didnt teach this to us

 

why they dont make maps with earth cube mapping i wonder

 

edit:

 

and your confusing the cube mapping i propose with the blender or 3dsmax cube mapping

 

blender cube mapping is done in volumes i propose to use a flat surface, no volume no movement just movement of skins and change of profiles lines

Edited by raaaid

Share this post


Link to post
Share on other sites

Cubic map is still distorted...fairly obvious.

 

A spherical object in 3D space can be mapped....wait for it...spherically  =  no texture distortion

You can't then flatten it out though without distorting things.

 

Any time you try to flatten a sphere you're going to get distortion...thus all the different projections attempting to circumvent this over the past hundreds of years.

Edited by Gambit21

Share this post


Link to post
Share on other sites

now youre wrong you get no distorition at all

 

as i explained in my video you divide earth in six planes WITH STARIGHT LINES, staright lines on earth are call orthodromics(the shortest path between two points

 

and the thing you might notice is that if i subdivide all those 6 planes into smaller planes theyre gonna get subdivided as well by orthodomic lines

 

you could get orthodromics of 1 m and they would match perfectly in the map than in earth

 

you assume as 99.9% of people we live in an euclidean space

 

i assume nothing just supposing we are in the matrix how it WOULD work the most effciently

 

think about it what if parmenides was right and motion is an illsuion

 

do things as they move APPEAR to change size

 

or things change size and they APPEAR they move ☺

Edited by raaaid

Share this post


Link to post
Share on other sites

you see a mesh currently is defined by each xyz of each of its surface points

 

what i propose is define a mesh by 360(one for each degree but you can use more)equation of profile lines

 

plus

 

360 equations of profile lines in the other axe

 

any middle position profile would be calculated by operating both

 

this lines would serve as limit of the cape where the other cape starts, what defines a profile

 

actually reality outside this simulation has to be cartoon like

 

i hate 3d cartoon i miss the old flat ones

 

Edited by raaaid

Share this post


Link to post
Share on other sites

edit can you guess how i did this animation of a cube?

 

 

https://www.ancient.eu/article/60/zenos-paradoxes-the-illusion-of-motion/

 

edit:

an animation of a 3d cube with aminimum resources ☺

 

 

edit:

 

the begining of our hobby:

 

http://torinak.com/qaop#!starstrike

 

it could have perfectly taken another evolution

Edited by raaaid

Share this post


Link to post
Share on other sites

here how i define a 3d space from a flat one in which theres shrink around 3 axes of space:

 

3d.jpg

 

i got inspired for this work by this video:

 

 

edit:

 

as you can see when you simulate rotation in this rendition is move the xyz vanishing points accordingly and an object doesnt really move but change flat size in either axe

 

edit:

 

the reason why its odd i find about cubic projection and then cubic projection its discover in 86 it because chosing as well a cube

 

actually the most logic aproach is pick a volume made out of staright lines with the less sides:

 

this:

 

zUjqj.png

 

from where you build a simulation of an infinite space by this infinite lop:

 

Triangles.jpg

 

eache vertice of these last picture is an axe of space arount to which move the camera over, a well of infinite shrinkness

 

who says you cant have a wide view with no staright lines?

 

just increase the camera frame around that flat skin of 3d space

Edited by raaaid

Share this post


Link to post
Share on other sites

edit:

 

you may wonder whats different this from euclidean space im describing an object with its xyz coordenates

 

yes but the size coordenates i can compress it and save it for its encoded in space nature

 

i just have to pick an object  at any xyz coordenates and move it with this algorythm to xyz 111 and that will be its real size

 

while on euclidean space you have to describe its size as well as the xyz

 

personally i dont understand how newton got away with the concept of spot masses

 

so basicallly on this space the size of an object is determined by its coordenates i just say this object is of this size i can touch with my hands by moving it to the coordenates 111

 

so i dont need to put into the renderitation the size of an object blus its position for in my opinion the SIZE AND POSITION ARE IDENTICAL THINGS 

 

if a ballon doubles size as it raises up with this space has moved on 3 axes at the same time, i really dont mind if its gotten bigger or closer cause actually you can not tell the diference unless you have binocular sight

 

what if i recover sight depth perception for i see flat just from one spot and i find out without doubt that the illsusion is not the change of size but the motion, with binocular view you can know

 

Etri2.jpg

Edited by raaaid

Share this post


Link to post
Share on other sites

I thought this guy was supposed to be forever banished to the free subject section? I clicked on this thinking it was going to be something serious...

Share this post


Link to post
Share on other sites

im serious and i refrain from my usual jokinf in the fs section

 

i have the basis im a nautic engineer and i got excellent grades on spheric trigonometry and geometrical drawing

 

our tech is litle evolved, 3d rendering is a right brand new technology in his infancy

Edited by raaaid

Share this post


Link to post
Share on other sites

compare what i propose:

 

TRIANGLULATION_12b.jpg

with which we currently use:

 

Azimuthal_equidistant_projection_SW.jpg

Share this post


Link to post
Share on other sites

edit:

 

if your navigating your plane and use radiogonometry to obtain your position you need at least 3 spots to triangulate from

 

Measuring+Earthquakes:+Triangulation.jpg

 

so id define an space as the distance to 3 spots on the plane that are set in a closed infinite loop like that farctal triangel i posted before

 

the main advantage i see for this rendition engine is you can have more than 180º fov keeping staright lines starigh, no fish eye

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...