#51 2016-01-22 20:40:52

Hypernova^
Member

Re: Rival: Cube/tesseract mod

Hi Chase,

  Thanks for your reply. Yes there are a few things I'd like to do with the particles. The current version of my game is based off the Justice Edition port of SauerEnhanced, which has some cool particle effects. In particular I'd like to replicate the glow particles and time-based expansion of particle sprites. I've been too busy with work to get back into fun coding for quite a while, so I might be forgetting some things xD But this should be a good starting point.

Offline

#52 2016-01-23 15:15:00

chasester
Member

Re: Rival: Cube/tesseract mod

ya size is an attribute it will be able to be modified, this really isnt that complex with the current particle system. Cube particles are very powerful in general, It just lacks any ability to easily modify the attributes. That system itself isnt bad.

chasester

Ps: Im not 100% sure if the release will be done by the 26 Im gonna be working hard on it, I ran into some trouble with loading and saving from the map file. As soon as I resolve some of the issues Im having I will be doing a release.

Last edited by chasester (2016-01-23 15:16:36)

Offline

#53 2016-05-13 10:14:59

VegemiteToast
Member

Re: Rival: Cube/tesseract mod

Very much hope this is still in development.

Offline

#54 2016-06-07 21:09:31

n0obGuy
Member

Re: Rival: Cube/tesseract mod

HEAR HEAR! I will endorse this in the form of a purchase should this take off. I absolutely LOVE the cube engine... I'll buy two copies to get a friend and I. I've always wanted to make a game in the cube engine, but if you can make one that is on par with Unity/Crytek in terms of intuitive UI it'll be THE deal maker.

Offline

#55 2016-06-08 10:57:29

chasester
Member

Re: Rival: Cube/tesseract mod

lol oh god i hope it doesnt have as "good of a ui as unity :p but ya thats the idea.

Developement is going slow, but it is still in production. I just got to the point of being able to serialize objects and save them back to a file and then reload that data back into new objects. Still working out a few bugs, but this will allow you to save any angelscript object and then bring it back, setting all the data from the stored data in the file. This file uses the same structure as the cube ogz files so it will be encrypted and compressed (which is a major improvement over unity :P).

Future goals are to allow for development across computes (basically coop but also allowing scripts and asset sharing etc. so that people can work together to build games,  or show there work to potential investors, across the room or across the world).

Also improvements with the animation system, posibably support for newer animation types, an editor that helps with ragdolls, animation editing and setting, and animation state machines.

Shader node editor allowing simple shader editing using a node system (obviously this will be a big feet saying the way that cubescript and glsl have this wierd unholy union in cube right now. But it will eventually happen).

Obvious improved gui and ui for edit mode (goes without saying). Similar concepts to a inspector in unity, allowing to change variables at runtime. Most things that are in command line will have a gui implementation to improve simplicity and the learning curve.

Improvements on the gui menu system. Allowing windows and modules etc. Just adding more features and more customizability. Probably work with a system with the html/css/js system. Tho it will probably be a mix between cubescript and angel script.

as for a licence/buying etc. It will probably be open source with a royalty similar to unreal. But Obviously its far from that at this point, it may be full open source. This will just depend on support and fan base etc.

I appreciate people are still interested
chasester

PS: I will probably come out with another demo video soon. Hopefully I can get more then just saving and loading done. But this backend stuff is very confusing to me ;P and I've ran into so may bugs I had to work around. Its seemed easier in theory :)

Offline

#56 2016-06-22 03:05:43

chasester
Member

Re: Rival: Cube/tesseract mod

New Video

More details Over the next few days :)

thanx for your interest
chasester

Offline

#57 2016-07-02 20:01:26

n0obGuy
Member

Re: Rival: Cube/tesseract mod

Will it be possible to select actual in-game NPCs? I was thinking about making an RTS using this new engine, or even if possible making it in Tesseract as it stands right now. Firstly I'd like to know how to select in-game NPCs, maybe the engine can highlight the hitboxes of the selected character to illustrate when I the user have selected an individual bot.

I was thinking of a 3rd mode, something like editing mode but I can only select NPCs or player entities (not cubes, textures, or spawns) and by some command like right-clicking, the bot would obey a directive to move to the location to which I right-clicked on.

Offline

#58 2016-07-03 06:11:35

Shikijo
Member

Re: Rival: Cube/tesseract mod

Hello! Not sure if this is where I post this but I was wondering if it would be possible to add a "zipline" feature in game, like jumppads and stuff..
for eg.  like in this bioshock infinite vid starting at 8:10 https://youtu.be/qbYH5mjBFKY?t=490
I think it would be a nice gameplay element..where it be in fragging or racemods.. *-*

Last edited by Shikijo (2016-07-03 06:12:27)

Offline

#59 2016-07-03 23:17:17

chasester
Member

Re: Rival: Cube/tesseract mod

@noobguy
RTS IN THIS ENGINE IS A WASTE. USE AN ENGINE THAT IS OPTIMIZED FOR RTS GAMEPLAY. (If you want I could go into the couple hundred reasons why this engine would not be good for an rts game). (As much as i dont like unity, unity would do this well). But as far as selection yes objects will have hit boxes that are set by physics and these hit boxes can be used to code various things like selections (or like shooting ppl, changing the cursor if they are aly etc).

IF FOR SOME UNGODLY REASON YOU WANTED TO STILL MAKE A RTS USING TESSERACT (OR RIVAL).
What you would do is just simply set the look at vector to the center of the screen. then do a free rotate that rotates around that center point (by modifing only yaw and pitch keeping the look at vector in the center of the screen.). Then disable the renderavatar (this stops the hudmodel and playermodel from rendering or casting shadows). Then simply run a raycast using interceptcloset(fpsent o, vec ray, vec o) where as o is where you (camera1), ray is the normalized vec of where you are looking, and fpsent is a new fpsent. Then test if(fpsent o && o->type == TAG_WHAT_YOU_WANT_ABLE_TO_SELECT) then simply add it to the selection group if hit.

That should give you a good start.

@Ziplinning
A zip line is fairly easy to implement but it would be like declaring "ZIPLINE" you would have to take over the player object and assign it to a physics control box that is attached to the said rail. then make the rail have lil or no friction, (so you just glide). Then just allow the user to jump. It will be possible at some point with a lil work. When I get to this point I will probably make a video showing something similar to zipline implementation.   

Chasester

Offline

#60 2016-07-26 05:01:51

oneiro
Member

Re: Rival: Cube/tesseract mod

is the physics code networked?

Offline

#61 2016-07-26 18:48:50

chasester
Member

Re: Rival: Cube/tesseract mod

i sware I ansered this yesterday ;P idk what happen.

No Its not, the server-client code will be rewritten, but not till most all other elements have been fixed. You Probably could do it your self by making all Physics objects send there N_POS then use something similar to predictplayer() to interposition the objects between server updates. IT will look perfet as long as ping is 500<. But this is more of a question to ask on the bullet forums, im sure they have a decent structure on how to implement multiplayer physics.

Offline

#62 2016-07-29 02:53:23

chasester
Member

Re: Rival: Cube/tesseract mod

vid

Added some new features:

-Multiple controllers per object. This means you can have a light, and a model share the same node so that it can render together (say in the case of a light on the wall.

-Fixed some weird physics errors with the default settings for the rigid bodies. This just means better out of the box simplicity.

-Added lights allowing the world to keep tract of the lights in the scene and add them to the light renderer.

-Added in some back end stuff to make cool stuff happen. Many in the scenegraph.

Added editing of node variable (properties) through the command editnodeattr "varname" int(number). Sadly this only works with variables typed as ints (this will be fixed soon).

chasester

Offline

#63 2016-07-30 13:23:31

catalinux
Member

Re: Rival: Cube/tesseract mod

Hi chasester,

I just wondered before implementing the bullet physics if you knew about the intencity engine implementation. Looks good.

https://github.com/kripken/intensityeng … bullet.cpp
https://www.youtube.com/watch?v=1vvsVkdJiSQ
https://www.youtube.com/watch?v=DHrfLF_RatE

Last edited by catalinux (2016-07-30 14:25:38)

Offline

#64 2016-07-30 22:25:32

RaZgRiZ
Moderator

Re: Rival: Cube/tesseract mod

Well this brings back some memories. RIP good old mods like Syntensity. So full of premise :)
Personally that's the first i've heard of it using bullet physics. Interesting. I guess they might have been necessary in some way to implement that car racing mode :P

Offline

#65 2016-07-31 07:36:35

catalinux
Member

Re: Rival: Cube/tesseract mod

And in this video the rockets influence the model physics:
https://www.youtube.com/watch?v=L02dNMDKGok

And mesh mapmodel collitions:
https://www.youtube.com/watch?v=DHrfLF_RatE

The racing is pretty neat :)
https://www.youtube.com/watch?v=kXqbdwFirJA

Cutscene (with bezier curves)
https://www.youtube.com/watch?v=8QpYGb_k2-k

Flashlight:
https://www.youtube.com/watch?v=Pdc80qSUrIk

I guess octaforge is the successor of Syntensity. If so I wonder why it doesn't include the bullet impl and all the other features.
One downside of this mod is that it uses javascript through V8. Plain cpp would be much lighter (no 3rd party add-on) and easier to compile. But on the other side easier to develop something new (no need to recompile), cut cubescript is sufficient.

Last edited by catalinux (2016-07-31 07:47:11)

Offline

#66 2016-07-31 20:26:11

chasester
Member

Re: Rival: Cube/tesseract mod

c++ would be simpler but you have to keep in mind that every time you wanna change something you have to go to your compiler and recompile it. Honestly Im not that impressed, its just a rehash of the attr5 system with more options (like collision box size and etc).

The beauty of AngelScript is that it runs as fast if not faster (with jit Which i have yet to implement) as any other script language out there, for these type of calls. This allows me to define an infinate number of variables (properties) and subset them inside of controlers. My node has 4 elements:
vec o, rot, radius; //36 bytes
vector<asIScriptObject>; //(length*4) +4 bytes;
=
40 + (length *4);*

compared to cubes old (i think his is 10 shorts)
vec o; //12 bytes
short attr[10]//2 *10 bytes
extentity *attach; // 4 bytes
vec radius; //12 bytes **
I asume he has this because he has to define a colision box some how?
=
48;*
*bytes on a x86 machine

this means I can run way more nodes in my world with less memory usage, combind that with the ability to create custom properties and reuse controlers in various forms and you basically have an infinately customizable object.

Say he wants a platform to run allong a path. He would have to find a way to create a new ET_ ent and attatch it to another new ET_ path, (assuming he even has attachment or parenting). It is a quick easy solution and work for many cases but if you want something that isnt "out of the box" you have some serious problems, which involve changing the source recompiling the project and then debuging it over and over. I can (in the current build) write code in my *.halo file, change it and debug it while im in game and see the results live. This is why scripting is king and people prefer using language elements rather then binnaries. Ya binnaries be faster but scripting is much easier, and simpler to level designers, who are normally not programmers :)

But its nice to see octaforge is still in business. I visit there website occasionally.

@Bullet physics implementation
My build can do the same, save the mesh based collision (only cuz ive not added the option yet, (the voxal terrain already uses this call)). And the projectiles have not been added to phsics because Im going to add them to angelscript first before adding them to physics. This should be added by the next build or 2 (well see what problems I run into).

chasester

Last edited by chasester (2016-07-31 20:35:23)

Offline

#67 2016-07-31 21:18:53

RaZgRiZ
Moderator

Re: Rival: Cube/tesseract mod

Curiously enough, make an angelscript schip8 emulator, i want to see how fast it will run :p

Offline

#68 2016-07-31 22:58:12

chasester
Member

Re: Rival: Cube/tesseract mod

check the forums

Offline

#69 2016-08-06 19:01:28

ImNotQ009
Moderator

Re: Rival: Cube/tesseract mod

OctaForge is not using JavaScript. It was actually using Lua (and then a custom superset of it) but nowadays it's being rewritten to utilize a custom build system to work with native C++ API.

It doesn't have Bullet because it's not an ideal solution for this engine. The developer says he's more eager to implement ODE instead.

Offline

#70 2016-08-10 03:02:58

chasester
Member

Re: Rival: Cube/tesseract mod

Ya I agree bullet may not be best, but Ive not intergrated to deep with it, it would be fairly easy to switch engines at some point. I think the biggest hang up with bullet is its non-predeterminism. Which when adding multiplayer can lead to problems with server and client prediction. And cause "popping" errors, if there are lag spikes. So I may change to an engine more designed for 3d games in the feature. Bullet was just already set up and an easy out of the box solution, which works well for right now.

chasester

Ps: I hate lua its sytex is very archaic and seemingly abstract. It would be nice to have a more c++ api, one of the major reasons I like as, there is litterly no learning curve it work almost perfectly like c++ save * is @ kinda :P but thats just cuz you cant do true pointers safely :)

Offline

#71 2016-09-10 02:02:53

daleth
Member

Re: Rival: Cube/tesseract mod

chasester,

I just recently found this post about your "Rival" engine and was really interested in it. I have always wanted to make or see someone make a version of the Cube Engine that is viable for a commercial use.  I have been a member of the Cube Engine community since it's beginning back in the early 2000's. From what i have read and seen in your YouTube videos, you have done great work with the engine and enhancing it.

I had some questions regarding Rival, forgive me if these have already been asked. Also sorry for jumping in randomly and throwing all of these questions at you, this is just something I was excited to know more about.

1. Do you have an official website of any sort for your work on Rival?
2. Do you plan to keep the cross-platform nature of the Cube Engine?
3. If so do you plan to publish the binaries for the different OSes or just offer the source for others to do so?
4. Do you have your work available anywhere in a public github or repo of some sort?
5. Do you have any updated binaries? The links on the OP are down.
6. Do you have a plan for when you will release a official Beta or v1.0 of your new engine?
7. How is development going overall?
8. Do you plan to just make an engine or do you plan to make a game using the engine? (Obviously you will have to make some sort of demo content / game so that the engine's features can be displayed, but I was wondering if you plan to try and make anything beyond that.)
9. Since you are doing what appears to be a massive rewrite of several aspects of the game, do you plan to also attempt to have game console support (or is that something too far our to even consider at this point)?
10. Also since you are doing a massive rewrite, do you plan to try and launch the engine on Steam or try to incorporate stuff such as Steamworks, or anything else like that?

-
Daleth

Offline

#72 2016-09-13 04:15:26

chasester
Member

Re: Rival: Cube/tesseract mod

@Daleth

1. No. At this point I dont see the point investing that much time into building one.
2. Everying I have used is crossplatform (angel script, bullet) and everything I will add will remain crossplatform. Ive seen some android builds of cube and tesseract and plan on adding mobil support later on. But for now I only have binaries for windows, but I know one person has built it on linux with no issues.
3. Source code will probably be open source in the feature, with some kinda non-commercial license, but for right now I dont want to have 50 mods of the code making it look convoluted so im keeping the source closed.
4. Same as above, if I got people interested in helping me add features to the engine I would upload it, But for now there is very little interest, ive given the source to 2 people so far just to play with.
5. The camera class i designed is messing up I could build some for you in the next couple days if you'd like. really nothing to special to look at, ive not added much functionality to angel script yet.
6. I am hardly in a pre-alpha state so beta will be some time off.
7. Slow, I get very little time to work on it, and most of that time comes all at once. I am swtich jobs in the next few months so hopefully that will allow more of a concistant pace.
8. I dont plan to make a game. I feel thats a poor way to approach a marketable engine. Commercial companies do a poor job at that so I doubt I will have any success at that. Secondly I have very little skill with various elements of 3d art so It would be hard to make a good game.
9. Everything so far is written to be multipatform. Basically yes. To suport a game consol would be no more than changing a few various lines of code. SDL does a fairly good job at making things cross-platform, I dont think much work is involved. Angelscript does a good job at making there stuff work on almost everything (there are a few notable errors, but the developer is very keen on getting stuff fixed fast and very much supports his community). And bullet is bullet idt It uses any thing that would make it not multiplatform.
10. Steam is a 100% what I will be shooting for. I will probably have steam functionality already written into the game engine so people can easily add achievements, workshop, etc from the game engine. So the idea will be to have a game engine that can offer and easily writable edit mode that will allow people to use the rules of the game, assets and prewritten script, along with custom scripts, game modes etc. Allowing a game to have a built in editor that can be used with almost no development. This edit mode would be multi player, and would allow developer the ability to set custom edit parameters or nodes etc to make it easier for the in user. Basically custom maps galore. I may release the engine on greenlight to help market it, but that will be at a later point. I really dont need to market right now, I dont have anything that is usable to make a game. In about 6 months I should have the very basic ground work for a game engine so that simple fps style games can be made.

Personally I would like to speed up development by getting more people, but that a double edge sword. I am very particular on how I want things written, and on how the flow works. Cube is a very clean engine, very fast, very effective. But bad coding conventions, weird memory allocation, and poor memory usage can lead to problems fast. On top of the fact that people that arent being paid dont listen or work well ;) but you know ive been there done that.

chasester

PS: sorrie it took so long to reply all these bots in here make it hard to notice actual people :P

Last edited by chasester (2016-09-13 04:19:03)

Offline

#73 2016-09-13 17:04:11

Hypernova^
Member

Re: Rival: Cube/tesseract mod

Hi Chase,
   I think the best way to open development to others is to create detailed tickets for tasks and allow people to submit branches for code review. This would make you like a manager over the project where everything must be reviewed to be allowed into the main Rival source base. Might be useful for bug fixes or simple boring projects that you don't want to do yourself. Just an idea.

Offline

#74 2016-09-13 21:22:40

chasester
Member

Re: Rival: Cube/tesseract mod

This would be very helpful in the future but for many of the core foundation I want to make sure there is a good ground work. This is the stuff that gets called 100+ times a frame cycle. So it needs to be effective efficient and basically perfect. If you have code you wrote that you think would help the project definitely post it here or just message/email me about it. I will definitely add it in the next build, When I get into an alpha state I am going to create some kind of content creation standard, guidelines, and things that need worked on. My biggest fear is having commands like python/pygame etc where many of the internal calls have weird and unstandardized calls making it hard for the users to easily guess or learn the coding standard. Many of these things can be avoided by simply making a modularized system for submitting code, making your own namespace (or just class depending on the type of code), and using standardized names when allowing out commands for as and cube scripts.

This will just make it easier to document and to add in new code, and make overall development easier and more clean.

But if you have code that could help the engine please submit it. (i may change variable names and syntax), But i appreciate all contributions no matter how small. Cubes code base is very large and you may know part of the engine better than I do, so it maybe a five min change for you where it may take me several hours to find the write code to manipulate.

chasester

@update
I will be coming out with an update in a month or 2, Ive been trying not to come out with too many petty small updates, The next update will probably be almost game read (well like the very basics) so you will have camera and player controls, along with triggering system, (so things like teleports, spawning monsters, picking up crates etc) will all be posible (may need a little coding on your part), along with other physics improvements like cylinder, and mesh based physics, all 9 of the options for physics controls being allowed. And movable platforms, based on any various variables you'd like to set. A large number of math functions will be added as well to help with the movement of these vectors to better allow basic platforming. Ive been spending a large time rewriting the serealization code to better improvements and to facilitate almost every type of variable that you want to add (things like strings arrays etc) along with internal classes like cameras, lights, models, physics interfaces, etc. Also I am working on serializing actual physics objects (which is done by bullet but still need stored) to allow for instant save files. (Something cube has nv done), meaning that you can save a moment in time and go back to it the exact way it was, (aka saved games).  I may not have this last feature fully implemented but most of this is written from the ground up, or modifying code that was used for a slightly different purpose. The release after should pull me into a pre-alpha/alpha state. I will mostly be working on gui and edit functionality which I may need help writing, or at least will take some assistance making it easier for the in user, though it will all be written in cube/as scripts so it can be changed 100%. So the goal is to make it as fluid as possible allowing almost any type of interface the user wants to write.

Thanx for all the support :)

Last edited by chasester (2016-09-13 21:32:31)

Offline

#75 2016-09-13 23:23:53

Hypernova^
Member

Re: Rival: Cube/tesseract mod

Yes I understand there could be a lot of optimization done (as with Cube 2). I like that you are shooting for a minimalistic and efficient approach (vs some open source engines which are way too massive and heavy). And yeah I totally get your concerns for code contributions, this is why I suggested creating descriptive tickets and doing code reviews before letting anything in.

I am currently working on Mac and Linux build files with gcc and gnu make (for most portablity). It's taking me forever cause I have about 0% free time atm (work and studying) lol. Eventually I could maintain Mac and Linux support for this project if it was needed.

I've been in the Cube and Cube 2 source since I was 13 (5 years) and written many extensive mods and other crazy stuffs when I was young, so when I get more time I really want to help with this project (although I have my own fork of Rival in mind). Probably my first step will be networking the physics objects, and also adding multi-material effects and upgrading sound (some of this I had written to you in email). I'd be happy to backport any engine-level sharable features I will add.

@update: Sounds great, I like that you are going the full mile with AS support. This will definitely set your mod apart from the others which exist around Cube 2.

<joke>
  Q: Why are ASM programmers always soaking wet?
  A: Because they're below C-level.
</joke>

Last edited by Hypernova^ (2016-09-28 13:56:52)

Offline

Board footer