[test] some sort of RPG engine ( calling for Blink, Strille, Andree ? :))

[luxregina]

Hey ...

Been taking a short break from Sylvaniah, and wanted to play with a old idea of mine



http://grafikdezign.free.fr/flashkit/engine3.htm
EDIT : I've put the latest version : engine 2
EDIT : I've put the latest version : engine 3


It is based on Blink's engine, that he came with in the Percepea thread on the Mario Kart engine...

I was thinking this could be used for other purposes than just racing games :

This engine could be use for the outdoors, and this one http://grafikdezign.free.fr/strille/strille.htm for the dungeons / indoors.

It doesn't need to be fast paced, as it would be used in the context of an adventure / exploration game - i feel it could lead to a fun game

However, here are my questions :

- Does anybody happen to have the latest Blink's file ( the one with the racing opponent, the little Elvis ) : i searched, but the latestest link i found was dead : i would like to see how he handled predefined paths

- I would like to study if the opponents deplacement could be triggered server side : that way, as in Andre Michelle's game ( the racing one ) other players could join and navigate on the same map : i wonder if it would be possible : any multilpayer guru point on view would be welcome

- Would like to ask Blink if he ever came with some Z-sorting in his engine ...

Well ...that's it thanks for your feedback

[WilloughbyJackson]

I think I have the Blink Elvis engine...I'll take a look around for it when I get home.

It's looking very nice so far...

-pXw

[tomsamson]

hey there Lux,
nay,i´m not strille,andré or blink,but well,i´d like to reply anyway
yeah,its looking nice so far.
z sorting ain´t so much the problem and yeah,updating player positions could be managed too with the fitting server,what i see as more problematic is the map size.
Unless you want to keep the map really small (for the kind of game you want to make it is small at least,for racing games it´d be ok),you´ll have to quite improve the engine.
as far as i know all games using that engine concept up to now did it pretty much the same way,meaning they use one big graphic for the map.
that of course sets a definitve limit on your map size (because filesize explodes with bigger maps and huge map graphics can´t even be handled well with a fast cpu and much ram powered pc).
therefore you´d have to extend the engine to work similar to a tilebased engine in the way it plots several smaller graphics to one map.
don´t know if it can be handled smoothly (together with the masking/skewing effect) with the current flash.
don´t want to sound negative,i think you´ve got a cool start there and it´d could become something great,just wanted to let you know about the downsides of the engine concept ahead.
(on a sidenote as far as i know blink is on vacation for a week)

[luxregina]

Hey

WilloughbyJackson > Thanks, that would help a lot

Tom : he he, of course you can reply : the list of potential people that could give feedback was too long, so i picked the ones that actively worked on that engine, to my knowledge

- For the maps, i studied a bit Strille's engine ( wich is not the one present here ) and he seems to handle the "connection" between several maps quite seemlessly, so yes, i think I can hope for a bigger "world" ... dynamicly loading the contiguous maps could do the trick on the filesize

- As you highlighted, this is just a start : i'm toying with the idea for now, and try to see if I can come with something satisfaying conceptwise, and then, if it works, i'll be starting talking about a Game and it will certainly be here ...

Just wanted to take a break for GFXs and deal with some code, for a change i know i won't be too successfull in that, but, heh ...

[luxregina]

Could you PM me Blink's email ( if it's not a problem ) ? I would like to let him know that i'm using his engine, and ask him few questions ...

[tomsamson]

oh,okies,i didn´t try that thingy by strille =)
anyway,wish you good luck with it =)
as with any of your projects this has great potential =)
i´ll pm you blink´s mail (though as i said don´t expect a reply within a week)

[lapo73]

very nice prototype Lux! It has some very good potential. And with the soon to come rendering speed enhancements, maybe it's the time to start thinking 3D seriously.

Adding a multiplayer side to this game would be a really interesting study: the basic interaction between characters and environemnt wouldn't be too difficult but it would be interesting to see how to handle other AI-based characters.

In an exploration game like this you could meet various characters/entities who may give you important informations about the adventure, sell items etc... but also they could try to fool you or start a fight.
It would be interesting to develop the AI on the server side and let the server handle the stories.

Maybe you'd need a database describing the world and have a smart client side system for maps to allow smooth transistions between each zone. Once you have loaded the current map portion you could start to silently preload the adjacent zones so that they are ready when the main character enters a new area.

The server would run the complete world simulation, running the AI and updating the clients by sending only the informations related with the area in which they are currently located. In other words each "cell" of the game world would be just like a server room and you can have as many of them as you want.

I am picturing this with our SmartFoxServer PRO (pimp mode )in mind but that would be doable with other server as well. In this case a Java based server would be very helpful as you could take advantage of full OOP coding for the server side extension.

Both SFS and Electroserver (maybe Unity doesn't do that) support Actiosncript on the server side also but I wouldn't reccomend it. Java, C++ or C# would be the best choices here.


Lapo

[Kianis]

Originally posted by luxregina
i would like to see how he handled predefined paths

Your project looks very interesting
I think blinkOk drew the path and used the motion guide to tween an mc along it. Then he took it's coordinates and translated them onto the 3d map.. At least that's what I recall, but I could be wrong.

[tomsamson]

Kianis suggestion is a good one,don´t remember if blink did it that way but it could be done that way for sure. an improvement on that concept would be if you placed target points on the map and then the enemies would move towards the next target point in their own manner,then you wouldn´t have totally stiff parts.
i´d suggest doing the pathfinding,collision detection etc tilebased though and then translate positions to the actual skewed map

[mclelun]

very cool man

[luxregina]

Yeah, it is likely what he did since the whole coordinate system is based on the 2D map ; that's the beauty of it : even a Noob like me can toy with it

The thing that is important in Blink's source, that i tried without too much success so far is that, if i remember well, he successfully integrated the SuperSamourai "fake 3D" method that allows to "emulate" the different facets of an object as your point of view change

There are objects like the trees or the grass, that will stay pure 2D, but i would also like to incorporate more complex objects that the player can go around : I will save that "3D" for the important details to not use too much processor ressources

Lapo > You are reading my mind but, then, it would be talking about a MMORPG ( *run away scared* )

[strille]

It looks like an interesting idea.

I do feel that Flash is a bit too slow to successfully handle z-mapping and raycasting. I know it kind of works in tech demos, but not for a real game IMO, at least not yet. Or maybe it's just my crappy computer...

[tomsamson]

yeah,i kinda agree with strille,when all which makes up a game comes together (game logic,graphics,sound etc) its so much more cpu demanding than in the tech demos. i know there are several games out there which use the engine concept and we worked on one,too,but yeah,they basically show the same bottom line,you have to really minimise the game window a lot and still watch out for performance a lot then, to allow smooth gameplay, so i think its not the time to make such a huge project in this engine style yet. with the next flash i´d be well up for it (if f8 holds up to the rumors at least a bit) but yeah,with the current release it´d be way painful i imagine.

[VENGEANCE MX]

Nice work. I got an fps of about 7 whilst moving, and 9 when static.

[adit_ya_sharma]

Very nice engine. I would like to see a fighter plane game through this engine in which you can see the ocean downwards which can keep scrolling in a loop and you kill the other planes, can that be done?

[luxregina]

I don't know, i get a solid 30FPS when testing on my computer, and i had it tested by several personns, and the average feedback is around 25 Fps ...wich isn't too bad so far...

- There are certainly lots of optimisations to be done on the maps, or the size of the gfxs, so i think it is safe to say that it could be faster. Also, the speed/smoothness issue will certainly be less vital than, for example, a racing game, or a FPS ...

- I think to this point that i would be ready to specify higher machine specs in order to have a final product a tad different from other Flash Rpgs : if the final product is good, and looks good, i feel players will undertand higher specs

- If we want to be ready for Flash 8, i think we should start working now ...

In few words : I really believe in the future of this type of engines ...

[tomsamson]

yeah,right now i get a solid fps,too (i guess vengeance´s system is a bit old) and yeah,sure you can tweak it here and there to get better performancewise in its current state but yeah,i still have the opinion it won´t run well in even half the screen size it has now once real game action is going on there (with the current f release).
i have a bit different opinion on some things you said,for example yeah,speed isn´t that important for a gametype like this but smoothness is,i don´t like imagening a npc or player/player battle at a 8 fps framerate (on average machines).
i also don´t think that people playing such a game would accept high system requirements easily i mean yeah,we as flash developers know what can be easily achieved smooth on low end machines in flash and what requires a high spec machine no matter how much we optimise it,but that doesn´t count for the average web gamer i think,he will just wonder why he could play even 3d games smoothly on his 3 years old machine and this game runs at 8 fps.
but don´t get me wrong,as i said i see big potential in your idea and when its done with the plan to do it for f8 right away i´d be into it,just don´t see sense in doing it for f7.
i agree there´s a cool future for this type of engines and sure,and as all flash developers i strongly hope f8 offers the expected performanxe boost so we can work on some of our dream projects like this so i wish you good luck with it =)
as i know you´ve started projects like 2k and brought a flash rpg as far as very few others,i think if someone does it,you´ve good chances to be the one =)

[VENGEANCE MX]

30 fps?! Sigh... I guess I've developed low standards as a Mac user. 7 is quite good for something like this on my machine. On the original engine, in Flash Player 7 (not playing in a browser window), I could get about 12 whilst moving, and just under 30 if I shrank the window to about a quarter of its size.

I've got a 450 Mhz G3... 512 MB of RAM, 8MB of VRAM. Ouch. On a PC with equivalent specs, it would run A LOT faster (Damn you, Macromedia...).

Has anyone announced a release date for Flash 8? Because I'm buying it with my own money the day it comes out.

[Santuria]

Mine never dropped below 32 and im running a 3.2 pentium 4 processor with 1gb of ram.

[Hasufel]

well, well ...