Beginning development on a Suikoden-inspired game engine

[thcrock]

Another idea! The randomness/time delay Suikoden III introduced to some treasures could be cool.

1. Certain treasure chests could be somewhat random, by putting weights on each item possibility. So if the game designer wanted to introduce randomness to certain chests, they could. Or just set an item's probability at 1.00 if they didn't want any randomness.

2. Time delay/ compounding . Remember the healing items/corpses in Suikoden III that could get stronger the longer you didn't pick them up? Or items that you pick up and come back after a delay.

[Xelinis]

I can attack both of those points with scriptable item discovery!

I was originally planning on enforcing a property list where each chest was mapped to an item. However, even if the scriptable approach means a bit more work for the game designer, it means a lot more flexibility and, if done correctly, less coding for me!

[Antimatzist]

Support Chars? And if you do that, enemies should have support chars too (not everybody, but some of them like soldiers and bosses)

[Raww Le Klueze]

2. Time delay/ compounding . Remember the healing items/corpses in Suikoden III that could get stronger the longer you didn't pick them up? Or items that you pick up and come back after a delay.

I found that more annoying than anything. Herbs I never bothered with since there wasn't any way of knowing what you'd get until you got it and you still got more reliably from the Boss Chests.

There's also that whole thing about an alarming amount of people that keep dying in the same exact spot in secluded areas. I can't get my head around what idea they could possibly have had for respawning corpses with random loot.

[thcrock]

2. Time delay/ compounding . Remember the healing items/corpses in Suikoden III that could get stronger the longer you didn't pick them up? Or items that you pick up and come back after a delay.

I found that more annoying than anything. Herbs I never bothered with since there wasn't any way of knowing what you'd get until you got it and you still got more reliably from the Boss Chests.

There's also that whole thing about an alarming amount of people that keep dying in the same exact spot in secluded areas. I can't get my head around what idea they could possibly have had for respawning corpses with random loot.

Well, that's down to a game design issue. A good designer could probably come up with a more plausible way to explain that. The ability to put in random locations, so the corpses show up in different places, would be one way. And boss chests being more reliable sources of herbs is definitely a game design issue too. The herbs just needed to be more numerous and perhaps recharge faster. On "not knowing what you'll get", that's a good point. They might need more visual cues, and even a 'There is a Medicine x5 here. Take it?' prompt could allow you to wait until its stronger.

Fully scriptable would probably be my preference, so the designer could put in arbitrary rules based on the player's stats. Like if they have taken an inordinate amount of fire damage recently, a fire protection accessory shows up. Or if they are broke, a bunch of money. A designer could put a lot of thought into item drops, if the script had access to a bunch of stats like that. If your battle system is modular, you couldn't put stats like that directly into the engine - the battle system they put in might not even have any concept of fire damage, for instance. But you could always just provide the game developer with a data structure to hold any stats they want, and make that structure accessible from the item discovery scripts.

[Xelinis]

To keep things modular and easier to debug, the battle systems will not be able to access the global data structure, which will most likely be in the form of a sqlite database. However, I'm taking a cue from Cocoa and enforcing that the entrance function of any battle system accepts a dictionary of property strings mapped to values as a parameter. This will allow the battle system designer to be able to define and work with any number of parameters while keeping the code clean and flexible.

[dorky106]

just let us know when it is far enough along to get a copy of this engine and start screwing around with what we can do with it to see how well it works.
Think of all of us as your beta testers

[Thiamor]

Not sure if this will be posted in again, but I wanted to share my idea(s).
I work with RPG Maker, so I know I'd not be able to do this, unless I did a lot of events, which connected into battle, as well.

The longer the battle goes on, the more exp and gold you get. Every 15 minutes for little monsters and such, and for bosses it depends on their own stats. If they are strong, it is less time it takes.

Also how Suikoden battles were, mainly 1 and 2, they had 6 members in that odd angled perspective. Well the same, and each person has their own 'squad' so to speak. You can either attack with who you got, or click (Squad) and it then changes 5 people to the person's squad. Each person would have their own. So all in all 36 people can be taken into battle.

I also feel 'runes' could be mixed with other runes, may it be TRUE or otherwise. Also instead of the set 'stars of destiny' any NPC can be recruited over. Even if it was an evil boss, after every fight you can ask them to join until they do. It'd make it more interactive. Maybe even add a thing for Item mixing, Armour Mixing, and Weapon Mixing. So then each Armour can have a different outcome for ANY person. Then imagine sharpening those weapons.

If possible to code for this engine.

[Xelinis]

I've begun to investigate the possibility of Lua support. The language itself is fairly powerful and easy to integrate; more details as they come.

Not sure if this will be posted in again, but I wanted to share my idea(s).
I work with RPG Maker, so I know I'd not be able to do this, unless I did a lot of events, which connected into battle, as well.

The longer the battle goes on, the more exp and gold you get. Every 15 minutes for little monsters and such, and for bosses it depends on their own stats. If they are strong, it is less time it takes.

Also how Suikoden battles were, mainly 1 and 2, they had 6 members in that odd angled perspective. Well the same, and each person has their own 'squad' so to speak. You can either attack with who you got, or click (Squad) and it then changes 5 people to the person's squad. Each person would have their own. So all in all 36 people can be taken into battle.

I also feel 'runes' could be mixed with other runes, may it be TRUE or otherwise.

Again, battle systems are going to be purely modular components here. The only module I'll provide by default will mimic Suikoden II's battle system; anyone else experienced with C and SDL will be more than welcome to program and upload a module of their own.

Also instead of the set 'stars of destiny' any NPC can be recruited over. Even if it was an evil boss, after every fight you can ask them to join until they do. It'd make it more interactive. Maybe even add a thing for Item mixing, Armour Mixing, and Weapon Mixing. So then each Armour can have a different outcome for ANY person. Then imagine sharpening those weapons.

If possible to code for this engine.

By creating this engine, one of my goals is to limit the designer's degrees of freedom into something manageable. Adding features like that will vastly overcomplicate things and make the engine more and more unwieldy. With me adding Lua into the mix, it may be possible for someone to implement these features on a script level, but as far as making it into a standard engine feature goes, I'm afraid that the answer is no.

[KC_MCDOHL]

How are you gettign on with this Xelinis?

[Xelinis]

A nearly full one-subject notebook filled with diagrams and a couple of new book purchases. I'm scrambling to finish up my degree requirements right now so all personal projects are on hold. The past three months alone have been the most intense academic period of my life, but it should all be over soon.

[KC_MCDOHL]

Oh wow, good luck with that then. Thats is, both your degree and your other projects. Cant wait to see what you come up with

[thcrock]

Good luck, and know that when you do get back to working on this I'll be watching closely. I'm actually doing an engine + game (not Suikoden-inspired, completely unrelated...tower defense-ish, actually) in C++ and SDL right now as well. Not putting nearly as much thought into the engine, though, just trying to get to the part where I make the game

[Tsubaki]

hey i dont mean to sound rude but i was wondering what experience you have? this is obviously a big job and hard too, so how come your doing this project and not an original RPG?

Im just a bit sceptical about people online sayin this sort of thing you get lots of crazys saying they like games so why not make one then all the story is lame and rips off other games..
I am in the process of writing an original RPG but its hard to find others online to help out that aint mental..

[Xelinis]

I'm nearly finished with my Bachelors in Computer Science from RIT, have 10+ years of experience with C and C++, have completed five (soon to be six) internships with all but one being at Fortune 500 companies (whose names I should not mention), and have academic experience working with computer graphics techniques as well as animation algorithms utilizing OpenGL. I'm a strict practitioner of MVC from the Cocoa perspective and I have a reputation amongst my peers for being OCD about code cleanliness (despite my unfortunate tendency to put documentation off till the last minute).

As for why I'm doing this, I've always wanted to learn how to design game engines, but my academic schedule unfortunately left me no room to take those classes. As my degree is upon me, I've figured that the only way for me to learn how to design a game engine is to just do one in my spare time. I've always admired the little details in Suikoden II's game engine so I decided to base my design around that.

Man, I feel like I just did a job interview. Anymore questions bridgekeeper?