Character Generator Software

[thepugilist]

I know this has be brought up in the past but I thought I'd practice some thread necromancy and bring it back to the forefront.

With the style of character generation, it seems like it would be an ideal time saver to have a simple "click and choose" character generator at a nominal fee, perhaps with low-cost expansions that add datasets from new expansions.

Is anything in the works for this or for the other common processes (star, world, ship gen, etc.)?

Thanks!

 

[hhawk]

I know this has be brought up in the past but I thought I'd practice some thread necromancy and bring it back to the forefront.

With the style of character generation, it seems like it would be an ideal time saver to have a simple "click and choose" character generator at a nominal fee, perhaps with low-cost expansions that add datasets from new expansions.

A character generation, UI would be tedious to write, and would be difficult to make flexiable enough to be useful.. Besides the process is meant to be used be a group of people to get right. A NPC generator would be a more reasonable request.

Is anything in the works for this or for the other common processes (star, world, ship gen, etc.)?

Thanks!

The easiest utilities to write would world generation, encounter generation, trade qualities. The type of programs that perform chart lookups quickly and efficiently. I might write a trade items generator for my campaign to because my player are going to want to do a lot of trading.

 

[GypsyComet]

I know this has be brought up in the past but I thought I'd practice some thread necromancy and bring it back to the forefront.

With the style of character generation, it seems like it would be an ideal time saver to have a simple "click and choose" character generator at a nominal fee, perhaps with low-cost expansions that add datasets from new expansions.

A character generation, UI would be tedious to write, and would be difficult to make flexiable enough to be useful.. Besides the process is meant to be used be a group of people to get right. A NPC generator would be a more reasonable request.

Mechanically, a Traveller character generator is pretty simple to write, even with all the ins and outs of MGT character generation. The point about character generation as a group is far more germane.

 

[hhawk]

I know this has be brought up in the past but I thought I'd practice some thread necromancy and bring it back to the forefront.

With the style of character generation, it seems like it would be an ideal time saver to have a simple "click and choose" character generator at a nominal fee, perhaps with low-cost expansions that add datasets from new expansions.

A character generation, UI would be tedious to write, and would be difficult to make flexiable enough to be useful.. Besides the process is meant to be used be a group of people to get right. A NPC generator would be a more reasonable request.

Mechanically, a Traveller character generator is pretty simple to write, even with all the ins and outs of MGT character generation. The point about character generation as a group is far more germane.

I was thinking about implementing the event tables, special alien rules, psionics, and other special cases. Each special case is easy to handle individually, but together they become very tedious and complex, and it becomes difficult to make sure that all the character creation paths work.

I remember the MT computer games; they had very nice character generation system, but they were limited in the type of character you could generate.

 

[kristof65]

I was thinking about implementing the event tables, special alien rules, psionics, and other special cases. Each special case is easy to handle individually, but together they become very tedious and complex, and it becomes difficult to make sure that all the character creation paths work.

With today's object oriented programming languages, it still doesn't seem all that hard to write. A Traveller pc-generator actually looks much simpler than a D&D3/3.5 one, even with the sheet number of choices and special events.

I'm not much of a programmer, but I've written a pretty detailed CT npc-generator in AmigaBasic, and more recently a WFRP 2nd Ed generator in VisualBasic for myself. I was thinking that an MgT one would be a good exercise for me to expand on my programming skills and try out one of other Visual Studio languages. If I feel like taking it on, I have no doubt there are programmers out there who could handle it with ease.

I do agree with you that designing a UI for true Player Character generation could be tedious. But it's been my experience that most character generation programs are used by GMs for NPCs, which means you can streamline a lot of the actual UI.

 

[hhawk]

I was thinking about implementing the event tables, special alien rules, psionics, and other special cases. Each special case is easy to handle individually, but together they become very tedious and complex, and it becomes difficult to make sure that all the character creation paths work.

With today's object oriented programming languages, it still doesn't seem all that hard to write. A Traveller pc-generator actually looks much simpler than a D&D3/3.5 one, even with the sheet number of choices and special events.

I agree that a lot of system is very easy to write. They provide the basic algorithm in the rulebook. The main difficulty will be handling the event tables in a standard manner that allows you add new careers and expansions easily without having to know how the entire program works. If you design the exception handling well, the rest of the work will be much easier.

Handling the exceptions within the rule system is what is going to make creating a useful chargen program tedious, error prone, and a lot of work. Whatever paradigm you use handling all the exceptions is going to be the majority of the work.

I have built similar rule based systems like these before. They always seem like less work then are really are. I am not trying to discourage you or anyone from making a chargen program. I am just saying that it will be significant amount of work

 

[GypsyComet]

There used to be a fine tradition of new gamer programmers tackling some utility for their prefered game as an early self-directed project.

Sadly, programming at the commercial level long ago ceased to be a small group or solitary endeavor, and that has rubbed off on the hobby side. Hobby "programmers" are now working within some common productivity package instead of writing executables, if they bother at all.

As such, I will not discourage anyone from trying. As long as you are using a real programming language and not one of these trendy-but-subtly-crippled scripting languages, the logic structures required for MGT character generation are hardly insurmountable.

Of more concern is that the OGL doesn't cover software, so by being a "more friendly" licensing scheme, it also helped squash the game utilities community.

 

[Deniable]

Of more concern is that the OGL doesn't cover software, so by being a "more friendly" licensing scheme, it also helped squash the game utilities community.

There are issues of providing source / access to the rules, but it doesn't forbid software. The TLL doesn't allow software, but the OGL, which covers the contents of the SRD, does. Wizards have provided an FAQ.

 

[hhawk]

Some ideas:

Don't try to make the system "universal" but rather add the individual careers as new modules/classes that link into a common framework

It is making the common framework so that it does not break every time you add a module is the trick.

The framework handles the "basic stuff" like dice-rolling, display, character storage/retrieval etc, the add-on classes do the special stuff hardcoded.

The framework has to handle the basic algorithm of character generation as presented in the book. The basic stuff you are taking about is the supporting library.

Using an interface/abstract class for the implementation should be able to handel it. After all the RESULT of the roles is always the same, it's the rolling and tables that differ. So "public abstract LifepathEvent doLifepathRole()" should work

;-p which is called within all instances of public abstract ChargenEvent doGenerateCharacter() ;-p

As for the group stuff, that only comes in late in CharGen and can be handled by a "assemble/finalise group" screen.

The overall problem is that players want to role their own dice. They want to interact with the group. A chargen program's usefulness is mostly NPC generation.

As a GM, I would not let a player use a program to generate a character. That is as bad as telling them to roll their own character and bring the character to the next game. They might as well use the point system, otherwise all the other player are going to make fun of their overpowered character that they rolled up alone ;-p

 

[kristof65]

It is making the common framework so that it does not break every time you add a module is the trick.

The framework handles the "basic stuff" like dice-rolling, display, character storage/retrieval etc, the add-on classes do the special stuff hardcoded.

The framework has to handle the basic algorithm of character generation as presented in the book. The basic stuff you are taking about is the supporting library.

If you break it down though, even the events/mishaps aren't really doing anything "exceptional" outside of the basic character generation algorithm. Regardless of the text or choices they present, hey do the same basic things - add/subtract to a stat or skill, or give a DM to a later roll or add a benefit or contact.

That is a sharp contrast to D&D's 3.5 edition where Feats are chock full of exceptions in what and how they affect stats and skills.

The overall problem is that players want to role their own dice. They want to interact with the group. A chargen program's usefulness is mostly NPC generation.

As a GM, I would not let a player use a program to generate a character.

As a player, I wouldn't want to use a program to generate my character. I see character generator programs as a GMs tool for NPC creation - which means a lot of the choices that have to be made within the events tables don't ever have to be parsed back to the user for input - which simplifies things.

Overall, I think MgT's Character Generation method lends itself better to a program than other RPGs do.

 

[hhawk]

It is making the common framework so that it does not break every time you add a module is the trick.

The framework handles the "basic stuff" like dice-rolling, display, character storage/retrieval etc, the add-on classes do the special stuff hardcoded.

The framework has to handle the basic algorithm of character generation as presented in the book. The basic stuff you are taking about is the supporting library.

If you break it down though, even the events/mishaps aren't really doing anything "exceptional" outside of the basic character generation algorithm. Regardless of the text or choices they present, hey do the same basic things - add/subtract to a stat or skill, or give a DM to a later roll or add a benefit or contact.

I was mainly thinking of stuff like psionics and naval academy that can throw a wrench into the basic system. I agree though, a lot of it can be systematically automated. The main issue is that you pretty much have to encode all the event tables into the source code so even if you or I wrote the program it is unlikely that we could distribute it.

 

[kristof65]

The main issue is that you pretty much have to encode all the event tables into the source code so even if you or I wrote the program it is unlikely that we could distribute it.

Agreed. I've got a couple of ideas that should allow you to compile the event tables into a fairly "standard" format that should be seperatable from the main program, but then someone still has to either type all the stuff in, or you have to find a way to legally distribute it. With that restriction/issue involved, if I were to write anything, I'm going to go ahead and modify any character generator I create with my house rules anyway, which makes it less than desirable for others, I'd guess.

IMO, the ideal solution is either Mongoose does a character generator, or licenses someone to do it. I wouldn't even object to an online generator with a small usage fee (either monthly subscription, or per character generated cost).

Chargen has too many benefits for Player Character Generation while it is IMHO useless for NPC generation.

Ok, I understand your benefits for using a program as a PC generator, but why do you feel a character generation program is useless for NPC generation?

 

[GypsyComet]

As for the framework:

All that takes is a software engineer that knows it's job. I have designed modular systems that are more complex in my line of work than the basic modular concepts needed here. Granted, not a job for a Private Homepage Project "expert" but any serious developer can do the stuff.

The right education will provide most of the data structures and the tools to build the rest. Where a professional level will come in real handy is in the meta-tool coding, those part of the program that make it readily expandable to suit later releases.

Note that I've encountered some "professionals" who couldn't code their way out of a paper bag, but a guy I knew in High School was experimenting with his own compilers and doing initial development of his own programming language...

 

[hhawk]

The main issue is that you pretty much have to encode all the event tables into the source code so even if you or I wrote the program it is unlikely that we could distribute it.

Agreed. I've got a couple of ideas that should allow you to compile the event tables into a fairly "standard" format that should be seperatable from the main program, but then someone still has to either type all the stuff in, or you have to find a way to legally distribute it.

I have some ideas on how to implement the table too ;-p Use function references in the table to invoke the proper events on the character object. Need to modify the advancement DM for the next advancement roll add a IncAdvDM(4) reference in the table data ;-p

It seems like a good job for python ;-p It just seems like a bunch of typing.

 

[Alric]

It wouldn't be hard to write a character generator for Traveller using ProFantasy's "MetaCreator". Once the development tools for HeroLab are finished it should be able to handle Traveller as well.

 

[Arakor]

It wouldn't be hard to write a character generator for Traveller using ProFantasy's "MetaCreator". Once the development tools for HeroLab are finished it should be able to handle Traveller as well.

Hate to point this out but "MetaCreator" is a product of AlterEgo Software, not ProFantasy who actually publish Campaign Cartographer 3.

 

[AKAramis]

@k65

NPC's are a GM's tools. No need to build them as characters IMHO, just give them some stats/skills that "feel right"

Easiest ay to get them to feel right: use the CGen rules.

 

[Gee4orce]

I could code this - I've thought about it (and if I did, I'd do it in something cross-platform, and open, like Ruby or Perl) - BUT...why ?

Rolling up characters is great fun in Traveller, MGT especially. Why automate something that's fun to do ? I can appreciate the need possibly for an NPC generator, but there are precisely Trillions of randomly generated Traveller characters that are mostly compatible with MGT available, so I don't really even see the need for that.

 

[hhawk]

@k65

NPC's are a GM's tools. No need to build them as characters IMHO, just give them some stats/skills that "feel right"

Easiest ay to get them to feel right: use the CGen rules.

To much work IMHO. Either adapt some from older stuff or wing it is the way to go. Chargen due to it's mostly random nature, is to long winded and boring for that and the "point buy" figleave is a bit to small for my liking.

Nope, IMHO Mongoose should pay someone the 40-50 man days to write the stuff.

What MgT needs is a good system for generating NPC based on archetypes. I want to generate I nice booklet of possible NPC (with random names and organize the lists by archetype. That way when an encounter happens I just flip to the right page and circle the character name and use their stats. You could even organize random encounters this way. It will add better variation to otherwise generic people in an encounter.

 

[kristof65]

Doesn't "760 Whatevers" fill that place?

Nope. It provides brief character/situation descriptions and possible options, as well as suggested or required stats/skills/equipment.

No NPC stats or names.