One other major new feature was "synchronized" melee combat. As shown in the pic on the right, I created code which would match the fighters' movements to each other in a flexible way (basically using the anim names to do this, but with a level of randomness to take advantage of the large amount of anims that Tommy had done). This resulted in  engaging combat that was fun to just watch. A "grappling" feature added even further variety. This carried over to flying melee as well, as shown on the right. Yes, that feature long-wished-for by fans of the FF games was working in my game. With this, resistances had not yet been implemented, but otherwise melee combat was largely complete. Additional features like stun and knockback would be added in future updates, but I wanted to get cracking on ranged combat as well as creating object and map editors. The Cat's out of the bag ... As I mentioned in the last Dev Diary, it was around this time in 2010, that news of the project slipped out on the FR Forums. I personally had not been active on FR in a number of years, but Tommy did still post there and he shared some screenshots. I really had no problem with this, as it wasn't like I was actively keeping a secret, it was more that it was just something I was working on and hadn't even thought if others would find it interesting, especially if it never got finished! Well, it turned out there was plenty of interest and lots of good questions were asked. Tommy and I did our best to answer what we could. The publicity was actually kind of nice, after having worked on it for so long without much feedback other than a few people. I started these Dev Diaries to share a bit about the history of the project and the current status.

Even more monumental (well it was to me, anyway) - I actually got to meet Tommy face-to-face in person (3D even!) for the first time. It was a blast to finally get to speak to someone I'd only known over the net for years. We had just a few hours, but spent a great evening chatting about ideas for the game and where we needed to go next. With things out in the open, we agreed that we needed to perhaps start introducing more of our original creations into the game, so as to not be seen as infringing on the Big 2's (DC & Marvel) stuff. We had been testing with familiar characters as it was fun, but this was always intended to be an original game with original characters. Tommy showed some of his on the FR Forums: Sgt. Shield, Green Goddess, and Granite Girl. I quickly whipped up "work-in-progress" skins for Hyperwave and Pulsar (see pics on left), some characters I'd created years ago but never used. Backstories for them? Well, I have a few things, but we haven't gotten to the big storyline portion of this project yet, so I'll leave that for another day. Our good friend, Lightning Man, also gave us permission to put his original characters in there somewhere and he provided a nice splash screen. But then ... Break #2!!! Work continued steadily - Tommy was optimizing the use of textures on his meshes and continually tweaking the many animations. I added a simple animation viewer so he could test. In addition, I was doing a lot of code cleanup and optimizations. Really boring stuff actually, but necessary for the game to work. My immediate goal was to get into ranged combat (which was going to be tricky) and the object/map editors. Maybe even release a quick demo or character viewer so people could check things out. But then ... it happened again - an unplanned break. It was October 2010, and just like the previous year, all work slowly came to a halt. Again, it wasn't any one thing, but various ones: Tommy started having issues with his hard drive and had to rebuild his PC (it's possible he may have lost some of his work, I'm still too afraid to ask how much). I got really involved with my job and family, as often happens to me at that time of year. Maybe we also were feeling burnout again. We had accomplished even more than during our first stab at this in 2009, but we did not keep going.

Days and weeks turned into months passed and before I knew it, it was 2011. At some point (I don't recall exactly when), the FR forums got hacked into oblivion and so we had even less reason to keep at it. The Panda3D forums also got kinda quiet and I lost interest in reading them for new improvements and bits of information. It seemed that there was no real interest from anyone anymore. I honestly did not think about the game for a long while. Not until Nov 2011, well over a year later. A new hope? A few things happened around Oct-Nov 2011 that pulled me back into the project. First of all, I completed an important project at my job that I had been responsible for and working on since 2009. Completing this gave me not only more free time, but a general feeling of "wow, I have my life back". Secondly, DCU Online went free-to-play. I had been interested in this game since I heard about it (even got on the beta), but could not justify the expense or time to play it. With my work project done and no money barrier, I dove in.

Turns out I really enjoy DCU Online. I've played other superhero MMORPGs, but this one just worked for me (doesn't hurt that I'm a DC fanboy, probably). I any case, as I played, I started noticing little things that reminded me of things I'd been working on for my game. The way that you pick a combat-type in DCU (brawler, martial artist) was very much like what I had in mind for my own game. The character customization was something that Tommy and I had discussed a bit and seeing (and liking) DCU's made me feel like getting mine completed. Finally, I discovered BiL (BeardedInLair's) posts at the FR forums on his experiments with Unity3D. His excellent stuff inspired me to take a look at what I had done again. (and I also realized that the FR forums were back up again thanks to BiL!) I actually started looking at Unity. It seems like a fantastic tool, but, at least right now, I can't get the hang of it! I read a post from someone who said "Unity provides a really amazing art pipeline, and an adequate programming API. Panda provides the opposite: a really amazing programming API, and an adequate art pipeline. So, whichever one of them looks better to you will depend on whether you're looking at it from an artist's point of view, or from a programmer's ." At the moment, I tend to agree. And I'm a programmer not an artist, so I started getting my Panda code running again. SuperSquad Rebirth It had been over a year since I'd even looked at this stuff. So I started slowly - installed the latest Panda SDK and changed my game font to an opensource one. Looked good! I started a reorg of the Character Editor (see at left) and started squashing some coding bugs I'd left a year ago. I'd left off as version v0.3.1, so I started a new one, v0.3.2. In just the couple of weeks I've been back at this (which included Christmas break), I have made the latest version a lot more stable. Mainly by adding more classes to my code (boring techie stuff, but it's made a difference). I've also started programming the game UI in a lot more detail. Inspired by DCU, but even more by Warcraft III and World of Warcraft. The latest UI can be see in the pic at the left. I am glad to see that the Panda forums seems active again and a major new SDK version (1.8) is just around the corner. My immediate plan is to stabilize the UI by adding action butons (I know all the power keys, but no one else would) and after that maybe even releasing a pre-pre-pre-alpha version for people to play around with. Stay tuned ... !

I started spending more time rediscovering the guitar and the Beatles' genius and less time coding Super Squad. I started jamming with friends and even with people on the internet. I started writing a song a day. Honestly, I think I was also feeling burnt out on the Panda and coding. I needed a break. Interestingly, Tommyman was also feeling the same , I think. Without really saying anything we just stopped doing as much. Even though we still chatted about other topics, the game was not mentioned as much anymore. Until Jan 5, 2010 ... That day,  posted that I had learned of a new update to Panda. Also the Panda site itself had been transformed to be more user-friendly and the documentation had been updated (to make more sense IMO). I continued to look and learn with some trepidation, as I was insanely busy at work (and still doing music).

Finally, in early March I took the plunge again and by Apr 8, 2010, I had coded a title screen and had the beginnings of a character editor. (well ... at least I could show a list of hero files from a folder, pick one, and have the model pop up). I worked to add a screen with all the fields in the hero file to the editor , so you could .... y'know ... edit them. I also reminded the few people who knew about this that my target date for finishing this game was still 2012. (Hmmm... that's kinda close now) By the end of April, the editor was working, along with Resistances, the ability to save, copy, and delete characters (all of this was trickier than it sounds). To do these screens, I was learning the Panda 2D GUI interface objects - something called DirectGUI. Which is a massive pain in the ass. Really. I love so much about Panda but this specific part of it is hard to work with - it uses a different scale than 3d objects, there are very few built in objects, you have to construct things out of parts.For example, in order to have names pop up that I could click in a scrolling list. I had to write a routine to add button objects for the names. And routines to scroll a page at a time, etc. Not fun! I tested it with up to about 2,000 hero files. It got really slow after that. Made a note to optimize it more in the future. But it got done!

A note on game design: I had mentioned this previously, but I designed it so you have direct control over character Resistances. This has several advantages over FF (IMO): - You can customize defense very specifically (more combinations than FF) - no wasting attribute slots (I call them "traits") to bump up your cold defense, etc. You just set it here exactly how you want it. - No passive defenses at all, freeing up power slots. - "0" is standard. You can go 5 above and 5 below to choose whether you're very invulnerable or vulnerable to a specific attack. (Again more levels than what FF offered) A reinvigorated Tommy started producing tons of art - new meshes, improved meshes, new animations, skins, and objects. By August 2010, I had gotten back to the main part of the code and had greatly improved the AI as well as cleaning up my code by removing unneeded functions and cutting down its size.

I added the feature that made me "famous" in the FF world in the first place (and a personal favorite) - "watch-mode" by pressing Ctrl-W, All the characters would now just go at it until one side (heroes or villains) were completely defeated. All melee punching of course, nothing fancy. But all the characters who could  - were flying and running faster than others (remember all their hero files had stats which the game was using for speed, strength, etc. - only resistance was not taken into account yet) So it was a battle to behold! I had no limits on squad sizes (still don't) so I was watching 8 vs 8 or 15 vs 1, whatever I would set my map file to. And Tommy kept cranking out more and more meshes. Every time he did, a new warrior joined the fray. At this point, I added the capability for characters to grow, shrink, and turn invisible. In the pic at the left, you can see examples of this. All these meshes are the same size (and visibility) normally. I simply wrote code to have them grow, shrink, or go transparent when the player clicked the right button (And if the hero file had the right trait in it).

Tommy was innovating plenty himself. He took a break from the characters and whipped up some buildings. By editing the simple map text files to place the buildings, he created a quick city that made the battle scene even cooler. Tommy also figured out the lines of Panda code that could make the game go into slo-mo mode. Perfect for capturing some great superhero action shots with the screenshot function I added. At this point, I had a rough roadmap written down. All the goodies to date comprised version 0.3 of SuperSquad. Version 0.4 would add: - Basic (very basic) projectile/beam combat - Clean up the hero editor a bit (more pulldowns, power editor) - Add a "Set up city battle" screen - you pick your squad , the foes and then you fight. (rather than having to manually edit the map file - Add energy costs to powers and replenishing

Version 0.5 would add: - Bug fixes (we had plenty of those) - Better AI - better tactics (use power properly, AI editor) - Shield powers - Keep implementing traits (had fly, grow, shrink, invisible, and fast healing so far) Story modes and missions would follow after this. This Dev Diary has gone on for quite a bit, so I'm gonna try to wrap it up and promise to do another one before too long. Yes, I will not wait another year!! One last feature to mention.  On Aug 30, 2010, I was able to add dynamic speed trails. As you can see in the pic on the right, a character running with this feature turned on leaves a "trail" behind him , much like what you see in comics. This is done completely with code I put together, There is nothing special about the mesh, no special art file required. You simply give a character this trait and they do this when they run. Coincidentally, this was around the same time that our little project was "outed" on the FR forums. Suddenly, everyone knew what Tommy and I had been doing. Is that good or bad? Time will tell. And with that, I bid you adieu. Until next time ....

January 14, 2011

To get the new year going in the right way, KennX is back with a dazzling array of new stuff. A big 25 new custom meshes with skins for characters you never thought you'd see! Been waiting for Wonder Woman in a wetsuit? Kenn's got you covered. Need some Milestone characters? Kenn's your man. Been working on that Crest toothpaste vs Snow Men mod and waiting for someone to make Crest toothpaste meshes with toothbrushes? Kenn heard ya! (really!) Enjoy!

September 28, 2010

A cool dozen for y'all today! KennX unveils his latest ... Enjoy!!!
        
        
        

September 20, 2010

SuperSquad: Developer Diary #3 Getting Somewhere!

In the last dev diary, SuperSquad had been setup with 8 different meshes and it was possible to move them around with WASD controls. Tons of enhancements followed over the next few weeks: Hero files Added a whole subsystem to read "hero" files from a specific folder and use the information in them to know what mesh to load, etc. SuperSquad's hero files are at the moment, simple CSV files (delimited by ";" rather than ","). Still adding to this, but here's where it is: NAME - character name MODEL - 3d model (or "mesh") used SKIN - skin folder used ANIMS - animations folder used (more on this later) STRENGTH - to be used to calc melee damage and other things SPEED - to be used to calc char movement AGILITY - to be used to calc % to hit for melee (offense/defense) ACCURACY - to be used to calc % to hit for ranged ENDURANCE - to be used to calc Hit points ENERGY - to be used to calc Energy points TRAITS - a list of traits a character has (ex: "Flight", "Shrinking") RESISTANCES - a list of 9 resistances vs attacks (Acid, Energy, Heat, Mystic, etc.) POWER1 - Still being defined NOTES - Freeform comments about your character (ex: "Capt. Hero was born in Alaska...")

As you can see, the hero files are quite familiar but not quite the same as FF's. Major differences include: - Energy rating controls amount of Energy, not energy recharge rate - Resistance ratings allow you to directly tweak how a char responds to different attacks like Heat and Cold. No need to use Attributes or Materials to set this up - Traits are like Attributes in FF , but a) you can have 10 instead of 5 and b) there are no Traits to control resistances to attacks as those are all covered elsewhere. - There are other differences. Camera controls Panda's default 3d camera control is, IMO, not usable for a RTS-style game. So code was written to make it closer to what you would expect. To be honest, I'm still not 100% happy with it, but it's getting there. It's patterned after FF, since I am so used to that, but with a few differences. Will continue tweaking this.

Collisions
Panda supports collision handling of 3d objects (which is good - otherwise it would be quite the pain to write a game like this). Added a little code at each character's creation, so they would no longer walk through each other. Instead, they push each other around.

AI
It was now around April 18, 2009 and I decided to take the plunge into AI. The first time I had ever done anything in Python was way back in 2002 when I wrote a little mod for FF called EZ Danger Room. Basically, it made characters chase each other and make decisions like should I fly or not? I actually dug up my old EZ Danger Room code to look at how I had done it and believe it or not, I was able to adapt it to SuperSquad without too much of a hassle. (Did have to learn about Panda intervals first).

Suddenly, things were in motion for real as the enemy characters started running after your squad. A few more revisions later (we were up to SuperSquad 0.1.8 at this point), and the foes were intelligently deciding if they should fly (taking off  and landing on their own). One simple change later and suddenly your squad members could do everything the foes were doing. I'm making it sound easy, but there was plenty of hair-pulling to solve glitches and bugs (rotation issues, animations, etc) Clicking Characters One simple thing you're gonna need to do for a game like this is to click stuff - click where you want to go, click someone to fight, click an object to pick up, etc. Panda fully supports this, but to do so, you start getting deep into the collision system. There's a whole section of the Panda manual on collision handlers that I'm only now starting to fully understand, but by version 0.2, I had added the ability to select a character by clicking them.

Melee By May 2, 2009, I was sick of the characters chasing each just to say "hi", so I added basic melee. You could now punch foes by clicking on them and eventually even knock them out. Plenty of bugs, though: - after 1 punch, couldn't seem to switch back to idle mode - have to fix the combat text , right now it keeps writing on top of itself until it's unreadable - you could attack a foe just by clicking them, it doesn't check if you're near enough or for collisions - you can cheat by clicking the mouse button repeatedly (without waiting for the anim to complete) and it still counts as an attack It did use STR to calculate damage and you and your foe's AGL to calc % chance to hit and subtracts dmg from foe's HP correctly. So it was a major step forward. This is one of the times (it's happened a few times before and after as well), that Tommy famously remarked, "it's starting to feel like an actual game!"

When I code, I really make an effort to use as little code as possible (less code=less bugs is an old programmer's mantra). Also, always trying to write code so it can be reused. So I was very satisfied when just adding 1 line of code suddenly made the foes capable of fighting back! Version 0.2.4 added basic flying melee. Tommy is hard at work as I write this to add some new anims to this, but the basic logic is fully functional. BTW - our test scene, even now, is JLA vs Avengers as those are the sample meshes and skins that Tommy made. Maps At this point (May 23, 2009), also added the concept of "map" files. These are plain text files that are read by a function in my code to create the scene. Obviously, they will be created by a still-to-be-made map editor eventually.

For now, I just created a quick map in notepad:
#TERRAIN;name;model;pos;hpr;scale
#OBJECT;name;model;pos;hpr;scale
#SKYBOX;name;model;pos;hpr;scale
#LIGHT;name;type;color;pos;hpr
#HERO;name;model;pos;hpr;scale
#FOE;name;model;pos;hpr;scale
T;map;world;(0,0,0);(0,0,0);(1,1,1)
S;skybox;skybox;(0,0,0);(0,0,0);(800,1600,1000)
O;dustbin;dustbin;(30,20,0);(0,0,0);(1,1,1)
O;dustbin_lid;dustbin_lid;(30,20,13);(0,0,0);(1,1,1)
O;lamp_post;lamp_post;(20,20,0);(0,0,0);(1,1,1)
O;pave;paving_slab1;(0,0,0);(0,0,0);(1,1,1)
L;light_1;ambient;(0.3,0.3,0.3,1);(0,0,0);(0,0,0)
L;light_2;directional;(0.74,0.7,0.7,1);(0,0,0);(0.7,0.2,-0.2)
H;superman;superman;(100,0,0);(0,0,0);(1,1,1)
H;batman;batman;(75,0,0);(0,0,0);(1,1,1)
H;wonder woman;wonder woman;(50,0,0);(0,0,0);(1,1,1)
H;green lantern;green lantern;(25,0,0);(0,0,0);(1,1,1)
H;flash;flash;(0,0,0);(0,0,0);(1,1,1)
F;naked guy;male_basic;(100,-400,0);(0,0,0);(1,1,1)
F;hulk;hulk;(75,-400,0);(0,0,0);(1,1,1)
F;iron man;ironman_21;(50,-400,0);(0,0,0);(1,1,1)
F;captain america;captain america;(25,-400,0);(0,0,0);(1,1,1)
F;hawkeye;hawkeye;(0,-400,0);(0,0,0);(1,1,1)
F;thing;thing;(-25,-400,0);(0,0,0);(1,1,1)

Movement Switched to point-and-click movement like a traditional RTS, which required more collision system code. WASD controls are still in there, though. Pausing Yeah, RTS games can be hard, so of course, a Pause feature had to be added. Not hard to do, but like much in Panda, you have to code it yourself.  Remember, it's an SDK , not a "game creator". Skins We definitely wanted for skins to work like they do in FF, where you can choose a skin folder for a model to use. This turned out to be quite tricky in Panda as models have multiple image files mapped to them, etc. Eventually, got some code working that did the trick and this is now called when loading your character, to set the skin to the folder specified in the hero file. At the right, is an early test pic of the Supes model using Cap's skin. Doesn't look bad, actually...

Art Throughout this, Tommy was creating tons of new animations (remember we wanted everything in the game to be original) and adding an occasional character to the roster. He also did lots of great research into Physics and LOD (level of detail). Things we may implement later. Other stuff After fixing all the melee issues, I added various miscellaneous features, usually just because I had just heard about them on the Panda forums. The picture at the left shows an option I added to show the scene in wireframe mode. I also added options to remove all textures from the scene, and take screenshots. But Then.... It was now June 9, 2009, I had started to look at adding title and menu screens (no idea how to do it :) ). But then ... all work on SuperSquad stopped. Why? stay tuned ...

September 3, 2010

SuperSquad: Developer Diary #2 Fast and Furious!

Welcome back ... at the end of the last installment, Tommyman had discovered plutonium ... that is, he'd successfully converted one of his meshes to Panda format. A flurry of activity occurred over the next several days as we poured over the Panda documentation and learned all we could. Tommyman converted more meshes and started adding animations. He gushed over the capabilities of the Panda engine - bump maps, blending animations, half-body animations, sharing animations among meshes. All things that we knew FF couldn't do, but could be achieved with Panda. I started looking at how to code this thing. The Panda documentation is all online and I just started working through the introductory tutorial. In 2 minutes and using only Notepad, I had a panda walking across the screen. Something I hadn't done in 2 years with other engines.

Tommy threw his Hulk and Thing meshes into one of the Panda sample programs. It was great fun discovering how easily we were seeing amazing things on screen. I played around with the code that controlled camera movement to make it work more like FF. Tommy and I both set out to start building our dream superhero games, agreeing to share art assets (which was very nice of him as I had none :) ), and help learn the coding side (ah, I can help there ... :) ) As I think he's stated elsewhere, his was more action-oriented, i.e. Marvel Ultimate Alliance or even fighting-game like. Mine was Super Squad, more like Freedom Force, RTS-style. Remember, I had designed all this in my head as a turn-based game years ago. FF just made me see how a real-time version could work.

The Coming of Super Squad
On April 5, 2009, I coded the initial version of SuperSquad. And here it is:

FILE- MAIN.PY

from supersquad import *

loadMap("world")
hulk = loadChar("hulk", (0,0,0), .5,.5,.5)
thing = loadChar("thing", (5,5,5), .5,.5,.5)
run()

This code also requires a supersquad.py file which I did:

FILE- SUPERSQUAD.PY

import direct.directbase.DirectStart
from direct.showbase.ShowBaseGlobal import *
from direct.interval.IntervalGlobal import *
from direct.gui.DirectGui import *
from direct.showbase import DirectObject
from direct.actor import Actor
from direct.task import Task
from pandac.PandaModules import *
from direct.showbase.PythonUtil import *
import random, sys, os, math, glob

def loadMap(mesh, pos=(0,0,0), hpr=(0,0,0), scale=(1.00,1.00,1.00)):
print "Loading model: " + mesh
map = loader.loadModel("art/objects/" + mesh + "/" + mesh)
map.setPos(pos)
map.setHpr(hpr)
map.setScale(scale)
map.reparentTo(render)
return map

def loadChar(hero, pos= (0,0,0),h=0,p=0, r=0, scale=(1.00,1.00,1.00)):
char = Actor.Actor("art/characters/" + hero + "/" + hero)
char.setPos(pos)
char.setHpr(h,p,r)
char.setScale(scale)
char.reparentTo(render)
return char

My thought was to write a library of functions (supersquad.py) that "issue" files would call.

I'm kinda embarrased to post this as it doesn't do much - just makes 2 characters appear on a flat map. Plus, it is quite sloppy code (too many imports, too many parameters as I would learn). Still it showed me that I could start getting the hang of this. I quickly added loading of animation files and a moveTo function to get these dudes moving around.

Around this time, Tommy went Panda-exclusive (stopped doing FF stuff for a while) as he was learning how to best export meshes and creating all-new geometry, animations and skins. We agreed that we wanted everything in the game to be original, that is created for the game itself. We didn't know if there would ever be a finished game, of course, but we wanted it to be something we had done. That being said, moddability, was (and still is) high on the list of things to deliver. We want people to do stuff with it in the future.

I quickly did a new version of Supersquad that read little text files (i.e. "hero files") from a folder and used the information in them (name, mesh, skin, stats) to know how to load a character and their animations. I added WASD controls to walk around, put up little portraits (can't click em, but they look cool), and you could press the number keys 1-8 to choose which squad member to control. Yes 1-8! Unlike FF, I wanted there to be the possibility of having big squads. A 7-man Justice League needed to be possible, dammit! You can see the fateful initial screenshot of the original lineup to the right! As you can see, Tommy had converted more meshes and animations. And it was now April 11, 2009. The first week of Panda excitement had come to an end. TO BE CONTINUED ...

August 30, 2010

Since the personal project I've been working on was revealed recently on the ol' freedomreborn.net forums - I decided to post this to share a little more information about the game development work I've embarked on (embarked on quite a while back, but more on that in a bit).

With this one major step forward - we were off and running ....

TO BE CONTINUED ....

April 1, 2010

It's come to my attention that there are rumors of an April 1 update here at the Fortress. Y'know ... like the one we did last year.

Yeah, that was a good one ... but we've all had a busy year and haven't gotten around to Freedom Force stuff recently.

So sorry... no update this time around. Happy Easter, though.

Proud winner of 2 Golden Skinny Awards
 

ARCHIVED UPDATES
2009     2007     2005     2004     2003     2002