Making THE game part one. Whats my engine need?

Hi,

Well, it's been a little while (I am sure some of you do not feel it's been long enough but, that's fine.) I want to start this blog by working through a couple of things that you are probably wondering.

1. You removed ALL of my previous posts... Did you loose your mind?

Answer: No, I was never mad, pissed off, or interested in social perceptions. Most of you do not know me as a person. However, when I decide to do something I do it. I don't ask permission and I do not care about 'public' perception. I pulled all my old posts because 99% of the images and info in them was dated and I REALLY needed to redo my full website. See, I released EVERYTHING I have done over the last 2 years as public domain. (Yes, it is available at my site 3TDstudios.com). When I initially released it, I did NOT think of the existing licenses that were included or the text that was on my old site. So, I pulled everything, redid it and re-released it. This stuff is up and available now.

2. You stated you were leaving.

Answer: Yep, and I have. I am NOT developing for T3D and I have no personal interest in seeing T3D live or die. I am a hobby developer that shares his stuff! (thank the gods!). I do not care if T3D, Garage Games or whatever splinter group forms or dies out. (EDIT: Let me explain that last line, I mean my game will not DEPEND on Garage Games or any other team or project. I have been at this site and been with the community for a long long time, I am very happy that Garage Games found a model that works for them and I wish each of the them the best. However, I am interested in making MY game and that means I need to take full responsibility to make it happen.)

I am here to share some of my experiences with you because, well, honestly, I like to share. It's probably the college professor in me. I take pride in making most of what I do available for the next generation.

So, take my experience with this stuff or leave it. It matters not to me and I will not address old business further.

-------------------------------------------------------------------------

Let us begin on this new Project.

Building THE game: Part 1 - the Engine (WTF, I thought that would work!)

Obviously, I am still using Torque 3D. This is an engine that I like to think I know and understand (though if you read further on, I discovered more than a few things I was fooling myself about!) I can make models and add them to the engine with ease, I can script a bit, I can code a bit in C++ and I started out thinking my 'game' would be something I could pop out in a few weeks. Well, it's been a few weeks and WOW, I have learned a TON of new things and I now understand SO much more than I did when I made my initial plans.

THE ENGINE:

So, without revealing anything about my project, (NO it's not the undersea adventure for Rift that I was planning, and it will become clear as to why that project is on the back burner for now). I started out with TONS of experiments. Seriously, I have over 25 different T3D builds on my local hard drives (keep in mind, I have 2+ terrabytes local storage of which 100.2 gigs include T3D builds.) using TONS of features, both paid products and MIT resources that I created and combined to 'test and play' different 'game features'. I think this is probably a common thing for us as developers. We see a 'new' resource or we are interested how a resource will interact with a set of scripts and code we bought (yes, I have bought a number of things.... to include ART packs! GASP!!!) I can honestly say at this point that I am not an expert by any means. Even on the art side of T3D there are things that confound me. This is NOT a bad thing, it just means I have more to learn! Nothing wrong with that and I am happy to admit this.

However, I have 25+ builds and in each build ALL of the features I tested work (for the most part, OBVIOUSLY there are issues and tweaks to be made but, I can and have noted them and in the end I am thinking I have my 'perfect' build..... So, I start.... I add resources, I add scripts, I add in EVERYTHING I need!!! HA HA HA... 'F everyone' I have the perfect build.... but, wait.... WTF???? My build won't compile in visual studio???? Wait it compiles but hangs when loading data blocks! AAARRRGGGGHHHH! (to quote Charlie Brown!).

So, I have all of these builds, I have all the source changes and I have all of the info to make these products and systems work. Guess what??? EPIC FAIL! None of my builds worked on first compile!

Back in the day, I was a regular Torque worker, (I do not mean Torque DEVELOPER because I spent my days incorporating other people's resources into my personal build of TGE. I spent hours and hours and hours making builds of the Torque Game Engine (1.42 and even MORE in 1.5.2). I ran into the SAME FREAKING issue...

NO ONE IS BUILDING MY GAME ENGINE!

As the header says, there is not one developer or contributor (except me) who is building my 'perfect' engine. So, my builds crash, my scripts hang with the dreaded 'loading data blocks', forever error.... etc etc. This happened back in the day and I was really hoping that it was not an issue anymore. I was WRONG. The difference now.... PERSONAL GRIT.

WTF is PERSONAL GRIT?

Grit, in this instance, is the ability identify a problem or error, understand why the error is happening and come up with a way to make the error go away (aka, FIX IT!). Lastly, GRIT is NEVER giving up on the problem! Now, this is NO small task since in T3D I need to ID errors within both C++ and Torque Script or GOD FORBID... both! Lets not also add ART and ART scripting errors (of which there has been MANY!)

SO... WHY DO WE HAVE THESE ISSUES?

Honestly, I sat down one day and thought about this, I came to a simple conclusion, these problems are BOUND to happen. The graphics guys are working on the port to DX whatever or OpenGl version (include number stuff here), the AI and Path finding dudes are working their part as well as the sound guys. The modelers are trying to make their stuff work, etc etc etc. Issue is... none of these guys talk or know what the other is doing. This is where a GAME DEVELOPER comes in!!

SO, GAME DEV... WHAT?

Game developers are the guys that can look at all of these dis-jointed pieces of code and make them work in sync. THIS is our job.

Currently I am able to make these assorted functions work as an Engine:

T3D 3.5.1 (or whatever the current MIT version is...)
Verve port (thanks J0linar... with some 'adaptions by me')
DX9c Refactor - HUGE Pain in the Ass to get to work with all the other stuff but, MORE than worth it. (seriously, if this is NOT a part of the NEXT release, T3D may as well GIVE up. I can ACCURATELY account for nearly a DOUBLE the Frames per-second improvement. (I have SCREEN CAPTURE and Video Capture proof if needed.)

UAISK. Weapon Scripts were hosed. (Not due to TJ's efforts! Just my personal twists...)

Climbing Resource (90% ladders are now OK....)

Dark UI (thanks Nils Eikelenboom!) By the way... WTF I had to fix the readout for the metrics (fps) (gfx) stuff so I could see it :-) Just kidding! Yes, I had to fix it but it was not complicated at all. Good work!

IronSights. With tweeks and new animations... oh and I figured out a crappy hack for render to texture stuff.... for scopes.

Motion Blur 2.0 - running in reverse does not 'f' up the view now. I need to adapt this for object based movement before it's ready for public consumption.

Toon Shader - standard deal, Outline (you can assign the thickness) and poster effects combined in one shader.

Vignette - thanks J0linar. This is still better than my original solution.

TONS of PostFx crap... I am NOT going to name it all here (GI is GONE... IT KILLS THE Frame rate unless you have VERY controlled circumstances, (aka interiors ONLY). REMOVED!

UAISK with RECAST and a custom "cover object" system thanks (those that helped... there are MANY! Love the community!)

TONS of new animations (Thanks Mike! (he is my mo-cap model...More to follow on that.)

DSSDO... Thanks to Lukas for this lead. He did not share any code with me but, I understood where he went with it... based on his assorted posts. (no, I will NOT release anything public... gotta respect my 'commercial' T3D brothers.) I will leave out my WLE gripe for this post.

'Bullet time'... its a cheat but, it works for now and in my limited situations, it works.

So, I started out to make a 'cool' AAA diving adventure game. Well, I can't (not because of the engine, but because of my OWN limits in knowledge and prowess in script and c++ and hey, the stock SWIMMING animations STILL SUCK!) I will get back to it eventually. For now, I am working on another game.... Something I deem 'simpler' and 'easier' but, as par-for the course.... not true. I still have run into a large number of 'incompatible' engine changes and add-ons. So, will I quit? No.... I will figure these things out and make them play nice with each other.

In the end, there have been a GREAT number of improvements and public add-ons that are essential to making a 'modern' game lately (within the last month or so). They may not all play together well initially but, with some 'hacks' and some effort, you can make them all work as an engine. If you have the stick to it attitude and the never give up mindset.

This is the core of engine and game dev tools development. (really, I actually just finished an animations needed document for my 'simple game' it includes DOZENS of new animations that I need for about 12 or 14 characters. (needless to say, I am working on a pipeline for animation capture.)

SO WHAT ARE YOU TRYING TO SAY HERE?

It's pretty simple. Torque is NOT an 'out of the box' answer to your (or my!) needs. It was NEVER intended to be that. However, since we have the source code and are able to adapt the engine to our needs we CAN accomplish nearly anything. Assuming that we have the drive to FORCE the engine to our will. (Hell, two years as a 'game ready' artist and I can tell you this... I have come up with DOZENS of ways to do my first product better than I did and I can even IMPROVE on my LAST commercial product.) THAT is the nature of professional growth, that is the nature of Game Development.

As usual for me. There will be more info to follow. Hopefully, with some new screen shots and perhaps a dev video. The plan is to make this game development process as transparent as possible. I am not trying to make the next COD or WOW killer here. Just trying to make a decent FPS with SOLID game play and some solid game mechanic foundations. After that, who knows... I guess it's up to my imagination and willingness to force Torque (or whatever engine I choose) to do what I want. Alas, that is for all of us to decide. Compromise is NOT failure unless you totally give up! (you can quote me on that)

LASTLY, Hack Versus 'fix'.... at this point.. WTF? Who cares, I will hack a piece of programming till the end of time as long as it works.... On top of that. I HAVE hacked them. No idea if what I did was right, it just works so I move the 'F' on..... I know this seems stupid but, in the end, nobody gives two SHITS about HOW a product was shipped, they just want a shipped product. (YES I can back this statement up). SOOOO, hack away. Make it work. In the end, you can mess around with 'proper' code or you can ship a freaking game. (to me, this is a no brainier).

I hope you got something of value from this post. Like I said, I will continue to post these intermittently. I am also planning to video record some of this and post it. Again, take this or leave it. In game dev, the choice is (how grand!) yours!