Tell me how you did it, and why

#1 I started my "torque-iness" with TGB (T2D back then) during EA.
The docs were mid at that date, TDN still not present and the tutorials outdated most of the time ^^

So I mainly toyed around with it to get used to torque script and some of its specific behavior and the way it interacts with torque in general.

So the way I started out with TGB is definitely not beginner friendly, but I've been using Dark Basic (and Pro) as well as Blitz 3D / Blitz Max for around years now, so I know the basic programming stuff and restrictions of 2D in 3D.

I haven't produced anything playable yet. Part due to missing media part due to missing motivation to work even more after 80-90 hour study weeks.
But I've 2 toying projects (one is angelic layer which is fully mouse controlled) and the other is an AI and genetics sandbox RTS I'm developing in the hope that I can do some semester thesis on AI basing on it. (and naturally I do it because I love RTS but not the way most of them work with their mouse skill instead of tactics and intelligent units that do not love to commit suicide like in Westwood / Blizzard games)

#2 I started with TGB 1.0x or something like that. I'd already done programming in lots of different languages (including scripting), so learning Torquescript was painless.

If you don't have any programming experience at all, it's probably better to start with Java or something, though you could pick up programming lessons from the TGB tutorials and examples if you're commited. Manual pattern matching isn't that bad ;)

I've done a whole bunch of prototypes/small games and found that if you make a whole bunch of random stuff, you'll get a hang of TGB very quickly. Just reading through TGB Reference is enough to give me a bunch of game ideas now ;)

#3 I started when I came in to intern at GG. I had met Jay at GDC and he invited me to come down. They just kinda thew it at me and told me to make a simple car physics demo. I went through the two fish tutorials and then jumped right into making the demo. The next day it was pretty much done.

To be fair, I've had experience with several different programming languages over the past eight years or so and I've worked with a couple other engines, so the transition to TGB was extremely easy. Even with the state it was in back then, it was a dream to work with TGB compared to Unreal Engine 2, Panda 3D, and Ogre.

The best thing you can do to facilitate learning to use TGB is brush up on your programming skills and game tech knowlege. It's not neccesary to know how all the stuff works behind the scenes, but it helps a lot. It's always good to be up on general "best practices", too.

#4 I started with TGB after struggling with other free engines (such as HGE, PTK, Clanlib).. and two years work on my own 2D engine. I finally caved in and realized I was slaving over a game engine when I could be working on games. Lucky for me T2D was in Early Adopter. After trying the free engines and a few commercial ones, I decided T2D was the way to go. I've been here ever since. I'm a self taught programmer, and I use mostly C and Lua.

#5 I started with TGE 1.x back in the day, and was a bit overwhelmed. So I thought I'd try a 2D game instead and grabbed TGB! It's been fun.. I have a day job as a senior multimedia engineer for a company that makes cell phones (heh) so I can't put as much into it as I'd like, but what I've done so far is a blast!

I've written a couple of game engines in the distant past, and frankly I'd rather work on gameplay logic than the engine at this point.. so TGE/TGB is perfect for my needs.

I have to get a few more goodies (tools, mainly) but TGB has everything I need for the simple games I'm trying to make. :-)

#6 I started with TGB in late August of 2005 -- a friend mentioned that he was working on something using TGE and it peaked my interest, when I went to the GG site I saw them announcing TGB -- I immediately jumped on that and said "I gotta have it".

I have been programming since I was 12 (now 27), but have never touched or even seen TorqueScript before. My only other experience with developing "graphical" (OpenGL/DirectX/VGA) applications was with Python and PyGame (I created an HTPC App called "pyHTPC"), and some ModeX Stuff I did way back in the early 90's when I first started experimenting with programming.

I picked up on TorqueScript fairly quickly, had a few issues with TGB's Level Builder at first though, then came 1.1.2 and then 1.1.3 and now I'm sitting on a gold mine of useful tools ... ;)

I've prototyped quite a few different things, have not really "Completed" a single thing yet (though, I'm a hobby coder, so its ok for me to jump around and not finish things ;p).

I tried some of the demo tutorials, didn't really find them useful (I'm not much of a tutorial learner), I looked at the example scripts, I read some forum posts, I did some code surfing through the engines C++ when I licensed the engine ... and now I feel fairly confident I can do most everything I'd like to do with TGB and have just now started working on actually completing something ... what, I'm not sure yet ... ;)

#7 I started with TGB 1.0 (I missed the early adopter price by a week >_<) and, first, I got through the fish tutorial first, then the shooter tutorial, then the old plateformer tutorial (with some things to bypass or to rethink by myself), and finally begun my main project.

The main tutos (the one you can access by selecting help > documentation in TGB editor), are pretty much all you'll need to begin learning how to make a game using TGB. They are really outstandingly well made and simple to understand. Maybe it wasn't the case back in the Early Adopter days, but right now they're a great start point, even for people like me who weren't coders at all before.

#8 I started with TGE (1.2 I believe) in Feb of 2004. I had checked it out for a couple months (scripting Realm Wars) before that. I used it for a couple months and then dropped it. Roughly 4-6 months after that I picked it up again and dug into it and forced myself to learn it. It was very painful, searching the old forums, very minimal TGE docs (a lot of it outdated) resource scavenging until I went mad lol. In the end I learned TGE fairly well.

I started with TGB on the first day ("stealth release" as Josh W. called it) of TGB Early Adopter. Back when there was the single shooter tutorial, that's it, no other documentation whatsoever. I learned it quickly (my knowledge of TorqueScript and TGE definitely helped) and since Melv M. and Josh W. were the only devs back then I started to try and answer people's questions on the boards (with only two of them deving they were crazy busy, especially while in EA, so wanted to help out). In the end figuring out the answers to peoples questions was the best way to learn the engine, especially learning many aspects of it (some I wouldn't naturally learn by just working on my own). I ended up compiling the most frequent questions into a "TGB Unofficial F.A.Q.". With this TGB now had two docs lol. After that I started to pump out some of my tutorials (the first one was about 88 pages and covered TGB, the Gui Editor, and some basic TorqueScript concepts)... keep in mind this was all waaay before the LevelBuilder existed. You could only script your game back then.

I ended up making about 10 full tutorials and about 1,000 - 2,000 posts in the TGB forums during that time... I did some minor TGB contract work, graduated, and then came to intern here at GG (which was a blast btw, I highly recommend interning to anyone). I worked on an RTS tutorial (which is now horribly outdated on TDN) for the first couple weeks, then slowly got a chance to work into TGB dev, Adam Larson and I both were interns doing some dev (yeah interns at GG get a chance to do real work too, along with some intern tasks of course). Josh W. was Project Manager, Justin D. (now Project Manager since Josh moved up to CEO of GG) was Lead Programmer, and Adam and I were devs. This is when we started thinking up what has now become the Level Builder... lol it has come quite a ways (I still shudder when I see old alpha and beta pics of it). I also had a few days to crank out some tutorials.... I did the Fish Demo and Fish Game tutorial in 2-3 days... created the Checkers Demo in 2 days and created the tutorial in 1 day after that (I'm still waiting to get a chance to completely revamp the tutorial.... soon).

Ended up getting hired and working more on TGB dev, though as the engine matured we have started to put a lot more focus on the docs, which became my responsibility. My official title now is "Lead Documentation Engineer" (fancy name for me being responsible for TGB docs) and now I have been working more on the framework for the docs, efficiency tools for our creation of docs (a Doc Builder) and more recently doxygen integration with TGB to fully generate the reference (no more out of date reference, well at least a much more up to date reference :).

I still love taking time and just playing with TGB. Since being hired the only personal project I've had time to do in TGB is "Fire and Healing" (that simple networking proof of concept)... though I absolutely love the direction the engine has taken. I also am very passionate about uping our level of documentation. I'm trying to take good and useful concepts from other types of doc releases for other tools while not stifling my own creativity and innovation (such as the method and glossary hot-linking in the last set of docs). So hopefully you will see not only innovation in our engine (providing an easy to use platform that doesn't limit your power), but also see innovation in how we share our knowledge with you.

#9 ADD: to answer your last question I would say the way I started wasn't the best... though mainly due to TGB not being around when I started TGE. I would highly recommend starting with TGB, moving up to TGE (if you want to make 3D games), then moving to TGEA (if you want to make more advanced 3D games).

#10 I got started with TGB because I have a TGE licence and getting started with TGE was a bit overwhelming at times also I did not always have the time to dedicate to reading and experimenting, a few weeks might go by and I forgot everything I had done. I have the 3d All in one book and that is what I was going by and of course the resources here, before this I had no experience with Tourque Script, but when it comes to languages I do have you normal pick of the mill web languages, html, php, css, cdml, java and also in college I learned C++ Visual Basic and COBOL (COBOL was dropped from the school's requirements the next year after I took it X-P ). So far working with TGB and torque script I feel that it is a lot like working with PHP which I enjoy a lot more than C++

TGB came along and it seemed like it had the potential for more instant gratification and rapid development., the only resources I have for TGB are what comes with it, what I can get and ask for from the forums and TDN. At first I just played around with it and would put it up, then play around with it some more. The goofing around I did was stuff just to see if I could do it, like mount a thruster on a plane and have it turn on only when the thrust button was pushed and turn off when the key was let up, I also played with the side scrolling game, I did the tutorial then onto experimenting to see what else I could do.
This went on from when I first got TGB from the early adopter program up until the Adventure kit came out, then I had some ideas for a few games and their AI seemed to fit the bill for what I wanted. To date I am still trying to extract the AI from their pack so I can adapt it to my use, however I have been working on a real game, the single player has not been going so well because a lack of AI, but I have been keeping busy with other parts of it, such as menu development, hit and collision response and for fun I decided to create a little two player part to go with the game that kind of reminds me of the tank game from Combat on the Atari

I can not say for sure if this is a good way to start out, but for me short of learning it in a school/college environment I think it is the best way for me.