Plan for Toby W. Allen

So.. I've been developing with Torque since.. umm January really; what do I currently make of it? Well it's a powerful engine, and very much a cool sort of rapid prototyping engine. However, I find the support and information side of things insanely disorganised and off-putting. So much so that many members from my society, some with no experience and some with a small amount, want to dump Torque.

Why would they want to do that? Well let's look at it from this perspective - You're taking on a new engine, you might or might not know how to go about this, so you start reading information about the engine, documentation and the "tutorials" on the website. You hit a main problem - These reveal barley anything about the engine, it's capabilities or how to actually use them.

Let's take an area that's been a serious problem for our game - the modelling side of things. While modeling static objects works well and easily. Character modelling is completely mad - The documentation is broken (images) and shows no real structure pipeline for anything other than a basic character with no discussion of how to add props (items, guns, armor, lights, fx, etc...). While forum discussions provide easily solutions to some of these issues, none of this information is added to the tutorials or put together into something that resembles a repository of information for this topic. Looking at the recent Valve SDK, when selecting a topic, you see a great list of specific areas covered - all in depth and up to date.

Similarly the GUI definition is not something that's explained on in any great depth, which is why videos such as the smdLabs stuff have been so great. The visuals of the GUI are barely explored, which is crazy, as the GUI editor in itself is a really sweet tool, one of the best and only tools as such, but the actual look of the windows and such isn't barely explained. Why aren't users provided with the photoshop files for the GUI? When you have for example the smdLabs guys saying well essentially you need to test things by breaking the scripting and seeing what it does, this is good and can teach you a lot, but why follow this approach when some basic tutorials covering the information in a good amount of depth will clarify the pipeline for this area.

Taking a look at mapping; without having searched the forums, I wouldn't have been updated to the release of the mapdif exporter. But wait.. I search a little more and now I get various threads with various mapdif releases.. none are properly detailed in their versioning, and resulting with various tools none of which I'm sure what to do with - This is what I have to sort out with my members, whom when they try to work with things are confronted with such issues which shouldn't happen. If the community creates a fix for the engine's tools, this should be released as part of a tool package that should be regularly checked.

Similarly with the scripting side of things - The example games provided with Torque are great, but they aren't able to really show off anything that's completely non-FPS like. The Racing mod is still greatly based on the FPS scripting. For example, how would you create a particle emitter which uses models as source and attach it to the actual weapon. Well you'd look for the scripting references in the FPS kit, but these don’t give you the big picture of how to properly use them.

Furthermore, the website lacks some much needed features for the team development - If a team is working with Torque, one of their main focus points will be the GG site. With this in mind, if someone like the art lead or lead programmer finds a good resource, he has no method of forwarding that to his other programmers without having to send a separate email. Think about the BBC new forwarding.. simple and many people use it.

If you look at the frontpage, where can you get the latest news? It's not clear, and when even a simple thing such as trying to look for the lastest news about the whole site isn't easy to find, this rings bells.

It's a series a issues like such that are hard for me to get members willing to spend time to look into and cause a great amount of confusion for newcommers. I would prefer seeing less content packs, and more actual decent tutorials and pipeline explanations for different styles of games and genres, to really show off the abilities hidden within torque.

--

I don't want to leave on a totally negative note - Torque has provided us with a solution that is very affordable, but the feeling of being let loose into a giant unknown is very present, specifcally for members. We've been able to get some good work done, but while having devathons, our members were often faced with problems that stopped production and education purely because of the lack of information on many subjects, which is where my call for better collated documentation come from. I understand that GG is working on the TDN, but it would be nice to know that this is heading into providing us with more depth to the documentation.

The critiques here are aimed to be constructive and informative...