GarageGames Announces Torque 2, Transparent Development

[Stephen Zepp]

This might get moved, but it made the most sense to me to place it here. The main reason I wanted to post it is because I know of several f13 members that are using Torque in some form, so I thought they might want some liniks:

Discussion thread on InstantAction, some Torque 2 info -- this is a public thread. WARNING: Excessive Serek Dmart content! The upside is you get to see me manage him 

Discussion thread on Torque 2 -- this is a private thread currently, and requires a TGE licensed account to view.

This is an announcement of intent--we've not yet finished the web infrastructure to execute on the strategy, so there isn't much additional information other than what is in these two threads available yet.

[Mrbloodworth]

Um, yeah....

I'm going to let the cynic out of me.

So, when are you guys going to complete the Documentation for TGE...before or after this? Lots of people have been waiting for more official, and complete documents and examples for TGE on par with what other "budget" engines have available.

TGE suits my needs, and i have done really well so far with it (imo, starting from nothing), and i have paid my dues using forums as "Documents" and outdated, unsupported "Resources" and a TDN no one ever updates, or edits for sanity's sake.

From the sounds of it, Torque 2 does not (again) meet my needs.. TGE still does, but working through the above issue has been the main hurdle.

That, and are you guys EVER going to ... "Improve" (Not going to say fix, because you guys do not feel its broken, even if it has no brain for sorting...ever) Transparencies?

I wish you luck with the new engine/platform... But i personal wish that things that were promised, and most likely should have been...were done before new stuff. It always seems as if TGE is abandoned by GG.

I know we have had this conversation before.

EDIT: also, Collada? Your killing me here.... (Lightwave user)

EDIT 2 : Gota say, after further reading...Seems Garage games needs to change its name to "Brick and mortar games"... I'm not sure the hobbyist, or small indi is going to be able to keep up with the requirements of T2... The tools that support what you are stating as new formats ETC... are not cheap.

[Stephen Zepp]

Not blowing off your questions, just need some time to sit down and answer, and I'm late for an afternoon of meetings--will get back to you tonight!

[Mrbloodworth]

Not blowing off your questions, just need some time to sit down and answer, and I'm late for an afternoon of meetings--will get back to you tonight!

Np, just keep in mind, in the 3 or so years i have been using torque, i have learned a lot, and improved a quite bit, and it has always been worth it. best 100$ (Well, i think i have spent about 500+ so far) i have ever spent, and has gotten me that much closer to my dreams.

But, this is the stuff i have been bitching about for a while now, and hear many a promise...but never see a correction and mostly get them passed over.. (Such as transparency). 

I AM looking forward to playing with torque 2. But i'm not sure it will suit someone in my position. Time will tell.

[Stephen Zepp]

Ok, couple of responses:

--We hear the request to "get better documentation", "make better tutorials", "give us more info on xxx" a lot, but unfortunately, we honestly have never gotten much of a response when we ask for specifics, except for "get better documentation", "make better tutorials", and "give more info on xxx". There exists a complete set of API documentation on every class in the game, including in many cases not only theory of the system (if major) as well as theory discussions built in to the dOxygen documentation that can be either viewed online, or self-generated locally for every project. There exists a complete set of online reference docs for all available console functions, and even an in game ability to display a list of all console methods for a particular object. Finally, there are also three books on various aspects of Torque, and a fourth (source code based) in development. In answer to your specific question, in your specific circumstance, the effective answer is probably (all truth be told), "won't happen in the way you are asking". All of the documentation generated was "after the fact" with regards to the development of the engine itself, and reflects that process flow (in some cases in a negative way). For the record, I've talked to quite a few commercial developers that have used not only Torque, but Unreal 2/3, Gamebryo, and other commercial engine products, and more than 80% of them indicate that the documentation of Torque is on par or in some cases much better than the other products.

We agree that some of the engines that are currently in development, and/or came out after Torque have better tutorials, and in some cases better reference documentation as well--and after extensive research, we've determined that this is in no small part due to two things: a competitor watching the negatives regarding Torque and actively focusing on correcting those issues during the engine development, and an increase in both customer expectations as well as a more comprehensive approach to developing an engine, not adapting an existing engine to a market.

One of the biggest complaints we get regarding Torque (other than "documentation sucks!") is that the technology is old/dated--and it's true, which is why a majority of our efforts is going in to the next generation of technology, including all aspects of proper productization. It's important to keep in mind that when Torque was released as a product, it was literally a stripped down version of the engine for a production game--and of course those are never "fully documented" for a purchase market. With Torque 2, we are actively reversing this critical flaw in the development process--not only are we going to expose systems as they are developed to the general public for review and feedback, but we will also be building documentation with a comprehensive strategy during engine development.

I want to point out a couple of things regarding resources, and TDN as well: both are community services, unrelated to GG. We don't write resources--community members do, and TDN is not "our" documentation, it's "your" documentation. That being said, we agree that they both suffer from not being updated, and for Torque 2 (and specifically the transparent development process), we are actively addressing the infrastructure and tracking issues regarding why they may or may not be updated.

I'm extremely curious as to why you got the impression Torque 2 "will not meet (your) needs". If you wouldn't mind, either private or public information as to why you got this impression would be awesome! You mention "collada" specifically, but it's again extremely important to understand that the component architecture of Torque 2 allows you to use whatever external model format you like, simply by writing a new loader component that translates to the internal representation.

--transparencies: Difficulties with transparency are a factor of having to be backwards compatible all the way to OpenGL 1.0--we're stuck (in our current products) with the way OpenGL 1.0 does sorting. Torque 2's GFX layer is render device independent (and has no backwards compatibility requirements with regards to older Torque products), and each of the render devices will use whatever underlying device requirements and capabilities are proximate and appropriate.

Ultimately, it comes down to "do we continue to focus on older technology and products that continue to become more and more outdated almost daily, or in the long run is both the community and our company better served by moving forward", and ultimately the example we refer to internally is how Apple dumped OS-9 and moved forward to OS-10. It was bumpy along the way, but no one ever really says "I wish you would have stayed with OS-9 and just fixed a couple of things".

[Margalis]

When I messed around with Torque the big problem I saw was a lack of communication. At that time GG was fairly new, a lot of people were asking for things like shader support and getting no answer. Not that the answer was no, just that there was no answer at all.

There were a few people making tweaks and such (Melv May IIRC though I might have that wrong) that were not rolled back into the engine, and people who offered to work on "Torque 2" were rebuked.

Now, realistically none of that stuff really mattered, nobody is making games with Toque where shader support is going to make or break them, but it was a perception problem.

The actual engine itself was nice, well-written, I had no problem changing or adding to it. It probably depends on how familiar you are with C/C++ and 3D engines in general. I don't remember ever looking at documentation. "The code is the documentation" was my motto. (LOL)

It's kind of in a weird place because the engine is not enough to create AAA titles but it isn't simple either, you could easily make a CounterStrike style game with it. I think the reality is that there aren't that many people who have the skills to use the engine to make something other than simple games and also don't have the skills or desire to have an industry job.

[Stephen Zepp]

When I messed around with Torque the big problem I saw was a lack of communication. At that time GG was fairly new, a lot of people were asking for things like shader support and getting no answer. Not that the answer was no, just that there was no answer at all.

There were a few people making tweaks and such (Melv May IIRC though I might have that wrong) that were not rolled back into the engine, and people who offered to work on "Torque 2" were rebuked.

Now, realistically none of that stuff really mattered, nobody is making games with Toque where shader support is going to make or break them, but it was a perception problem.

The actual engine itself was nice, well-written, I had no problem changing or adding to it. It probably depends on how familiar you are with C/C++ and 3D engines in general. I don't remember ever looking at documentation. "The code is the documentation" was my motto. (LOL)

It's kind of in a weird place because the engine is not enough to create AAA titles but it isn't simple either, you could easily make a CounterStrike style game with it. I think the reality is that there aren't that many people who have the skills to use the engine to make something other than simple games and also don't have the skills or desire to have an industry job.

We love users like you! ("the code is the documentation") Hehe..all joking aside, You might find my latest blog interesting: Transparent Development, Torque 2, and You!" (yes, corny title--it's intentional!).

[Margalis]

Good stuff Stephen, sounds like that addresses what I was talking about.

The butterfly example you use is a good example of how deep object hierarchies can get out of control, especially in C++ with multiple inheritance. Very difficult sometimes to mix and match different characteristics. I tend to use a lot of delegate classes rather than inheritance to get around problems like that.

[Stephen Zepp]

Good stuff Stephen, sounds like that addresses what I was talking about.

The butterfly example you use is a good example of how deep object hierarchies can get out of control, especially in C++ with multiple inheritance. Very difficult sometimes to mix and match different characteristics. I tend to use a lot of delegate classes rather than inheritance to get around problems like that.

Heh...older gamers might remember Deer Hunter from Dynamix, which was made with the Tribes 1 core engine. For about a week and a half, deer had jet packs--was pretty funny as well.

We're very psyched about components---some really nifty stuff when they are designed well. Torque X was our first run with a fully component driven game object system, and it really makes a big difference, so we're pulling the trigger and shifting completely over to that type of system.

[Mrbloodworth]

"The code is the documentation" does nothing for artists.

[Stephen Zepp]

"The code is the documentation" does nothing for artists.

We know man--it was a humorous response :)

[Mrbloodworth]

"The code is the documentation" does nothing for artists.

We know man--it was a humorous response :)

I know =)

I really don't sit around and bitch about torque, in fact its gotten me close to my dreams than any other engine i have played with.... But that doesn't mean it doesn't have things that irk me. And art docs are a main gripe of mine...Not the C++.

That being said, i read we can use any type of model format we want. Sadly, this is again something that means jack all to artists working with torque... One of the main reason i picked torque long ago was the pipeline. It had exporters for my 3d app (Even if i had to figure EVERYTHING out because there was NO documentation on art assets and requirements), world building , a scripting IDE , everything... T2 sounds like it will have none of that, while its cool to "make your own X", my focus has always been on design, and art. I don't want to, code an exporter or importer, in fact...I know i can't.

Anyway..thanks as all ways for reading.

If it wasn't for "lightwave" Dave, i would not have used torque. Just saying.

As far as collada, i already went through this with FBX... Something light wave DOES support...Now here we are again (3d artists) with yet another baby jesusss of a format.. I know you can never account for all apps, but a small 3 would be fine. 3ds, Lightwave, and milkshape (Don't forget the poor people).

EDIT: Just want to say, I'm not afraid of hard work, and yes, i know games are hard work, i hit the realization a long time ago, and my designs, and work i have done for people of GG community reflect this. I just know my limitations, and what is worth it to me to devote my time to.