What would you like to see addressed in Torque 2D?

#101 I wanted to share this link with the team -- this thread isn't 100% appropriate for the link, but bear with me.

http://measuringmeasures.blogspot.com/2010/01/you-release-late-and-infrequently.html

The summary is that you have to release early and often in order to get something working into your users' hands. From there you'll know what people want and need.

The opposite is waiting until something is "complete" after a long dev cycle, and releasing something that users don't want.


I'm really concerned about Torque2D because of this. You guys have been working so hard trying to make something that can kill the market, but when you release it'll be a piece of software that no one has actually used.

The fact is that we don't know what we want until we see it. Actually having a piece of software to tinker with is a whole different universe than just imagining in a thread what would be nice to have.

Have you considered releasing betas in chunks?

#102 Torque 2D is not finished and not usable so it doesn't make sense to release it. Neither are we just tweaking or trying to get it perfect before we release, we're writing large portions of the codebase right now. When it's ready for early alpha/beta, we'll announce that.

We may do a private beta on the API and not the builder, not sure yet. Either way, when it's in a state where we can get sensible feedback, we'll look at that step.

We've all got a lot of experience in how to release products so we really don't need to "learn" about that. It's totally impractical to release a game-engine using agile methods but I understand that everyone wants to get their hands on it and we'll be looking at trying to get out at least release a schedule ASAP.

Right now, we're not releasing late, we've not announced any date for that very reason. We'll be doing a beta release just like T3D.

#103 solid response melv

I also wanted to request the ability to easily "Deploy" a build. i.e. on mac, build the bundle and properties properly and present what you've worked on in a nice package. Not sure what you'd do for windows outside of coming up some kind of Torque bundle format accessed through T2D api calls.

thanks for your rapid responses on this thread, it's really encouraging

#104 No problem David, I wish I could say more!

I'm definately interested in what people want from a 2D engine. It's easy to just look at the "GG Machine" but I wanted to say that 2D is being worked on by only a few people but those that are working on it are doing it with passion! I love waking up in the morning and working on it.

I've spent a lot of time going over the many issues I've experienced and have seen reported on the forums with TGB. TGB is a fine product but has its flaws, some inherent to the 2D engine itself, some from the underlying TGE. Nothing is perfect.

My initial work, when it was decided to do a ground-up T2D, was to type up all the major issues I knew about. I'm not talking about bugs, I'm talking about usability issues and recommendations. I scanned the forums to remind me of many as well. I spoke to many people about it. I knew of many myself. When I first started I had a crazy-long list and the task looked insurmountable but slowly after many weeks of organising it into something more structured I started to evolve a few solutions.

Eventually I came up with a central concept that fused together all my individual solutions into one concept.

One of the foundations to this concept makes T2D *highly* extensible and decouples much of the internal "objects". Another huge change is that the extensible systems are not just for end-users, the stock features are written using them e.g. "I'm eating my own dog food".

What with the xmas period, GGs move to Vegas and just being so busy, we've not had the time to do another blog. We'll try to rectify that ASAP. I must say that I'm really sorry that the T2D dev is taking so long but several things aligned to cause it and I'm scrambling to catch up. Because of this we've shied away from putting out a public schedule. Now that things are settling down, we'll see if that's possible.

I worked quite a bit over the xmas period and I'm doing around 10-15 hours a day on T2D and it's really paying off and I must say that I'm thoroughly enjoying it!

@Pete: My only regret is just how slow it can be to get information out there. Bring back the days when I could blog every other day about what I was doing now! In those days I had plenty of feedback because it was a totally transparent process.

#105 Not sure if this has been mentioned or already addressed, but I'd like to see some changes that are more OO-friendly, like true polymorphism and inheritance that isn't one layer deep.

#106 I have the privilege of SVN access and can confirm that the Torque 2D team are complete animals :)

Quote:My only regret is just how slow it can be to get information out there. Bring back the days when I could blog every other day about what I was doing now! In those days I had plenty of feedback because it was a totally transparent process.

I was just thinking this regarding some of my recent work and how important feedback loops are to keeping development fun, interesting, and on track with expectations and needs.

#107 It's a strange situation. In my consultancy I have often recommended, even for closed-source projects, that companies provide public read access to the repository for as much of the source as possible. This allows power users, who are generally a subsection of the evangelist users, to see into the process and see what's going on. It doesn't present any credible risk to the business because source code does not a business make, and only a tiny fraction of users know how to pull code down from a repo and build anyway.

In this case however, your target market are programmers who can probably pull down the code. That would tend to cannibalize your sales, I suspect.


In any case, I have a prediction. We've heard about a particular big announcement. I predict the big announcement is that the team is tired of maintaining a scripting language that has no advantages over the many robust languages already available. I predict movement to an established language like Lua or Python, or maybe a system that makes T2D language agnostic.

#108 Hey, we share the same dream, Pete :)

#109 We already have that situation where there are a number of associates that have full access to both Torque 2D and Torque 3D, not just employees. Because source-code is a value added item to developers, we can't make it public to the community which is essentially what you said but I'm just backing that up.

I wish I could talk openly about the T2D feature-set but unfortunately I cannot. I had a long video chat last night with GG and I discussed that subject and we'll be looking at how we can get more information out quicker to everyone. I am pushing! I agree that if we can talk openly about what's going on then we can get very useful feedback. We'll always get crap from folks who want to hold you over the fire because you apparently "promised" something but I'm personally not worried about that although marketing departments always seem to be scared of that issue. If we remove something we discussed then it'll be for good reasons, not just because we're being mean. :)

I personally hate "big" announcements, they tend to get hyped up too much, often land with a whimper and are not always that big to everyone. I do prefer lots of smaller feature announcements regularly.

I can assure you that now that GG have finished their move and xmas is well over, Eric and the crew are looking at how we can communicate better for products like this.

The lead developer wants to spill all the beans, I just need to get permission and do it in a way that won't hurt the product development or subsequent marketing!

#110 +1 for hating big hyped announcements (this is counter productive 9 times out of 10).

+1 for lots of smaller feature announcements/discussion regularly.

+1 for not caring about grumbling over "promised" features when strongly messaged as in-development potentially not making it into a 1.0 release... and not being used to generate "hype"

+1 for empowering power users, evangelists, source licensees with more frequent development snapshots (which aren't supported, may have code added/removed) in addition to the current alpha/beta/final binary installers.

+1 for TorquePowered.com, Eric and the crew :)

#111 +1 Listening to Josh. :)

#112 You know, I think one thing that would make a lot of sense is to integrate Torsion into Torque 2D/3D or TP/GG provide an editor with the same level of power. Make a deal with Sick Head and get it done.

#113 Another thing I'd like to see in the editor is the ability to hold down shift when you move items around and have them lock to a single axis like in photoshop. I'd also like to be able to copy items by alt dragging.

In regards to the scripting language, I've grown fond of torquescript. I wouldn't mind them changing to something else, but hopefully it won't be too much of a learning curve. I'm not a professional programmer so I don't know those other languages you mentioned.

#114 @Xanthor - Torsion won't be integrated into T2D, but i think you'll be very happy. ;)

#115 @Tom - I'll take your word :D

#116 So do I take the time to setup a project in TGB and learn all that there is to learn about it and start to write a game, getting frustrated that I used an engine that will be replaced in 2 months -or- do I wait 2 months and use the newly announced and released T2D -or- do I wait 2 months and get burned with the newly announced T2D to be released by the end of the year / early next year?

I know, I know, no schedule to be spoken of, but a nice ball park of a release in Q2 or Q3 or Q4 or Next Year type thing would be nice.

This is a big deal even more so as the T2D is not backwards compatible and there is not a planned upgrade path for TGB projects.

#117 having events scheduled by an object in the scene be paused if that scene is paused

example:
I have a ship that recharges its shields by scheduling a "recharge" event every 10 seconds
I would like that when the scene the ship is in is paused, and say the "recharge" event has 5 seconds before execution, I would like the "recharge" event to execute 5 seconds after the scene is un-paused, rather than executing regardless of the state of the scene

hopefully my example wasn't too TGB focused to be of use for T2D

thanks!

#118 would torque 2d support elevated tiling?

#119 A few things:

1) First, an easy one:) I wish ".5" was a valid float. I have to type "0.5" for the torquescript compiler not to freak out. Yeah, it's minor, but I've probably wasted a few hours tracking it down over and over again.

2) Ability to stamp a decal onto a sprite (permanently). Hard to explain, but Flash has bitmap objects which are more or less like sprites, and they have the ability to draw (on themselves) any other drawable object and that image goes permanently on that bitmap image (see the "draw" method of the bitmap object in Flash CS4 - http://livedocs.adobe.com/flash/9.0/ActionScriptLangRefV3/flash/display/BitmapData.html#draw%28%29). The main use I see would basically be to splash decals (say, scorch marks, bullet holes, blood splatter, whatever) all over your background without huge cost. e.g. see the Flash game Boxhead - after a few dozen levels the ground will be just completely covered in thousands and thousands of blood spatters. Unless I am mistaken, doing the same in TGB would require keeping thousands of blood spatter sprites alive and mounted to the background.

3) Basic drawing commands: Circles, splines, etc.

4) Masking of sprites.

5) "Screen" blend mode in addition to "Mult" with setBlendColor. Currently there's no way to make a sprite flash brighter, only darker, since it's a mult, and the color is clamped to 1,1,1.

5) Movable pivot points.

#120 Blending modes a la photoshop is a great idea.