Alpha2 1.1 is live

#21 I don't see what anyone is complaining about, this is a good engine even at 1.0.2 and I definately feel that I got my $'s worth. Bugs should be expected in any software. I doubt if they are so bad they make the engine unusable..and if so I'm sure a patch is available to remedy it.

I've been working on 1.0.2 and I'll stay with that version for a while before I move on. They're doing an amazing amount of work on this and for me it's easier to wait until things are solidified.

#22 Chris, T2D is still in early release.
The 1.0 release is not the final product.
The final release, the traditional 1.0 Release in software, is still a bit away.

Quote:
Early Adopter Release
This is an early adopter release. Torque 2D is not yet complete and major components may change as it gets improved for the final release. It is an advanced stable platform available for developers who want to get an early jump on 2D game development.

If the final relase of T2d is considered version 1.0 in tradiotional software, you can probaly consider the current 1.0 release as version 0.7

#23

Quote:My priority list is really just having a bug free release of 1.0 and packaging utility. That's really all that I paid for. I guess what bothers me most is that there's allready work being done on a 1.1 alpha, while 1.0 was never released in a bug free version.

I think what the problem is comes from the expectation with the version notation "1.0". Now, it was probably not a good idea for GG to call T2D's initial release 1.0, as such a version number typically indicates a certain level of completeness that is at odds with the widely unknown concept "Early Adopter". However they did warn that it wasn't finished.

Despite the (general) honesty about version numbers in the open-source world, GG operates under a more marketting/profit-driven approach. They have a product to sell, and they'll sell more copies of it if it's called version 1.0 than 0.5. It's a bit of marketting sleight of hand, but it's pretty far from what I would consider actual deception. Like I said, they gave you fair warning, even if it was a bit of a mixed message.

Personally, I think GG needs to do a better job of informing people of what they're buying before they get it. The combination of the definition/terms "Early Adopter" and "version 1.0" can easily be misinterpreted as both Christopher and I did. To some, this combination translates to "stable Beta", which in plain English means:

"This project is feature complete (all the features we want to do in the release called 1.x are there) and it is reasonably stable. It can be used, but expect to run into non-trivial non-outlier bugs every now and then."

This was, by the admission of the T2D developers, not what T2D's 1.0 release was. It would best be descibed as:

"This project has a useful subset of the features we want in the final release, and it is reasonably stable. Later revision will be adding, removing, and/or modifying features quite significantly. It can be used, but expect to run into non-trivial non-outlier bugs every now and then."

I think the problem is just one of effective communication. Or, a lack thereof. If you have terms that are never used outside your circle of experts (like "Early Adopter". The rest of the world calls these pre-Alpha) and combine them with other terms that are used outside of your circle but mean significantly different things (like "version 1.0"), you have to expect some confusion.

In short, Christopher, it was never the intension of the T2D development team to take the featureset of T2D 1.0 and completely bugfix it. That featureset was never "what you set off to do in the beginning"; it was just a useful subset of the eventual featureset of the product.

Quote:I remember hearing a while ago, when we were first considering using t2d, that there would be a final version out in the summer of '05

I was around during those days, and I heard nothing of the sort. The best-case estimates I had heard was a true "release" by the end of the year.

I'd say the current timeline of losing "Early Adopter" status is sometime in the next 2 years. Maybe halfway, maybe only a quarter of that. But sometime in the next two years, they'll have a full version of T2D (which should be called 2.0).

#24 Early Adopter can't last another 2 years..... I think it will go live with 1.1 being finalized by summer.
2.0 should be a paid upgrade like next version of TGE. 2.0 can be shader enhanced T2d :) particles and tiles could have shader effects. That might take 2 years... but not the base SDK from EA IMO**.

#25 I have to say i totally agree with smaug.

but... i honestly dont see any part of t2d that is so imperfect that it holds me back from developing a game. Sure, video driver support sucks, performance still isnt that great, there are some physics and other bugs... but is any of this blocking a developer from writing a game? no. it may pose problems trying to polish the game for release, but my bet is that the people complaining of 'unfinished' t2d status are nowhere close to finishing a game. They are just getting bogged down in idiosyncraticies of the t2d engine...

news flash people: if you cant work around the bugs in t2d right now, you do not (nor will ever) cut it as a developer. stop now and take that marketing job you always hated.

#26 Since I am the only person really "complaining" in this thread, I guess Jason's comments are directed at me.

Quote:

but... i honestly dont see any part of t2d that is so imperfect that it holds me back from developing a game. Sure, video driver support sucks, performance still isnt that great, there are some physics and other bugs... but is any of this blocking a developer from writing a game? no.

If your idea of a game is Gold Fever or Fortune tiles, then fine. As far as I can tell, these are the only game(s) (isnt Fortune Tiles just a re-skin of Gold Fever?) that are being released using T2D's current incarnation. This isnt the type of game that we set out to make when we bought torque though. If we wanted to make puzzle games, we would use Blitz.

Quote:
but my bet is that the people complaining of 'unfinished' t2d status are nowhere close to finishing a game. They are just getting bogged down in idiosyncraticies of the t2d engine...

1) We have spent over 6 months working on a game using T2D over here. We are waiting on a fully documented and bug free release of the engine so that we can release the thing. All animations, level design, music, sound effects, game logic, etc. are finished. In the meantime, we have moved on to other projects. I am sure there are other developers out there who are waiting as well. David Grace comes to mind with his "Cloud Burst" project.

2) Have you checked out the demos that come with T2D? Using the space shooter demo as an example...when you click outside the scene window the whole scene EXPLODES! Likewise, if I get an IM notification in the background while my game, or the T2D demos are running, the scene also explodes. This is NOT acceptible behavior, and is not simply an "idiosyncracy" of the Torque engine.

#27 @Christopher

Quote:
If your idea of a game is Gold Fever or Fortune tiles, then fine.

What's the type of your game? And what are your problems? I'm curious.

Quote:
1) We have spent over 6 months working on a game using T2D over here. We are waiting on a fully documented and bug free release of the engine so that we can release the thing.

Haven't you eliminated the bugs that occured to you? I spend much more time cleaning my own code from bugs than searching them in the engine. I hardly ever find bugs really. No that the kind of bug that makes the program crash, more the kind of bug that setEnable() is not promoted to mounted objects or things like that. But these are fixed quickly and you will mostly find solutions for them here.
Physics would be another thing. If physics was some integral part of you're game and it T2D physics system is not strong enough I can understand that you wait with releasing the game.

Quote:
2) Have you checked out the demos that come with T2D? Using the space shooter demo as an example...when you click outside the scene window the whole scene EXPLODES! Likewise, if I get an IM notification in the background while my game, or the T2D demos are running, the scene also explodes. This is NOT acceptible behavior, and is not simply an "idiosyncracy" of the Torque engine.

I am sure this one can be controlled somehow. Maybe there is some event fire when the game window loses focus. You could use that to just pause the game.
Actually I just solved this problem by commenting out 2 lines in Platform::process() in winWindow.cc. It is not a bug that the game slows down if it loses focus, it is desired behavior and it can be change quite easily.

With the IMs, I don't have any experience. I never noticed anything like that.
.

#28 Hello Michael,

Our T2D game is a sidescrolling shooter. Most of the problems we are having are physics and timing related. Like mentioned in my previous post, we have a lot of problems when a background process interferes with T2D. Our scenes sometimes simply blow up, which results in enemy objects being catapulted at the player object.

There are some other random tidbits....one of our stage bosses consists of 3 animated strips, which are animated seperately and then layered on top of one another. The strips are all of differant sizes, but the individual cells are the same size across all strips. We set all of the strips to the same scale but ,despite that fact, we get weird seams where the strips are layered.

Another problem we have had was regarding the loading of OGG files. Sometimes T2D flat out crashed when loading certain files, none of which are larger than 34kb.

There are probably some other issues, but I cant think of them off the top of my head. I really havent looked at that code in a few weeks now.

I should also note that all of these problems were with the "early adopter" version of T2D. I am waiting until the final version of 1.1 before I port our 9000 or so lines of TorqueScript over. In the meantime, we have been working on another game concept.

By the way, thank you very much for taking the time to research the issue pertaining to the slowing down of the engine when the scene window loses focus. Which lines in particular did you look at? Also, it is not just a problem of slowing down....all of our objects which are using timers and any physics functions go absolutely wacky when the window is out of focus.

#29

Quote:If your idea of a game is Gold Fever or Fortune tiles, then fine. As far as I can tell, these are the only game(s) (isnt Fortune Tiles just a re-skin of Gold Fever?) that are being released using T2D's current incarnation.

Hmm. By that logic, the X-Box 360 is utterly incapable of playing RPGs. After all, there aren't any RPGs out, so obviously it can't be used for it. And there are quite a few FPS games on it, so that must be the extent of the system's capabilities.

If you want to understand what a piece of code (or hardware) can do, you don't look at the products that are already released. You study the code (or hardware) itself. Just because some people don't use it in published products doesn't mean that they are impossible.

#30 Out of curiosity, have you enabled the setScenePhysicsFPSActive(true) in your game? That decouples framerate and physics and solves a lot of these kinds of problems.

Incidentally there's also King Kong, Little Gods released, neither of which are quite as static and puzzle-y as Gold Fever and Fortune Tiles (Little Gods no doubt uses some degree of physics in it). My own project uses a modicum of physics and collision and it seems to work great too, hopefully well enough to release before the end of January.

#31 Smaug-

You totaly misunderstood what I said. Jason Swearingen said that the current, and unfinished, incarnation of Torque2d is sufficient to release commercial games. I said that developers are using T2D, in its current state, to release puzzle games. Fortune Tiles does not use collisions or physics, and it only uses the most basic of animations. This game hardly takes advantage of what makes T2D stand above other indie game engines. I am not denigrating Fortune Tiles by the way, as it is a very fun game. I am merely saying that Fortune Tiles could have been made with just about any other engine.

Are there any commercial T2D games available which DO make full use of T2D's features? Not to my knowledge. In fact, the fact that there are so few commercial T2D games should tell you something about how developers as a whole are approaching T2D in general. It is a great concept, but it is not finished yet. When it is, I'm sure that more developers will warm up to it and at that time we will see a glut of awesome T2D derived products.

#32 @Christopher, I actually didnt say that.. what i said is more along the lines of "most people who complain are just stuck on feature X and would beter spend their time moving on to other features and finishing their damn game"

and yeah.. you are the only person complaining in this thread.. but you are by no means the only person complaining :P

I guess YOU have finished your game more or less, which in itself takes you out of the realm of the people i'm bitching about.

The whole reason i went off on the subject is because so many people complain about feature X when it's not really stopping them from moving on in their development.

@Christopher, the main issue you seem to be raising is performance.. and yeah, especially in 1.0.2, i raised this as being a huge problem (i did tests showing that 300 staticsprites causes huge lag) But i'm writing tools, and havent even started writing a 'real' game yet, so that's the extent of my complaining (reporting it as a big bug) Since you havent yet ported to 1.1, i would strongly suggest you do so, as I have noticed that a lot of perf issues have been resolved (but i havent run any more stress tests yet)

and yeah, if your game is pretty much finished, i think your bitching is more desrving... though maybe it would be better to post your problems on the fourms and try to get them fixed in a one-off nature.. just in this thread it seems like you got a few answers to some of your problems....

#33 Christopher,

Quote:here are some other random tidbits....one of our stage bosses consists of 3 animated strips, which are animated seperately and then layered on top of one another. The strips are all of differant sizes, but the individual cells are the same size across all strips. We set all of the strips to the same scale but ,despite that fact, we get weird seams where the strips are layered.

This is a well established problem with the solutions posted many times on the forums. Presumably you are aware of the new image-map system that will solve issues such as this? Assuming of course this is a known problem and you're not just keeping it under wraps hoping it's known. ;)

Quote:we have a lot of problem when a background process interferes with T2D. Our scenes sometimes simply blow up, which results in enemy objects being catapulted at the player object.

Again, this is a well known problem with posted solutions, essentially where forces were being applied in a frame-rate dependant manner. Again, this problem has now been solved to the best of our knowledge. In-fact, it was solve early in the year with details on the forums. Putting aside background proceses, don't forget that when tasking-switch away, the default mode for the underlying TGE is to update the simulation at a much slower rate (so as to reduce its CPU usage as a background process) resulting in a very coarse time-base for the physics therefore exaggerating the problem outlined above. This can easily be changed.

Quote:Another problem we have had was regarding the loading of OGG files. Sometimes T2D flat out crashed when loading certain files, none of which are larger than 34kb.

This is core TGE and I am uncertain as to why this is happening. Lots of people use OGG files without problem. I know very little about the audio side of things but I'm sure the community can help here or I can pass it along to those in GG "in the known". Presumably you've posted this problem and had some kind of response? Maybe there's an improvement in TGE v1.4? Have you tried your OGG files in any of the TGE demos?


I will say that there will never be a "bug-free" release. You'll only get one where we've done our best to remove the ones we know about. ;)

Seriously though, unless you intend to patch the known problems in v1.0.2, you'll have to recreate lots of your saved objects such as particle-effects and tile-maps. This was one of the major leaps between 1.0.2 and the alpha releases. As from the alpha releases, we're ensuring that saved-stuff doesn't break from release to release.

Lots of the games mentioned previously were v1.0.2 and they have gone out to major portals but these have had the appropriate patches applied to known issues. They've also added their own fixes for problems that they saw as a potential problem. We've also helped them resolve certain issues just so they can get their product out the door. Some of these changes have been fed back into T2D and are now part of the alpha-releases so it's a very useful two-way process for everyone.

To comment on the current games released with T2D; I don't think you can infer anything, especially as you're not aware of games that are being developed that I know about but cannot discuss. These are most certainly not simple casual games. It's like saying that the reason I haven't got a Lexus is because I can't afford it. I haven't got one because I choose not to because I'm not motivated to.

The ability to quickly put games together has specific appeal to those people writing casual games. That's not to infer that these games are necessarily easy to write but rather that one of the benefits of T2D is that it can make the job much easier and we'd like to think that the time-to-market can be reduced.

I understand your concern over bugs but using an early-adopter to make a product is a risky business and whilst I know I'm preaching to the choir here (sorry about that), it does come with a commitment from yourself that you'll probably have to upgrade or fix/patch the problems yourself.

With all this said, I am more than willing to help you or anyone else in resolving your problems, especially if it means that together, we can get your product out with a good level of confidence. If you wish to contact me privately then please do so and I'll see what I can do to get you moving in the right direction.

- Melv.

#34 @Christopher:

Here is what I've done to Platform::process:

//--------------------------------------
void Platform::process()
{
   DInputManager* mgr = dynamic_cast<DInputManager*>( Input::getManager() );
   if ( !mgr || !mgr->isMouseActive() )
      CheckCursorPos();
   WindowsConsole->process();
   
   if(!ProcessMessages())
   {
      // generate a quit event
      Event quitEvent;
      quitEvent.type = QuitEventType;
      
      Game->postEvent(quitEvent);
   }
   // if there's no window, we sleep 1, otherwise we sleep 0
   if(!Game->isJournalReading())
      Sleep(gWindowCreated ? 0 : 1); // give others some process time if necessary...
  

   /* The following code seems to be responsible for the frame drop. */

   HWND window = GetForegroundWindow();
   if (window && gWindowCreated) 
   {
      // check to see if we are in the foreground or not
      // if not Sleep for 100ms or until a Win32 message/input is recieved
      DWORD foregroundProcessId;
      GetWindowThreadProcessId(window, &foregroundProcessId);
      if (foregroundProcessId != winState.processId)            // commenting out this and the next line will turn off this feature. 
         MsgWaitForMultipleObjects(0, NULL, false, 100, QS_ALLINPUT);
   }

   /* But its better to comment out the whole block */


   Input::process();
}

However, this is a quick hack. I would try to find a cleaner solution.

I am doing a kind of sidescrolling shooter too and had similar problems back with 1.0.2 with physics. Porting to the latest alpha would be a good idea. They are running quite stable and I don't think you'll have to port you're script code a second time for newer releases (someone correct me if I am wrong!).
Oh and please post some screenshots in the show off section if you are okay with this! I'd really like to see some non-static T2D games.

-Michael

#35 You won't have to convert any saved objects and any change that breaks existing code isn't won't be done unless there is a very good reason so you should be confident that any code working with the alpha will work in the final release with only minor modifications.

No guarantee but there are definate controls to try our best to ensure that this will be the case.

Yes, post some images, the more stuff in the show off forum the better. :)

- Melv.

#36 Hey guys,

Thank you for all of the replies.

We are still a bit reluctant to port our code over to the current alpha because of the potential changes that could still arise, particularly with the upcoming TGE 1.4 merge.

@melv

We greatly appreciate your offer to help us get this thing out. After we finish up our current project, we'll get in contact with you about our T2D project.

#37 Not a problem, it benefits everyone to get products out the door.

I look forward to hearing from you.

- Melv.