Will Constructor be fished or droped -- ?
by Adam Wilson · in Constructor · 10/19/2007 (2:55 pm) · 34 replies
I really hope as much time that has been spent on it and we have spent on it that Constructor will get the major bug fixes worked out and so we can finish our projects and get the beauty out of it that we have hoped for. The only thing that I can see is the Portals and transparent Backed DTS files not rendering outside dif's, dts's and terrain. I have not tried Collision meshes with the backed DTS files exported out.. but I think that is relay the only things that are stopping Constructor's Export to DIF from making Constructor a finished working great tool.
About the author
#2
11/03/2007 (5:00 am)
Leak Points is also a big problem of Constructor.
#3
11/06/2007 (9:36 am)
I was wondering... Will constructor be usefull for Torque 2? 
#4
---
Nebulagame Development
My personal blog
11/07/2007 (2:07 am)
I don't think that is an issue really. constructor generates map files. Unless for Torque 2 it is decided that the file formats have to change drastically, there shouldn't be any problem. The possibility of Constructor in its current state not supporting several features of the engine (shaders or other nifty stuff) is big however.---
Nebulagame Development
My personal blog
#5
12/10/2007 (2:15 pm)
Has this been drooped? or will be get another version with more fixes..
#6
12/13/2007 (9:34 am)
I'm sad to say, that as far as I know, it's been dropped, for now at least. The future is uncertain but it is not being worked on or supported at all anymore. We have to work with what we have and not expect any more updates. 
#7
I would imagine that Even if it continues to be supported for awhile longer, that it will at some point be dropped. I do not believe there will be a need for such an editor with T2, and since eventually thats the only engine per say there will be, I can understand to a degree.
12/13/2007 (9:51 am)
Though Matt recently did provide a patch for it fairly recently (as far as I know it is only linked to in one of the forums) to fix collisions and other things.I would imagine that Even if it continues to be supported for awhile longer, that it will at some point be dropped. I do not believe there will be a need for such an editor with T2, and since eventually thats the only engine per say there will be, I can understand to a degree.
#8
I would hate to think of all that good code going to waste in a repository somewhere.
12/13/2007 (3:03 pm)
If this is the case, I'm wondering about the possiblity of having the source for it provided?I would hate to think of all that good code going to waste in a repository somewhere.
#9
12/13/2007 (5:21 pm)
@Gustavo and David regarding Torque 2 and Constructor, I *believe* they're dropping DIF altogether, with the idea being to use DTS and polysoup for everything. The reason being that the BSP code takes up more CPU time than just rendering out a DTS and handling polysoup collisions.
#10
12/14/2007 (7:22 pm)
Yes but right now we need the ghosting fixed for the portals that would be one fix that I think would help us all out....................................................................................................... man that relay isn't what I needed to hear. 
#11
12/14/2007 (11:27 pm)
If it does get dropped would it be possible to release the source code? I'd pay for it.
#12
To my knowledge, this is correct concerning T2.
12/15/2007 (12:16 am)
Quote:@Gustavo and David regarding Torque 2 and Constructor, I *believe* they're dropping DIF altogether, with the idea being to use DTS and polysoup for everything. The reason being that the BSP code takes up more CPU time than just rendering out a DTS and handling polysoup collisions.
To my knowledge, this is correct concerning T2.
#13
Though Torque products are not expensive, but almost nothing is completed for final use.
Unreal Engine has many completed tools for final use, but it's too expensive.
Hope GG can provide a pruduct line that can make things from the beginning to the end.
12/15/2007 (7:53 pm)
When can GG provide a completed thing?Though Torque products are not expensive, but almost nothing is completed for final use.
Unreal Engine has many completed tools for final use, but it's too expensive.
Hope GG can provide a pruduct line that can make things from the beginning to the end.
#14
12/17/2007 (7:28 am)
All I can say is "MAN !" 
#15
1. Hey what do you expect for 150 bucks?
2. That's the community's responsibility to fix, not ours, after all we are *busy* with other projects.
3. Hey it was Free! you don't expect us to fix something we gave you for free do you?
4. Quit buggin us, we have made tremendous sacrificies for the community, you should be happy with products that only half work.
There seems to be a pervasive attitude that since TGE can be *made* to work, if you put enough time and effort and work arounds into it, that the customer base has no right to expect bug fixes or improvements to the product.
GG seems to be always busy on their next big thing, never on the stuff they have already marketed and sold to us.
When I bought TGE, it did not say there were known bugs in the engine that GG decided not to fix and that I would only find out about them via the community.
When I bought TGE, it did not say anywhere that the attitude of GG would be that since it was so cheap, it would be unreasonable of me to expect it to work perfectly.
When I bought TGE, it did not say anywhere that the art pipeline was so difficult to use that Constructor was the best reasonable option, but that Constructor itself was bug ridden, but after all, what do you want for free, double your money back?
When I bought TGE, it did not say anywhere the product had been officially declared dead, and was in a semi-useful state, and doomed to stay that way, because of the next project (whatever it might be, there always seems to be one)
When I bought TGE, I did not buy it because it was cheap. I bought it because after reading all the features and the comparisons to different engines, it appeared to be my best choice. I personally would have rather paid more money (I know plenty of other people are not in this position) for a product that was supported.
It was GG's decision to sell the product to the public at 100 or 150 bucks, whatever it is. It was not the community who set the price of the product. The original product was not given away free. Therefore, I think it is not unreasonable for the community to have expected that the product would be supported, and that it was in a working condition.
The product is NOT marketed as something you MUST be a C++ programmer to use, but if you aren't you are very quickly going to be in trouble.
When I go and buy a refrigerator from Best Buy and it is an open box deal so I get it cheap, I still expect it to work just like a new one. If the people at Best Buy were to tell me the refrigerator is broken, you can bet I want to pay even cheaper, but at least I would know up front what I am dealing with.
If torque / constructor are broken, and GG does not want to do bug fixes and/or improvements why would anyone want to pay 2 cents for it, much less 150 dollars? Possible reasons are: 1. they didnt know what they were getting in to. 2. Torque is still the best engine available in the price range.
It is reason number 2 that makes me sad. There is no doubt in my mind that if GG would be more responsive and support/fix/upgrade their products that Torque would take over the known Universe of game engines. I just don't understand the attitude I appear to see. It is like TGE is a bastard child that no one wants to take responsibility for. The next big thing is always taking up GG time.
What would it take for the community to change GG's attitude toward TGE, TGEA and constructor? Money? Empassioned pleas? Begging?
I will continue to use Torque until something better comes along, and I may be waiting a long time. I recognize the fact that Torque is an incredible piece of software and that the people at GG have given the community a real opportunity. I also recognize that it is a labor of love for the GG people and they may not look at it like a normal business, but the customer who buys the product does not know any of this when he buys. As long as the product is sold and not given away with the understanding that the user is on his own (like Quake engine is), I believe the responsibility of supporting the products (and I mean bug fixes, and a reasonable ease of operation, let alone improvements) lies squarely on GG shoulders, and should be not be fluffed off on to the community.
12/18/2007 (7:43 am)
I agree with Sun Yu. Although I am relatively new to Torque I have read many thousands of posts. It is apparent to me that the people at Garage Games have taken a "take it or leave it" attitude. Every time someone complains about the state of ANY GG product here is GG's response (whether it be stated or implied):1. Hey what do you expect for 150 bucks?
2. That's the community's responsibility to fix, not ours, after all we are *busy* with other projects.
3. Hey it was Free! you don't expect us to fix something we gave you for free do you?
4. Quit buggin us, we have made tremendous sacrificies for the community, you should be happy with products that only half work.
There seems to be a pervasive attitude that since TGE can be *made* to work, if you put enough time and effort and work arounds into it, that the customer base has no right to expect bug fixes or improvements to the product.
GG seems to be always busy on their next big thing, never on the stuff they have already marketed and sold to us.
When I bought TGE, it did not say there were known bugs in the engine that GG decided not to fix and that I would only find out about them via the community.
When I bought TGE, it did not say anywhere that the attitude of GG would be that since it was so cheap, it would be unreasonable of me to expect it to work perfectly.
When I bought TGE, it did not say anywhere that the art pipeline was so difficult to use that Constructor was the best reasonable option, but that Constructor itself was bug ridden, but after all, what do you want for free, double your money back?
When I bought TGE, it did not say anywhere the product had been officially declared dead, and was in a semi-useful state, and doomed to stay that way, because of the next project (whatever it might be, there always seems to be one)
When I bought TGE, I did not buy it because it was cheap. I bought it because after reading all the features and the comparisons to different engines, it appeared to be my best choice. I personally would have rather paid more money (I know plenty of other people are not in this position) for a product that was supported.
It was GG's decision to sell the product to the public at 100 or 150 bucks, whatever it is. It was not the community who set the price of the product. The original product was not given away free. Therefore, I think it is not unreasonable for the community to have expected that the product would be supported, and that it was in a working condition.
The product is NOT marketed as something you MUST be a C++ programmer to use, but if you aren't you are very quickly going to be in trouble.
When I go and buy a refrigerator from Best Buy and it is an open box deal so I get it cheap, I still expect it to work just like a new one. If the people at Best Buy were to tell me the refrigerator is broken, you can bet I want to pay even cheaper, but at least I would know up front what I am dealing with.
If torque / constructor are broken, and GG does not want to do bug fixes and/or improvements why would anyone want to pay 2 cents for it, much less 150 dollars? Possible reasons are: 1. they didnt know what they were getting in to. 2. Torque is still the best engine available in the price range.
It is reason number 2 that makes me sad. There is no doubt in my mind that if GG would be more responsive and support/fix/upgrade their products that Torque would take over the known Universe of game engines. I just don't understand the attitude I appear to see. It is like TGE is a bastard child that no one wants to take responsibility for. The next big thing is always taking up GG time.
What would it take for the community to change GG's attitude toward TGE, TGEA and constructor? Money? Empassioned pleas? Begging?
I will continue to use Torque until something better comes along, and I may be waiting a long time. I recognize the fact that Torque is an incredible piece of software and that the people at GG have given the community a real opportunity. I also recognize that it is a labor of love for the GG people and they may not look at it like a normal business, but the customer who buys the product does not know any of this when he buys. As long as the product is sold and not given away with the understanding that the user is on his own (like Quake engine is), I believe the responsibility of supporting the products (and I mean bug fixes, and a reasonable ease of operation, let alone improvements) lies squarely on GG shoulders, and should be not be fluffed off on to the community.
#16
TGE is not dead, officially or unofficially. There have been updates and bugfixes to the engine, though a good number of the community features and resource have not been integrated into the core. Most of them will not be integrated into the core since the vast majority are gameplay additions rather than core engine additions. There are a number of issues that are known, however. Jeff Faust has been a very active bug stomper for TGE and TGEA as he has worked on AFX. Other community members, associates, and employees have fixed a large number of the bugs that have plagued Torque from its early days. But you're right, it's not perfect, nor does it claim to be. There will always be bugs, though they should be stamped out as early as possible. One of the difficulties with some of the bugs is TGE's integrated nature. Everything touches something else, which is one of the beauty designs behind Torque 2 and TorqueX's component design. But T2 is a long way away and TorqueX is bound to XNA and C# as a complete engine rewrite without some of the core Torque functionality.
You can do a lot with Torquescript. There are four books and tons of resources that demonstrate that. Of course, to work on the level that most people want to modify the engine to make their dreams come true, C++ knowledge and tearing through the engine internals is a must. The "Is TGE Right for Me?" link off of the product page denotes an uphill battle for people learning C++, Torque, and a toolset, but it doesn't (nor can it really) depict the level of difficulty that new users can experience.
As for using Constructor, sure it is a good option even with its problems, but you are not constrained to using it. You can use GTKRadiant (now under the GPL), QuArK (also GPL), or 3DWorld Studio (commercial) or any other editor that can save Quake or Valve220 map files. Map2Dif will not compile bezier curves, however, so you should not use them in QuArK or Radiant when creating maps for Torque.
Compared to a number of other engines in the "indie" market, Torque's artpath is antiquated. Rather than supporting asset manufacture in formats like Collada (which Lightwave still does not have support for, dammit) or FBX (which LW does have support for), they use a Torque standard without the bells and whistles. Of course, if you know how to model and setup your nodes and teach your art team the specifics, it is an optimized workflow that gets them all on the same page. It's not easy by any means, but it is optimized. The key problem is training artists to use it effectively. This is especially a problem for new users who are learning their applications or are purchasing content from Turbo Squid in hopes of getting it right into the engine with no issues. Of course, most of them would do the same thing in Unity and wonder why it either crashed or reduced their framerate to mush because they imported a high-poly mesh designed for procedural rendering in Poser instead of a mesh optimized for realtime rendering. So I don't really know how to "win" in this. Torque 2's planned support for Collada is a nice step in the right direction, but for TGE/A licensees, they will need to work with the current artflow and exporters. A developer starting with the 3DS resource (and the memory management optimizations at the end) could go a ways with integrating other formats to use with TGE/A. I know a couple of developers who have done that, but not the status of their integration.
There are a number of problems on different levels that a number of us discussed at IGC concerning the community and community interaction. While we try to cover a lot of bases on the product pages and the forums, there's a lot of room for improvement.
12/18/2007 (9:24 am)
Very thought-out post. Just like to cover a couple of points.TGE is not dead, officially or unofficially. There have been updates and bugfixes to the engine, though a good number of the community features and resource have not been integrated into the core. Most of them will not be integrated into the core since the vast majority are gameplay additions rather than core engine additions. There are a number of issues that are known, however. Jeff Faust has been a very active bug stomper for TGE and TGEA as he has worked on AFX. Other community members, associates, and employees have fixed a large number of the bugs that have plagued Torque from its early days. But you're right, it's not perfect, nor does it claim to be. There will always be bugs, though they should be stamped out as early as possible. One of the difficulties with some of the bugs is TGE's integrated nature. Everything touches something else, which is one of the beauty designs behind Torque 2 and TorqueX's component design. But T2 is a long way away and TorqueX is bound to XNA and C# as a complete engine rewrite without some of the core Torque functionality.
You can do a lot with Torquescript. There are four books and tons of resources that demonstrate that. Of course, to work on the level that most people want to modify the engine to make their dreams come true, C++ knowledge and tearing through the engine internals is a must. The "Is TGE Right for Me?" link off of the product page denotes an uphill battle for people learning C++, Torque, and a toolset, but it doesn't (nor can it really) depict the level of difficulty that new users can experience.
As for using Constructor, sure it is a good option even with its problems, but you are not constrained to using it. You can use GTKRadiant (now under the GPL), QuArK (also GPL), or 3DWorld Studio (commercial) or any other editor that can save Quake or Valve220 map files. Map2Dif will not compile bezier curves, however, so you should not use them in QuArK or Radiant when creating maps for Torque.
Compared to a number of other engines in the "indie" market, Torque's artpath is antiquated. Rather than supporting asset manufacture in formats like Collada (which Lightwave still does not have support for, dammit) or FBX (which LW does have support for), they use a Torque standard without the bells and whistles. Of course, if you know how to model and setup your nodes and teach your art team the specifics, it is an optimized workflow that gets them all on the same page. It's not easy by any means, but it is optimized. The key problem is training artists to use it effectively. This is especially a problem for new users who are learning their applications or are purchasing content from Turbo Squid in hopes of getting it right into the engine with no issues. Of course, most of them would do the same thing in Unity and wonder why it either crashed or reduced their framerate to mush because they imported a high-poly mesh designed for procedural rendering in Poser instead of a mesh optimized for realtime rendering. So I don't really know how to "win" in this. Torque 2's planned support for Collada is a nice step in the right direction, but for TGE/A licensees, they will need to work with the current artflow and exporters. A developer starting with the 3DS resource (and the memory management optimizations at the end) could go a ways with integrating other formats to use with TGE/A. I know a couple of developers who have done that, but not the status of their integration.
There are a number of problems on different levels that a number of us discussed at IGC concerning the community and community interaction. While we try to cover a lot of bases on the product pages and the forums, there's a lot of room for improvement.
#17
12/18/2007 (2:39 pm)
Can you do one more thing to Constructor before you allow it to gather dust.. Fix the bad zones / portals. or give the hole thing over to the community so they can fix it and TGE / TGEA user can use this could be great tool to get there games done before they have to go buy T2.. THANKS 
#18
Its also a sad thing that GarageGames took such a large dump on us, and then did not bother to flush....
I am not a receptacle for GarageGames excrement!
12/18/2007 (3:35 pm)
Its a sad thing Constructor have come so far. Only to die before actually growing up into a fully usable tool....Its also a sad thing that GarageGames took such a large dump on us, and then did not bother to flush....
I am not a receptacle for GarageGames excrement!
#19
That said, if work on Constructor has stopped, and the current product is the best it gets, then this is just really disappointing. We had converted our project over from Quark to Constructor at an important time in development, partly because of the problems with Quark and partly because we anticipated that Constructor models would play nicer with TGE. We're at a point now where we really need to get portals working properly, but we're just banging our heads against the wall.
It's unfortunate, because I think a lot of people are going to feel really abandoned if this is the case.
12/18/2007 (7:47 pm)
I still haven't seen or heard anything official from GG -- not sure if David speaks for them, really -- so it would be nice to hear something along the lines of "Yes, further development on Constructor has stopped" or "No, we still plan on finishing Constructor" from someone in the company.That said, if work on Constructor has stopped, and the current product is the best it gets, then this is just really disappointing. We had converted our project over from Quark to Constructor at an important time in development, partly because of the problems with Quark and partly because we anticipated that Constructor models would play nicer with TGE. We're at a point now where we really need to get portals working properly, but we're just banging our heads against the wall.
It's unfortunate, because I think a lot of people are going to feel really abandoned if this is the case.
#20
12/18/2007 (11:54 pm)
@Mike: and Mac Users are feeling much more abandonned since issuewith Leopard. (a minor one: can't save or export! doh)
Adam Wilson