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
#22
12/19/2007 (6:56 am)
I'm voting on "fished". I think that's mostly because I don't know what "droped" means. Personally, I kind of like fishing.
#23
12/19/2007 (12:27 pm)
Aaaaaaaaaaaaaaaa? I don't need to know rock soiled yes or no but something would be nice. Something that could help teams decided on how much time was wasted in thinking that what they have created was going to be used. they way I see it GG lost out on $$ in games / content packs that could have been done by now and sold in there store that is if Constructor was more of a finished tool..... ( sigh as I type) 
#24
Got it. At least I understand now why Constructor was released for free.
12/19/2007 (12:43 pm)
@Stephen: That's fine, and I think many of us understand where you're coming from after seeing so many of these types of threads elsewhere around here. Mostly we are just looking for some sort of general position on Constructor and its future, not necessarily a commitment to a feature list or release date. So best I can tell, GG's position is "We don't know yet, we're not committing to anything, and we'll let you know when we decide."Got it. At least I understand now why Constructor was released for free.
#25
--Torque 2 will not use BSP (Constructive Solid Geometry) based models, so Constructor as it stands right now is not appropriate for the long term future (years) for the Torque asset pipeline. It's possible someone in the community may back port or add in the Interior class as components, but it's not on our planned list.
Constructor remains viable for current projects in TGE and TGE-A, and there is an internal build that may make it out to the public sometime in the future. I can't guarantee that it will come out at all, and I certainly can't guarantee that it will fix every issue there is with every work flow.
--There is only a subset of users that have issues with Constructor, and as best as we can tell, it comes down mostly to work flow--something that is pretty much impossible to debug and make completely robust. We have literally dozens of professional artists that use Constructor daily with great success, and while Matt has found a few more minor bugs (and one somewhat major if I remember correctly), it's more of a work flow training/documentation issue--specifically with the proper use of detail brushes and portals--but again, since we can't really watch people make their art with Constructor, it can be difficult to confirm.
I realize this sounds like a "it's not our fault, it's yours!" argument, but that's not my intent here--I'm just trying to describe the issues that remain on the table as we can best tell, as openly as possible.
--Constructor is not going to receive any major feature adds, updates, or anything of that nature for the future. Bug fixes may or may not be released as appropriate.
--Releasing the source code is really not an option. Constructor is built on core Torque code, and releasing the source code would effectively be releasing the engine. I'm not taking it completely off the table, but it's not in the plans right now.
12/19/2007 (1:51 pm)
Here's what I can confirm right now:--Torque 2 will not use BSP (Constructive Solid Geometry) based models, so Constructor as it stands right now is not appropriate for the long term future (years) for the Torque asset pipeline. It's possible someone in the community may back port or add in the Interior class as components, but it's not on our planned list.
Constructor remains viable for current projects in TGE and TGE-A, and there is an internal build that may make it out to the public sometime in the future. I can't guarantee that it will come out at all, and I certainly can't guarantee that it will fix every issue there is with every work flow.
--There is only a subset of users that have issues with Constructor, and as best as we can tell, it comes down mostly to work flow--something that is pretty much impossible to debug and make completely robust. We have literally dozens of professional artists that use Constructor daily with great success, and while Matt has found a few more minor bugs (and one somewhat major if I remember correctly), it's more of a work flow training/documentation issue--specifically with the proper use of detail brushes and portals--but again, since we can't really watch people make their art with Constructor, it can be difficult to confirm.
I realize this sounds like a "it's not our fault, it's yours!" argument, but that's not my intent here--I'm just trying to describe the issues that remain on the table as we can best tell, as openly as possible.
--Constructor is not going to receive any major feature adds, updates, or anything of that nature for the future. Bug fixes may or may not be released as appropriate.
--Releasing the source code is really not an option. Constructor is built on core Torque code, and releasing the source code would effectively be releasing the engine. I'm not taking it completely off the table, but it's not in the plans right now.
#26
Constructor being available for any GG user, I understand your point - and agree with it -, but aren't GG customers (ones with paid licenses), legitimate candidates to access this source code too?
12/19/2007 (2:43 pm)
Quote:Constructor is built on core Torque code, and releasing the source code would effectively be releasing the engine.Aren't some of us supposed to be TGE/TGEA code licenses? meaning full access to the engine code?
Constructor being available for any GG user, I understand your point - and agree with it -, but aren't GG customers (ones with paid licenses), legitimate candidates to access this source code too?
#27
12/19/2007 (2:55 pm)
Constructor as a product (even free) is designed specifically for the large majority of artists that don't have source code licenses.
#28
12/19/2007 (8:07 pm)
If you are not going to release the source code then how about releasing some documentation. There are huge blank spots in the TDN on constructor. In particular I would like more documentation on how to use the light entities (spotlight, runway light, point lights, etc.). 
#29
12/20/2007 (12:31 am)
Even if they do decide to release the source code for it, their most likely in no position to even consider it right now. 
#30
12/20/2007 (12:37 am)
Is there any chance to at least get access to the precompiled scripts so one could potentially attempt to fix the glitches on that end? 
#31
I guess a solid Constructor pipeline would be nice for someone to post please - please?
I just don't want get stuck 3 weeks into make a building and find I built it wrong to work with Constructor.
12/20/2007 (1:34 am)
Hey did every one forget that they gave us Constructor for free??? Instead having to use QuArK? I'm happy with that - GG go ahead and do what ever you like. Torque tech is such good value, there is no reason to complain. I guess a solid Constructor pipeline would be nice for someone to post please - please?
I just don't want get stuck 3 weeks into make a building and find I built it wrong to work with Constructor.
#32
www.garagegames.com/mg/forums/result.thread.php?qt=66044
and this happens on the simplest of modules and the most complected of modules and is is always the X axis in constructor portals..
12/20/2007 (8:18 am)
One thing that I can't get away from is the bad ghosting that the portals render that of all things I wish were fixed the rest I think is up to the artist to tweak and fix like all other modules of this tip but the ghosting of the portals is something that isn't the artist and I wish this would get fixed.... thanks www.garagegames.com/mg/forums/result.thread.php?qt=66044
and this happens on the simplest of modules and the most complected of modules and is is always the X axis in constructor portals..
#33
Its important that the player can not see across multiple portal zones. If you have a place like a long hall, dont put portals along the long hall.
Portals MUST be square or rectangle, you can NOT make a functional portal if any of the portal's contacting walls do not meet at 90degrees. (Unless the opening is small, seems trapezoid portals work sometimes when they are less then 10 Torque Units tall.)
Multiple Portals along one plane will often lead to problems. Think of a multi level building with a wall of windows. In this case ONE big portal 'may' work best.
Think long and hard before placing portals, portal misuse, overuse will not help your performance, and might lead to hours of extra work.
Sometimes portals will NOT work no mater what, be sure all near geometry is clean and simple. And sometimes a portal works even when you TRY to build it wrong, often if the BSP project is small with only ONE portal.
Thats my findings to date, hope it holds true for others....
12/20/2007 (9:50 am)
I have invested alot of time Constructor debugging, and investigating BSP as TORQUE see it, here are some of my portal findings... Its important that the player can not see across multiple portal zones. If you have a place like a long hall, dont put portals along the long hall.
Portals MUST be square or rectangle, you can NOT make a functional portal if any of the portal's contacting walls do not meet at 90degrees. (Unless the opening is small, seems trapezoid portals work sometimes when they are less then 10 Torque Units tall.)
Multiple Portals along one plane will often lead to problems. Think of a multi level building with a wall of windows. In this case ONE big portal 'may' work best.
Think long and hard before placing portals, portal misuse, overuse will not help your performance, and might lead to hours of extra work.
Sometimes portals will NOT work no mater what, be sure all near geometry is clean and simple. And sometimes a portal works even when you TRY to build it wrong, often if the BSP project is small with only ONE portal.
Thats my findings to date, hope it holds true for others....
#34
I have seen friends working on Unreal Engine, every tool is ready for final use. What they did is only change a tiny bit of the source code, and put their main time on arts, scripts.
Until today, TGE and TGEA had a quite bad support on multi-language support, if GG can not handle the utf-8 encoding and Chinese/Japanese/Korean/Thai input in the engine, why does not GG use the way as same as other engine did? Like Half-Life 2, Unreal, Call of Duty, Company of Heroes, etc.
Maybe Torque is only suitable for making simple and small games, if you want to make one game simular as Call of Duty, it will cost you a lot of time and make you very painful because there is no suitable tools as same as the ones provided by expensive engines.
If someone thought the cheap price should be meaning nothing and that is the position GG would like to be, I have no word to say.
12/20/2007 (8:51 pm)
The thing is not about it's free or not, I would rather it's NOT free. GG always did things half or not compitable for the final use, then they went on to develop the next version.I have seen friends working on Unreal Engine, every tool is ready for final use. What they did is only change a tiny bit of the source code, and put their main time on arts, scripts.
Until today, TGE and TGEA had a quite bad support on multi-language support, if GG can not handle the utf-8 encoding and Chinese/Japanese/Korean/Thai input in the engine, why does not GG use the way as same as other engine did? Like Half-Life 2, Unreal, Call of Duty, Company of Heroes, etc.
Maybe Torque is only suitable for making simple and small games, if you want to make one game simular as Call of Duty, it will cost you a lot of time and make you very painful because there is no suitable tools as same as the ones provided by expensive engines.
If someone thought the cheap price should be meaning nothing and that is the position GG would like to be, I have no word to say.
Torque 3D Owner Stephen Zepp
When we have accurate, solid information, we will discuss it with you..
Right now, anything at all we say is pure speculation/best guesses. Do we want Constructor to work on every platform, for every user, in every workflow, every time? Sure we do.
Can we promise exactly when all of that is going to be done, if ever? Not with any amount of consistency or accuracy.
I could come out and say that we are currently working on XXX different upgrades/ports of various product lines simultaneously, with varying degrees of accuracy as to expected release dates--and if we miss even one of them we would get nailed for it--it's what has happened in the past.
For the record, David Blake is the official Community Manager for GarageGames.com. Yes, he speaks with a "GG approved" voice on the forums. We do our best to keep him up to speed on what's going on (he works remotely), as well as what he can, should, and should not talk about/confirm. Sometimes he's going to get something wrong (just like all of us have from time to time), and /or not have the latest information from us, so please don't pester him unnecessarily!