I have to say, I'm not all that impressed so far
by Kurt Koller · in Torque Game Builder · 04/19/2007 (12:29 pm) · 33 replies
We have some developers working on a game that are on Mac and PC. The target for the game is MacOS. This is an installation type project and it has to deploy on Macs.
1. TGB's interface is really really really clunky.
2. Every time we load a project on the Mac, it loads a window, fades out, loads up another differently sized window, then fades out again and loads up a third window, finally with the game in it. sort of.
3. After loading on the Mac, the project is ALWAYS in a hugely wide resolution, and we need to change it to something reasonable, like 1280x800. Once we save and quit, it doesn't remember this on load.
4. To build for the Mac, on the Mac, choosing build does nothing. We have to do "analyze" which takes about 9 minutes on a dual-core 2.33Ghz MacBook pro, and THEN doing a build works. Wow, that build interface is klunky.
5. The Mac version sure crashes a whole lot.
6. The PC version LOOKS like it will build the Mac version, but fails with an "internal error." Not that big a deal.
7. On the PC version, when we copy the project from one machine to another, there appears to be virtually no way to get the project to appear in the project list. On the Mac, we copied it into the games directory with the install it shows. On the PC, we tried that, doesn't work. we ALSO tried copying it to that weird path under C:\Users\[user]\AppData\Local\VirtualStore\Program Files\TorqueGamebuilderPro\games\[game] - it doesn't show up in the project list. So we can't copy this to other developers on PCs. This really sucks for us. New projects that we CREATE show up here, so I'm not sure why this isn't working. Wow, my empire for a freaking browse dialog.
8. When running full screen on the Mac, going into editors it is sometimes IMPOSSIBLE to back out because buttons aren't on the screen. This happens at a resolution like 640x480.
There are a slew more, but I have to say this tool seems really really clunky.
The support is also nonexistent. I tried calling 4 times with questions about licensing, pleading for someone to call me back, and never got a single reply.
1. TGB's interface is really really really clunky.
2. Every time we load a project on the Mac, it loads a window, fades out, loads up another differently sized window, then fades out again and loads up a third window, finally with the game in it. sort of.
3. After loading on the Mac, the project is ALWAYS in a hugely wide resolution, and we need to change it to something reasonable, like 1280x800. Once we save and quit, it doesn't remember this on load.
4. To build for the Mac, on the Mac, choosing build does nothing. We have to do "analyze" which takes about 9 minutes on a dual-core 2.33Ghz MacBook pro, and THEN doing a build works. Wow, that build interface is klunky.
5. The Mac version sure crashes a whole lot.
6. The PC version LOOKS like it will build the Mac version, but fails with an "internal error." Not that big a deal.
7. On the PC version, when we copy the project from one machine to another, there appears to be virtually no way to get the project to appear in the project list. On the Mac, we copied it into the games directory with the install it shows. On the PC, we tried that, doesn't work. we ALSO tried copying it to that weird path under C:\Users\[user]\AppData\Local\VirtualStore\Program Files\TorqueGamebuilderPro\games\[game] - it doesn't show up in the project list. So we can't copy this to other developers on PCs. This really sucks for us. New projects that we CREATE show up here, so I'm not sure why this isn't working. Wow, my empire for a freaking browse dialog.
8. When running full screen on the Mac, going into editors it is sometimes IMPOSSIBLE to back out because buttons aren't on the screen. This happens at a resolution like 640x480.
There are a slew more, but I have to say this tool seems really really clunky.
The support is also nonexistent. I tried calling 4 times with questions about licensing, pleading for someone to call me back, and never got a single reply.
#22
www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=9586
05/23/2007 (7:22 pm)
Here's a .plan from Justin that somewhat explains what's going on during the analyze phase:www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=9586
#23
Unity knows whats in the project, Unity also pre-compiles all the scripts when they are created
TGB does not know what is in your project and does not try to assume it does either, TGB also re-compiles all your scripts at Build time to ensure they are clean ...
Is Unity faster ... maybe ... for a really large project, or a project with a lot of scripts ... I would NOT suggest using the Analyze and Build process in TGB ... you'd be better off using a custom build system you write yourself ... the analyze process is not necessary ... as you, the developer, know what is in your project ... the Build portion ... take a look at the source for it ... see what it does ... generate your own 'project file' for TGB ... OR ... just pre-compile all your scripts with Torsion and remove your source from the directory and call it a 'release build' ;)
05/23/2007 (7:27 pm)
The major difference in the build times for the project between Unity and TGB are this ...Unity knows whats in the project, Unity also pre-compiles all the scripts when they are created
TGB does not know what is in your project and does not try to assume it does either, TGB also re-compiles all your scripts at Build time to ensure they are clean ...
Is Unity faster ... maybe ... for a really large project, or a project with a lot of scripts ... I would NOT suggest using the Analyze and Build process in TGB ... you'd be better off using a custom build system you write yourself ... the analyze process is not necessary ... as you, the developer, know what is in your project ... the Build portion ... take a look at the source for it ... see what it does ... generate your own 'project file' for TGB ... OR ... just pre-compile all your scripts with Torsion and remove your source from the directory and call it a 'release build' ;)
#24
A 9-minutes single level analysis seems long. you're scanning files for strings that might be files and checking to see if they exist, right? So for a handful of source files that should take 9 minutes on a 2.33Ghz Core Duo? Me thinks something's off there, I dunno.
David H - yeah I think you're right, just precompile, then copy and remove the source. Good idea.
05/23/2007 (7:48 pm)
Dan - sorry, that doesn't make any sense.A 9-minutes single level analysis seems long. you're scanning files for strings that might be files and checking to see if they exist, right? So for a handful of source files that should take 9 minutes on a 2.33Ghz Core Duo? Me thinks something's off there, I dunno.
David H - yeah I think you're right, just precompile, then copy and remove the source. Good idea.
#25
Dan, yeah that was a previous show. The new one has a whole different feel. I'll upload a link to the photos of the new show once we have them.
Here are some screenshots:
flickr link
05/24/2007 (5:01 am)
I got an email about Dan posting a new comment to this thread, but I don't see it here.Dan, yeah that was a previous show. The new one has a whole different feel. I'll upload a link to the photos of the new show once we have them.
Here are some screenshots:
flickr link
#26
now I need to make another game based on this and release it publicly.
05/30/2007 (5:20 pm)
Hey guys, client loved the game, thanks for the help all of you.now I need to make another game based on this and release it publicly.
#27
05/31/2007 (10:16 am)
In the beta (not sure if it's in 1.1.3) the build step has been re-written, you may want to check that out if you haven't already. There were definitely some areas of the analyze that repeated scans that increased build time dramatically.
#28
06/19/2007 (6:20 pm)
Very cool looking screenshots. I really like the custom arcade-style cabinet/kiosk. 
#30
Really liked the screen shots of your game. :)
How much time and people it took you to create the game ?
I am going to buy TGb in few days( after 1.5 comes out), especially for making such games. Your feedback will give me a rough idea how much time and effort I need to spend developing such games.
Hope to see more games from you soon.
Regarding the analyze timing problem, I don't think its that a big deal . I am used to sitting many days for rendering 3d animations, 10 mins or even an hr of compilation is not that bad for me. I prefer a good high quality game to be created and every 2d engine I have used could not provide me the speed and non laggy game play torque can provide.
06/20/2007 (8:30 pm)
Hey KurtReally liked the screen shots of your game. :)
How much time and people it took you to create the game ?
I am going to buy TGb in few days( after 1.5 comes out), especially for making such games. Your feedback will give me a rough idea how much time and effort I need to spend developing such games.
Hope to see more games from you soon.
Regarding the analyze timing problem, I don't think its that a big deal . I am used to sitting many days for rendering 3d animations, 10 mins or even an hr of compilation is not that bad for me. I prefer a good high quality game to be created and every 2d engine I have used could not provide me the speed and non laggy game play torque can provide.
#31
Honest (not sarcastic) question/request:
Given how things worked out in the end, is there anything that you want to change, or reinforce, in your original post here?
We really do value feedback, prefer it constructive, but nonetheless want to hear it if it's honest, and it sounds as if your project wound up a success--it would be good for us to hear how things wound up "after the fact" regarding your TGB experiences, good, bad, and indifferent.
06/20/2007 (9:54 pm)
@Kurt:Honest (not sarcastic) question/request:
Given how things worked out in the end, is there anything that you want to change, or reinforce, in your original post here?
We really do value feedback, prefer it constructive, but nonetheless want to hear it if it's honest, and it sounds as if your project wound up a success--it would be good for us to hear how things wound up "after the fact" regarding your TGB experiences, good, bad, and indifferent.
#32
06/21/2007 (11:59 am)
I'll second what Stephen said, any additional constructive feedback you could give would be invaluable.
#33
06/21/2007 (12:08 pm)
Absolutely. I will write something up over the next few days. 
Associate David Higgins
DPHCoders.com
Kurt, Analyze does just that ... it anaylzes your code and looks for dependencies ... scripts, images, shapes, particles, layers, etc ... it sifts through all the code, finds the references ... then includes them ...
This allows for a good amount of 'oops, it left important stuff out' issues to arise, especially if you build your level file names dynamically, for example ... as it can't interpret that ... so 'levelXXX.t2d' files would be left out if you just simply referenced them as "level" @ %level @ ".t2d" ...
Could TGB handle this better? Sure ... will it? I'd be willing to bet it's on the 'todo' list ...