1.4 Status/Problem solving Thread.

#81 After using search and coming back empty handed I decided to post an inquiry. I wasn't quite sure where to post this but it seemed as though this would be a good place to do so.
On to the problem at hand; I recently downloaded the 1.4 release from CVS and decided to rework my prototype using the starter.fps since my original project was lost and the fps starter kit is the closest thing to what I need. I created a new mission and made some changes and attepted to save it as I would normally would have done with the previous release but when I click on "save" or "save as" I get a pop up with the message "Save Function Disabled", "Save Mission function has been disabled for this demo walkthrough.", "If you want to save make changes to a mission and save them, try the FPS demo.", this was while I was in the FPS demo so I am a bit confused as to what I am doing wrong. I was wondering if anyone could tell me what I am doing wrong, if there is a workaround for this (i.e. what was changed), or if I should setup a new folder and work from there instead of working with a copy of the example folder much like you could do with previous releases. I appologize if I missed something and thanks in advance.

Nevermind think I fixed it myself without much effort forgot to check the EditorGui.cs in creator\editor instead of common\editor

#82 @Caelumira: Glad you got things fixed!

@Everyone else: Thank you so much for the bug fixes! 1.4 is under code freeze now, so it's just bug fixes that I'll be doing for the next week or so.

#83 Ben, I just checked out the latest CVS and notice that no bug fixes for the getSteeringAngle method in the AIWheeledVehicle code have been included. I just want to call your attention to that again as I use it and it's badly broken in about four different ways, as I recall. I posted about this in this thread and then made a resouce out of the changes I did that seem to work well and address the problem in a fundamental way. There's also a graphic visualization tool included in the resource (which is what the resource was originally about) but the important part is the steering code changes. The broken code caused me some head scratching so it would be nice to see it fixed in the release, if AIWheeledVehicle.cc is to be a part of that release.

#84 I don't know where the error is, but I heard that you would be fixing the waterblock bug when a map has a square size other than 8.
Is that true?

Also, some kind of compiler that comes with torque would be great, I am using TBE and whenever I edit any of the source code with a resource it doesn't compile.

Just my 2 cents.

#85 Something bad with setOrbitMode
I created AIPlayer and set serOrbitMode to this AIPlayer like this
%this.camera.setOrbitMode (%aiPlayer, %aiPlayer.getTransform (), 1, 10, 5, 0).
When I set %aiPlayer.setMoveDestination (x, y ,z), aiPlayer move but with same strange vibrate, when I set %this.camera.setOrbitMode (%aiPlayer, %aiPlayer.getTransform (), 1, 10, 5, 1), and set setMoveDestination aiPlayers move ok, but PlayGui strange blink

#86 Ben, how about putting the fxshapeReplicator bug fix into 1.4 or next HEAD, rather than leaving it as a resource.

#87 I agree with Duncan on the replicator bug fix!

Looking forward to 1.4.....

#88 I have a problem with the latest release I got from the HEAD. I know that it is in this version because I went with a clean build completely free of any changes I had added and the same problems persist. I also tried it with just the demos you release along with it.

The latest OpenAL32.dll causes a message that says "TheOpenAL system failed to initialize. You can get the most recent OpenAL drivers here." The word "here" is a link that just brings me to the buy Torque screen.

I happen to be positive I have the latest drivers for my system.

#89 Oh yeah and putting in one of my older versions clears the problem right up.

#90 I'm having the same problem.

;)

Eric

#91 It uses the local OpenAL32.DLL, not whatever drivers you have installed - AFAIK. It seems that CVS likes to clobber binary files (it's done several images too), so I'll probably get to spend some time hacking that up. Right now I'm working on getting the demo & installer built.

I'll definitely check out the fxShapeReplicator bug fixes. We're pretty much in code freeze mode but there might be time for one or two more bugs.

#92 @Ben,

Have not read the entire thread; but for linux we do not use the OpenAl.DLL, instead we haev our own loading code that looks for libopenal.so

@All,

Well here is the issue with Linux OpenAl, the current code looks for libopenal.so in 3 places
1. CWD/lib
2. CWD/tlib
3. CWD
4. the system (normal linux way of loading libs)

all 4 paths fail .. 1 - 3 fail because there is not libopenal.so in the mentioned directories and 4 fails because (atlease on my system) there is no libopenal.so, however there is a libopenal.so.0

I resolved this by copying libopenal.so from torque/lib/openal/LINUX to torque/example


The loading code can be found in engine/platformX86UNIX/x86UNIXOpenAl.cc line 207 - 266

This can be resolved in one of two ways for 1.4.0
1. copy the file to the example dir and check it into SVN sot hat it is pushed with CVS
2. reference libopenal.so.0 so that it can be found using standard lib paths under linux (usualy resides in /usr/lib)

-Ron

#93 @All,

1.4.0 Linux update:

First and foremost I want to give a shout out to Xavier and Gregory, they realy helped me get the Linux code base up to snuff!!

I am putting the final touches on the linux code base as I write this (its actualy compiling ;)), going to-do some quick nUnit testing (what little I can on this cruddy laptop which gets a whopping ~10fps!!!!)

the change list is rather long and extensive (thanks to Tom B and his stinking Unicode ;) Bloody Hell you wanker!)

I'll be pushing my changes into SVN this evening when I get home from work (had to go into the office to get some 'alone' time to get this puppy done!) and building the installers Sunday (want to-do some testing, maybe Get Ben G or Xavier to help me seeing this laptop blows chunks)

Have questions, comments or concerns e-mail me or post here


@Ben/GG:

Sorry this has taken so long to get done, been a b!tch findign time with the wife working from 7PM - 2AM, work busting my nads and makeing me work til ~6:15 nightly and oh yeah kids... *sigh* I have had a blast working on this project and hope we can continue working together :)

-Ron

#94 I am a WindowsXP user. I had a problem with the openAL driver not initializing after using WinCVS to get the latest version of TGE. I went to the openAL site and downloaded the Windows installer. It seemed to fix my problem immediately.

http://www.openal.org/openal_webstf/OpenALwEAX.exe

#95 @Edward, I'm also using XP (Home) but TGE 1.3 and the old 59k size Openal.dll was working fine for me. I just tried the link you provided and that file (84k size) only plays the Gui sounds and does not play in-game sounds. I have not installed TGE 1.4 so can't comment on that.

I don't know if any audio changes were made from 1.3 to 1.4 which could be causing the problem. I look forward to seeing the change list.

#96 @Ron does the current code in CVS compile on Linux under GCC4? Also want/need help with Linux, I do have a little experience in that department, would be glad to volunteer.

#97 @Dreamer,

From the feedback I have received TGE compiles like a charm using gcc >= 3.4.*
I will keep you in mind next time we have to jump through some hoops for a release ;) it sure has been a circus this weekend trying to get things in order and the installer made/tested.. think I made a total of 5 installers this weekend *sigh*


-Ron

#98 @Duncan

I apologize for not investigating my solution more thoroughly. You are right. I don't have in-game sounds either.

However, I went to the console and discovered that I was missing two functions for the openAL audio. So, I went looking for a new openAL32.dll.

http://www.dll-files.com/dllindex/dll-files.shtml?openal32

After I downloaded the new dll and replaced my original openAL32.dll file, I am now hearing in-game sounds. I did notice that the original and the new files were different by 20kb (give or take). I also read elsewhere in these forums that the CVS server has been known to sometimes corrupt some binary files.

#99 Hey Ben,

The VC6 project is still not up to date. I just got a completely new HEAD build from CVS ('cause I wanted to start with a fresh slate) and all the Unicode, Vorbis path, and imm32.lib linking errors still seem to be there. Obviously I realize the true fix to that is for me to stop using an outdated IDE (which I plan to do this weekend) but it would be nice if we could get these things updated. On the otherhand maybe I'm just getting the wrong revision or something. If this is the case could someone point me in the right direction.

-Allen

#100 Hi Allen,

I was in the same boat as you until I finally gave in and upgraded to VC 7.1. It's been smooth sailing ever since.

Good luck,

Aaron E.