DotNetTorque Update, Status and Release Info
by Vince Gee · 05/09/2012 (9:08 am) · 6 comments
Good Afternoon GG Community!
I know it has been some time since I last blogged on the state of DotNetTorque. Do not misinterpret the fact that I have not blogged as a sign that development stopped or slowed. For the most part, I have been working through scripts which are not exactly the most interesting types of files, and it has become a chore to work through them.
Right now, I have 2 files left in the PostFx folder. Once these last two files are converted to CSharp I will have just one small change left to do to be finished. Then finally, I will be able to say pass one is done.
I am currently happy with performance, the CSharp structure on the other hand makes me frown a bit. To understand why it makes me frown, you need to understand the situation I was in when I wrote it.
I had two choices when I started to re-write the FPS Starter. Choice one was to rewrite everything using an MVC model, using the Torque DLL as the View and the CSharp as the Model and Controller. The second choice I had was to make it look identical to the TorqueScript structure.
Simply for the sake of familiarity for existing TorqueScript coders, I chose to follow the Starter.FPS torquescript structure. This way, someone who picks up our product could open the Starter.FPS inside of Torsion and open up our project in MS Visual Studio and compare them side by side to see a direct translation of TorqueScript to CSharp.
For the most part, the code is very similar to the TorqueScript except where I used some stock CSharp libraries which performed some operations faster than the Torque DLL implementation.
I plan to finish the last two scripts this week, get the demo server online and the download up this weekend. It will be a public download which anyone will be able to grab. The purpose of this first release is for load testing and bug reporting, as well as letting the community see how everything performs.
I am hoping that I will be able to stomp out the bugs and issues over the next month or so and we will be looking at a true release coming around next quarter. Once this build is stable and documented, I will then be able to move on to re-writing everything into the MVC model for a dedicated server and client. This will be the build which the OMNI engine will start from.
A lot of cool stuff is happening at Winterleaf Entertainment L.L.C. from coding to artwork. I cannot express my pride in being a founding member of the organization enough.
Have a Great Day!
Vince
I know it has been some time since I last blogged on the state of DotNetTorque. Do not misinterpret the fact that I have not blogged as a sign that development stopped or slowed. For the most part, I have been working through scripts which are not exactly the most interesting types of files, and it has become a chore to work through them.
Right now, I have 2 files left in the PostFx folder. Once these last two files are converted to CSharp I will have just one small change left to do to be finished. Then finally, I will be able to say pass one is done.
I am currently happy with performance, the CSharp structure on the other hand makes me frown a bit. To understand why it makes me frown, you need to understand the situation I was in when I wrote it.
I had two choices when I started to re-write the FPS Starter. Choice one was to rewrite everything using an MVC model, using the Torque DLL as the View and the CSharp as the Model and Controller. The second choice I had was to make it look identical to the TorqueScript structure.
Simply for the sake of familiarity for existing TorqueScript coders, I chose to follow the Starter.FPS torquescript structure. This way, someone who picks up our product could open the Starter.FPS inside of Torsion and open up our project in MS Visual Studio and compare them side by side to see a direct translation of TorqueScript to CSharp.
For the most part, the code is very similar to the TorqueScript except where I used some stock CSharp libraries which performed some operations faster than the Torque DLL implementation.
I plan to finish the last two scripts this week, get the demo server online and the download up this weekend. It will be a public download which anyone will be able to grab. The purpose of this first release is for load testing and bug reporting, as well as letting the community see how everything performs.
I am hoping that I will be able to stomp out the bugs and issues over the next month or so and we will be looking at a true release coming around next quarter. Once this build is stable and documented, I will then be able to move on to re-writing everything into the MVC model for a dedicated server and client. This will be the build which the OMNI engine will start from.
A lot of cool stuff is happening at Winterleaf Entertainment L.L.C. from coding to artwork. I cannot express my pride in being a founding member of the organization enough.
Have a Great Day!
Vince
About the author
www.winterleafentertainment.com
#2
Great work guys
05/09/2012 (3:13 pm)
Will you be able to run torquescript and csharp scripts along side each other, or will all scripts need to be converted to csharp?Great work guys
#3
05/09/2012 (4:10 pm)
the way that DotNetTorque is written, youll be able to run Torquescript and C# (or VB.NET for that matter) along side each other. 
#4
I have started looking through the TS code to determine what is really needed and what is not. Even that is a daunting task somewhat, so I cannot imagine redoing the whole engine like you have.
Hey, there is one really cool use you can now use C# for. I realized my code would do it as well. Since you really can control the engine externally with a very nice language, you should be able to do testing on pieces of the engine in a very systematic way. I am not sure if this would be called "unit testing", but it does allow for more control of the engine interactions for some benchmark and stress tests. That is part of the reason I want to control startup so much is to build a test suite for all the different pieces. Also, by doing so I will be forced to learn the inside and out of Torque.
Good job on the progress of your C# interface. I am guessing you are playing a lot of Queen songs: "Another one bytes the dust...and another one does, another does..." as you go through the remaining files!
05/09/2012 (6:33 pm)
Awesome! Can't wait top see the results. Will it be a binary we can run to see how quickly it runs? That is what I want to see. Kind of give me a bench mark to compare results for stock torque versus a C# rebuilt Torque. I have started looking through the TS code to determine what is really needed and what is not. Even that is a daunting task somewhat, so I cannot imagine redoing the whole engine like you have.
Hey, there is one really cool use you can now use C# for. I realized my code would do it as well. Since you really can control the engine externally with a very nice language, you should be able to do testing on pieces of the engine in a very systematic way. I am not sure if this would be called "unit testing", but it does allow for more control of the engine interactions for some benchmark and stress tests. That is part of the reason I want to control startup so much is to build a test suite for all the different pieces. Also, by doing so I will be forced to learn the inside and out of Torque.
Good job on the progress of your C# interface. I am guessing you are playing a lot of Queen songs: "Another one bytes the dust...and another one does, another does..." as you go through the remaining files!
#5
Yeah, the beta we will be releasing will just be like a released game of starter.fps, source will not be released. The next beta will be source code.
Also, Frank, I know your busy on your own project, but if your interested on getting your hands on this a bit earlier then everyone else, ping me and I will get you the full source for you to look over it. (You'd need to sign a NDA) I wouldn't mind someone with your expertise looking through the code.
Let me know,
Vince
05/10/2012 (5:34 am)
@Frank,Yeah, the beta we will be releasing will just be like a released game of starter.fps, source will not be released. The next beta will be source code.
Also, Frank, I know your busy on your own project, but if your interested on getting your hands on this a bit earlier then everyone else, ping me and I will get you the full source for you to look over it. (You'd need to sign a NDA) I wouldn't mind someone with your expertise looking through the code.
Let me know,
Vince
Associate Ross Pawley