@machine: I already have the banking transitions figured out (I think) as part of the elevation profile of the track. Those will be modeled as Bezier curves. Since variable-radius corners seems to be in high demand, I'll see what I can do about implementing the Euler Spirals. ---------------------------------------------------------------------------------------------------------------------------------- As a quick progress report, I've been spending the past few weeks implementing DirectX as a DLL for the track editor. Modularizing the code has taken second priority (a distant second) to getting the rendering engine working.
Several things have come up recently which have forced me to put this track editor on hold. I'll post back as soon as I can resume progress on the track editor.
Okay, I just found this thread on this forum and have just recovered my account. There has been an update to the progress of this track editor on YouTube, here. I will need a few days to get myself reoriented to this new forum and the changes that ISI has made to the system concerning rFactor 2.
Welcome back! I will only say that you are my latest hope in order to get a proper track building tool. I already lost all my faith in RTB. Lack of control is what better describes it. Well, that and lack of support and user orientation. Today using BTB is still my option if I start a new project. You have all my support if you need it.
Oh man. I hope you recover/ had copies of your source code. The most devastating sim editing thing I had happen was a hard drive literally going up in smoke way back in the day. I lost all of the source code to my GPL and N2XXX editing programs.
Don't worry, I do have a backup from when I first transferred the data over to the mac. Don't have any backups of the progress made after that (hence the "regress" report), but I still have a significant amount of progress saved on a USB stick. I got the computer back on the 6th and I'm still working on getting everything transferred back over. Also, right before the computer died, I discovered several things that'll greatly speed up development (hopefully).
Another update: I somehow managed to kill the Windows 10 partition (specifically). Fortunately, I was able to pull the track editor off that partition. Now I need to reformat and reinstall Windows 10. Before Windows 10 died, I was able to do some form of restructuring within the track editor code, and I have around 600 compile errors that I'll use to determine what I need to change to accommodate that restructuring.
I finally got my backup reloaded into Visual Studio and I've started tracking down those 600-odd errors. Down to 138 compile errors. And I have absolutely no clue if my optimizations will work once I start running the editor. On a side note, I discovered that WINE really only supports DX9 or lower. So I'll be changing the graphics API from DX11 to DX9 to make it run better on Linux and Mac systems. It's also the case that the virtual machine software tends to go haywire with anything above DX9. And since I'm now developing the editor on Windows 10, I need to use virtual machines to test the editor on Windows 7 and Windows XP. Also, DX9 is friendlier to older computers and is what rFactor 1 and NR2003 use. So by using DX9, I'm increasing the chances that whatever the track editor displays will be exactly what the simulator displays. I will not be able to test how accurate the track editor's rendering is for rFactor 2. EDIT: Down to 16 compile errors plus 4 false positives. Plus I have to find another means to get the Scene Explorer to communicate with the Properties Window.
I'm down to ZERO build errors on the track editor. But my changes have introduced a CTD bug that happens whenever the Scene Explorer is accessed in any way (eg: right-click menu, adding objects, starting projects). I must now track down the source of the error, but I have a guess as to where it is. EDIT: My guess was DEAD! WRONG! EDIT 2: Well, I fixed part of what was causing the CTD. Now I see that there is a problem with the old rendering engine (which is going to be discarded anyways). EDIT 3: Fixed the CTD with a temporary fix. I'll eventually put in a more robust, permanent fix later on.
Quick update: Apparently, I lost more data than I thought I did when the computer crashed. The "new" rendering engine is the un-optimized version of the rendering engine. I'm re-optimizing it now.
Hi everyone, new guy. So, with all your new issues, there's no eta on a beta? Your work looks so promising. Keep up.
Thanks. Actually, it's been a policy of mine not to post an ETA on the beta or release because I don't want to disappoint anyone if I miss it. But yes, the issues have been quite the set-back. Speaking of which, on the application I use to test the rendering engine, I had to rewrite it from scratch because the API I had been using decided to suddenly not work.
Will it be possible to do point to point tracks like rally stages? If yes , With timings ?, or like btb that needs to edit manualy the scn file and rename others like xfinish xpitin etc..?
I have actually thought about rally/drag tracks. For NR2003, such things are not possible. But for just about any ISI-based simulator, adding support for such things should be trivial.
I have successfully replaced the old DX11 rendering engine with the new DX10 rendering engine (no, that is not a typo) and it is working within the editor. Framerate unknown, but the test app gave me ~3,600 fps on a blank screen. I will now work on implementing the long-overdue mesh generator.
