I kinda think of Mr. Bouduin as THE engine file "guy" in the train-sim world.
The fact that he created physics for these models of his own volition means a
great deal to me. The only trouble now is making sure the M10000 is worthy of
his efforts.July 18th, 2007
Perhaps the biggest news, for me at least, is that Bob Bouduin sent me a set
of engine files he created for the M10000. I had been planning to ask him if he
would mind doing so. Instead he just kinda showed up with the files already
prepared. This is the second time Bob has bestowed this honor on me. The DL109
series all have Bob's engine files.
I kinda think of Mr. Bouduin as THE engine file "guy" in the train-sim world.
The fact that he created physics for these models of his own volition means a
great deal to me. The only trouble now is making sure the M10000 is worthy of
his efforts.
The other piece of news is that I had the opportunity to run MSTS on one of those "new fangled" dual core computers. What a difference!
Now I'm dreaming up some way of collecting enough pennies to replace my tried and trusty Compaq SR1220NX. While I'm at it, maybe I should replace my 21" monitors with those nice panel displays. Better performance and more desk space in one package.
The City of Salina keeps progressing. I have the cafe car and the coach complete. The interiors of both cars are well under way, but not as slick as I think they can be. Since the interiors can actually be a separate shape file from the rest of the train, I'm hoping I can add a lot more 3D detail. Something that looks close enough to the real thing to make you feel that you really might be riding on the M10000.
I've done car interiors before. The observation cars built for Paul Fowler had pretty detailed interiors considering they were my first. Right now the M10000 interiors are about the same level. I hope to do better this time around.
I'd like to some depth to the walls for one. Show the banding as separate parts, make more realistic seats, and include the overhead lighting. Maybe I can use a brighter material to represent the pools of light on the ceiling. I'm not sure how far it can be taken. I guess that's one reason this is fun, finding out what's possible and how creative you can get within the limits of MSTS.
One thing that has plagued me is the nose of the Motor Car. It's gone through several revisions. I think this is the fourth incarnation. Each version has little flaws that bother me. On the latest manifestation the "cheeks" of the cab, the transition area under the rear cab window, look pretty good to me. There is a kind of parting line visible in the shape just ahead of the red stripe. I need to smooth that out a bit.
There is also that shadow ahead of the upper part of the door. This is caused by using polygon smoothing on a part that contains both curved and flat surfaces. The flat parts end up being rounded visually and the shadow is part of that. The same thing happens if you create a cylinder with flat ends and then use the polygon smoothing. The ends will appear rounded, despite the fact that they are perfectly flat. To cure it I just make flat parts separate parts without the smoothing.
The other thing that still isn't quite right is the curved piece just under the cab windows. It should run directly into the line of the roof that slopes down from above the floor. On the real M10000 this line is actually a distinct curve. On the model it's quite a bit flatter. I'm not sure I can duplicate it without adding another few hundred polys to the model. I'll settle for having the roof line run a little smoother into the line under the window.
The big issue coming up is how to include all the rivets present on the real train. I'm not sure how to go about creating them. I'll probably have to use full side and roof textures, something I used only on the original BETA model. Most of this model is textured with little squares of color. The only well defined areas are the circular patterns at the front and rear of the train. It's a different approach to texturing that was used minimally on my earlier models. The Baldwin and DL109 series uses it a bit more for some areas. Here on the City of Salina it's probably overused. It might be fine for a model of fully welded equipment, but the only way I'm going to capture the rivet detail of the real thing is to fall back to the more conventional technique used with the BLW and ALCO diesels.
The hard part will be getting those rivets and other details to look right on the curved sections of the cars. Keeping the distortion to a minimum may prove to be a challenge
This is the first model I've created a cab view for. I read up a little and found that the original KuJu cab views were all created using 3D models of the cab. So I took a shot at this method. The interiors of these models are not as hard as it might look. The body shell is just a copy of the exterior parts with all the polys flipped and retextured. It would be pretty cool to be able to use these in a manner similar to the interior vies of the cars. Unfortunately it doesn't work the way it sounds.
The cab views are screen shots of the detailed 3D cab interiors, not much different from using a 2D photo of the real thing. The first shots I used I created using screen dumps fro Shape Viewer. This has a drawback however, because it tends to view the model from the origin point of the model. It's kinda hard to get the right position.
My second attempt was a free hand 2D estimate of what the cab might look like. It was basically a big green wall with square openings, a far cry from anything realistic.
I found myself wishing that I had used 3D Canvas to create this beast. In 3D Canvas you can change the position of the virtual camera used to view the model. This means you can actually position the camera in the very point in the cab were the engineers head might be. It was then that I remembered that TSM has the same type of feature. The "camera" is positioned at whatever point you have indicated as the "center" of the view. So I set the approximate point where I thought an engineers eyes might be and took a screen shot.
As you can see it has a little more character that a green wall with holes. I should probably move the camera closer to the window. Someplace that I can see the right and left cab windows from. I can just turn the view and take another screen shot for each direction in the cab. After that it's "just" a matter of detailing and texturing the cab and shooting it again.
I found my biggest obstacle to creating cab views was the MSTS Cab Editor itself. I can not get the hang of the thing...Actually after the first few attempts I never tried again. I found it MUCH easier to work with the CVF file in a text editor. After years of troubleshooting bits of code and more years of looking at MSTS ENG files it made a helluva lot more sense than the goofy Cab Editor.
A quick look at the opening section of the CVF file will explain what I mean...You can see that there are entries (or sets of entries) that describe size of the view, the ace file to use, where it is positioned, and what direction it is turned. The general structure has a lot in common with the other parameter files used in MSTS.
Tr_CabViewFile (
CabViewType ( 2 )
CabViewFile ( CabFront.ace )
CabViewWindow ( 0 0 2048 1536 )
CabViewWindowFile ( ../../GP38/cabview/AcWndFrn.ace )
Position ( 0.5 3.5 9.3 )
Direction ( 11 0 0 )
CabViewFile ( CabLeft.ace )
CabViewWindow ( 0 0 2048 1536 )
CabViewWindowFile ( ../../GP38/cabview/AcWndFrn.ace )
Position ( 0.5 3.5 9.3 )
Direction ( 6 -73 0 )
CabViewFile ( CabRight.ace )
CabViewWindow ( 0 0 2048 1536 )
CabViewWindowFile ( ../../GP38/cabview/AcWndFrn.ace )
Position ( 0.5 3.5 9.3 )
Direction ( 6 86 0 )
EngineData ( cnj78 )
CabViewControls ( 16
Dial (
Type ( SPEEDOMETER DIAL )
Position ( 118 161 5 25 )
Graphic ( ../../CLWCabviews/RF16Cab/needleB.ace )
Style ( POINTER )
ScaleRange ( 0 120 )
ScalePos ( 204 150 )
Units ( MILES_PER_HOUR )
Pivot ( 17 )
There are still a few items I have to work out but that's the gist of "CVF Files 101"
Well I think that's about it for now...Thanks for listening.