Spam thread for new NAVS content

[Erick_Cantu]

Some preliminary alpha work gives an idea of what the shed windows will look like:

[Erick_Cantu]

For those of you who haven't been following on the FizzBizzle, I have been in the process of moving and, as such, have been working on a more low-effort project: a prototype for a new series of generic boxcars optimized for photo textures. These feature simplified models, a slightly different mapping layout (with both sides of the car mapped to the same spot), and a paint kit without any car-specific details on it. A whole lot of optimization work was done on the brake parts in particular, and I like the revised brake rods and levers so much that I might actually incorporate them into all of my new models.

The first car in the series, which is serving as a prototype while I work out best practices, is a modern Plate B car with non-terminating ends and a peaked roof. For the moment, I am using temporary photo textures from an old Curphey car while I build the paint kit. The paint kit is structured differently than prior projects because this model is optimized for photo textures. The shadow layer's primary purpose is just to ensure that the right shadows are cast from the model's parts. It has details for the brake layers and the underframe, but the roof, ends, and sides of the carbody are pretty featureless. The highlight layer is set up similarly.

There are 3 roof options, StanRay X-Panel, StanRay diagonal, and P-S. These are pre-rendered layers containing both highlight and shadow. They are set to hard lighting so that a single layer casts highlight and shadow on anything under it. I can get away with this because nothing in these layers needs to be pure white or black - a single layer set to hard lighting has a much lower dynamic range than a separate shadow and highlight layer. I tried combining the base shadow/highlight layers into a single hard lighting layer this way, but the bright parts weren't bright enough and the dark parts (e.g., the black areas simulating the coupler pockets) weren't dark enough.

There is an optional metal layer for the roof. I might supply a few different metal layers. Since it can be a pain to get good photos of car ends, I am going to pre-render some end corrugation options the same way. The sill extends pretty far down on the model, so you can edit it with the alpha channel as needs dictate. It's low enough to do giant SIECO or ACF sill profiles if need be.

Ultimately, the painter just needs to paste photos below the shadow and other layers, using the included wireframe layer for reference. Each car uses a single texture map. These models are designed for a more "conventional" approach where the painter copies the model, renames it, decompresses it, edits the texture names, and recompresses it. Each car uses a single 2048 x 1024 pixel map. Originally, I planned on not having truck options, but lots of Plate B and Plate C cars were offered in 70 or 100 ton configurations, so I'll probably make the trucks separate anyway, with multiple physics files to serve the painter's needs.

[Erick_Cantu]

These shots show what the car looks like with a grey base layer instead of the photo test layers. Note how the sides and ends lack detail. I used the P-S roof layer here and hid the metal roof layer.

[eric]

The test version of TSRE v8.005 should support the rectangular texturing.

Sent from my SM-S911U using Tapatalk

[CSRX]

I personally cannot wait to be able to convert some of my plate C boxcars over to this new model. I have some really good photo textures cars of Plate C I would hate to see lost forever.

[rdayt]

Nice idea Erick.
This is going to find a lot of uses in the hobby.

Randy

[johnnyboy99]

for Canadian National's "multilingual" boxcars and reefers and CP's multimark being on the B end of both sides wouldn't a separate texture for the other side work?

[Erick_Cantu]

Where are you going to put it? There isn't any room on the map. Obviously, this model won't work for every case. Most of the time, people doing photo paints only have one car side photo to work with, so it made no sense to make the usual compromises on resolution elsewhere (e.g., the frame) to map the other side separately.