Page 1 of 2 12 LastLast
Results 1 to 10 of 12

Thread: Bridge set being started

  1. #1
    Join Date
    Jun 2013
    Location
    Minnesota
    Posts
    1,615

    Default Bridge set being started

    I'm starting a set of various types of bridges for use in routes. The set is in very early stages (three types are started: glulam and sawn-lumber road bridges, and a concrete trestle rail bridge, plus an old wooden pile trestle I never released), but I want to get some opinions on how much detail to go for in these things. There are specifically two things I'm not sure if I should include:

    • Fasteners on wooden bridges, since many are visible. Also rivets on steel bridges. I may paint them on, or use alpha'd planes to mark them. No way will I 3D-model every rivet on a plate-girder span...
    • Is it worth making steel beams 3D? By this, I mean, if a beam is made out of 3/8" steel, should I make the ends of the beam 3/8" thick, or should I just make it flat (no thickness, like sheet metal). I suppose the thickness plays a part in the answer, since it would be absurd to flesh out an 1/8" beam, but it would be a given to do a 1" beam (if one exists).


    Of course, here is a picture of the glulam shape in gmax (the textured one), and the sawn-beam bridge (not yet textured)
    Glulam.jpg
    sawn.jpg

  2. #2
    Join Date
    Dec 2005
    Location
    .
    Posts
    15

    Default

    Regarding your 2nd bullet, if the steel beams are I-beams they should be continuous at the pier(s) and the ends would be invisible at the abutments. However the top and bottom of the "I", the flanges, would have enough width to be worth modeling to near true scale (width goes up with web height, which goes up with unsupported span). And when approaching the bridge, the bottom flange in particular would be visible. Based on similar objects' behavior in the sim, would experimenting be needed to see if on the approach this flange has a stairstep look? It could be that the appearance is better with the flange as a 2D object instead of a thin, 3/8" height. Like some freight car stirrups that have been modeled, the flange might need to be thickened to not "strobe"; or just be left 2D. I thought I'd reply since it's an opportunity to vote for realistic replacements for default U.S. highway overpasses. I've hoped there could be a swap-in for those someday.

  3. #3
    Join Date
    Jun 2013
    Location
    Minnesota
    Posts
    1,615

    Default

    I haven't done much work on road bridges lately, but there are several styles planned. So far I've got two railway bridge types mostly finished: reinforced concrete, and steel I-beam. The latter was an opportunity to test out my idea for rivet lines: a transparent sheet with the rivet pattern on it placed just above the riveted surface. I think it looks promising, but the real test will be plate girder spans...*shudders*
    Br_1.jpg
    Br_3.jpg
    Br_4.jpg

  4. #4
    Join Date
    Dec 2005
    Location
    .
    Posts
    15

    Default

    Those look really good! The concrete bridge in particular looks like modern construction, and both bridges appear to have walkway grates, supports and handrails like those on several CN Bala2 bridges. The steel bridge would have splice plates which are suggested by your "taller" columns of rivets in the girder webs over the supports. Your "shorter" columns of rivets between supports suggest that there are cross members between girders and we're seeing just the connectors used for the angles (also on the inside and invisible) that attach those members to the web. The overall look seems very plausible and the rivet pattern could also be assumed to be the high-strength bolts that are used today. Or an older bridge with riveted connections could still be in service with older primary members and a modern walkway retrofitted to it. The railroads often stretch the service life of an aging structure past the point when a state highway department would have replaced it. Not to get into the economic reasons why, or trying to suggest that a 49 year old bridge is safe whereas a 51 year old bridge isn't! I just think what you've shown is convincing looking and not particularly limited to one era, so these structures could work for many routes.

  5. #5

    Default

    Keep in mind that many of those state highway bridges being replaced probably have 50 years of exposure to salt and snow plows from winter use, which severely weakens things over time. Go down to places like Texas or Oklahoma, where they're just as likely to use sand and not have snowplows, and you'll see highway bridges from the 1930's.

    Railroad bridges don't take that type of chemical abuse, which might be why so many 75+ year old bridges remain safely in service....

    Travis, as for the detail.... It's tempting to go overboard on detail, but remember most of that won't be visible in the game, so are the draw calls worth it to have an anatomically correct I-beam, or is it just as useful to simply have a representation?

  6. #6
    Join Date
    Jun 2013
    Location
    Minnesota
    Posts
    1,615

    Default

    I think I'll keep any steel thickness less than 3/4" or so as flat and only 3D the big dogs. So far I think the biggest shape (10m I-beam with handrails on both sides) has only two drawcalls, and that's only because of the alpha'd rivets. The abutment pieces have two because there are two textures. The piers, so far, only have one. The goal is to get as much detail as I can while going cheap on drawcalls

  7. #7
    Join Date
    Sep 2006
    Location
    .
    Posts
    2,747

    Default

    Some eastern railroads engaged in perishable meat hauling out of the midwest meat packing district did experience deterioration of bridges from melting salt brine during the ice reefer era. When time came to single track these railroads, often the westbound track was retained and eastbound track was pulled up on account of better condition of the bridges on the westbound (empty direction) side.

  8. #8
    Join Date
    Jun 2013
    Location
    Minnesota
    Posts
    1,615

    Default

    Here is a version of a deck plate girder bridge I just finished. This is kind of an oddball, and will only be available with a 10m length and a bare steel deck, so a more common version will also be made. There will be versions in 5, 10, 20, and 30m lengths, with open and ballast deck options, with or without walkways (on one or both sides)
    40A-1.jpg
    40A-2.jpg

  9. #9
    Join Date
    Mar 2007
    Location
    Phoenix AZ
    Posts
    470

    Default

    Very nice!!

  10. #10
    Join Date
    Jun 2013
    Location
    Minnesota
    Posts
    1,615

    Default

    I've got a standard deck plate girder shape mostly finished (at least, the open-deck version), and I think it looks pretty great.DPG_2.jpg

    There's just one little problem that I can't seem to lick.
    DPG_3.jpg

    Can you see it? That white creeping into the bridge ties? That is coming from a portion of the texture at least 30 pixels away on a 1536px2 texture, which houses alpha'd out areas where my rivet patterns are. The texture map ran even closer to that area, but when I moved it back it was still rather prominent. What really irks me is that you can't see this effect in Shape Viewer, or in TSRE, or in GMax where I made it*, but it is painfully obvious in OR, both monogame and vanilla. The only way I could reliably get it to stop was to turn off antialiasing entirely, though I'm not sure I like OR that way.

    *not that that says much - GMax's in-house rendering could hide an army

    So the question I have is: How does one go about preventing this? Does one prevent it?

    I may end up pulling the nuclear option on it and just moving the alpha bits to another texture sheet entirely. I probably should have done that anyway

    The joys of modeling...sigh...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
-->