Author Topic: Bridges (Read 62411 times)

The Hood · « **Reply #70 on:** January 03, 2012, 01:59:25 PM »

And now an iron bowstring girder bridge, loosely based on Barnes Bridge and again using AEO's parts.

Milko · « **Reply #71 on:** January 03, 2012, 02:15:22 PM »

Hello

Superb!, you're revolutionizing the look of the pak!

Giuseppe

wlindley · « **Reply #72 on:** January 03, 2012, 02:57:19 PM »

Splendid! ...the challenging part, will be picking a variety of different building-costs and operating-costs... will there some bridges that are less expensive to build, but much more expensive to maintain?

jamespetts · « **Reply #73 on:** January 03, 2012, 03:16:05 PM »

Have a look at the Experimental version - there are a whole range of bridges there (albeit sharing the original bridge graphics temporarily) with different costs, weights, maximum lengths and heights.

The Hood · « **Reply #74 on:** January 03, 2012, 05:45:27 PM »

And another - a wrought iron arch bridge.

jamespetts · « **Reply #75 on:** January 03, 2012, 05:51:32 PM »

Splendid! I work near Blackfriars Bridge, and this is a rather wonderful reproduction.

sdog · « **Reply #76 on:** January 03, 2012, 10:04:06 PM »

the wrought iron arch bridge is quite a beauty. The colour looks very similar to brick however, i'm not sure if a sliglty different colour would be possible and make it more obvious it is not a brick bridge with ornaments. That was what i was thinking at first glance, it reminded me of the Patrician's brick viaduct.

I've never seen such a iron latice bridge, i always thought they would need a lattice work on top too for static stability. Similar to tubular bridges. I don't know a lot on bridges however.
I just looked at a photo of meadowbank bridge, which looks very similar to yours. It has no lattice above, only some struts at regular intervals. Those might have been to support the catenary, if the bridge was electrified.

On height restrictions,

Arkadiusz · « **Reply #77 on:** January 04, 2012, 12:16:58 AM »

Indeed, splendid work once again.

Talking about design, I have some issues with the bowsting grider. The supporting cylinder looks odd. It is like this, because on one hand the girders are holding the construction on the sides, while the cylinder supports only the center part. Central cylinder support columns are fine for concrete counstructions, but rather not here.
I have never seen Barnes Bridge IRL, so I googled some pics and they confirmed what I was thinking... A rounded cuboid would look much more better or maybe two or three cylinders in one line could also do the job.

I might add some costs and appearence comments later if you wish so.

The Hood · « **Reply #78 on:** January 05, 2012, 09:05:10 PM »

@ARkadiusz, I think you're right about the bowstring girder. The screenshot below shows it using a brick base pillar instead. What do you think?

@sdog - the colour of the iron arch is player colour, set to brown in the example above. Green shown below. Iron lattice bridges like that are quite common in the UK, especially in our cities for railway viaducts built in the 1800s. The bridge is loosely based on Fulham railway bridge, which has the struts as decoration. Not sure what you meant for the rest of your comment re height restrictions either...

jamespetts · « **Reply #79 on:** January 05, 2012, 09:36:16 PM »

I wonder whether the bowstring bridge is a little high? I appreciate the problem with the repeating tiles (or else it'd be the same height with a longer top part), but, perhaps, to compensate, one can make it lower, so that it does not look vertically stretched?

AP · « **Reply #80 on:** January 05, 2012, 11:07:38 PM »

The wrought iron arch bridge is very nice indeed. Hope the speed limit's sufficiently high that we can have an excuse not to replace it too soon in-game!

Fabio · « **Reply #81 on:** January 06, 2012, 12:06:29 AM »

Hey folks, do you mind if I steal some of your bridges and adapt them to pak 128? I would obviously credit The Hood in copyright strings...

The Hood · « **Reply #82 on:** January 07, 2012, 01:32:14 PM »

Fabio, that's absolutely fine. Which format suits you best? I have .blend, .xcf and .png (the last of which, as always, is available in SVN for the graphics I have already uploaded and others will appear soon). Perhaps I could trade you for some citybuildings - did you get anywhere with the pak128.Britain textured versions of your blocks?

AP · « **Reply #83 on:** January 07, 2012, 02:21:27 PM »

Just thought I should report a few in-game oddities with the new railway bridges:

Tubular box girder bridge - intermediate piers. The piers for this type don't occur between every tile, which looks good when the game draws them 'correctly', but that doesn't seem to be all the time. They often seem to occur asymmetrically and off-centre. e.g I'd expect to see pier spacings of 2-2-2 (tiles) or 1-2-2-1 or 1-3-1 but I often get more like 1-4 which looks a bit odd (though granted not impossible).

The Tubular Box Girder bridge and Steel Beam bridge prices are presumably placeholders - construction 0.01c and maintenance 0.04c is rather favourable!

Lastly, some of the bridges seem to have length limits associated with them again - I recall this being disucssed previously and removed e.g from the brick viaducts. Not sure if it was deliberate they've been included?

Edited: typo.

The Hood · « **Reply #84 on:** January 07, 2012, 03:07:52 PM »

Pier spacing is done by placing a pier every x tiles, 4 in the case of the tubular box girder, counting from one end. It knows notginf of symmetry. It may be improved by lower spacings but that is the price you must pay.

Re lengths, we need a way of differentiating between the different types with costs, height and length limits. Some types will be limited, others won't.

ӔO · « **Reply #85 on:** January 08, 2012, 10:50:23 AM »

I guess it would be possible to make a grid chart for the cost vs. length of each type of bridge.
looking at the various wiki pages (although this means take with grain of salt) and some other stuff I've read in the past.

There are a lot of variations with each type of bridge, so these are just general figures.

Bridge Type	Maximum span	Strength	Material efficiency	Design effort
	(between pillars)
Rolled steel girder	30m	Light to Med.	Poor	Low
Plate girder	100m	Light to Hvy.	Poor	Low
Concrete girder	50m	Med. to Hvy.	Poor to Fair	Medium
Steel box girder	100m?	Light to Hvy.	Poor to Fair	High
Concrete box girder	100m?	Med. to Hvy.	Fair	High
Tubular Box girder	140m?	Med. to Hvy.	Poor to Fair	High
Masonry arch	140m+	Med. to Hvy.	Poor to Fair	Low to Med
Steel arch	550m	Med. to Hvy.	Fair	Med. to High
Concrete/Steel arch	330m?	Med. to Hvy.	Fair	Med. to High
Truss	26m?	Light to Hvy.	High	Medium
Pratt Truss	76m?	Med. to Hvy.	High	Medium
Continuous truss	400m	Med. to Hvy.	Fair	Medium
Truss Cantilever	549m	Med. to Hvy.	Fair to High	Medium
Concrete Cantilever	100m+	Med. to Hvy.	Fair to High	Med. to High
Cable Stayed	1088m	Medium	High	High
Suspension	1991m	Medium	High	High

Design Effort, I would guess, is how much planning and work must be necessary to construct the bridge. Longer spans, total and between the pillars, obviously require more effort.

The box and plate girders are what the modern elevated ways are made from.

The Hood · « **Reply #86 on:** January 08, 2012, 01:59:23 PM »

Helpful, AEO. What we need to do is translate this from engineering terms to simutrans variables, namely speed, weight limit (?), length limit, height limit, capital cost and maintenance cost.

I suppose there is the issue of total length of bridge vs total distance between pillars. Clearly a brick arch viaduct has a short span, but can be very long in total and I intend this to remain unlimited in the pak. Is it worth having total length limits at all? Height limits I will include, because some bridge types have pillars which do not exceed certain heights.

Strength - this can be related to weight limits. Am I right in thinking standard now incorporates something along these lines? If so, how?

Material efficiency/Design effort - can these easily be related to capital cost and maintenance cost?

jamespetts · « **Reply #87 on:** January 08, 2012, 02:08:23 PM »

It is, in my view, worth having total length limits, since some bridges don't have supports. In Experimental, I have differentiated between bridges that can and cannot have supports, and made those that cannot have supports less expensive, but have shorter spans. I intend to integrate your lovely new bridge graphics into the next Experimental release, and try to harmonise the way in which the bridges have been set up in Standard, but there might be some useful parameters in the Experimental bridges (largely sharing graphics at present) which might be helpful to implement: see here for the bridges.

ӔO · « **Reply #88 on:** January 08, 2012, 02:53:33 PM »

@Hood
I would guess a high design effort would cause an increase in initial cost, because an expert engineer must be hired.

Material efficiency is pretty easy. More efficient will mean cheaper to build for any given length.
Strength can determine maximum speed and weight. It's usually lighter and faster or heavier load and slower. For the purpose of standard, "medium" would be something like 110km/h for road vehicles. Mostly all rail bridges would fall under heavy. I probably should have made a "Very Heavy" for high speed rail

AP · « **Reply #89 on:** January 08, 2012, 03:27:49 PM »

Quote from: jamespetts on January 08, 2012, 02:08:23 PM

It is, in my view, worth having total length limits, since some bridges don't have supports.

Which types don't have supports? And do you mean don't have them in-game just because they haven't been drawn? I ask because, in reality, I don't think any type has a length limit on potential design, if you can provide intermediate piers.

I think the focus on capital cost, maintainance cost, and weight / speed limits, is the way to go.

jamespetts · « **Reply #90 on:** January 08, 2012, 03:38:34 PM »

Ahh, I think that the position is more subtle than that: the basic types of bridge are indeed capable of having supports, but supports cost money. In Simutrans, the bridge cost is a fixed per tile value, without taking into account whether there are supports or not. In order to have realistic costings, one needs a bridge without supports to have a fixed maximum length, and the supported version to cost more, but have no such maximum.

Arkadiusz · « **Reply #91 on:** January 08, 2012, 04:35:22 PM »

@The Hood

This now looks a lot better

Though, I still feel there is something to be improved on the bowstring girder. The base on which rail tracks are laid suggests to be thick and solid looking on the top of it (as it is the same texture as for rail tracks on ground), yet it is very thin from the side perspective. Maybe it would look better after changing the texture on the top to something that is partially transparent... as many such bridges do not use any stone underlay and tracks are put straight on the skeleton structure of the bridge.

The Hood · « **Reply #92 on:** January 08, 2012, 05:43:21 PM »

@jamespetts,

Shorter spans still need the abutments, which would cost a fixed amount regardless of the span. So for longer bridges of the same type, the additional support cost would be comparable (logically at least, no idea about numbers in real life!) to abutments. So I'd say it's probably OK to have a single cost per unit length, and avoid having too many bridges to choose between in the menu. Certainly in standard I will be having just a single type and I'm erring more and more towards no length limits.

@Arakdiusz,

how about this? This is just using the same deck used in the steel truss bridge, but everything else is the same.

Arkadiusz · « **Reply #93 on:** January 08, 2012, 06:53:38 PM »

Yeah, that made the difference. Thumbs up.

The Hood · « **Reply #94 on:** January 08, 2012, 07:21:36 PM »

Back to drawing now: iron lattice bridge...

jamespetts · « **Reply #95 on:** January 08, 2012, 07:37:18 PM »

I really like that one!

wlindley · « **Reply #96 on:** January 08, 2012, 10:01:33 PM »

Thread should be subtitled: Isambard Kingdom Brunel...!

The Hood · « **Reply #97 on:** January 08, 2012, 10:41:37 PM »

Thanks. Finally (for today) the cornish-style wooden viaduct with stone pillars - a work in progress as I'm not happy with the pillar textures yet, but a preview nonetheless:

jamespetts · « **Reply #98 on:** January 09, 2012, 02:35:43 AM »

These are extremely impressive!

ӔO · « **Reply #99 on:** January 09, 2012, 04:03:29 AM »

I think that newer railroad deck is excellent for the truss type bridges.

I've found these documents that describes costs and maintenance of bridges.
http://www.iti.northwestern.edu/publications/utc/tea-21/FR-4-6-Krizek.pdf
http://www.esc.vic.gov.au/NR/rdonlyres/98A7F47C-5ABC-4691-AB18-38FD41D073CE/0/MaintenanceCostBenchmarkingFinal.pdf

A more mathematical approach to calculating maintenance cost.
http://users.encs.concordia.ca/~hammad/papers/C13.pdf

AP · « **Reply #100 on:** January 09, 2012, 10:46:52 AM »

These are really good Hood! Would a lighter timber colour on the cornish style viaduct perhaps allow the shape of the fan to show more clearly? (ie greater difference between highlight and shadow) Just a thought.

jamespetts · « **Reply #101 on:** January 09, 2012, 08:45:34 PM »

Of note to this topic, have a look at this very recent commit Standard, which permits elevated ways to be used over shallow water (with a depth equal to 1). This possibility may wish to be considered when implementing bridges.

Edit: One thought has occurred to me on the utility of maximum length restrictions, which had not occurred to me before: if I understand correctly, the view so far taken, I think, has been that there is no need to have different types of bridges with and without pillars, since pillars are simply added automatically where and if they are needed, and, since a bridge with pillars is usually theoretically unlimited in length, there is no need to have a length limitation for any bridge.

The difficulty that has just occurred to me with that is this: there is also a maximum height parameter. Bridges with no pillars, obviously, have an unlimited height (as it matters not one whit to the structural integrity of the bridge how far the drop is provided that it is secured effectively on its banks). Conversely, bridges with pillars, while having a theoretically unlimited length, have a limited height. So, pillarless bridges have a height limitation, and pillarful bridges have a length limitation. Putting the two together is not the same as having a bridge with neither limit, since a particular type of bridge might need to be too long to be built pillarless, but over too high a gap to be built with pillars, making the gap uncrossable without using a different type of bridge.

See also here for my extension request in relation to bridge heights.

The Hood · « **Reply #102 on:** January 09, 2012, 10:11:36 PM »

Hmm, I get your point but the thing is that in Simutrans you can only change elevation by 1 per tile, so in order to have a truly deep gap to cross it must also be quite long (e.g. 8 tiles or more). Longer, in my opinion, than any pillarless bridge to look good in the game. I'm still to be convinced that single span bridges without pillars, with a length limit and no height limit have any role to play either, given what length they would need to be for the lack of height limit to be important. I'm generally for giving a good range of bridges visually, with some differentiation along cost height and speed lines.

prissi · « **Reply #103 on:** January 09, 2012, 10:33:23 PM »

You can change height by two tiles with the slope tool though.

AP · « **Reply #104 on:** January 09, 2012, 11:24:05 PM »

Quote

You can change height by two tiles with the slope tool though.

I tried in-game, yes, it's possible, but quite tricky to gain height at a faster rate than 1:1, and certainly the game doesn't seem to generate terrain steeper than that.

Quote

there is also a maximum height parameter

So I've just discovered - none of the bridges in game will build over a nice 32-wide by 16-deep gorge I created as a test. Can i suggest the bridge types should tell us these h and l limits in the menus, to aid choice?