News:

The Forum Rules and Guidelines
Our forum has Rules and Guidelines. Please, be kind and read them ;).

Good luck James!

Started by Spike, February 10, 2012, 11:36:11 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Spike

You had better luck in forking off a variant of Simutrans than I had, and you are very successful with experimental. I think I said it someplace already, but I want to say it again - thank you for creating it, and good luck in your efforts to stabilize and expand the branch.


jamespetts

Thank you; that's kind of you to say. Maintaining a fork is quite a lot of work, especially merging in features from Standard (and the coders there do have a habit of adding some lovely new features and making improvements regularly), and I have been greatly assisted by many of the coders who have taken part in development over the years, and also the Standard coders who regularly very kindly give me tips on how to deal with some or other problem in the code.

Very best wishes, in turn, for Iron Bite - it is interesting to see how you are progressing with the GUI there.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Spike

Thanks also. I hope some day there will be more than just UI, but my skills don't allow that at the time being. Maybe I'll try better signal positions next, that is something that looks important to me. Also the idea of tunnel resistance looks interesting, I know that TTDpatch has it, and Experimental, but I don't know if Standard has it too. Maybe there can grow some fruitful exchange between Iron Bite and Experimental.

jamespetts

What do you mean by "tunnel resistance" here? There has been discussion of different costs for different sort of tunnelling, but no conclusion reached yet, and nothing implemented. I am considering at least having much more expensive tunnelling underneath the sea, although have yet to reach a decision about what, if any, other sort of tunnelling limitations to impose. One possibility that I had considered was a depth restriction on tunnels (as suggested in the "cost of tunnels" thread), which would allow, for example, a "cut and cover" tunnel type with a depth limit of 1 or 2. If you have any thoughts on the subject, they would be welcome.

As to fruitful exchanges between Experimental and Iron Bite, that is always a possibility in open source projects such as these, but I do wonder whether the implementation of code translation in Iron Bite might make that somewhat difficult.

Incidentally, when you write of better signal positions, what do you mean? Do you mean the position of the signals in the display, relative to the track, or are you referring to some functional aspect of signalling? I am very interested in a better simulation of signalling in Experimental, although that has to take second place in my attention to bug fixing and balancing changes.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

sdog

QuoteMaybe I'll try better signal positions next, that is something that looks important to me.
that's been something i was hoping to read here since ... i started to read here. (as new user this was the first thing that annoyed me, sorry to be so impolite.)


QuoteAs to fruitful exchanges between Experimental and Iron Bite, that is always a possibility in open source projects such as these, but I do wonder whether the implementation of code translation in Iron Bite might make that somewhat difficult.
can't we extract the translations from a diff, and forge some regular expressions out of it. those we can easily apply to do the translation with a script.
Hajo, it'd be easier though if you'd do the translations in a script directly. Else integration into experimental would also make it rather difficult to inherit changes from standard unless they'd translate too. (i have the feeling, that if there was a working script and it shows it can translate without breaking compilation it would most likely be accepted. there's enough pressure demanding a sollution from intl. irregular devs.)




One more thing, Hans, what do you think on experimental using euklidian distance instead of manhattan distance? (diagonals are shorter) This was rather controversially discussed for standard a few years ago.

jamespetts

Strictly, Experimental doesn't use the Euclidean distance any longer, but an orthogonal distance calculation invented by Knightly last year, which calculates the shortest distance possible when turning only in increments of 45 degrees.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Spike

Quote from: sdog on February 11, 2012, 06:12:24 AM
One more thing, Hans, what do you think on experimental using euklidian distance instead of manhattan distance? (diagonals are shorter) This was rather controversially discussed for standard a few years ago.

In the old days the diagonals didn't work very well and their use was discouraged by me. So the manhatten distance matched with the proposed rectangular layout. Since the diagonal work well now, I think it was just consequent to change to a new distance measure which resembles the layout better.

"Tunnel resistance", I meant the higher air resistance in tunnels. I don't know if it is a liked or wanted feature, but the realism fans might like it. It should be easy to code and I thought that Experimental had it (TTDPatch had it, I know that from an old discussion about train physics).

jamespetts

Interesting - I was not aware of this "tunnel resistance" thing. It could be done, I think. Do you happen to know what the differential air resistance in tunnels is?

We do have air resistance modelled in Experimental, and this can be set per vehicle so as to simulate streamlining.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Spike

I'm sorry, I have forgotten most of it. I just know that the reasons was that in tunnels air can barely be pushed sideways by the train and therefore gives a higher resistance. The train must push the air ahead, and this can even be felt when a train or metro enters an underground station. Kind of an air-wave heading into the station just in front of the train.

jamespetts

#9
Ahh, yes, I knew that they pushed air around - the "piston effect" - which is indeed noticeable on the London Underground (especially as the trains fit very snugly into the tunnels). This was blamed for fanning the flames in the 1987 King's Cross fire.

I was not aware that it made a significant difference to air resistance, however; I'd be interested in implementing this if anyone can find me figures.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

kierongreen

According to: http://www.ce.metu.edu.tr/~ce439/Slideshows/ruling%20grade.ppt

For a narrow tunnel (I'm guess more likely the case in the UK, and especially single track as modeled in simutrans) resistance seems to be about three times that of the open air.

How air resistance compares to other friction - well, the same source suggests at 50km/h in the open air approximately half of the overall resistance is due to the air, therefore in a tunnel 75% is. As speed increases the effect does significantly however...

jamespetts

That is very interesting - I could well look into adding this when I have time. Thank you!
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

isidoro

Wouldn't that depend on the shape of the locomotive and, to a lesser degree, on the tunnel network?


Spike

A sci-fi novel proposed evacuated underground tubes for highest speeds ;)

sdog

#14
not just a novel:
http://de.wikipedia.org/wiki/Swissmetro

(in english only a brief summary)
http://en.wikipedia.org/wiki/Swissmetro






here's an interesting link UK house prices in relation to distance to london measured in minutes on a train:
http://darkgreener.com/train-times-v-house-prices-graphing-the-commuter-belt

isidoro

I didn't understand evacuated, but I didn't dare ask.  Now I understand...  or nearly... is the train sucked?

sdog

in some of the older vac train concepts the train is pushed by the higher pressure air behind. In the swiss concept the train is driven by a linear motor. (the project is dead [now -- if it ever was alive])

dustNbone

Evacuated in this sense means that most of the air has been pulled out of the tube, leaving it significantly below atmospheric pressure.  Basically less air means less air resistance, but I'm thinking keeping an entire tunnel system under significant negative pressure has some pretty substantial costs associated with it. 

isidoro

So that is something not so difficult to implement in the game engine: special tunnels, with high maintenance costs, but that allow trains to go faster that the nominal top speed...

I wonder how a train makes the transition from outside to inside of the tunnel...

sdog

it would have been a completely independent system.

isidoro

That makes sense, but makes it more difficult to include in the game, on the other hand...

dustNbone

Yeah it would require a reasonably well sealed system, this would also include stations, or at least the platform areas.  Airlocks would be required between the system and the outside atmosphere.

sdog

i remember some reports from the 90s that there were tests that suggested that evakuating a tunnel was not trivial at all. i can't find sources though, so don't trust me on this.

if you want to read something on research and development in tunneling here's a source:
http://hagerbach.ch/en

jamespetts

Hmm - and the passengers would presumably need oxygen suits...?
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

isidoro

Space suits, I'd rather say...  With very low pressure your body does pop out, like pop corn...

dustNbone

Hmmm yeah I guess passenger areas/trains would need to be seperated, and kept somewhere near atmospheric pressure.  So that lungs would remain inflated, etc.

ӔO

#26
as long as the cabin can be pressurized, although there needs to be some special consideration for the articulated joints between cars, it wouldn't be too different from an airplane. I would guess that something like ball valves can be used to keep different zones at different pressures. Also, depending on how tight the fit of the train is inside the tunnel, it too can act like a partial valve.


regarding the air resistance for flat faced vs. aero faced trains inside tunnels.

85km/h for KiHa 187 series
http://ja.wikipedia.org/wiki/JR%E8%A5%BF%E6%97%A5%E6%9C%AC%E3%82%AD%E3%83%8F187%E7%B3%BB%E6%B0%97%E5%8B%95%E8%BB%8A

130km/h for HOT7000 series
http://ja.wikipedia.org/wiki/%E6%99%BA%E9%A0%AD%E6%80%A5%E8%A1%8CHOT7000%E7%B3%BB%E6%B0%97%E5%8B%95%E8%BB%8A

That's a good 35% reduction in speed. But there is more to this, because these speed restrictions only apply to tunnels with small diameters. It doesn't affect lines with wide tunnels designed for double tracks.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

sdog

the original US concept of vacuum trains had trains that snuggly fitted into the tubes, The trains were one piece, the tracks completely straight. The trains would then work as pistons compressing gas in front of them, for the next station. Doors on the arriving side would be open in the station. When the train is there they would close. When the train is about to leave it would enter the tube again the doors would be opened and the train would be pushed by the overpressure in the station back into the tube.


The swiss thought about a high speed train, possibly maglev, inside a conventional low pressure tunnel. with enough room between tunnel walls and to let air flow through. The pressure would be considerably reduced to reduce air resistance.

My own thoughts: Stations could easily be pressurized, either by bulkheads at either end (dangerous) or something similar to London underground, where a wall with automatic doors seperate pax from train. Especially with maglev you could easily push a train against some fittings at this wall.

For a modern ICE tunnel the incline seems to be notieable. On the Nuernberg -- Muenchen track the northbound train reaches it's top speed in a tunnel for the first time on that route, while the southbound train loses it's top speed. With full traction it still falls from 300 to 270 km/h. The tunnel has an incline of 1.8%

wikipedia:
QuoteAufgrund der starken Steigung können die Züge Richtung Ingolstadt/München ihre Höchstgeschwindigkeiten in der Regel nicht halten. So verlässt ein mit 300 km/h am Nordportal eingefahrener ICE 3, selbst bei maximaler Traktion, das obere (südliche) Portal mit ,,nur" etwa 270 km/h. In Richtung Nürnberg fahrende Züge erreichen im ,,Irlahüll" dagegen in der Regel erstmals ihre Höchstgeschwindigkeit (beim ICE 3: 300 km/h).

dannyman

First off, James, good luck and thank you for all your efforts on Experimental.  Truly a Herculean task!

To the discussion at hand: the idea of scaling up pneumatic tubes to an underground railway was explored in the late 18th / early 19th century and a prototype was built in London on New York, but such a system never saw commercial service.  Once you invent electricity the problem of running trains underground gets way way easier.

There are some 19th-century technologies that I would love to see in SimuTrans for 19th century urban operations:

1) Elevated railroads.  These were big in New York and Chicago.  Small steam locomotives pushed/pulled wooden streetcar passenger cars above city streets.  Later, electric driving cars were added, which could also run in the subway.  Over time, the wooden streetcar cars converted to steel-sheathed passenger cars.  Most elevated tracks in New York were torn down in the earlier 20th century as service moved to subways, but Chicago (my home town) still has the famous elevated loop marking the downtown.

Check out: http://warofyesterday.blogspot.com/search/label/Sixth%20Ave%20El

2) Cable Cars!  These were invented in San Francisco in the late 19th century, and were very popular in Chicago.  The original inspiration was seeing a horse car lose its traction on a hill and pull a team of horses to the death.  Bury a cable underneath the street and loop it through a steam engine at a power house, to pull the cable along at 20 kph.  A grip is dropped from the car through a slot in the street above the cable.  At first the grip slides along the cable, with the car picking up speed as the grip improves.  The RoW would be like a very expensive streetcar track, and the rolling stock would be modified horse cars.  Cable Cars can climb a hill better than anyone!

The cable car system in San Francisco was mostly destroyed in the 1906 fire, along with much of the city.  Anyone who enjoys Simutrans, though, would enjoy a visit to the Cable Car Museum to see how they made a streetcar system work before electricity.  Cable cars were of course no match for electric-powered streetcars.

sdog

"Cable cars were of course no match for electric-powered streetcars."
except for steep gradients, still.

there are also new cable car systems being built. (a company in Sterzing, not to terribly far from my home develops them) Such a system was also considered for street level operations recently. There's a lot of experience from regular/suspended mountain cable cars in Tyrolia.


tracks are still much cheaper than linear motors, but carts have the same weight advantages.

rsdworker

ahh vacuum trains - the trains would be pressurized similar to space ships which they have own oxygen systems like air flows around space ship
simllar ones used for smaller transportation around buildings - a postal tubes or cash tubes which are vacuum tubes