Topic: bridgewater-brunel.me.uk - Simutrans-Experimental - Pak128.Britain-Ex 0.9.1

[DesiroUFOSound]

A causeway with road on it is the problem near the middle of the map stopping ships pass  having a knock on effect for other players profits

[jamespetts]

Can you give me the co-ordinates?

[DesiroUFOSound]

A block at 1179 938 another at 1129 1026 and another at 1212 913  if i see more i will post co ords

[jamespetts]

Thank you for that - I think that I have got them, and also locked the account of the player who was doing this, as this was clearly intended to be malicious rather than just thoughtless.

[VOLVO]

Technically DrSupergoods are partly operating my company. I understand thia is a personal thing that one just don't want to share companies, but unregulated competition only make a mess ofthings especially when some players has already been too large and the small one just struggles. It also causes extend service overlap, that's why the "chocolate express" eventually caused such a big problem London Transport Board is eventually formed to regulate bus services. Competitioms like we jave in the map just doesn't happen in real life(at least in the modern world).

[jamespetts]

More or less unregulated competition prevailed for most of transport history, and modern authoritarian state regulation is really a very new thing, although it was a creeping advance through much of the 20th century.

The causeways should be impossible to build in the next version, and the one thing that we do not allow on the server is action that is malicious (that is, intended primarily or solely to harm another player's company) in nature.

[VOLVO]

Frankly I haven't seen such action happening. But the fact is that competition can lead to such action. I.E. I'm a really large company and build a same route alongside the existing one simply I can afford to lose some money to some route. This action is simply to completely eliminate competition by brankrupting rival companies. Doea that count as malicious?

[jamespetts]

Frankly I haven't seen such action happening. But the fact is that competition can lead to such action. I.E. I'm a really large company and build a same route alongside the existing one simply I can afford to lose some money to some route. This action is simply to completely eliminate competition by brankrupting rival companies. Doea that count as malicious?

This is an edge case: this sort of thing can happen in reality in many contexts, although it is sometimes prohibited by regulation. It would be very difficult to work out whether any given loss making route is intended as a feeder route or intended to harm, I think.

[DesiroUFOSound]

My underground railway made me a massive loss so i sold it all replacing it with buses   now waiting for 1980s buses like Leyland National 

[DrSuperGood]

The problem was no one was playing the server. As such I stopped playing to slow the advance of time. Eventually all our companies passwords were stripped and so abuse could begin when someone eventually did join. I re-locked my company yesterday but the others were left unlocked.

If we could find more players it would be great. It is pointless me playing alone as I might as well be singleplayer.

Also the pakset itself has died by this time. In a mostly water based map there are no more viable ships to come. Without such ships managing the number of goods is impossible.

[VOLVO]

Please could you do this for me please. I shall be back soon. What year is this now by the way, looks like I missed a lot.

[jamespetts]

Yes, there are balance issues in the current version (town growth, possible issues with ships that I will investigate) that limit the playability, which may have been responsible for the demise, but the next version is some time from being ready, so I might as well keep this going for now.

Can someone elaborate a little more on the issue with ships? I note that it has been mentioned before, but I cannot recall the details, and I cannot recall whether this has been addressed in the development version of the pakset or not.

[Sarlock]

I still log in every few days to check on things.  With no one playing, however, the clock is barely moving forward.

The core issue for ships is that after the Windjammer, which has a storage capacity of 900 piece goods + 2450 of other cargo (3350 total) in its holds, does not have a viable upgrade after it expires.  This was discussed some time ago and if I recall it may have been addressed: the holds of the next generation of ships may have been updated already (but not in the pakset used for the server game).  I have several thousand windjammers currently plying the seas -- when I have to upgrade to the 800 hold version (though faster than the windjammer), I will have to swell my flotilla by a factor of 3 or 4 just to equal what I am currently serving.

This issue aside, the main reason the game has gone stagnant for me is that there isn't anything to do.  Every new industry is a coal power plant, and given that power supply seemed to be causing a lot of desynchs, we can't even servicing them.

Population is still slowly growing, and creates an opportunity to expand local services a bit, but other than that the map is saturated and there isn't much to do.  With the only viable cross-map pax/mail transport being underground rail, which is locked at 120 km/h max speed until new tunnels arrive in 1959 (and then, only up to 135 km/h) we're pretty much finished there too, until high speed, high volume air travel opens up in the 50's/60's.

The sheer size of the download is a negative drawback as well, especially with the odd desynch still occurring.  The transfer/load and initialization process takes a good 5 minutes, and when the game desynchs 5 minutes later, it's hard to find the desire to want to do another 5 minute transfer/load.

[jamespetts]

Thank you - that is useful feedback. It is hard to tell, because running an experiment on this scale is not easy, but I am hoping that more restrained town growth will result in a far less intensively worked map and in consequence a smaller download/loading time.

[DrSuperGood]

I am also concerned about the viability of rail late game in maps of this scale. Where as ships can fit nice and compactly in routes over water, trains need massive amounts of rail all over the place with 50 tile long stations to perform similar routes. Where as in the water map there are massive shipping lanes, if this was a land map they would probably be 20 wide parallel railway tracks. Since double track scale is not an option I would recommend scaling down the length of trains a tad. If that 50 tile long train is just 15 tiles long things are much more manageable and the world need not be covered in rail.

As for passengers, I would recommend trying to only spawn possible journeys (no journey should be impossible with the current available transport). Early on most passengers viewed most journeys as "too slow" even if you had a pretty efficient connection set up (cross map boats). Maybe have some sort of average journey speed and journey multiplier statistic that can vary with age. When I set up some stage coach routes within towns it was at the stage where for a reasonable town I needed >50 stage coaches and still could not move enough people around. In the early 1800 there should be very few journeys made in total as people hardly travelled between towns let alone across the map.

[Junna]

I relocked mine a week ago. Has it been opened again?

[jamespetts]

I am also concerned about the viability of rail late game in maps of this scale. Where as ships can fit nice and compactly in routes over water, trains need massive amounts of rail all over the place with 50 tile long stations to perform similar routes. Where as in the water map there are massive shipping lanes, if this was a land map they would probably be 20 wide parallel railway tracks. Since double track scale is not an option I would recommend scaling down the length of trains a tad. If that 50 tile long train is just 15 tiles long things are much more manageable and the world need not be covered in rail.

Where does the idea of 50 tile long stations come from: what exactly would these trains be serving? The viability of rail and ship should be realistic to each era. The capacity of rail freight wagons are closely based on the capacities of real rail freight wagons, as are the capacities of ships on real ships. Industry output and demand, however, has been recalibrated in the pakset since 0.9.1, which may affect things somewhat. Other than industry demand and output, the thing that is not realistic in this map is the urban density or sprawl, which is as a result of the city growth system that needs to be fixed: what we have at present is the equivalent of the entirety of England nearly covered with one enormous, converged metropolis. This is likely to affect the viability of rail, as, at present, it is very difficult to build any rail lines anywhere.

As for passengers, I would recommend trying to only spawn possible journeys (no journey should be impossible with the current available transport). Early on most passengers viewed most journeys as "too slow" even if you had a pretty efficient connection set up (cross map boats). Maybe have some sort of average journey speed and journey multiplier statistic that can vary with age. When I set up some stage coach routes within towns it was at the stage where for a reasonable town I needed >50 stage coaches and still could not move enough people around. In the early 1800 there should be very few journeys made in total as people hardly travelled between towns let alone across the map.

The passenger generation system is redesigned for the next major release, but not in the way that you suggest, rather in a way that should be even more realistic. The reality is that far fewer people travelled in eras past, precisely because transport was slower. Many people will only make a given journey if it can be done in a short time. People who live near the seaside visit the seaside a lot more often than people who live far from it. The new model, which I believe that I have discussed in detail elsewhere, will in time be linked to a new model of city growth that will mean that successful passenger transport (which will be linked to successful freight transport by means of successful passenger trips being able to be made to consumer industries only in so far as they are provided with goods), which should drive a far greater degree of realism. The system for passenger generation that exists in 11.35 is based on distance ranges and produces quite a number of anomalies which will hopefully be eliminated in the next major version.

Junna: I cannot remember whether your company is unlocked or not.

[DrSuperGood]

Where does the idea of 50 tile long stations come from: what exactly would these trains be serving? The viability of rail and ship should be realistic to each era. The capacity of rail freight wagons are closely based on the capacities of real rail freight wagons, as are the capacities of ships on real ships.

Their capacities are, but the lengths are not. Where as a 4 tile long ship can hold 2000 tons of cargo, a modern train needs over 18 tiles. Where as for the ship I could use a 4*4 odd station with storage buildings, the train will need 1*18tiles and possibly more if high throughput. If Food is transported raise that to 20 tiles long. On top of that only 1 train can occupy a section of the track at a time meaning that where as a single tile wide shipping lane can move near infinite amounts of goods, the trains will need multiple tracks in parallel. Sarlock shows this with his cross map underground food trains where he has large town sized stations to accept all the freight traffic.

[jamespetts]

The town sized stations are necessitated because of the excessive demand, though, are they not? It is quite fundamental to the way that Experimental is set up that the vehicles have realistic values for all of the data that can be specified for them.

[Sarlock]

I love these discussions 

The town sized stations are necessitated because of the excessive demand, though, are they not?

Sort of.  A 24 tile station is 3km long.  A 50 car train is probably only 4 tiles (500m) long at scale... so it really only needs a 4 tile station with respect to scale.

Balancing real world statistics against a game that has adjustments to scale (size and time) makes it very difficult to reconcile.  Realistically, a 1x6 station is 750m long and 125m wide---which is probably enough room for 5-8 platforms and can handle all but the longest of trains.  Such a station in reality would be able to handle a throughput far higher than it can in the game.

One could reconcile this difference by increasing train capacities by several-fold.  But then this breaks their continuity to reality... so, somewhere, something has to give in the balancing equation.

When you consider that modern freighter ships can carry a good 50,000t or more (much more, for some of the largest ones) this gets even more complicated.

[jamespetts]

The real issue seems to me to be how limited that space really is in the game. On this current game, it is far more limited than it ought to be because of excessive town growth. This should hopefully be less of a problem in future games. Any alteration of the carrying capacity of rail vehicles beyond realistic values would have never-ending adverse consequences for realism, such as a reduction in the tare weight to payload ratio, difficulty in balancing capital cost of wagons using realistic values, inaccurate track wear (and getting even more complicated if the same is applied to road where the fourth power law gives an exponential relationship between loaded weight and wear). Every decision made in the design of Experimental and this pakset is premised on the use of as near to real life figures as possible.

[Sarlock]

Town growth itself isn't an issue - a well established player can easily afford to bulldoze as many city buildings as they need to make space for additional platforms and ways.  Where it gets challenging, or in some places, impossible, is where areas get hemmed in with unbulldozable objects (factories, attractions, etc) and other player's ways and stations (most of whom have long since gone inactive).

Fortunately we've been able to expand significantly underground on this map and could find the room to make these large stations.  Doing it above ground would be almost impossible.

I understand the implications of changing capacities, and wasn't necessarily advocating this change, but I was highlighting one of the issues.  We're only simulating part of the economics, so it ends up heavily skewing things to certain choices gameplay-wise.  Balancing to gameplay needs is more important that modelling reality, if we want to create a balanced game.

When we reach air travel in the game, wait until you see the monstrous sizes of airports we'll need 

[DrSuperGood]

I have been locked out of my company, either that or forgotten my password. Seems server activity is picking up again so might want to play.

I think it is time to raise the issue of "will Sarlock go broke before the end of the server"? By this I mean from overflowing his company's money counter resulting in him having an impossible debt for a while which underflows. This already is happening to the public service provider which is stuck in oscillation between impossible debt and insane wealth. The big issue is if his company will be subject to removal during the oscillation.

[Sarlock]

Interesting to find out!

I haven't been on in quite some time and seem to have been locked out of mine as well.

[jamespetts]

I notice that the server had gone down owing to a mutant logfile eating up most of the hard drive space. It is back running now at the beginning of November 1942; a small amount of time progression may have been lost owing to the failure. Sorry for any trouble.

[jamespetts]

I think that this has been disused for some considerable time.

However, this seems to be the pertinent thread in which to note that I have just upgraded the server to increase from 3Gb to 4Gb of RAM, and to increase from 1 core to 4 cores (which actually costs me less than the previous arrangement owing, I suppose, to advances in technology). The increase in number of cores in particular should improve loading/saving speed, which is multi-threaded and can be one of the more annoying things to have to wait for.

This is running what is technically still the current release version, but it is very out of date. I wonder whether there would be any merit in running the latest development builds for testing purposes?

[Junna]

I would say that sounds like a good idea. Certainly is why I haven't been on it for a very long time.

[DrSuperGood]

To be honest I thought the server died about a year ago. The server listing server died so knowing it was operational was near impossible.

I would recommend using it for nightly tests. Maybe with a smaller map, maybe specific conditions that need testing.

[Vladki]

You can test stuff at server.exp.simutrans.com.
It is not updated every day though.

[jamespetts]

I was hoping to test with a larger map; the other server may be better for testing specific issues. Dr. Supergood - the announce server is not run by me. If nobody else is able to host it, I could run it on my server, but I do not know how to configure it.

[Vladki]

I can run larger map (or even multiple servers). I don't know how to run listing server, but can try it. Is it part of simutrans sources?

[jamespetts]

I do not think so - I am not sure how it is set up.