News:

Want to praise Simutrans?
Your feedback is important for us ;D.

Balancing notes from the online game

Started by jamespetts, January 22, 2012, 10:21:11 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

ӔO

#70
6-PAN rear motor car has the wrong power.

even if the number was correct, it's very awkward to use. It's main problem is sluggishness, but putting that aside, it's just not very good when you can use SR V class on the same set of tracks.

8 car set of 4-LAV is much better in just about every aspect. It fits in 6 tile stations (6-PAN takes up 5), has more power, has better acceleration and more capacity.

there's no pantry car with the 4-LAV, but the 6-PAN is not good for long haul anyways.


There needs to be another road mail van, other than the austin 7. Either that or it should be able to hold more.


more bugs

SR bulleid corridor set, catering missing from dining car
LMS open set, also missing catering from dining car

balance

SR V class, LMS 5XP, maintenance too cheap compared to power and speed. They are so good, in fact, that you sort of have to wonder why you would use GWR King and LMS 7P

GWR King at somewhat of a disadvantage compared to LMS 7P. Same tractive effort, but more power in 7P while you still have to pay for the 150t tracks for both.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

AEO,

thank you for your report. I have now corrected the 6-PAN rear motor unit's power. When things are balanced properly, it should be more economical to run by a good margin than any steam train. As to the 4-Lav, don't forget that it has lower comfort.

Can you elaborate in respect of the mail van - why does there need to be another one?

I fixed the missing catering levels, and made interim adjustments to the running costs of the SR Schools class and the LMS Black 5, but a full rebalance will be necessary before these values are properly in line.
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.

ӔO

the capacity of the austin 7 mail van is inadequate for the amount of mail generated.
it is the same problem encountered with early steam train to horse carriage transfers.

it is not as pronounced, but still leaves a bit to be desired.


the morris 8 that comes after becomes the mainstay, but it can barely keep up with the demand.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

Hmm - the problem might be that the demand is too high, not that the capacity is too low. I am planning on reducing the passenger factor for the next version, so it might be that the van then becomes adequate.
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.

ӔO

#74
LMS mail carriages are slightly shorter than the LNER, SR and GWR carriages, yet have the same capacity. This gives them a slightly unfair capacity advantage

also 4-wheel and 6-wheel brake mail carriages have an overcrowded capacity, which probably shouldn't be there.

Morris eight van and Morris Minor van could possibly use a touch more traction. like 1 or 2kN more each.

BR Mk.1 coaches have inconsistent speeds between set.
- Mk.1 restaurant buffet, 175km/h
- Mk.1 TPO, 175km/h
- rest, 160km/h


BR Mk.1 and Mk.2 coaches are really confusing to use. possibly because their next/previous is not set correctly.
for example:

Mk.1 mail gangwayed brake cannot be last. - therefore you cannot make a mail only convoy out of Mk.1 coaches.
Mk.1 GUV allows anything else after, not just mk.1/2 coaches
Mk.2 must end with brake coach, but not start with it. - aren't there different batches of these? some that work with Mk.1 and some that don't?
Mk.2D, two different versions exist with the same name
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

AEO,

thank you for your feedback. May I ask you to elaborate on the last point - what do you mean by confusing to use?
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.

ӔO

@james

with the mk.1/2 coaches, the way you can assemble them is just bizarre, to say the least. With timeline on, it's not too bad, but try assembling them with timeline off (or if you have a lot of rolling stock left over in the depot) and it becomes very difficult to try and assemble two convoys that with the exact same sequence of cars. This is because the next coach possible selection moves around so much and because you can only add some cars after some other cars which also relies on some other car being there first.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

Can you be more specific about this part:

Quote
This is because the next coach possible selection moves around so much and because you can only add some cars after some other cars which also relies on some other car being there first.

I should also note that this pakset is intended to be played with the timeline on; however, it is also intended that Experimental have the ability to buy/sell second-hand vehicles, so this needs some consideration.
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.

ӔO

okay, well, if you start off with...
- mk.2d buffet. There is a nice selection to choose from.

then you select...
- mk.2 standard open. Now where did the carriages go?

then you select
- mk.2 tourist open. Oh, there they are again.

next
- mk.1 brake open. I can still see the other mk.1/2 carriages, but wait, why did the mk.1 restaurant buffet disappear and in its place I now have mk.1 gangwayed corridor brake?

next
- mk.2 brake open. Okay, now I can select nearly everything else that has existed in the past.

next
- mk.1 gangwayed brake. now I can't add any mail carriages after it.

next
- mk.1 GUV. now I can add anything from the past, and even the mk3/4 carriages

and well, if you play around with it, you will see that the selections move around a lot too.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

Ahh, I see. Will have to regularise this...
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.

ӔO

#80
some of the bridges in 1960 have cheaper maintenance than the equivalent speed and weight ground level track.


example:
concrete supported rail bridge (225km/h, 150t). $837.50/km to build, $20.00 maintenance
vs.
concrete sleeper heavy (200km/h, 150t). $720.00/km to build, $39.60 maintenance


same problem with road bridges.

---


BR class 71 is too cheap in maintenance and cost compared to class 73, which has less power.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

ӔO

#81
the airplanes are quite horrid to use.

they are fast, so they get a lot of pax flow, but they are low capacity, so you need a lot of them. The problem with using a lot of them, is that they have poor turn around times, which jam the airport. Compounding the problem, they only like using one runway because there's no one way signal in the pak to route the airplanes to use specific runways.

this usually results in a lot of pax that get dumped and decide to use the trains. However, there is usually around 8000~12000pax that get dumped at once, so the trains can't handle it.

also, stopovers are hard to replicate. Currently, it's like a forced layover every time the airplane lands.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

Milko

Hello AEO

Quote from: ӔO on March 01, 2012, 06:44:41 AM
the airplanes are quite horrid to use.

they are fast, so they get a lot of pax flow, but they are low capacity, so you need a lot of them.

I'm working on the development of the aircraft, I am currently trying to have aircraft available at any age but are not yet able to give more variety. The number of passengers of the planes that I have done is aligned to the number of passengers that the aircraft had in reality. I still have to draw all the larger aircraft, such as the 747, the Airbus 330 and 340, the latter are characterized by 300/400 people on board. These innovations should reduce, but not zero, the problem that you're exposing.

Giuseppe

ӔO

@milko

I should have added "in the 1960~1970" range.
there are  only the BAC 1-11-200 and comet in this era.
---

309/1 and 309/2, when used as 4+2+4 has more power, more capacity, faster turnaround and is cheaper to run than a class 86 with mk1/2 coaches with similar length. The only thing that the 309 is disadvantaged in is comfort, but not by much.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

As Giuseppe said, the aircraft capacities are based on reality. The capacity problems in game come about through the passenger factor being far too high, I think. I was looking at the railway lines in Thurmouth last night, and noticed fully loaded express trains departing every two or three game minutes from just one of the many, many lines coming out of that town. That is far too much, and will produce unrealistic results as you have found with aircraft. Larger aircraft, such as 747s, are intended for long haul (thousands of kilometres); the short haul aircraft should suffice for what are in the game in essence short haul routes.
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.

omikron

Couldn't there be some kind of penalty factor for plane travel, such as for example an enforced waiting time at the airport before being able to board a plane. That would make plane travelling less likely.

More importantly, however, I think there should be some more variation in the routing of individual passengers. Instead of everybody going from A to D only going via B, some percentage might instead choose to go via C. In that case, plane travel could make more sense.

omikron

jamespetts

Quote from: omikron on March 01, 2012, 10:12:09 AM
Couldn't there be some kind of penalty factor for plane travel, such as for example an enforced waiting time at the airport before being able to board a plane. That would make plane travelling less likely.

There already is - see the "min_wait_airport" setting in simuconf.tab.

QuoteMore importantly, however, I think there should be some more variation in the routing of individual passengers. Instead of everybody going from A to D only going via B, some percentage might instead choose to go via C. In that case, plane travel could make more sense.

This would not, unfortunately, be possible without a massive overhaul of the routing system, which is not currently feasible, and would in any event likely make the routing system too computationally intensive. If one day in the distant future one were to have divergent routes, however, it would not make any sense for the divergences to be entirely random: there would have to be some realistic economic basis for the divergence.
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.

Milko

Hello

Quote from: omikron on March 01, 2012, 10:12:09 AM
More importantly, however, I think there should be some more variation in the routing of individual passengers. Instead of everybody going from A to D only going via B, some percentage might instead choose to go via C. In that case, plane travel could make more sense.

There was a thread already open about where you can find some more information.
http://forum.simutrans.com/index.php?topic=6227.msg60003#msg60003

Giuseppe

el_slapper

IRL, medium-sized planes were enough in the 60s. Therefore, they should be enough in the game at the same era.

Maybe it' a question of planning? Air companies are using hub, up to a certain point, yet there are also many lines that do not go through non-hubs. I can fly from Paris to Katowice or Wroclaw, without changing in Warsaw/Munich/whatever big airport. I'll have less planes to choose from(2 per week for the Paris-Beauvais/Wroclaw line), but I'll have.

It is possible that passenger factor is too high, but even in that case, it should be up to the player to adapt : many metropolis today have 2, even 3 airports. If one airport, as big as it may be, is not enough, build another one at the other side of the town. Be also sure to make more "direct lines" to reduce pressure upon your main lines, by transferring it upon secondary hubs.

It is very tempting to have only a few hubs & giant lines between them, but there comes a time when it's no more enough. Then, you have to manage an increasing complexity. IMHO, it's realistic. Atlanta is the main hub in USA, yet I can take direct flights to other US towns. That's the joy of simutrans : you're always looking for a good line balance, yet, town size increases makes the right balance always shift.

Lyon is not the biggest french airport. By far(though it's a big town). Yet, it already links to many different towns, including a lot of non-hubs.

dustNbone

I was poking around at the journey times on the server map, and many of the journey times (with waiting times included) aren't that much quicker than the train journeys in many cases.  Combined with the fact that the airplane journey will probably end much further from the actual destination than the train will, there is definite overlap in which is the "better" service.   I think some way to make pax that choose air travel tolerate longer waiting times before rerouting would help. 
 
  Once the map gets to the point where it is now, those multi thousand passenger bounces really make a mess elsewhere on the network, sometimes taking a very long time to correct themselves.

Carl

#90
Can I ask -- by 'bounce', do you mean the rerouting of passengers?

Am I right in thinking that this only happens when the route they have chosen is no longer the fastest in the database? Unless I'm mistaken, I don't think that passengers reroute just because they have been waiting a long time. Either they will board any convoy to their next immediate destination, or they will be discarded (if max_wait_time is exceeded).

jamespetts

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.

Carl

So one way you'd get passengers ditching the plane for another mode of transport (though not the only way) is if there were an alternative direct service (e.g. a train) between the two airports.

One way to avoid this ever happening is to have your airports as separate stations which are nevertheless within walking distance (an on-foot connection) to your airport-train or your airport-bus stations.

This way, once passengers arrive at the airport they are committed to travelling by air unless (a) they are discarded through waiting too long or (b) a reroute occurs because the air route is no longer the fastest route.

ӔO

#93
I've ditched the airlines  for major routes and put them into service as branch service that was previously the job of ferries. (because I accidentally replaced the ferries and now there are no longer any ocean going ferries to use). I've estimated that I need about 60~100 airplanes for each line to operate smoothly.


- class 313 and 507 seem to have an unusually low comfort level compared to any of the previous commuter EMU trains, despite similar seating and overcrowded capacity. 3-sub, 4-sub, 414 and 415 all have around 70~80, but the 313 and 507 have 60. It's also unusually cheap to run.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

If you'd ever travelled on a class 313/507 (the interiors are the same) you'd know why I set the comfort as I did! A picture of the interior can be found here.
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.

The Hood

I'd agree. But not as bad as a pacer!

ӔO

ah, it's one of those "we don't have money, so we'll just use parts from buses" type trains. gotcha :)


Morris LD mail van (1952) is the best for quite a while. It basically takes the best of Austin Morris EA (capacity 34) and low load time of Leyland DAF 200 (0:30-3:20)

The other road mail vehicles are not very interesting to use due to various issues.
- Morris Marina Van -- not enough capacity for the loading time and low on power.
- Leyland Sherpa Van -- less power compared to DAF 200, yet same running costs.
- Austin Morris EA Van -- loading time is horrid compared to Morris LD Van
- Morris FG -- loading time is horrid compared to load it can carry, just 2 more than vans, but at least it's cheaper
- Leyland G -- poor power to weight, but might be interesting to use inside cities.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

ӔO

There seems to be a few previous/next lines missing for the BR class 221 super voyager set. You cannot add a middle car before or after a buffet car. Quite sure this is limited to experimental, because the 221 set doesn't exist in standard.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

ӔO

could use a 300km/h elevated railway. maybe with somewhere around 70t weight limit.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

ӔO

#99
I'm thinking maglev can use an overall reduction in capacity by around 33% and also have a maintenance cost increase by at least 50%. The 560km/h track can also use a maintenance cost increase by around 50%.

It's extremely spacious and profitable in its current state.


and I do realize I'm the one who thought up the capacity. :p

Should be 3+2 (1st class) or 3+3 (2nd class) seats per row, instead of 3+4 in a 24~25m car.
I mean, it's supposed to be wider than your regular BR stock, but not that much wider.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

jamespetts

AEO,

thank you for that report. This probably applies to Standard, too - perhaps you could mention this on the Standard Pak128.Britain forum?
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.

ӔO

#101
It would be nice to have something that runs on AC electric and is placed between the 225km/h EMU and 160km/h EMUs in the late game.

The 225km/h sets are $40~60 for 8 cars
395 - javelin is costly to operate and weak in corners and has poor comfort. It's only slightly cheaper than 373 in terms of maintenance, despite having half the power.
390 - pendolino is not bad, but slightly short on power due to improper configuration, leaving it short one motor car.


The 160km/h sets are roughly $10 for 8 cars
357 - best power to weight, but comfort is not as good as 375/6
375/6 - second best option in terms of power/weight/comfort
444 - best comfort, but terrible power to weight ratio
450 - same power to weight as 375/6, but clearly not as good in comfort

The only thing in between are the 200km/h diesel sets, which are pretty good in their own right, but don't have an electrical counterpart. Something that costs around $20~30 in maintenance with around 175~200km/h would be nice to have.
My Sketchup open project sources
various projects rolled up: http://dl.dropbox.com/u/17111233/Roll_up.rar

Colour safe chart:

kierongreen

New class 350 sets are on order and existing ones are being trialled to run at 110mph (say 175km/h) so could be introduced around 2012/3 time.

jamespetts

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

Heh, yes I know they are around already, some existing ones are being modified (but not most/all) and new ones are being ordered - same class different top speeds for different subclasses :p