The future of Trainz

[IsambardKingdomBrunel]

I read that on uktwatsim as well Vern.

They can shove it where the sun don't shine.

If it was actually an outstanding product and not a cobbled together mish mash, of a failed product ( Kuju RS ).

Then i would say fair enough, give them a chance.

But it is the same old spin from rs.com month after month.

3 yrs down the line and the AI still cannot navigate a route, without being lead like a blind man.

Still no AI shunting\switching and now there is this facile career mode.

Which has obviously been added to draw in the casual gamers and alienate the dedicated simmer.

RailFail it is and RailFail it will always be in my view.

IKB.

 

[boleyd]

RW seems to be following its own path - latest newsletter refers to introduction of quality certification for items sold through themselves/Steam. I'm sure it's being done for noble purposes but the cynic in me says they want to nudge out the independent (cheaper) payware and the freeware by at least passively suggesting that only their own "QA" add-ons are guaranteed trouble free.

This is a revenue enhancement project. Their major income is from addons. It is a clever way to extend their staff in developing Career Scenarios and Locomotives. Rail Simulations has totally committed to a gaming genre as described openly by their CEO whose background is the gaming industry.

Development of improved AI will only occur if the gamers eventually begin to realize it is boring being the only person in the train company. Mysterious AI trains whiz to and fro running in a nether world that they can never touch. But, lots of people are delighted by achieving goals thru adjusting the speed of a train and then braking to stop it all within bounds to gain points and the public's adulation.

If Trainz is moved into the game genre to attract new customers then it will disappear as well. Gamers cannot be satisfied by the simplistic world of just driving a train. That is why real world drivers are found to wreck trains while they text. It is a boring job and not very complex unless a valve pops on a steam engine or the whistle gets stuck at on.

For customers seeking a RAILROAD experience Trainz can easily satisfy that objective. The complexity of railroading is not in driving the train but operating the railroad. The railroad is a multi-dimension continuous operation. Properly stopping a train to pick up passengers has little challenge but operating a railroad is a real fun challenge. The current crop of tools and assets in Trainz comes very close to offering the platform to operating a railroad. There are several liabilities from past versions that have to be "gently" removed. As mentioned, third party developers must be welcomed not just by wow and awesome. They need the technical interfaces to the program to create complex functions and operations that emulate a railroad.

My pet peeve is the visual aspect of Trainz 2010. It still looks like 2004 to me. Something is missing. When I compare it to RailWorks it is night and day in terms of visual appeal. Try as I might, I cannot create a fully believable scene in Trainz. In RailWorks I can create a rural environment where you must look closely to see that it is computer generated. Despite its several shortcomings RailWorks had one or more artists create the vegetation which is the area where Trainz falls behind. The SpeedTrees, as an example, is mechanically good but visually bad.

Hopefully, Auran fully supports third party artists and knowledgeable railroad asset developers to round out what is currently the best simulation of a RAILROAD on the market.

 

[Vern]

Don't want to disagree Dick, but Trainz, MSTS and Railworks et al to me are about driving the train not about being part of a bigger picture, in the same way as M$FS put you as the pilot, but you didn't have to worry about looking after the ATC. I'm still partial to Zusi 2 because, even though the graphics are pants and on a par with early DOS flight sims, the realistic physics and AI/timetable operation grips you. Zusi does have the benefit that if you finish one run and the timetable is still running, you can jump to another train.

It is possible to use the various assets and textures in Trainz to create a lifelike scene and I'm really looking forward to getting back in once I've put this current RW route to bed. However it's the inability to not be part of the big picture - i.e. switching points yourself and having to give the AI a helping hand - which I still find the biggest bane of Trainz and does go back to the virtual model railway idea at the start of it all.

If Auran/Neverfail could give Trainz an autonomous despatcher and scenario editor, preferably as good as the one in Zusi 2 - and I got a better laptop to run TS2010 - Railworks probably wouldn't see me for dust. At the end of the day if I want a signalling simulator, I'll play SimSig, if I want to be a robber baron I'll play Open TTD or similar.

 

[boleyd]

Perhaps that is what I find so frustrating - the programs are designed to experience driving trains. I have not driven more than a dozen trains and then only to see how the scenery was arranged. I have made my own routes and complex AI scenarios but am probably wasting my time looking for some way to "run a railroad" in a realistic setting.

In Flight Simulator and X-Plane I do focus on "driving" the aircraft since it is a higher skill level, multi-dimensional and subject to external random forces such as weather and traffic/ATC. The most used airports do see some object and texture changes. I also use a lot of photo-scenery I import. However, I have no desire to "run an airline".

Guess I will go back to the aircraft:wave: and await the evolution of Open Rails.

 

[sniper297]

More disagreement

"It is possible to use the various assets and textures in Trainz to create a lifelike scene" except the switches, making realistic looking switches requires so much tedious tinkering I'd never finish a route, I'm an old man and ain't gonna live that long.

As for the AI traffic, primary problem I see with that is if during an opposing meet I enter a single track section first, an AI comes up to the trailing point on double track, stops for the red signal but then sets the switch for his track and locks it. It's nice to have the "driver command" menu so you can enter the "mini dispatcher" mode and give him a stop order so he releases the switch, but it would be better if he did something like check the switch, throw the switch, see that the block is occupied, throw the switch back to the other track and wait two minutes before throwing it to check again. That kind of logic SHOULD be possible to program into the AI, simple IF THEN ELSE commands. Phil and I are currently kicking back and forth assorted screwball workarounds, but it's such a simple thing and in my opinion the only thing that really needs to be fixed.

Advantage, already mentioned the mini dispatcher mode to force him to unlock the switch, the good part is he only locks ONE switch. With MSTS the switches are set and locked X number of signals ahead based on the signumberclearahead statement in the signal script files;

http://forums.trainsim.com/vbts/showthread.php?t=239517

Anywhere from four to six for a value gives reasonably good results. Still requires some tinkering with heavy traffic, and has the usual bugs that need to be debugged while setting up the activity, but it DOES work.

Railworks is completely hopeless;

http://forums.uktrainsim.com/viewtopic.php?f=215&t=77530

They "fixed" that bug by making it completely impossible for the player to change ANY switch on the AI path even if the AI isn't scheduled to START his run for two hours! How any sane man can accept that kind of kludge is beyond me, but apparently the devs are happy with it.

For serious hard core trainsimmers I don't think Auran has anything to worry about at present, and as the casual RW users are finding out the hard way freeware content comes from people who demand more from the base game.

 

[NikkiA]

They "fixed" that bug by making it completely impossible for the player to change ANY switch on the AI path even if the AI isn't scheduled to START his run for two hours! How any sane man can accept that kind of kludge is beyond me, but apparently the devs are happy with it.

The rule of thumb for 'fixes' on RW appears to be 'it stays as it is, until it hinders RS.C's abilities to sell payware, THEN it might get looked at, and either a half-assed fix, or if it's something simple, truely fixed'. Until now Scenarios have never been RSC's strong point with payware, so limitations of the AI aren't a big deal.

Maybe this will change with the career system, but I doubt it, it's much more likely that they'll just hope that people don't notice - the career system scenarios are some of the most boring scenarios that RSC have made, and that is really stretching boredom.

I suspect the bottom line is that the payware jig[1] is nearly up for RSC - they've covered the major interests and now are left with more and more nichey locomotives/stock to model. So they're grasping at everything they can - make scenarios from 3rd parties seem '2nd class', then make 3rd party payware seem '2nd class', freeware will come next.

[1] Of course, Just Trains still think there's enough market to justify £12 for another set of reskins of their A4 - reskins that should have been included from day 1

 

[IsambardKingdomBrunel]

Trainz is not just a train driving sim.

That is where it beats RailFail hands down.

It`s about the management of a railway system, in all aspects.

I agree with your view on payware in RF, Nikki. The end game for RS.com will come a lot sooner than the time they have to fix the fundemental flaws in the game.

IKB.

 

[djt]

Gamers cannot be satisfied by the simplistic world of just driving a train.

You obviously have no clue what is involved with running a real train.


Try running 12,000 tons over a typical US sub division sprinkled with speed restrictions and reaching full capacity while complying to strict operating rules and proper train handling and then tell us how “ simplistic the world of just driving a train” is.


Running a train in any of these “train games” is boring because of one major shortcoming, toy train physics.

 

[aardvark1]

Trainz is not just a train driving sim.

....... snip.......

IKB.

Trainz is a game, nothing more, nothing less, and not a train simulator. It is a game that costs $US 25~50 dollars.

To see a real train driving sim, click the link below:

http://www.railsciences.com/pages_products/products_simulator.html

I would think the cost of a real train simulator to be far more expensive than the cost of the Trainz game

What the post by...djt...says is correct.

But, I do enjoy playing the Trainz game.

Have fun, .

 

[boleyd]

You obviously have no clue what is involved with running a real train.


Try running 12,000 tons over a typical US sub division sprinkled with speed restrictions and reaching full capacity while complying to strict operating rules and proper train handling and then tell us how “ simplistic the world of just driving a train” is.


Running a train in any of these “train games” is boring because of one major shortcoming, toy train physics.

If driving a train is so complex why does Trainz make it so easy? Driving a car is also complex dealing with steep hills, signals, speed limits and weather. I compare train driving to flying an airplane which I can do reasonably but in the safety of X-Plane. I tried driving trains in Trainz and the "other program" and it was much easier than flying an airplane. ATC at a big airport is one of the supreme complexities of transport. Thus rather than repeating the same route over and over with no dynamic changes I would like to run a railroad - that is complex.

It is the lack of dynamic changes that makes driving a train boring. In a flight simulator you have other AI aircraft running REAL commercial schedules and REAL controllers managing the airspace while you "fly" online in an environment of dozens of other people flying their simulators and add to that real weather. I agree that your heavy train transiting significant grades is one of the more challenging aspects of railroading but it is repetitive with no external un-scripted changes.

 

[Vern]

Some of the earlier text based and simple cab view simulations (like Train Driver 3) did throw unexpected problems at you such as speed restrictions, diverted across to the slow/relief line, signal failures etc. While you could specify the frequency at which these occurred, the game engine decided when and where they would occur at random. In TD3 there was also the possibility of your loco failing if you ran the ammeter too long in the upper yellow or red sector. While on the subject of running in the red, you should really get an overload trip - at least on early British traction - if you allow it for more than a few seconds (this is a feature on some of the diesel locos recently created for OpenBVE). RW is worse of all as you don't get overload trips, rarely have to manage wheelslip or for that matter loco overspeed trips.

I think in some respects all the "big three" lost their way with regard to the above and focused on being scenery and infrastructure simulators rather than the full gamut of what can happen when driving a train. I guess the problem is making that work in a fully interactive 3D environment as opposed to a simple graphics or OpenBVE where other trains are simply static scenery (as opposed to moving scenery in Railworks!).

All of which I guess are factors that Auran at least need to consider, when contemplating the future of Trainz as being a trainsimmer's No.1 choice.

 

[djt]

If driving a train is so complex why does Trainz make it so easy?

It's called seriously abbreviated physics. One of the most difficult tasks with running a real train is controlling buff and draft forces throughout the entire train, no consumer train game has even come close to simulating this.

It is the lack of dynamic changes that makes driving a train boring.

Again toy train physics are boring so it's not hard to figure out why running a train in a game doesn't hold anyones interest.

The lack of simulating anything that represents real railroad operating practices doesn't help either and just amplifies the fact that the developers of the game have very little knowledge of railroad operations.

 

[cascaderailroad]

Try diving in Cab mode downhill, on a 2% grade with 14,000 tons behind you (4 loco's head end, and @ 125 cars, and 2 more helpers holding back on the rear, in 1/2 dynamic braking). Sometimes the train takes quite awhile to get it under control, or to stop. In real life just throwing on the brakes or full emrgency would cause a derailment.

 

[djt]

Try diving in Cab mode downhill, on a 2% grade with 14,000 tons behind you (4 loco's head end, and @ 125 cars, and 2 more helpers holding back on the rear, in 1/2 dynamic braking). Sometimes the train takes quite awhile to get it under control, or to stop.

In TRS2009/TS2010's defense the physics in “cab” mode is far more realistic then what is found in RailWorks.


At least you can make somewhat realistic brake pipe reductions with the automatic and bail off the independent and have the train react in a somewhat realistic fashion. The functional Amp meter isn't bad either.

In real life just throwing on the brakes or full emrgency would cause a derailment.

Yes throwing a train in emergency can cause a derailment. The position of loads and empties throughout the train has a lot to do with it.

 

[martinvk]

Interesting to read how someone who drives real trains see things. But then the cost of just one engine would buy a lot of Trainz. And then there is the cost of the railcars and all of that scenery. The fact that Trainz does a lot with so little speaks volumes.

Yes that RSI trainer might be more accurate in many respects but can anyone just model any region, any era as easy as Trainz does with Surveyor. Are there thousands of objects available to add to a route? They both might be called train simulators but that's probably all they have in common. One highly specialized for a small market with huge budgets, the other for a much wider audience but on a limited budget. So when comparing, lets be sure we are comparing comparable things. I don't think many people really expect Trainz to simulate all the exact characteristics of a real train. Some things just have to be ignored or drastically simplified if an ordinary PC is to have any chance of running the program.

Getting back to the OP, I hope that the future of Trainz will continue to approach prototypical realism without sacrificing easy of use or requiring unrealistic hardware.

I hope that Trainz will have an improved central AI dispatcher more aware of the user's actions and with a little more common sense. Shortest path does not mean reversing through an industrial siding. Waiting for another train to clear a junction is not a cause to panic and try to reroute.

I hope that the Trainz community continues to be a fun and friendly place where drivers, route builders, modelers and enthusiasts are welcome.