Cascade Game Foundry

[GoGoran]

That brings up a question i have. what was so special about the first MSTS 2? I wasn't part of the community back then, (still trying to convince my father to install the original on the single home computer) and i looked up some of the photos of what i thought was production photos of msts 2. I didn't look that impressive to me. Minor adjustments, using the same software the first one was made of. Reminded me of what RW is doing.

MSTS2 really wouldn't require much creative work, though lots of solid programming. Just keep all the good of MSTS (the physics model and the easy editability), add the BinPatch features and then some, make route creation easier to use, bring the graphics engine up to date (incl. support for high-poly models, true-color textures, etc.), and keep full backwards compatibility with MSTS. No need to reinvent the whole game.

[railfan101]

MSTS2 really wouldn't require much creative work, though lots of solid programming. Just keep all the good of MSTS (the physics model and the easy editability), add the BinPatch features and then some, make route creation easier to use, bring the graphics engine up to date (incl. support for high-poly models, true-color textures, etc.), and keep full backwards compatibility with MSTS. No need to reinvent the whole game.

in this day and age, reinventing may be nessesary considering how old the original engine is

[GoGoran]

in this day and age, reinventing may be nessesary considering how old the original engine is

I mean the gameplay. Of course they would have to do the engine from scratch, more or less, or it would have as much in common with MSTS as Windows 8 does with Windows 2000 Pro (yes, there are some commonalities). But in the physics department there really is little to improve, aside from the knowns bugs which would have to be fixed. Anyone who has run a train on that Australian route with its super-length ore trains knows what a powerful physics engine the good old MSTS has.

[TrainMan_112]

The physics engine is (quite) far from perfect.

Just sayin'.

[NorthernWarrior]

The original stab at MSTS2 by Kuju didn't look vastly different from the original - I guess it was somewhere between MSTS and what they produced in Rail Simulator a few years later. The second attempt at MSTS2 by Aces looked a whole lot better and autogen combined with the World Of Rails would have given us a sim of sims. Sadly that too cruelly fell by the wayside. I still hope that MS are monitoring what is happening with Railworks - which has turned into a massive business for the company that bought it out, with the CEO talking (rightly or wrongly) 6 and 7 figure sums for route development - and deciding it's time they got their finger back in the pie. Whether what Cascade are doing could even compete with RSC, who as I say seem to be on a bit of a roll at present, remains to be seen.

[GoGoran]

The physics engine is (quite) far from perfect.

Just sayin'.

That is your opinion.

From my experience, it is a masterpiece. I have fine-tuned many totally different locomotives and EMUs/DMUs, all with prototypical data for weight, power, torque, etc., and by adjusting the friction value, I always reached prototypical behavior (acceleration, top speed, braking) after some experimentation.

I measured e.g. the time for a 300-ton-train (if I recall correctly) to accelerate from 0 to 200kmh, since I had the real world figure of the German BR 103 for this task, and adjusted the eng values till the MSTS model behaved as in the real world and was able to do it in 180 seconds.

I also got a test route with steady grades, e.g. 5%o over 20km straight, where I can evaluate the pulling power of locomotives and compare the MSTS models' behavior to the prototypical values. For many real-world locos, you can find such tables which say e.g. that the loco can pull a 600-ton train over 20%o grades at a steady 80mph - then you can use this value and check in MSTS whether it does that in the sim, too, and adjust the eng file values up and down until it works out around 80 - 82mph on the 20%o grade, steady.

It takes a lot of time and fiddling, about an hour for each locomotive or EMU/DMU, but if developpers would take this time and test and adjust their MSTS models thoroughly, they could deliver absolutely prototypical behavior.

With a few little compromises only. One such compromise being that MSTS does not simulate conventional electrics with DC motors properly, as far as the torque curve is concerned; but in practice, you will barely notice that. Another compromise is the lack of simulation for short-time maximum power, but there is a workaround which I already layed out in detail on this forum. And a few others, but overall, the MSTS physics model is a marvel and by far the best of all train simulators.

[TrainMan_112]

From my experience, it is a masterpiece.

That is your opinion.

It seems that your post solely addresses European ops (specifically EMUs/DMUs), and that is fine, but keep in mind that your particular field of interest is a mere sliver with regard to the vast spectrum of global railroading.

Here are just a few of the flaws that I've seen:

1.) Trailing units DO NOT bail off. When you have five or six (or more) trailing units, you get more than you bargain for when you make a reduction and the trailing locos set up hard and can not be actuated. The independent brake also does not control the entire loco consist, just the lead unit.

2.) Locomotives load instantly. When I advance the throttle to 1 or 2 from idle, I get 100% load of that particular throttle position in an instant. There are a few steep hills around here where it is common practice to swipe the throttle from idle directly to 8 (when stopped on the hill) and let the units load before knocking off the automatic. In MSTS if I were to go directly to anything above 4 or 5, the loco consist takes off like a dragster, I get instant wheelslip, and if my coupler breakage values were anything close to realistic, a coupler would likely be broke.

3.) Brake pipe propagation seems non existant. When an engineer makes a reduction on a 7000 foot long train in real life, the reduction in pressure is propagated to the rear; it is not instantaneous as "simulated" in MSTS. This is a big one for me.
In the defence of MSTS though, it seems that propagation and pressure transmission is simulated somewhat when the brakes are RELEASING, but not when they are setting up.

4.) Slack-action. It seems that trains in MSTS are in various states of "compression". To my knowledge, in real life, there is either 'stretched' or 'bunched' with regard to slack; not "very stretched" or "very bunched". There may be an exception with trains that have many cars with cushioned drawbars which (I think) would increase the amount of potential slack in each coupling. In MSTS, the train seems to "compress" heavily instead of just bunching up and remaining that way. Likewise when the train is stretched, and becomes extremely distended. This can be helped with various physics mods, but there are still some issues such as when extreme draft force/extreme buff force is exerted upon the train (pulling/braking a very steep grade).

In short, you may be satisfied with the European stuff, but as you can see, there is a fair bit to be desired regarding North American and comparable operations.




PS: Apologies for hijacking the Cascade Game Foundry thread. I'll stop.

[GoGoran]

That is your opinion.

Experience based on reproduceable scientific facts and observations is not an opinion. It is measurable and objective. Opinions are subjective and may differ from person to person.

It seems that your post solely addresses European ops (specifically EMUs/DMUs), and that is fine, but keep in mind that your particular field of interest is a mere sliver with regard to the vast spectrum of global railroading.

Wrong. I also do lots of locomotives. In fact the grade test route is only useful for conventional trains, since they are the only ones for which such information is widely available.

2.) Locomotives load instantly. When I advance the throttle to 1 or 2 from idle, I get 100% load of that particular throttle position in an instant. There are a few steep hills around here where it is common practice to swipe the throttle from idle directly to 8 (when stopped on the hill) and let the units load before knocking off the automatic. In MSTS if I were to go directly to anything above 4 or 5, the loco consist takes off like a dragster, I get instant wheelslip, and if my coupler breakage values were anything close to realistic, a coupler would likely be broke.

Sounds like your engine files are not properly set up. You can adjust everything.

If the lower throttles give you too much power, you can rate them down as you please.

Also think about the run-up time to max. power, which can also be adjusted at will.

Just to name two possibilities.

3.) Brake pipe propagation seems non existant. When an engineer makes a reduction on a 7000 foot long train in real life, the reduction in pressure is propagated to the rear; it is not instantaneous as "simulated" in MSTS. This is a big one for me.
In the defence of MSTS though, it seems that propagation and pressure transmission is simulated somewhat when the brakes are RELEASING, but not when they are setting up.

Again, sounds like improper eng and wag file setups. You can set them up so that they charge for ten minutes on a small 20-car train, if you deem such behavior realistic. Or find real prototypical charging times. Anything goes.

4.) Slack-action. It seems that trains in MSTS are in various states of "compression". To my knowledge, in real life, there is either 'stretched' or 'bunched' with regard to slack; not "very stretched" or "very bunched". There may be an exception with trains that have many cars with cushioned drawbars which (I think) would increase the amount of potential slack in each coupling. In MSTS, the train seems to "compress" heavily instead of just bunching up and remaining that way. Likewise when the train is stretched, and becomes extremely distended. This can be helped with various physics mods, but there are still some issues such as when extreme draft force/extreme buff force is exerted upon the train (pulling/braking a very steep grade).

This behavior can be changed at will by editing the coupler and buffer section in the eng and wag files.

Also take into consideration that some modelers might have their coupling sections taken from the wrong type of couplers - American or Russian automatic couplers behave totally different from European chain couplers and buffers.

In short, you may be satisfied with the European stuff, but as you can see, there is a fair bit to be desired regarding North American and comparable operations.

There certainly are some bugs to be fixed and some things simply amiss right now, but overall, there is no competitor which even comes anywhere close.




PS: Apologies for hijacking the Cascade Game Foundry thread. I'll stop.[/QUOTE]

[trainbrain]

Infinite Scuba is the first release from Cascade Game Foundry
http://www.infinitescuba.com/

[Hack]

Somebody mentioned months ago CGF might be working on a scuba title - I thought it was a joke.