Just bought my first payware addon

[alanch]

But then you get the problem sometimes, that changes are done to how signalling and pathing works, which breaks previously functioning scenarios... (e.g. the latest 'Career Mode' updates, which broke several scenarios from IHH and DT add-ons, and even at least one default scenario on Hagen - Siegen).

Would you rather have developments/improvements and a few broken scenarios, or no improvements and no broken scenarios? What do you need to do to make an omelette?

[AndyM77]

But then you get the problem sometimes, that changes are done to how signalling and pathing works, which breaks previously functioning scenarios... (e.g. the latest 'Career Mode' updates, which broke several scenarios from IHH and DT add-ons, and even at least one default scenario on Hagen - Siegen).

I'd prefer to see updates that help fix Railworks as a whole, if as a side effect a scenario or two gets broken then you've two options. Option 1 fix it yourself, the scenario editor is far easier now than it used to be, it's nothing to be scared of. Or take option 2 contact the payware provider be it IHH, Oovee, RSC, DT, etc, let them know that the latest patch has broken "x" scenario, and if they are a company worth dealing with they should come out with an official fix. Of course if they say, "well, it's not our fault it's now broken" or along those lines, then simply don't support their add on products anymore.

Payware should be a cutthroat business, if they can't be bothered to help out purchasers then simply stop purchasing from them even if it's something you desperately want, because it's the only way that they learn.

The more core updates the better as far as I'm concerned (Signalling / Ai / Sound / Performance improvements )

[transadelaide]

I'd prefer to see updates that help fix Railworks as a whole, if as a side effect a scenario or two gets broken then you've two options. Option 1 fix it yourself, the scenario editor is far easier now than it used to be, it's nothing to be scared of. Or take option 2 contact the payware provider be it IHH, Oovee, RSC, DT, etc, let them know that the latest patch has broken "x" scenario, and if they are a company worth dealing with they should come out with an official fix. Of course if they say, "well, it's not our fault it's now broken" or along those lines, then simply don't support their add on products anymore.

Keeping in mind the mathematical truth 1 + 2 = 3, can I suggest option 3 here? If you manage to fix it, instead of keeping it to yourself you could let the original provider know and offer them the use of your fixed version as the official update to be distributed. The amount of effort it would take to write a gracious email offering your solution is minimal, and the payoff for other end users would be fantastic. Some people (including me) do this on the forums but to be of any real use you have to reach more than the tiny minority of RW users who are UKTS members.

Oh, and do it graciously without any outrageous demands for crediting or royalties. That would quite rightfully be replied to with something along the lines of "we'll fix it ourselves, goodbye."

[AndyM77]

I'd go along with option 3 to a point.. If they accept the offer of a user submitted fix, then I'd hope that they'd at least test it fully on their systems before allowing everyone to download it, and also have the politeness to acknowledge the "fixer" in a readme file / on the website.

I'm still in the school of thought that if a payware provider expects us to purchase something, then we the purchasers have a right to expect full support on the product over a reasonable timeframe. Of course what a reasonable timeframe is up to the individual purchasers, but you've got to look at the Microsoft model, in particualar critical fixes are still being dished out for Windows XP which is well beyond it's current lifespan.

[transadelaide]

I'd go along with option 3 to a point.. If they accept the offer of a user submitted fix, then I'd hope that they'd at least test it fully on their systems before allowing everyone to download it, and also have the politeness to acknowledge the "fixer" in a readme file / on the website.

I agree on the testing part, especially as I remember when DT made the one-time mistake of outsourcing scenario creation to somebody who created dodgy scenarios. I would see any acknowledgement as being at their discretion and kept in line with whatever acknowledgement they offer their beta testers.

I'm still in the school of thought that if a payware provider expects us to purchase something, then we the purchasers have a right to expect full support on the product over a reasonable timeframe. Of course what a reasonable timeframe is up to the individual purchasers, but you've got to look at the Microsoft model, in particualar critical fixes are still being dished out for Windows XP which is well beyond it's current lifespan.

In the case of DLC for RW, I would see a change to the route being the definition of the end of a scenario's lifespan. If it happened within a month or so of release then it would be more reasonable to expect repair or replacement of affected scenarios. Once you get past that point if the track layout changed I would regard any updates/replacements as being a free bonus rather than part of reasonably expected support.

In the case of Windows XP, there are four things at play. The first is that an operating system as expensive as M$ sells (Vista 1.1 aka Windows 7 should have been a free service pack) should last for a damn long time, the second is that their next OS was a load of 5H17, the third is that people are still buying copies of XP and finally, they themselves defined it's lifespan as being 31 December, 2014. Apple's approach to operating system's is radically different, they release them with far fewer bugs, stop non-critical updates as soon as the next version retails and their OS upgrades sell for far more reasonable prices.

[FoggyMorning]

I'd go along with option 3 to a point.. If they accept the offer of a user submitted fix, then I'd hope that they'd at least test it fully on their systems before allowing everyone to download it, and also have the politeness to acknowledge the "fixer" in a readme file / on the website.

I agree on the testing part, especially as I remember when DT made the one-time mistake of outsourcing scenario creation to somebody who created dodgy scenarios. I would see any acknowledgement as being at their discretion and kept in line with whatever acknowledgement they offer their beta testers.

I'm still in the school of thought that if a payware provider expects us to purchase something, then we the purchasers have a right to expect full support on the product over a reasonable timeframe. Of course what a reasonable timeframe is up to the individual purchasers, but you've got to look at the Microsoft model, in particualar critical fixes are still being dished out for Windows XP which is well beyond it's current lifespan.

In the case of DLC for RW, I would see a change to the route being the definition of the end of a scenario's lifespan. If it happened within a month or so of release then it would be more reasonable to expect repair or replacement of affected scenarios. Once you get past that point if the track layout changed I would regard any updates/replacements as being a free bonus rather than part of reasonably expected support.

The difficulty with that is if a payware provider continues to supply scenarios reported as faulty to new purchasers. Whilst I suspect the majority of an add ons sales are gained in the period immediately after launch there continue to be residual sales long afterwards. In that situation it comes down to a choice between fixing scenarios caused by updates beyond the control of the scenario creator or removing the no longer working scenarios from the sales package

[paulz6]

My view is that if improvements to the AI and signalling means a significant amount of existing scenarios are going to be broken, then the AI and signalling is not actually improving at all. It is just changing from one inadequate model to another inadequate one.

[transadelaide]

My view is that if improvements to the AI and signalling means a significant amount of existing scenarios are going to be broken, then the AI and signalling is not actually improving at all. It is just changing from one inadequate model to another inadequate one.

So you mean that to be a true improvement there would have to be improvements AND backwards compatibility with any awkward work-around solutions used in scenarios?

[paulz6]

My view is that if improvements to the AI and signalling means a significant amount of existing scenarios are going to be broken, then the AI and signalling is not actually improving at all. It is just changing from one inadequate model to another inadequate one.

So you mean that to be a true improvement there would have to be improvements AND backwards compatibility with any awkward work-around solutions used in scenarios?

Perhaps you should think about it a bit more before rolling your eyes all over the place.

A lot of scenarios I have created have been on the Oxford - Paddington route based on real timetables. I'm telling an AI train to move from point A to point B and on to point C according to a schedule. Would you not want any improvements to be able to handle the same?

I don't see anything wrong with the current mechanism of giving an AI train a list of instructions to complete. If the tasks can be completed now, why not post upgrade.

Yes there are a lot of things that could change, improved physics may alter timing if the scenarios have been created using performance values. Hopefully though, an improved AI system would not fail to load just because it thinks things are scripted to be in the same place at the same time. They would cue up according to the signalling system.

Some scenarios may end up resulting in a blocked route system, but if a significant number of scenarios cannot handle a new AI subsystem then I would suggest it would be no better than the last.

[msmith4000]

Apple's approach to operating system's is radically different, they release them with far fewer bugs, stop non-critical updates as soon as the next version retails and their OS upgrades sell for far more reasonable prices.

What you say is right but Is'nt it a shame then there is hardly any decent software to run on Mac OS's.

[AndyM77]

My view is that if improvements to the AI and signalling means a significant amount of existing scenarios are going to be broken, then the AI and signalling is not actually improving at all. It is just changing from one inadequate model to another inadequate one.

If the signalling and AI was improved (somehow ) it shouldn't break any old scenarios because it would be improved!

Hehe, being serious now. If we (routebuilders) are building routes to be as prototypical as we can be in regards to signal placement (or at least choosing the correct signalling type for what lays ahead on a non-prototypical route), then any changes to the signalling scripts shouldn't actually break anything if the "core" programmers also follow the prototypical rulebook in regards to bi-directional and block working. On the AI front, if it was made more intelligent, or could work in a totally dynamic way, then any current scenario should also work as it won't care where 'x' train is, as it can work its way around problems.

With time and programming love, AI and signalling could be brought up in standards IMO without breaking anything current, and shouldn't require a bodge job. However, at present I don't see RSC wanting to spend time just working on signals / AI as expanding the simulator seems more important to them at the moment.

In an ideal world, I'd love RSC to take at least 3 months out from producing 'add ons', and use that time to fix the no sound from rear powercar issue, and other sound related issues, look properly into the "Something Bad Has Happened", although I'm not seeing that quite as much now and investigate into improving the AI / Signalling further without breaking current routes and scenarios. Sadly I know that this won't happen as they (RSC) need an income and the 'add ons' is the easiest way for this to happen.

[RSderek]

the guys who make DLC are not the same guys who fix bugs or update code...
so we can do both.

We are just working from a different ordered list of wants.

lets keep the threads lite, they do tend to go south fast for some reason.

best regards

Derek

[paulz6]

I don't think there is too much difficulty in creating AI drivers with their own personalities that drive according to the rules of the road in front of them. The difficulty is in the AI dispatcher/signaller/route controller (whatever you want to call it). In an open system like RW where routes can be designed in an infinite number of ways, the AI controller is going to have difficulty deciding what to route where and avoid deadlocking the system. There will be limitations in a dynamic system and the scenario creator would have to be careful not to schedule too many trains that may cause the route to end up being deadlocked.

[AndyM77]

the guys who make DLC are not the same guys who fix bugs or update code...
so we can do both.

We are just working from a different ordered list of wants.

lets keep the threads lite, they do tend to go south fast for some reason.

best regards

Derek

That's good to know, however... and I know you might not be able to say because you're an employee / contractual reasons, are the "Code" teams in the same office / building as the "arty" types? If not, is this a possible reason why things are percieved as not seeing "Core" items being fixed?

I'm certainly trying to keep the thread postive and I think or rather hope that the nastyness of recent months has got back down to normal levels now

[bigvern]

I'm certainly all tranquil... However after getting Zusi 2 reinstalled this afternoon and loading the excellent Rhine Left Bank route, I'm immediately presented with a complex infrastructure and signalling to go with it. There are numerous AI trains everywhere, as you would expect at the real Koln. However the despatcher is having no trouble sorting things out, keeping traffic moving including diverting trains to designated alternate routes. The sense of being part of a real railway is immense. And that's before you head south from Koln with "Merkur", Class 103 electric clattering through the points then climbing the super-elevated flyover line by the side of the depot. Discrete timetable window open at the bottom of the screen I check the mileage and see I can now open her up. Oops, too much power traction overload, out snaps the circuit breaker. Wind the wheel back, reset the CB and off we go a bit more cautiously, out on the straight, up to 158 km/h. Drop the power wheel back to Notch 20 and speed starts to decrease as expected. Stations come and go, all the time the other trains which are a part of the timetable going about their business seamlessly under the AI despatcher. And of course I know that any late running of my train will have a reaction on the others that are running - maybe I will be regulated to let that freight go first.

Maybe I wax lyrical but to me that is the essence of train simming, doesn't matter that the graphics in Zusi are not that special - the suspension of disbelief is far more real than anything Railworks can offer at the moment.