Yet another intractible pathing problem

[tads1970]

Hi James,if it wasn't for James testing,im sure there would have been a few people with less hair,up front and a general cause for concern goes a long way in my books,take yourself and the fine route you are making,and the amount of time spent,does it bother you that there are limitations with the software at hand?,and things may not be possible that you hoped were?

Lol yeah the video is funny,your right about the cpu's catching fire,one went on me when i was still wet behind the ears,didnt catch fire but almost,IIRC mine was a Pentium 3 550mhz,lol i have to look back and laugh now.

Daz

[Retro]

Thanks for the kind comments about my route Daz. Yes it does bother me that while Rail Simulator can allow me to build a very complex route with many interactive features the sort of Scenarios I could create for it are limited by the problems James and others have highlighted. The fact that a consist of just coaches are not recognised by the AI engine rule out quite a few Scenarios I had hoped to produce based on real life operations at one of my stations.
James since the start has done some excellent work in analysing the problems and organising them in such away that all are aware and I am sure that we would not be as far forward as we are with the program if he had not done this. In particular Ted has also done some great work in trying out and testing features of the program and many others such as our signalling Guru's have helped bring things forward with their help and suggestions. I thank all those who have spent time and effort in testing the various aspects of Rail Simulator for our benefit.
The one thing that does give me encouragement is the fact that RSDL are willing to listen and take action in these matters and that people on these forums are forever finding ways round some of the problems we have had.
I hope that these problems can be fixed in the future so that the many people who are contributing towards giving us the third party Items both payware and freeware in all areas of the Sim feel in the end that their efforts have been worth it.
Regards James.

[Klaabu]

It's nice to see some people have not given up even after such a weak Mk2 patch. Personally I bought Isle Of Wight route addon because hope was high RSD will do what they promised - most problems in distaptching and signalling system will be fixed. But I was wrong. Nothing happened. Both Adam and Derek have constantly reported here how important functionality of the RS is for RSD team. But reality talks something else - no any functionality improvements were made. Instead of that lots of energy was put to get 3-monitor view working. How many people can use this new feature? Few maybe. But most of customers were waiting for something else - to get RS function like simulator and not like any childrens railway game. Unfortunately its common practice that game developers are doing what they want and not what customers need.
After some testing and trying RS with Mk2 installed I decided to completely uninstall RS from my home computer (EU version) to avoid of getting angry. In my bussiness laptop there is still running US version and while having free moments I'm trying to analyse LUA scripts. Many trainsimmers here are talking about poor signalling or dispatching systems. But in my opinion the biggest problem is in poor interacting between signalling and dispatching systems. You can make very advanced signalling scripts and improve much dispatchers ability to find destinations using additional destination markers. But ... when dispatcher is ignoring states of signal links and don't react properly? Made scenario for IoW route where there green light was given for player loco (and red for AI accordingly) but junction was switched against player causing derailment. Absolutely clear - dispatcher was in total conflict with signal state. Only RS coders can fix this problem. How long it takes to fix it? Have we wait another half year? Can Derek or Adam answer to that question? Why didn't they fix those problems in Mk2? Because they don't have competitor. MSTS 2 is announced to be released in the end of 2009. So ... guys ... they have lots of time left to do it ... maybe in september 2009 we can see fixes in dispatching systems. But there is doubt our eyes can't see it at all.
For me personally it's not tragedy at all if they don't react to critics. Very interesting new simulator named World of Subways is coming out soon and Trainz 2009 seems to keep interest. Rail Simulator can be taken seriously only after dispatching system problems have been fixed. And all that is not in our hands.

[Sly401]

Well if I take it seriously .. what does that make me?

Yet more sweeping statements.. I respectfully suggest that until Blablabla is fixed YOU cannot take it seriously, which is entirely fine by me.

Now I don't usually bite but quite honestly some of the way opinions are put over as being actual fact is insulting to those that are trying to do something positive, and who are content at the way things are progressing... we all want more, and probably different things at that.

Your opinions good or bad are perfectly valid .. but that is all it is YOUR OPINION. to which you are perfectly welcome.


Sly

[AndiS]

The point simply is that you can do a load of things with a programme like KRS. This is why people have such diverging perceptions of it.

If your focus is on modelling and bringing your models into the sim, you get a fair chance. You need to learn a lot about blueprints, but once you got that (and you manage the 3D modelling programme you are using, which is more difficult in my opinion), then KRS is great for you.

If you are building routes, and you do not jump too many traps, then KRS can be a great programme for you.

If you love prototypical operations, you are shot. Simply that. You can try to fiddle with the 10% that work (scenarios where trains do not cross each other's paths), and try to advance to 20% (overtaking a train), but it is very frustrating. You clearly feel that this programme was not made for you. Still, you keep reading this thing about the most realistic simulation, it is really bad for your nerves, believe me.

However, I myself hope that the RSDL had enough moaning by now that they really put some effort into a solid solution of the problem. After long denying that it is the core AI which is faulty, and not just some Lua scripts, they met Kuju people, so lets hope they get some real code fixes going soon.

Furthermore, I would not say that there are really no improvements in operations. Signals defaulting to danger and dynamic brakes working on Diesel trains certainly is a step forward. Yes, you ask yourself why it did not work on release day, but that is the nature of any patch, you always ask "why now, why not earlier". As has been said many times, the cruel reality seems to be that real simulators are very expensive and for the price of a game you get a game, which is good enough for many, but disappoints the experts.

[jamespetts]

Furthermore, I would not say that there are really no improvements in operations. Signals defaulting to danger and dynamic brakes working on Diesel trains certainly is a step forward. Yes, you ask yourself why it did not work on release day, but that is the nature of any patch, you always ask "why now, why not earlier". As has been said many times, the cruel reality seems to be that real simulators are very expensive and for the price of a game you get a game, which is good enough for many, but disappoints the experts.

This, I am afraid, is rather too generous to Rail Simulator: as I have written elsewhere, it is not accuracy of simulation that is the main problem: it is basic stability. Game or not, one should expect the software not to crash or behave unpredictably. Besides, it is perfectly possible to have game priced items that are at least moderately realistic simulators: consider Microsoft Flight Simulator, for instance.

[Retro]

The point simply is that you can do a load of things with a programme like KRS. This is why people have such diverging perceptions of it.

If your focus is on modelling and bringing your models into the sim, you get a fair chance. You need to learn a lot about blueprints, but once you got that (and you manage the 3D modelling programme you are using, which is more difficult in my opinion), then KRS is great for you.

If you are building routes, and you do not jump too many traps, then KRS can be a great programme for you.

If you love prototypical operations, you are shot. Simply that. You can try to fiddle with the 10% that work (scenarios where trains do not cross each other's paths), and try to advance to 20% (overtaking a train), but it is very frustrating. You clearly feel that this programme was not made for you. Still, you keep reading this thing about the most realistic simulation, it is really bad for your nerves, believe me.

I agree Andi,
A pretty fair assesment at this moment in time.
As I have said previously.
Building a Route - No real problems. The Editor is very good and easy to use and once you have mastered Route Templates, Track Rule Blueprints, Marker Files etc. it is easy to create something and enjoy doing it.

Models - If you can again master the Blueprint System and use 3D programs and conversion tools and are a good 3D artist with technical knowledge you can produce Scenery items, Rolling Stock etc to a high standard as we have seen from the excellent stuff now downloadable for free or as in D.T's case both Free and Payware. Plus the Loco's and Stock available from RSDL.

Sounds - If you have got the correct recording equipment and access to the to the Loco's or other sound sources it is now possible to produce realistic Sounds. Richard for one has proved this with his HST sounds.

Scenario Work - Very difficult indeed for reasons outline by many people.

Puzzled also as Klaabu pointed out. Why do we need the 3 monitors abilty. There are much more important things that need doing.
To Sly's post. I take this program seriously as he is probably aware and find it very usable in the areas I have mentioned so I would agree with Sly here.

James points out about crashes and unpredictable behaviour. I have encountered these and the program will either crash to desktop or send to Microsoft. In the Route Editor the problems tend to occur when using the space bar to cycle through the track states or after placing Signals particularly when coming out of Scenario Mode after placing loco's to check the Signals are working correctly.
I don't expect other people to have the same opinions as I do but I feel they must be allowed to express them to enable things to move forward and for debate to take place.
Regards James.

[boleyd]

The debate has certainly become focused on the AI/dispatcher/pathing/signaling area. As a short-time user (few months) my first objective was to do a local spur which worked ok. Then I wanted to have a high activity rail-yard where trains come and go, as well as the creation and breakup of train's assets/wagons/cars/etc. This is where I got into trouble by assuming that the AI part of RS was the answer. After lots of wasted time I just settled on trying out various very simple things and also had problems (my msgs in RS site's forum). I was annoyed when I saw RSDL personnel tell us to learn LUA scripting. Eventually I began to look at it. In the mean time I became aware of the elements that others have seen to be faulty, such as in this thread. They looked like I was not going to be able to make my rail-yard even if I became a LU(la) expert.

RSDL has promised more frequent updates to follow MK2. However, the customer base has no way of evaluating the capability of RSDL to meet this commitment. As customers we also have no way of knowing which of our issues are acknowledged by RSDL as problems to be looked at. As customers we have no way of knowing which problems on the aforementioned list are scheduled for attention. Finally, as customers we have no idea of when any more patches will appear beyond a trust us promise. Unfortunately, the trust has been somewhat diluted given the lack of attention to these core issues in MK2.

We are therefore once again in the dark. This is "once again" very fertile ground for customer's speculation on a variety of subjects. As the clock ticks away with no information about the raised issues (on the list, to be fixed, when, etc.) addon developments, on some fronts, is at a standstill. Indeed a significant problem for a product that has some dependency for its sales on a flow of exciting addons, tools, etc. to keep the juices flowing!