Class 390 pendolino by nexusdj - problems in TS2009 SP4 Native mode

[shaneturner12]

Moderators - if this is in the wrong forum, please move it to the correct one.

The problem I am having is as such:

The Class 390 pendolino (updated), on the Download Station by nexusdj, seems to be having a lot of problems in TS2009 SP4 native mode.

I have had to make a small modification for it to appear in Native mode (adding class tags), but it is now behaving erratically.

When it is produced via a portal (portal used is maggs' Re-Rail Portal Basic) it de-rails after only two of the nine cars have produced, and the second car appears to run into the back of the first. There is also a PortalTunnel script error at the same time.

The pendolinos also appear to not be loading correctly at some stations (mainly AJS ones) - they start loading, but then continue moving at 11mph (the strange thing is that the class 321s, also by nexusdj, behave correctly)

After some detective work using Asset-X, the validation tool in Asset-X reported something about a.bog1.

Also, the TARDIS scripts run into problems when working with the bogeys table.

Has anyone got the train in question working in TS2009 (preferably SP4) or TS2010 native mode?

Apologies for the long post, but it is a complex issue.

I can provide more information if required.

Shane

EDIT: Click the following link for a screenshot of the derailing issue. (Due to the size of the image, I cannot directly add it to the post)

http://hostthenpost.org/uploads/084b045b92401e14c7dd11f2b9ce2a24.jpg

I have also confirmed that this issue happens after reinstalling the pendolinos, and that it happens at any portal that produces it.

 

[shaneturner12]

Apologies for bumping my thread, but I could do with an answer sometime soon if possible - it's also affecting my friend's layout (who also uses TS2009 SP4) and possibly other users' layouts.

Shane

 

[nexusdj]

Try removing the reference to the "pendotrailerpant" script from the car 3 and car 7 config files and see if that cures it ?

If it works then you'll also have to remove the references to the corona's for the sparks from the mesh table entries to stop them from showing permanently.

Let me know if it does and I'll look into changing it and uploading again .

Not sure on that a.bog1 fault as it works fine in compatibility mode and it also passed the uploading checks. Which coach reported the bogey fault or is it all of them ?

 

[shaneturner12]

The a.bog1 fault was reported on car 8, although the TARDIS script could not find the bogey tag on any of them.

I will have a try with your suggestions, and will report back in a few minutes.

Thankyou so far with your advice.

Shane

 

[shaneturner12]

Results are in - the issue still happens.

If it helps with finding the issue, the train is formed with First Class at the start of the train (basically, it's in number order)

Shane

 

[nexusdj]

I've just opened the car 8 mesh in the trainz mesh viewer and the a.bog1 attachment is there and is in the correct place , as is the bogey reference in the config file .

The only other thing I can think that is causing the problem is the scripts that make the cars tilt as they go round curves ?

Also I'm not knocking asset x as I've used it on numerous occasions with excellent results to upgrade content but it isn't bullet proof and you need to be careful as I've had 2 engine specs ruined by it as it removed all the throttle-power and dynamic brake entries.

EDIT : Try removing all the script references from each of the cars and see if that works , the only problem with that is that you'll loose the working coupler on the driving coaches ?

 

[shaneturner12]

I know what you mean - one reason why I'm using the version from the DLS for the testing (and used notepad to make the changes mentioned in the last post)

I am just having a look at car 2 - the max_coupler_gap figure is concerning me.

As far as I can see the problem is only affecting me in Native mode.

Shane

 

[nexusdj]

Just switched my 2010 over to native mode and built a simple test track route .
As soon as I placed the full 9 car set it flagged up a red dot error :

I'll try removing all the scripts and see what happens ?

 

[shaneturner12]

Same error I've been getting.

If you can, it may be worth trying to produce one from a portal and see what happens.

Shane

EDIT: Derailing issue still happens even after altering max_coupler_gap.

I will check tomorrow now, as I have to go to bed.

 

[nexusdj]

I've removed all of the scripts references except on car 1 .

No red spot error when full set is placed in surveyor .

So I've extended the test track and made it an oval , I've also placed 2 portals with a section of track between the 2 .

I've set up a session with 1 pendo going continuously around the oval with an additional pendo immediately coming out of the portal and travelling to the other every 15 minutes .

No errors or derailments

I'll add the script back to car 9 and try again . If that's okay I'll add it back to each car one at a time and re-test to see if I can single out the problem car .

********************************************************************************************************************************

EDIT : As soon as I add the script entries to car 2 then it generates the red dot error in driver mode and the pendo that comes out of the portal derails as soon as the 2ND car is generated .

So it's the scripting on cars 2 to 8 that's generating the problems . As a work around I could upload the train with the scripts removed from those cars but it may look weird as only the driving cars will tilt when cornering at high speed ?

I'll also compare the driving and trailer car scripts to see if there's anything missing between the 2 ?

 

[shaneturner12]

Thankyou for the assistance so far nexusdj.

I'm hoping a solution can be reached that maintains most of the pendolino's features (although I know a compromise will need to be reached somewhere)

In the meantime, I will do a bit of testing myself, and see if I can find the offending car.

Shane

EDIT: Results so far:

Car 1 - OK (even with scripts)
Car 2 - OK when script references removed (P.S. The game seems to stall a bit when producing the second and subsequent cars)
Car 3 - OK when script references removed
Car 4 - Currently not OK, undergoing testing.
...

 

[stovepipe]

I've had to do some fiddling with these assets before for Native TS2010.

If I recall correctly it is because some of the cars are not set up as a loco - "engine 0" in the config file - whereas the script declares it is a locomotive. It might be the other way round, but I've no time to check at the moment.

The fact you are getting a train.gs error certainly points to something like that. I will look later when I get in from work.

 

[shaneturner12]

I will also try your suggestion stovepipe.

I know that nexusdj will also probably be keeping an eye on this thread, so if I can get it to work, I will post back accordingly.

Shane

EDIT - PARTIAL SOLUTION

I have now managed to get car 2 to work (without removing it's script) by changing the isclass line in the .gs file from Locomotive to Vehicle.
I am still trying to get car 3 working by looking at the original class 390 pendolino by spanner/jivebunny.

 

[NikkiA]

Quite a trivial fix, really...

In cars 2,4,5,6,8 change 'isclass Locomotive' to 'isclass Vehicle'.

In cars 3 and 7, change 'isclass Locomotive' to 'isclass Vehicle', AND change line 61 to read 'Locomotive the_loco = null' instead of 'Locomotive the_loco = me' (you can't cast 'me' to a loco, since it's a vehicle, and it gets assigned to the frontmost loco anyway, which will be either car 1 or 9.


If you have msys or cygwin, you can simplify the process by checking cars 2 to 8 out, then executing:

find . -name \*.gs -exec sed -i "s/isclass Locomotive/isclass Vehicle/g" {} \;

Then just edit line 61 in cars 3 and 7 as above.

Otherwise its a bit of a slog going thru each car and performing one tiny edit.

Or you could just wait for nexus to upload a fixed version, which I'm sure he'll end up doing.


The scripts could probably do with some error checking, as they may well fail if people try non-standard consists (although a Thunderbird config should still work, since you'll still have a locomotive on the frontmost position, but I can see a situation with a broken apart consist being crashy, and before anyone says that would never happen, I've seen 47's dragging around half-HST consists that end with a 'doorway to hell' at the backend, it's not impossible the same would want to be done with a pendo, and besides, in trainz who knows what people might try, best to recover gracefully even in a silly situation.

 

[shaneturner12]

Thankyou NikkiA, you have provided the solution to both the issues mentioned in my first post, as well as the train.gs script error.

Just out of interest, is it meant to disable the sparks?

Shane

 

[NikkiA]

Thankyou NikkiA, you have provided the solution to both the issues mentioned in my first post, as well as the train.gs script error.

Just out of interest, is it meant to disable the sparks?

Shane

It's not meant to, I'm not entirely sure why it does, but then the sparks were broken for me before - always on - and I'd rather have always-off than always-on.

I suspect that whole spark_manager() function is a mess that needs debugging and rewriting in places.

Having a play now to see if I can get sparks working again, will update if I find a fix.

 

[nexusdj]

Many thanks NikkiA that's sorted it and the train still tilts

I did actual try changing locomotive to vehicle last night on the car 2 script but I got 15 errors listed in CMP (most of them about line 16) . The only difference I can see is that I didn't use a capital "V" on vehicle !

I've removed the sparker entries from the mesh table seeing as they don't work as I believe the sparker script relies on the fact that the vehicle it's on is a locomotive (I may be wrong with that fact!) .

If everyone can live without the sparker then I can upload the set as a higher kuid2 if not then I'll wait to see if someone can come up with a solution to the sparker problem .

 

[NikkiA]

Ok, that's... odd.

I did a basic edit to enable logging, well, ok, 3-4 edits because I hadn't put 2 and 2 together and realised it was using Interface.Log which no-longer goes to the message window in TS12, instead of Interface.Print. So I had 3-4 launchs until I realised that and got it printing debug info...

But on the final launch when I got debug info, I noticed that it IS sparking, but it's much much much dimmer than the original was - because on the original, it had 6 sparks in one place always-on, whereas with the script 'fixed', it only has 1 spark out of the 6 attachment places at a time.

Can you check with your version whether or not you're seeing dim sparks too? or if it's somehow something I've 'fixed' with enabling debug output.

If it is working now, but very dim, it probably needs a new corona that isn't quite so small/dim (but not a huge one either, just a bit bigger as you have to look very closely to see the ones I'm seeing)

Edit: also bear in mind that spark 'generation' is heavily curtailed below 8m/s (ie, 17mph, and no bloodnok, 10m/s is NOT ~35mph, it IS ~35kph though

 

[shaneturner12]

Another possible addition, which would also apply to the Class 321s, is possibly adding nightmode to them, so you can actually see them properly at night (mainly lit windows)

It is only a suggestion though - you do not have to accept it.

Meanwhile, I will see if I can find a solution on the sparker issue as well (I think that will be one for bloodnok, as I think it's his script)

Shane

 

[nexusdj]

Isn't the spark effect based on the power usage of the vehicle it's attached to ?
If it is then that vehicle wont be using a lot of power since it's been changed to a non powered vehicle .
The original script could of been set up by Jivebunny on the basis that this vehicle is powered which would mean the locomotive entry in the script was correct (or botched that way to get it to work )?