An open letter to N3V engineers

Before people start changing scripts, please think
-The script itself is just fine, SetHeadlightState(true/false) is a valid way to turn headlight off/on
the script writer wanted to turn ALL lights off in "case S_NONE" there is nothing wrong with that.
[...]
please dont correct/error stuff that is basicly ok

The problem is that the way it was set up is wrong. This only turns off the default headlights, which have little use on this locomotive, since custom coronas are in use, which are not turned off. Removing this line changes nothing about how the custom coronas work. The actual light "shine" effect would also be controlled by this, but since the custom coronas are still on, and alas, the actual headlights are still on it wouldn't make sense for the light shine to disappear either.

Furthermore the problem is that whenever the script changes something on the lights itself, it treats this exactly as if the user would change something as I've already explained two pages back, causing the game to freeze because the script enters an infinite loop of spawning more processes. So no, this is not okay.

Since I know that polish locomotives traditionally have rather weak headlights and the default light "shine" from the game is probably too powerful in comparison to this, I'll see if I can write a short script addition that authors of polish/eastern european locomotives can implement to improve the realism if they choose so by making the light effect at night weaker and closer to reality.. This is obviously up to the original authors and not CRG to add new features.


Greets, Mika
 
May I?

Even though the above mentioned script had its issues I think this might point a bit general situation in the trainz-developement process.

Let me show you an example.
There was this pretty nice loco with some well-scripted feature: http://www.ot2007.de/index.php?page=DatabaseItem&id=27
My personal favourite feature was the asymmetrical interior-view.
Using Alt+C for changing cab you were able to see the long (left side) and the short end (right side) of the loco in TS2010 and TS2012.
Then some magic happened starting from TANE and the very same loco installed from the very same cdp just started to act differently.

And I am not against the development and hopefully it is clearly visible how the graphic has developed between the versions (since we are looking at the very same model)

Not knowing the details so it is only a fictional guess from me:
- in the dev process an idea popped up to change something (for example degree to radian)
- meeting followed the idea, a decision followed the meeting
- basic test were okay, so everyone in the dev and QA team were happy

The only problem here is that there were some affected assets outside of the test-box.

And there are some of these small irritating issues where Trainz is just not fully compatible with its predecessor version. (see also: Speedtrees, or faulty DLS objects)

But such discrepancies would be avoidable if there were:
- more detailed changelog between versions
- up-to-date development guidelines (CCG) with version differencies

As my personal opinion: these are clearly missing.

It is easy to say that the faulty content need to be updated but there are lot of older content out there (from inactive members) that will never be updated.
And it should not be a surprise that dozens of players are using an older and more stable version (i.e TRS2009) instead of updating and fighting with the new versions in every 2-3 years.

Guys you are doing clear steps forwards with Trainz just please make it a bit more visible how we can stay on the same path.

Best regards,
Marc


26817313_f18f9ba62f8d0986744fb4294cb55cf6_l.jpg

Bigger size: https://img1.indafoto.hu/1/0/242722...7313_f18f9ba62f8d0986744fb4294cb55cf6_xxl.jpg
 
Last edited:
Thank you guys for your contribution! Mika, Santa Mika (Mika is more or less Nicolas, isn't it? We call santa just Nicolas or Santa Nicolas - I know, I know in Germany you have a Weinachts Man which sounds very serious in the end ;-)
 
Two thoughts about this. Firstly, great relief that Samplaire and his fantastic content will not be leaving us. Secondly, as this fix involves the same change to the same script in (presumably) the same folder in multiple assets, could it not be automated in some way and therefore speedily implemented and revised iterations of the assets uploaded? Or is that far too simplistic?!

Paul
 
I add a suggestion to this topic which is related to the way of recieving content

The Download system could be improved a little bit because I didnt find the Platinum content for Trainz Railroad Simulator 2019 at first, the Download manager included in the content store is not an intuitive solution to find the content, it could be named better like Content Download Manager.

Kind regards
Georg
 
Yes, the way the downloads are organised in the Content Store does seem haphazard. There is a filter on the top right corner where you can put in a search term, such as TANE or TRS19, to narrow the confusion down a bit but even then it can be difficult to find things.
 
Content management is kind of taking this thread off Samplaire's topic, suggest a new thread if people want to dicuss an unrelated topic rather than hijacking this one. ;o)
 
Now you have what you wanted to have!

There is a group of Polish map creators who were preparing a huge map for TRS19. Now the map crashes the game to desktop. About 6 guys worked hard for months and now with that small decision you ruined their work. Think, you are responsible for many people now, not only for the game to be as much efficient as possible. And this is not that everybody reads the thread, not everybody knows English...

OK, there are good people here who work now on solution for this armageddon. The script you sent me, Tony, works ok with that SM03 but the solution in it can't be just copy/paste to other locos. Polish lamp signals are not that simple to achieve in trainz and that is why the problematic script was introduced. Everybody works hard to have the game as realistic as possible. You also make mistakes in code otherwise no service packs would have been necessary...

Very good xmas present, yes it is!
 
What we want to have is a happy, thriving, growing community. I’m not 100% sure whether your post refers to things getting better or worse.

If there is a new problem, send a bug report and we will fix it. We do this on a regular basis.

you can always install an older version and not update if you prefer.
 
It seems you don't understand the size of the problem Tony...

Let me explain then. For the last 10 years we were making models using those scripts for improving realism/fun/possibilities of their eventual use. Yes - there was a hidden issue in the script that someone left their and forgot - I didn't even knew this exists because I don't have a clue about scripts. I trusted their author that all is ok and continued my work to bring more fun and possibilites for the community. For 10 years there was absolutely no problem with any of these locomotives, everything seemed to be cool - everybody got used to it (I have no idea how many people around the world used these models throughout that time). Many of these people never visited this forum. Possibly many others do not use English and are not aware of the problem at all. Many others bought TS19 and recently, unaware of the problem installed your last updates to this version... and suddenly everything blew up to smithereens -> game crashes, maps are being lost, content being damaged, angry people crying etc. Many users do not know what is going on - they only know that suddenly a lot of their work is gone or damaged. It's not even possible to locate and contact all people who might be hit with this situation.

Making such "surprise repairs" in places where for at least 10 years everything was working ok will not bring you a "happy, thriving, growing cummunity" but a lot of frustration instead - this is what Samplaire meant in his post.

Now we're trying to locate the problematic scripts, add the necessary changes and later plant them in all the locomotives. There's some unexpected issues though because the script is quite complicated and has a number of versions. It will also take a lot of time to plant and re-publish on the DLS. Since I am the author of most of these locomotives, I'll try to fix all that is possible and re-publish it. That's the best I can do to help with the situation. A piece of advice for the future though: if something works ok and does not cause trouble - don't "fix" it.
 
Last edited:
A piece of advice for the future though: if something works ok and does not cause trouble - don't "fix" it.

While I understand the meaning of that often quoted saying, it also means "don't improve or advance things, keep everything the same".
 
What we want to have is a happy, thriving, growing community. I’m not 100% sure whether your post refers to things getting better or worse.

If there is a new problem, send a bug report and we will fix it. We do this on a regular basis.

you can always install an older version and not update if you prefer.
Not trying to be rude, but ugh. This is a typical generic customer-service response, and I can imagine how unhelpful and annoying this is for some people.

My suggestion is this. You cant expect 10-year-old assets to work in newer versions of trainz, without upgrading the scripts. This patch may have broken some older content, but now after SP2, Trainz runs so much faster, smoother, and more efficiently (A plus for people who work with the latest, up-to-date content). I think trainz is just trying to move ahead, but in the process, (unfortunately) they have pushed older stuff behind them without really paying much attention.

So its really a trade-off between maximum performance, and maximum compatibility.
 
While I understand the meaning of that often quoted saying, it also means "don't improve or advance things, keep everything the same".

I'd phrase it thus, if the 'improvements ' pr 'advacements ' are more trouble than they are worth....leave well alone.

It took centrelink ( australian federal govt pensions welfare agency for those who arent aussies ) 5 months to sort out our pension , I eventually had to see one of their experts in order to get it sorted, turns out they had three separate data systems, all more or less doing the same job, in order to sort out our claim,he had to use the oldest version as both the newer ones would not allow him to enter the data, I'm sure in some aspects the new versions were very good with new features, but if they don't allow the user to do what they need to do, or could do previously , then they are not an improvement.

Its almost impossible, given the many, many assets made by so many different people with different levels of expertise using different apps to create the items, for trainz to work flawlessly in any new versions, its a bit like windows which has been patched and fixed and patched and fixed over and over again, making it work well is like trying to patch a leak with bandaids. This is one reason why I have stuck with TANE SP3,. It has flaws, but I know the flaws, I can work round them.


NV3 needs to keep making new versions of trainz to stay profitable, if users want their old stuff to work then best keep a version of the older software which you know it does work with essential items , rather than chase the 'improved' versions , as the updat may well be better in some areas, but a lot worse in others. I'm sure that NV3 deeply regret the fact that there is so much freeware out there which is not working well with their newer creations, but although the availability of those assets is one of the games best features, its also one of the games worst when it comes to updating the app and ensuring it works properly . No ones forcing you to update folks, if you want security, stay with the older versions.
 
Last edited:
Agreed. If an asset doesn't work with the latest version, that's not up to N3V to fix. Its harsh (esp with thousands of assets) , but it's up to the author to keep assets up to date and "healthy".
 
N3V should be fair enough to announce well described and pretty detailed list of upcoming changes to get the community ready in advance. Not to brake everything silently and wait for reactions.
 
N3V should be fair enough to announce well described and pretty detailed list of upcoming changes to get the community ready in advance. Not to brake everything silently and wait for reactions.

Yes they can do that, but its almost impossible to predict if changes will cause issues, as there are hundreds of thousands of assets, checking if all of them work with a new version is a huge effort, it would take dozens of people checking for months. I don't like the situation any more than you do, my 80 mile route I've been working on for three years will no doubt never work correctly in 2019 as it has many assets made in gmax and customised billboard trees as well , but there is no hope of remaking them in the near future in 2019 comptible forms so I have to stick with the older software , or , change the items so the routes will work. We cant expect NV3 to not release new items as they have to stay in business, I don't mind as long as they keep older versions available to download, but they havent done this with Tane SP3, if i were anyone using trainz, i would keep copies of ALL installers as an insurance policy so you can go backwards and use the older software on routes you want to keep running.
 
While I understand the meaning of that often quoted saying, it also means "don't improve or advance things, keep everything the same".

No. It means "improve or advance elements that are important and requested by a wider community, don't improve things that are not perfect but basically work fine".

Not trying to be rude, but ugh. This is a typical generic customer-service response, and I can imagine how unhelpful and annoying this is for some people.

My suggestion is this. You cant expect 10-year-old assets to work in newer versions of trainz, without upgrading the scripts. This patch may have broken some older content, but now after SP2, Trainz runs so much faster, smoother, and more efficiently (A plus for people who work with the latest, up-to-date content). I think trainz is just trying to move ahead, but in the process, (unfortunately) they have pushed older stuff behind them without really paying much attention.

So its really a trade-off between maximum performance, and maximum compatibility.

No my friend, you misunderstood. These are not 10-year-old assets - the script was created 10 years ago and throughout those 10 years everything was ok, nobody ever pointed out that there might be a flaw inside that in great theory can cause trouble in the future. The last locomotive based on that script was uploaded to DLS about 2 weeks ago and was working perfect before the last updates in TS19. So it's not some prehistoric content, it's fully up-to-date content - and yes, I can expect it to work in new versions because I made a lot of improvements in the models themselves to be up-to-date - including fixing config files and some solutions unwanted in the more recent versions of Trainz. There's a few hundred of these locomotives but they were upgraded throughout those years and most of them are up-to-date.
 
Last edited:
No. It means "improve or advance elements that are important and requested by a wider community, don't improve things that are not perfect but basically work fine".

What a strange statement. Because something "basically works fine", presumably for some users, does not mean it should be maintained as a "feature" indefinitely.

From your product timeline you probably remember the animated junction points created by Mike10 long before procedural tracks with their animated points appeared. His system was based on an undocumented feature (I believe it was actually a "bug") in the version of Trainz that existed back then. When the bug was fixed his animated junction points no longer worked and there were howls of protest from some sections of the community. The response from Auran/N3V was the same as you would get from any software developer - undocumented features and those based on bugs are not and should not be supported.

As others have pointed out, the cost of supporting poorly written scripts, badly designed textures and meshes, etc so that these assets can continue to be used, is poor performance for everyone.

As to what N3V should be doing about this, consider this scenario. Suppose you modify a new car by replacing some of the manufacturers engine parts with poorly designed non-standard replacement parts that have a feature or features that the originals lack. You then discover, after an engine software upgrade from the car manufacturer, that those parts no longer work. Is it the manufacturers responsibility to change their engine management software so that those non-standard parts can continue to work?

I can understand why this has annoyed you. You have obviously put a great deal of effort into your project and have been let down by multiple poorly written and modified scripts from third party creators. I would also be annoyed (and in fact have experienced that annoyance myself when a bug from N3V destroyed all my work in creating a session in one of the early trouble plagued TANE releases) but in your case complaining that N3V has a duty to either fix the problem or allow their software to run poorly written scripts (or other issues) is not fair or reasonable.

I think your only real option is to contact the original creators of the script(s) and ask them to update/correct their scripts. No-one should be able to understand the working of these scripts better than the original creators and that may be why the versions modified by others have also failed. Perhaps others in the community can help, as has already happened.

But I can see that we are probably never going to agree on this matter.

My best wishes anyway.
 
Last edited:
Gentlemen,

As I said before, N3V probably is not aware they (Auran, actually) made not just a piece of software but a platform which us the community develops along with them adding our talents for the current and future success of the platform! If they dont understand it or underestimate it they ruin it. I bet most of the people who buys new versions of Trainz are loyal followers. Hough!

They should cooperate.





The problematic script was not made because we just wanted it. It was made because plain loco script had lights just on or off which was not realistic. Saying that the script, perhaps not perfect, enhanced trainz possibilities! If N3V was perfect, there had not been any need for service packs. Nobody’s perfect in terms of coding, this is a multi dimensional process and they should understand it beforhand saying ‚the script is bad’.
 
Last edited:
Back
Top