Trainz - No longer casual user nor community focused?

[pware]

Despite having been on the Forums for six years (and having used Trainz for seven), I have only been on PC for four. And I'm as lost for content creation as ever.

Welcome to the club. Creating models (e.g. rolling stock, vegetation, buildings, etc) is a "dark art" but, in my opinion at least, it has always been the case. I can manage to do a reskin, a non-environmental texture or even a driver icon, simple stuff that does not involve GMax or Blender.

It just feels like there's way too much that needs to go into even a decent model, and while yes, technology gets better, technology gets more complex, I feel the bar at this point is so high it's difficult to make a beginner's guide without getting lost.

Unfortunately the trend to greater complexity has been largely consumer driven. Few of us (perhaps none of us) would be happy today using models created for the earliest versions of Trainz. Now we want our models to be more realistic looking. That means PBR textures baked in, LODs to reduce the overheads, seasonal variations (winter, summer, etc) included, etc. A lot more time, effort and expertise is now required.

As with a lot of things in life, we now expect professional results but at an amateur price.

Will I keep Trainz? Absolutely. It's the only good train sim that works on Mac, and just as a driving experience it's pretty good.

In my opinion it is also the best train sim for PCs as well. As for creating content, routes and sessions are now my forte. As long as they are still "do-able" for us amateurs I will be happy to leave the models to those who have the time and expertise.

 

[big_b]

Well I've been active in the forums and Trainz for just on 20yrs now I think.And in that time I can not remember a version that actually worked as intended and have been able to use without any problems.It's always been the next version will have better features and better problems to solve without actually getting anything fixed correctly.And to me it seems like were still on that path only the problems are getting bigger and more complicated to solve for most users (without the technical knowledge) to do so,even the experienced users sound like there spend a lot of time just doing repairs rather than using the game/simulator to run trains to keep things going.
I seen a post about why the lack of new routes on the DLS only about 290 for TR22 a lot of which look like upload tests not drive able routes, probable due to constant changes as were always in beta and routes are constantly being upgraded to conform with changes as we often get a message to install updates to a new build not the one where all our routes are under construction as it may corrupt them until things are sorted out.
Before we go much further I thing it's about time we had a decent survey to find out exactly where everyone stands with all the changes and see if we are going down the right path.

 

[Williamg]

Could it be N3V are looking at TSW and considering going down that road? A high level tool set that is either too complex for the average user and ultimately only available to selected "associates"?

I don't think its there yet, but I have a feeling N3V is going in that direction with Trainz which worries me.

The two features that are really missing in TRS are dynamic lightning for streetlights, what I really hoped would be in TRS22 since we are on PBR Textures for a while now. And a type of schematic track view (Like a dispatcher view for example that is being used in Run8).

 

[boleyd]

They could. By culling various data sources (digitally) a fairly decent profile of a community/area could be made. Then apply existing N3V train, and environment, elements for a reasonable presentation of a small town, and eventually a city. Upgrades would be adding various digital/visual elements for a price. I would not be surprised to see that such data is currently available for use commercially, climatically, medically, and the list goes on as better and better models were created from an expanding list of data sets. All N3V has to do is get the funding to exploit this rich data to create environments to depict a railway or a customer managing a bus line in a city.

I would bet that many companies are exploiting today's data to model environments to assess the economies of a market. Why not present that to the public for their dynamic visual and functional enjoyment?

 

[1611mac]

How does this day of (real life, not Trainz) AI affect this? AI can do anything now... write songs, duplicate artists, write novels, create photos and art, create apps, you name it. Why not create actual or fictional train routes?

 

[johnwhelan]

How does this day of (real life, not Trainz) AI affect this? AI can do anything now... write songs, duplicate artists, write novels, create photos and art, create apps, you name it. Why not create actual or fictional train routes?

You missed out the "real soon now."

It has it's limitations such as knowing what is historically accurate etc.

Cheerio John

 

[jjslll54]

I've been with Trainz since the beginning. I think I missed out on the community version. I was looking for an alternative to Microsoft's Train Simulator at the time. I could never figure out how to build routes in that simulator. I'm primarily a builder rather than an operator. After twenty two years I still don't know how to run a session. Not interested except for testing. Maybe the new living railroad will help in that if it comes to fruition. I'm a self taught Gmax modeler. This game inspired me to go to collage to learn professional modeling and I learned to use 3dsmax and maya. Too bad those two programs are too costly for a hobbyist. In the process of learning blender so I can start modeling again. As you can see I'm one of the more extreme users and I like all of the new upgrades and what I can do with them. I can sympathize with the users that want a simple game they can drive trains on. Trainz started out as a virtual way to build model railroad layouts where space and money were not an issue and the only limit was your imagination. You can still do that with the current version of Trainz. No need to make it any more complicated than you need it to be. My two cents worth.

Jack

 

[pware]

AI can do anything now... write songs, duplicate artists, write novels, create photos and art, create apps, you name it. Why not create actual or fictional train routes?

That is the common, but so far erroneous, impression we get from the hype.

It has it's limitations

Correct. The new Generative AI systems, such as ChatGPT, work by using what already exists.

From Wikipedia.

Generative AI models learn the patterns and structure of their input training data and then generate new data that has similar characteristics.

(underlining is mine for emphasis)

Examples:-
"Paint a picture of this (image supplied) house in the style of Picasso" - it would look at all existing paintings of houses by Picasso that it had access to and use the common style elements it found to modify the supplied image.
"Compose a song in the style of Bob Marley using this poem" - ditto for Bob Marley's musical style.
"Create a video clip of Abraham Lincoln addressing the United Nations General Assembly" - ditto again - the latest video AI systems can create a video based on a single still image.

An amusing (or frightening) scenario that I recently came across was when these AI systems start generating data based on new data created by other Generative AI systems that has been added to their learning database. It could be a bit like an old trick that was performed with the early photocopy machines - make a photocopy of a document, use that photocopy to make another copy, repeat multiple times. Eventually the copy become so blurred, as each copy magnifies the imperfections in its source, that the document is unreadable.

My thoughts.

 

[martinvk]

I've been with Trainz since the beginning. I think I missed out on the community version. I was looking for an alternative to Microsoft's Train Simulator at the time. I could never figure out how to build routes in that simulator. I'm primarily a builder rather than an operator. After twenty two years I still don't know how to run a session. Not interested except for testing. Maybe the new living railroad will help in that if it comes to fruition. I'm a self taught Gmax modeler. This game inspired me to go to collage to learn professional modeling and I learned to use 3dsmax and maya. Too bad those two programs are too costly for a hobbyist. In the process of learning blender so I can start modeling again. As you can see I'm one of the more extreme users and I like all of the new upgrades and what I can do with them. I can sympathize with the users that want a simple game they can drive trains on. Trainz started out as a virtual way to build model railroad layouts where space and money were not an issue and the only limit was your imagination. You can still do that with the current version of Trainz. No need to make it any more complicated than you need it to be. My two cents worth.

Jack

I thought I was reading something I had written and except for the part where you went to college to learn professional modelling, we have a very similar trajectory with Trainz.:wave: Also, I never switched to Blender, having started with GMax, it still does everything I want to do. Perhaps one day when i reach its limits. Meanwhile, I get to use all my civil engineering in Trainz, building stuff that I could never afford IRL

 

[nugget2225]

I think N3V is just trying to do what people have asked. Even though i don't like S20 over version one, I will eventually learn it. Alot of us are creatures of habit, And it takes time to change. I never really had any interest in HD scenery myself, But it does look much better. I am more interested in trainz a living world. My biggest issue would be all the aging assets. The bridges , buildings and roads, are just so outdated to me, And we have to constantly use the same few assets over and over. Track, Textures and functionality seems to be all going in the right direction to me. As far as the development side and building assets and scripting, I have no idea how to do any of that. I have always been interested un route building and playing session and doing sandbox playing. I would like to see more actual functionality in the trains themselves. I know Jointed rails and others make amazing models and cabs, But why not more functionality. Starting trainz from cold and dark, More switches. I simply can't compare one train sim to another because each one has good and bad about them. Thats just my 2 cents worth.

 

[JCitron]

To clear the roughly dozen warnings from the script the script needs to be replaced which means a dozen lines of config.txt file need to be added or changed. The original scripts worked fine in the original version of trainz. Second references to kuid2:101839:60002:1 need to be replaced with kuid2:101839:60002:3, this happens in about five places per config,txt file, Next all references to 3link_hanging.im need to be changed to 3link_hanging.lm. Note the associated lm.txt file only has the one .im entry. Same for 3link_coupled.im, and a few other .im files.

certain tags need to be removed such as shadow and light. The build number needs to be changed. Yes it is only a 430 line text file but you need to know your way around it.

Then you get to the reskins. I've spotted a dozen that need the same changes but I'm unable to make them as they aren't under my kuid and I doubt if the reskinners have ever touched the config.txt file in such depth before.

Doing one takes a few minutes then you need to check them for errors. Multiply that by a few hundred assets and swamp my beta tester with updated assets and you get an idea of the size of the problem.

Cheerio John

Not to derail this thread too far from the topic, there's a utility I use to bulk update config.txt files and other text files.

Try Search and Replace Master.

https://www.knowlesys.com/software/search-and-replace-master/

With this program, you can specify the strings you want to replace within the same text file located in multiple directories and is great for updating config.txt files that have the same string that needs replacing. To use this, you open the assets needing update for editing, set the parameters in the program and run it. There's a summary at the end stating how many were replaced out of so many strings found.

 

[lackoo11111]

Meet the new N3V product - TRANSPORT World.

We already have two games like that . Transport Fever and Transport Fever 2 .

 

[johnarran]

Not to derail this thread too far from the topic, there's a utility I use to bulk update config.txt files and other text files.

Try Search and Replace Master.

https://www.knowlesys.com/software/search-and-replace-master/

With this program, you can specify the strings you want to replace within the same text file located in multiple directories and is great for updating config.txt files that have the same string that needs replacing. To use this, you open the assets needing update for editing, set the parameters in the program and run it. There's a summary at the end stating how many were replaced out of so many strings found.

Nice !

 

[boleyd]

N3V is holding HD scenery version training sessions online. This suggests the HD patch is neither simple nor intuitive.
Over the last few years, I have concluded, Trainz is no longer for the casual route builder, at least to get to the level of the quasi-professional, Payware, routes that will now appear.
The skills bar is way past what the casual builder may be prepared to invest time into to match what will now be offered.
Ever joined a modelling club to find out it is a 'Rivet Counter' clique filled with competitive museum quality builders you have no hope in matching. (And soon left)
Not being able to get even near build quality promoted without a great deal of effort and expense, (Payware), will discourage and lose N3V customers early.
This will leave mainly 'Session' builders and 'Drivers'. Rolling stock, Trackside, vehicle and buildings assets etc., will only be built by Long Time Users, "LTUs" (the non-disgruntled ones), and quasi-professional (Payware) builders.
Is this the path N3V wants to take, a 'Payware' model by quasi-professional builders?
Is Trainz devolving to what Microsoft Railroad simulator was, a 'Driver' focused game?
Perhaps N3V needs a cutdown 'Railroad Builder' version focused on a 'reliable' inbuilt set of basic operations tools and assets? TMRS2004+ with PaintShed?
I am tired of having to spend days or weeks recovering from the latest version/patch of my aged route. Beta Testing done on perfect, latest version builds only?
I am tired of trying to come up with work arounds that can't fix all the problems, leaving yet another partially degraded railroad to my eyes.
I am tired of NOT having a utility I can run to highlight a list of what is now incompatible and providing possible alternatives. (non-destructive 'Compact' tool)
Blaming 3rd party assets is NOT a solution considering that the DLC, (3rd Party and free Assets), is used as a Marketing carrot.
I think N3V has forgotten that most LTUs, have invested years of time and have legacy layouts. My railroad has grown out of TMRS 2004 and possibly TMRS 1.3
So why would anyone now pay a yearly subscription for a game that regularly breaks older user created routes when patched, and has raised the Skills Bar and time input substantially?
You can still install different versions concurrently to access your favourite routes. (Assuming N3V will still allow concurrent versions + light route and session modding)
I think the better alternative now is to buy anew each? time and simply run inbuilt route/sessions and modify the occasional route/session from the builtins.

Trainz is a product for relaxation and creation That product no longer exits (for me) in the latest release. As time went on the Trainz product was a reliable source of casual enjoyment. That has changed. It is now a new learning curve that is far from casual. It is more intense than the one we had when we started. Evolution is a steamroller, out of control. Adapt or "just fade away" (Douglas MacAuther).

 

[pcas1986]

To clear the roughly dozen warnings from the script the script needs to be replaced which means a dozen lines of config.txt file need to be added or changed. The original scripts worked fine in the original version of trainz. Second references to kuid2:101839:60002:1 need to be replaced with kuid2:101839:60002:3, this happens in about five places per config,txt file, Next all references to 3link_hanging.im need to be changed to 3link_hanging.lm. Note the associated lm.txt file only has the one .im entry. Same for 3link_coupled.im, and a few other .im files.

certain tags need to be removed such as shadow and light. The build number needs to be changed. Yes it is only a 430 line text file but you need to know your way around it.

Then you get to the reskins. I've spotted a dozen that need the same changes but I'm unable to make them as they aren't under my kuid and I doubt if the reskinners have ever touched the config.txt file in such depth before.

Doing one takes a few minutes then you need to check them for errors. Multiply that by a few hundred assets and swamp my beta tester with updated assets and you get an idea of the size of the problem.

Cheerio John

Hi John,
Its more than a few hundred and in fact is several thousand assets some of which are yours. Many of these assets work just fine in TS22 providing compatibility is turned on. But one cannot cannot correct and upload to the DLS any asset that has obsolete functions. I suppose that is N3V's way of forcing changes albeit slowly.

Conveniently forgetting about LOD updates because they can be hard, the biggest problems are script updates. Some are relatively simple but some, and especially those that want to load other assets config or searches for objects, can require quite a lot of code update. I guess affected assets will get there one day but it won't be tomorrow, next week, or even next year.

Apologies for the late response but I was out for maintenance.

 

[johnwhelan]

Hi John,
Its more than a few hundred and in fact is several thousand assets some of which are yours. Many of these assets work just fine in TS22 providing compatibility is turned on. But one cannot cannot correct and upload to the DLS any asset that has obsolete functions. I suppose that is N3V's way of forcing changes albeit slowly.

Conveniently forgetting about LOD updates because they can be hard, the biggest problems are script updates. Some are relatively simple but some, and especially those that want to load other assets config or searches for objects, can require quite a lot of code update. I guess affected assets will get there one day but it won't be tomorrow, next week, or even next year.

Apologies for the late response but I was out for maintenance.

The LODs aren't a problem, most of mine had had them for a long time. There are a couple that have 503 polys on the lowest LOD that need fixing but that isn't a major problem. The major one on the scripts is switching out superscript and dropping in JAG which seems more robust and isn't encrypted. I standardised on JAG sometime ago and they seem fine. The first 400 are done, I have another 300 plus that should be fairly simple and a couple of dozen that that will need some TLC.

The reskins are a concern, it would be nice to have limited CRG capability so I could fix them up.

What would be nice is N3V to release a set of "standard" scripts. I seem to recall Paul Hobbs mentioning it would be nice as well.

Cheerio John

 

[textrain]

Trainz is a product for relaxation and creation That product no longer exits (for me) in the latest release. As time went on the Trainz product was a reliable source of casual enjoyment. That has changed. It is now a new learning curve that is far from casual. It is more intense than the one we had when we started. Evolution is a steamroller, out of control. Adapt or "just fade away" (Douglas MacAuther).

I agree with you. I cannot grasp the new surveyor. As long as I can use the classic surveyor I will continue to enjoy the game.

 

[ZyxwvU]

I came back to Trainz series after ~15 years in 2020, jumping into TRS19. On one hand, it was nice that Surveyor did not change that much, on the other I quickly recalled many of its limitations. I wasted many hours of my life doing trivial tasks (e.g. rotating every single lamp) that could take maybe 10-15 minutes, if not the limitations of the available tools that I could not understand. When I saw Surveyor 2.0, it was like "f i n a l l y". Yes, I had to spend some time to learn it, but it was a lot of fun to me to discover how different tools can be combined to achieve various effects with greater accuracy and reasonable time. Yes, the interface could be polished here and there (eventually this is what "T+" is also for), but the general philosophy has already saved tens of hours for me. And I prefer learn the tool once and have my tens of hours of life back, using them more productively on other things .

Regarding asset creation, I see two aspects:

1. "simplicity" of creating an asset is both a blessing and a curse. On one hand, yes, it allows many people making them. On the other hand, I see that many people also don't care about basic technical things that they really should care about when they release anything. Random stuff put into config files, sub-optimal performance... it would be fine, if those content creators accepted the fact that they make mistakes, that IT world works in a certain way and does not distinguish between "casuals" and "professionals". If they had the will to learn, fix and improve. But often it goes in a wrong direction of putting blame on everyone, except oneself, when things break, when Trainz performance suffers, etc... this is what I don't like.

2. I tried creating assets in an "old way" before learning PBR techniques and after some time I find making PBR textures... easier than normal textures with tools like Substance Painter. The only issue I see is the need to learn the tool and to buy it. Unfortunately, we have a good, free 3D modelling tool (Blender), but there is no equivalent of Blender for PBR texturing yet

Things which are hard to me in Trainz is currently operating the railroad in the Driver mode, e.g. doing some shunting, scheduling some trains. It really depends on the quality of the route - I often run into issues with waypoints, misconfigured signals, junctions, cars, lack of thinking about proper naming of objects, lack of visibility where different types of cargo could go etc... there's a lot of stuff that one must understand, check and fix manually to make the route actually playable beyond the predefined sessions.

 

[johnwhelan]

I came back to Trainz series after ~15 years in 2020, jumping into TRS19. On one hand, it was nice that Surveyor did not change that much, on the other I quickly recalled many of its limitations. I wasted many hours of my life doing trivial tasks (e.g. rotating every single lamp) that could take maybe 10-15 minutes, if not the limitations of the available tools that I could not understand. When I saw Surveyor 2.0, it was like "f i n a l l y". Yes, I had to spend some time to learn it, but it was a lot of fun to me to discover how different tools can be combined to achieve various effects with greater accuracy and reasonable time. Yes, the interface could be polished here and there (eventually this is what "T+" is also for), but the general philosophy has already saved tens of hours for me. And I prefer learn the tool once and have my tens of hours of life back, using them more productively on other things .

Regarding asset creation, I see two aspects:

1. "simplicity" of creating an asset is both a blessing and a curse. On one hand, yes, it allows many people making them. On the other hand, I see that many people also don't care about basic technical things that they really should care about when they release anything. Random stuff put into config files, sub-optimal performance... it would be fine, if those content creators accepted the fact that they make mistakes, that IT world works in a certain way and does not distinguish between "casuals" and "professionals". If they had the will to learn, fix and improve. But often it goes in a wrong direction of putting blame on everyone, except oneself, when things break, when Trainz performance suffers, etc... this is what I don't like.

2. I tried creating assets in an "old way" before learning PBR techniques and after some time I find making PBR textures... easier than normal textures with tools like Substance Painter. The only issue I see is the need to learn the tool and to buy it. Unfortunately, we have a good, free 3D modelling tool (Blender), but there is no equivalent of Blender for PBR texturing yet

Things which are hard to me in Trainz is currently operating the railroad in the Driver mode, e.g. doing some shunting, scheduling some trains. It really depends on the quality of the route - I often run into issues with waypoints, misconfigured signals, junctions, cars, lack of thinking about proper naming of objects, lack of visibility where different types of cargo could go etc... there's a lot of stuff that one must understand, check and fix manually to make the route actually playable beyond the predefined sessions.

Professional you mean N3V payware with none rotating wheels on the carriages?

Find a content creator who makes reason quality stuff and stay with them.

Cheerio John

 

[ZyxwvU]

I fully understand the pain of seeing the non-rotating wheels. I think I'm similarly annoyed when I see a house with night mode which looks like the same room has lights turned off and on at the same time . But in my comment, I was writing about a slightly different thing.

I'll explain more what I had in mind to be more clear. One day, I met a person who complained that Trainz is slow. That person was also an author of a couple of assets. In a general discussion about Trainz mentioned the fact that the number of materials used by assets affects the performance and that some assets use dozens of materials instead of 1 for no reason - then I got attacked by that person. The attack was focused on things like "in the past, things did not work that way", "how dare you rating assets optimized for older computers with current standards (???)", etc. The conclusion was that the person is not a professional graphics designer, so the game should not force him to think about optimization, and no matter what he does, it should work fast (and I'm the "bad guy"). This is not how computers (and graphic cards) work and such approach isn't very healthy both for that person and the community. N3V can't really make the game faster in this particular case. By the way, regarding the claim about the past behavior, I looked at some articles from 2005-2006 and the number of materials used by 3D model also mattered at that time.

Is the context of my comment more clear now?