Tsc - Future Of Train Simulator Classic Developer Update 15th March

Amazing stuff already! Really happy to see this game improving and seeing this sort of developer updates to keep us updates! Really looking forward for more

 

Good news! Could the TSC be on a cloud gaming platform like Geforce Now?

 

It’ll be great once the quick-drive consist editor works properly

main issue is with 3rd party content and reskins

but sounding great so far

 

ah so there is DirectX12 translation tool out there..... neat. i hope it possible in the future to upgrade Silverlining 3d cloud version 6. https://www.sundog-soft.com/2021/03...-better-3d-clouds-to-simulation-and-training/
also in version 6 they seem to fix that color banding issue.
Eliminating Color Banding in SilverLining - Sundog Software (sundog-soft.com)

i hope lights can be added to lofts in future update.

 

First; Really really great that you keep us updated, kinda like the roadmaps on tsw. It is a good and including thing.
One bug i really would like to see fixed, is page two in QD AI spawn points becomes inaccesible (probably when the scenario tracks.bin reaches a certain size. It’s like that happens at the exact point where the quick drive scenario no longer opens in the the 32bit version)

 

I'm a bit worried by the vagueness on backwards compatibility. IfTSC stops being a stable platform where I can be sure that old content will still work, I and many others will lose interest.

 

I love to see the proposed fixes to the error reporting. As a 3rd party dev of 2.5 years at this point, I am however saddened by the lack of any mention of the Blueprint Editor. It needs a good update since its current version runs tragically. It seems to take up as much CPU usage as the game itself! I feel like it also needs better error reporting, and I hope this part is also included in what you said. There's also a bunch of things in multiple blueprints that go completely unused, such as Heat Haze in RenderComponent (when has TS ever had such a feature?) Along with redundant blueprints and such things I feel it'd be good to "trim the fat" out of BPE. Another concern regarding development would be to give the Developer Documentation a once-over, because there's a lot missing. As for the game itself, CPU core usage (as I understand it at least) is tragically unoptimised, from what I've heard the game only utilises one core of the CPU? Also, one last little gripe, particles are rendered last! So the "post"-processing effects such as the black and white (or more noticeably, the sepia) filter completely ignore the particles. Glad to see you guys are picking up proper development on TSC, wish you all the best and I hope it'll have a legacy similar to MSTS!

 

Sometimes when you need reload scenario from saving you can meet problem with broken loco scripts (cut out SiFA, PZB, LZB), for example. Can your consider this issue? There is also problem sometimes with overspeeding after reloading of scenario from the save. Theese are very frustrating bugs for career scenarios.

 

hey would multicore improvement fix the flying underground slowness in editor?

 

This was a great read. Thanks for not shying away from being technical.

 

I haven't been playing TSC for a while now, because I switched to TSW, but this article looks very good.

I have a lot of experience in software development and the approach makes sense to me. It really feels like TSC gets a second life. It may be you need to stop supporting Windows 7 and 8. I think this is OK, they are no longer supported by Microsoft for a long time, and you cannot have it all. It may be a good idea to keep a frozen old version available for players that really do not want to upgrade.

Thanks for sharing this!

 

I'm scared to ask - and should not have to, but this is DG. Will the upgrade be free - or will it release as TS classic '24 (or something), requiring us to buy it again?

 

This sounds very promising.

 

You never had to rebuy TS for any update through all of the years since RailWorks. This is not TSW

 

This is very exciting. I appreciate you're probably not looking for feature ideas while you're still working on the foundations but I would be interested whether TSC will package together older RSC/TSC assets and those essential UKTS packs into the base game, so that anyone who wants to start a collection they're not downloading a load of packs then having to buy some really old routes and assets just to have the basics. Maybe add pre-2010 content to the base game?

The other question is whether it would be possible to change time of year/weather/maybe even time of day for scenarios? I realise that the driving train and all AI is set to a specific time but a way of adding 6 hours (for example) to everything wouldn't make a huge difference to coding but with different weather and at night the same scenario becomes a different challenge.

Updated routes too, ones that aren't in my idea for the early content starter pack. Hopefully we'll also see assets, traction in particular, made from scans and therefore hopefully a quicker release schedule as TSC's most modern rolling stock is becoming as old hat as the coding used in the original game. In the UK we have only classes 700 and 800 from the last 10 years' new rolling stock revolution.

 

I can’t answer you other than IMO it would be totally fair if this was a paid upgrade.

 

They mentioned SilverLining update - as nice as the AP weather is, it is based on the old system of only having 4 different daytime definitions each representing 3 months and it's only realistic for a certain latitude and involves almost everything to be set up manually. Would very much appreciate an update for this global sky system - or a better one, as found in various flight sims.

 

This is why it crossed my mind. I originally bought TSW-CSX thinking (misled?) it was the base game with a single route. Only to find I'd bought a DLC - then had to buy TSW. Since then there have been 3 (4?) new versions that keep charging for the base game and I've paid for CSX route at least 3 times now along with the base game. Clearly I'm a sucker, but could not bring myself to fall for it again with TSW3.

I admire your confidence, and am hoping you're correct, but what's to stop them doing the same here? I appreciate though, that it seems like one hell of a lot of work, with 1000's of errors to correct, but can't see DTG doing this for free.

EDIT: Release history of TSW - and a pattern of releasing I can see DTG going through again with TSC:
Train Sim World: CSX Heavy Haul
Train Sim World
Train Sim World 2020
Train Sim World 2
Train Sim World 3

All separate purchases and not all DLC's made transition from TSW2020 to TSW2.

 

I have to agree. It's not a given to get such a massive update (which I had never deemed possible a few months ago), I would pay for it too . It could even come as a separate game in Steam, cross-sharing the DLC, so you could keep the legacy version without tricking Steam

This will push me into buying a new rig, will then use my current Windows 7 system for Linux only. Fair enough

 

Sounds good as new ready-to-go global sky/weather system will be nice feature both for new and existing routes. Weather/sky blueprints are one of the most complex to investigate.

 

Technically nothing, practically some.

Up until now, the core game has been updated, so whoever purchased it once, did get all the updates. The revenue comes from selling DLC. I have 450, do the math, plenty of cash. The DLC bundles are actually quite cheap - a lure for new players and an offer for loyal ones. (They also did a good job funneling players towards specific content, which in turn empowered workshop.)

In case of TSW they have started adding new games.
But they also have to re-add all the DLC. TSC has nearly a thousand. Arguably, a great opportunity to rework some of them, but still.

Also, while many of us simply play the Steam version, many others have 3rd party stuff, and they expect those to be supported. As they said, losing one or two might be acceptable for the greater good, but that could mean having to keep multiple installations. And that's quite the cost, as flip-flopping (3rd party) content is beyond clumsy, so people just run installations over 500 GB. Using SSD. Big nope.
And then all the retired, but perfectly functional stuff. Workshop, that's a big one.

Technically, with a bit of folder magic, multiple versions can be kept for compatibility, similar to the x86-x64 choice.

What they can do, is similar to what I'm advocating, keeping the game a bit modular and locking specific features behind a specific purchase. Very similar to the US / EU asset packs.

 

It’s bad karma, but i am going to remind those who ridiculed me when i said:

Give us 64 bit. I was told it could not be done. We got it. Because turns out it was possible after all.

I said, give us multicore support. I was told it could not be done. Turns out it’s possible after all.

I said give us dx12…..well you know the score by now….

You know who you are ;-)

 

If you make significant updates to TSC - especially if backward compatibility is significantly lost - please release it as a separate Steam product so those of us who can't or don't want to update can still play our large and expensive legacy inventories.

 

If we get a new more efficient code i’d like this to be considered. Give us the the option to load tiles of assets from a greater distance from the player train than now. With a more efficient engine, it could be up to the user to decide what you want to “waste” processing power on. I really do not need more fps. But i dislike assets popping up at 800m distance and would gladly trade away even more fps for greater loading distance.

 

hey with DX12 could dlss 3.0 be supportive in future?

 

ya even track on few routes cut off at short distances then others. for example cn yellow head route track end at shorter ranger then Canadian mountain pass track witch the distances is greater. now for me using blender to build model the way i know how to make object that can been seen for mile distance is 1_2000_object name. but even the number has limited i think, 1_2500 and that how i make object load from far away distance for me anyway.

 

give me the winning lottery numbers for Saturday night

 

The article says "Physically-Based Rendering (aka PBR) and a hardcoded DirectX version upgrade (non-translation). We know these will ... likely to impact the workflow for everyone who makes content for Train Simulator making it much harder for people to make content"

What will be harder? Route building (will we keep the editor)? Reskinning? Loco creation? Something else?

 

Read again, those are things they are not doing.

 

Thanks for pointing out!

 

Keep breathing, TSC already does PBR rendering, just not in metallness-roughness, but glosiness-specular pipeline. TrainUberShader uses PBR techniques.

After all, if even all environment be migrated to PBR textures, all legacy pipeline should be kept for backward compatibility with existing content.

 

Although, if completely new TSC will be payware separate Steam product, I'll gladly adopt it.
But I will pay only for updated rail track system not based on cartoon tiles and rectangular rails, modern detailed environment and landscape textures, extended scripting API - and all of this with an optimisation better than in Trainz franchise, as N3V's game has all of this but with extremely high GPU load together with primitive train physics and player interfaces.

 

- Would a separate version of TSC be considered with features that would break backward compat? e.g. A version branded "TSX and included full PBR changes etc?"
- Anyway to fix the "Wheels still inching forward even when stationary" bug?

 

This is not a core engine thing - it already allows you to create the most detailed tracks if you wish to. It's a content creation thing, and not part of the core rework.

 

This would need further scrutinization. It does not happen on every vehicle/route, so it might be a blueprint issue / incorrectly setup assets.

 

I know, but this feature is still one of the mandatory points for me to pay for "new one TSC" one more time.

 

Let the new engine come out, and content creators will see how far they can go in terms of performance

I'm quite excited about all the stuff I've read so far.

 

Agree.
As content creation is still a corner stone of TSC for me. That's the main reason I'm still playing it.

 

DTG says in the Article
Current Progress
One thing we have found with the work we have done so far is that the software is much more stable than the version you are using right now. Though we will wait to celebrate that achievement until you all have had an opportunity to try it out.

Backward Compatibility
We have always been particularly proud that we’ve been able to get this far whilst maintaining compatibility with about 95% of the back catalogue of everything people have made over the last decade or more. It is something we continue to think about, and we can assure you it remains firmly in focus as we move forward

 

In my previous comment in this topic I forgot one thing. There was a remark on LUA and lack of backward compatibility. That is true, LUA has a few changes over time that are NOT backward compatible. I think, however that it should be feasible to automate the upgrade of these old LUA scripts, probably even on the fly. The changes are not big and you can decompile LUA scripts Not sure about the very latest LUA versions. If you really want to support LUA better, it is more important to create good testing tools in game. The lack of these and the lack of documentation cause the main issues for developing scripts that go behind simply showing messages. I also think the API is bugged, a number of things doe not work while they should.

 

Yes, as I said the wording is worryingly vague. What does the 95% figure refer to? Compatibility up to now, or compatibility with the changes they've made?

 

Upon release.
I don't think that number can be properly measured.

My understanding is that they make a few engineering decisions and also prioritise what to fix, what to test.
If I were the person saying that, I'd mean a combination of two groups.
First, of course it's impossible to test absolutely every product out there. Inevitably there is going to be a random asset out there which is bogus in a unique way and thus going to break unexpectedly. Like how the Springfield line did in x64.
Second, the engineers may decide to stop supporting clearly bogus assets. I suppose the milk bottles really have to say, but the road vehicles turning into big walls on All Aboard routes might not cut it. Or scenarios with a faulty xml. In some cases it's not about removing compatibility but adding verification to ensure that the scenario will play fine.

That's why I suggested to overpower (old, often retired) creators and enable the community to provide fixes. The main problem isn't some asset to stop working but the inability to get it fixed. Like how NS Coal District scenarios are still faulty, or how most US content has couplers prone to separate. The next level is authors denying the modification of content, causing stuff like All Aboard routes to be discontinued and deleted.

 

This sounds appealing, but I imagine that forking the product at this late date would be extremely confusing, especially for those that are already switching between TSC and TSW to play their preferred routes.

 

SCS have been doing it for years with no adverse effects even for us (as you seem to think) stupid end users, you just run a legacy build via the beta opt in menu.

 

Ad hominem aside, I do not think it's unreasonable to want all routes and assets to be in one place and accessible from one executable.

 

I hope you are able to do something about support for higher-resolution displays. We have processing power and resolutions not dreamt of when Rail Simulator was first released. My new laptop has a higher resolution than I am able to use with Train Simulator, but I am able to run full screen resolution and high graphics settings in TSW3. I have to use a much-reduced resolution and still suffer memory errors in TS.

 

They'll probably do it as they did with the 32-bit exe - just select the version to launch at startup.

 

I'm so happy to hear the progress, well done Dovetail! I am so happy to see this game getting some much needed love, as in my opinion it is currently the best Train Simulator out there and this will make it even better. What I want to see for one of my wanted upgrades is a personal thing, but I want to be able to select menu music from previous versions (Train Simulator 2014 and 2015 for example) for my menu music. I heard that possibly custom menu music will be available, but I am not sure if that means old menu music will be built in options or you will have to find a download for the menu music file somewhere on the internet. I am hoping it will be built in options as that would be much easier. I also don't know much about computer processors, etc but I have heard Train Simulator Classic performance suffers due to having something to do with only using a single thread from your CPU? So would be good to see if the game can use multiple threads from your CPU so we can get some lovely performance of the game. Thanks again Dovetail, I am looking forward to the future updates!

 

The music I dont know (or care) about so can't comment, as for the multi-core then this quote directly from Steve should answer your question...