Train Sim World 2 - Roadmap Update 08/02/2022

The problem with this, if it's similar to the TSC version of things (which I suspect it is) is that when you put a train in a scenario and tell it it's start and end points the game automatically works out where it will be at certain waypoints. If you then drop another train in, and that train has interactions with the first train the game works out the amended times at the waypoints. Excel won't do this for you
So it could be that TSW dynamically works out the waypoints, interactions and signalling to allow you to just say "start at London Victoria at 09:30 and head for Brighton" and it works out the interactions as you go along, but I would think it's unlikely, and more likely you'd end up with a signalling mess and a logjam

 

Glad to see Long Island Railroad got higher on the preservation crew update list, really looking forward to any improvements for that route.

Also happy to hear that the Peninsula Corridor fixes are finally close to release now. It was really about time, can't wait to drive the route again, as I havn't done so since the original TSW days.

Another small thing, but glad that "White Plains" was finally corrected to "North White Plains" regarding Harlem Line.

 

Cant remember any confirmation of shunting services too. But for sure they will add empty rolling stock on the sidings. Maybe the 363 or g6 get added static.

 

Hi
in few weeks they announce TSW 3 for summer release and stop support for TSw 2 leaving tsw 2 with game breaking bugs like they have done with TSw 2020.

 

I've never really used TSC but you've obviously more experience of doing this than I have, so I suspect you're right on how this sort of stuff works.
What I envisaged as a minimum was being able to somehow upload the same information that you can type in manually in the Scenario Designer user interface. Fair enough it won't have timetabled stops but I was at least hoping to avoid the manual user interface.

 

Not too much new on the roadmap, but I’m intrigued in seeing what rivet come through on the sursee route

looking forward to see what changed with peninsula corridor updates when they release hopefully fixed a few issues I’ve noticed with the route

 

Why is there such a bottleneck in releasing the Preservation Team's updates. Surely by now there must be enough contents from Adams team to fill a whole Download Slot of update patches by itself. At this rate routes already worked on last year will and still awaiting a slot with be out of date before their even released !!

 

Says who ??

 

I agree. A concise list or summary is what many people want, not endless banter.

 

It feels that long term issues such as the PS5 limit and profile switching have been going on for so long that I’m sure I had stabilisers on my bike when they first came up !
The Rush Hour problems just seem to go on and on and on surely these should have been fixed by now . I seriously wonder how many of us would have purchased these at the time if we had known about all the issues . It’s no good just putting them all on roadmaps for months on end and nothing happening with them .
Meanwhile here is another DLC for you all to buy . I’m sorry but the brand loyalty has to work both ways

 

What's this new journeys thing on there?

 

And they are gonna be free or paid?

 

Joe talked about it in another thread. He wanted to add shunting services but ran out of time. There are, however, static 363s.

 

We need more American trains, I feel that the German trains is getting so much love, than american

 

Not ran out of time more like stuck behind a red signal

 

Where was that even said have not heard anything like that.

 

Its what i think, its what they have done when they release tsw 2, stop the support to tsw 2020 leaving the game full of bugs and if they release in the future Tsw 3 you can say goodbye to tsw 2 support.

 

Why would they still support tsw2020 , it’s really outdated

 

I can't watch the stream live but have read the article/waffle. And still they say the updates will be released 'soon' Could DTG have the decency to explain what they mean by 'soon' The updates have been completed for months and we were told back at the beginning of December that we will see them 'soon'. Two months later and it is still 'soon'.

When I say 'soon' I mean in the next few days or weeks but obviously that is not the case with DTG. They know when they are coming but seem to think if they keep us on a hook it will be good for business. I suspect it might comeback to bite them on the arse. All it will take is one good Train Sim to be released and a lot of people will ditch TSW2 imo.

 

Soon basically means it could be a week , a month or 6 months . It’s pot luck

 

Exactly, is a little more clarity too much to ask?

 

You need to speak with your dealer mate, he selling you some rubbish stuff

 

As of right now, I don't hear Matt saying anything about making TSW 3 which I really do hope it's not because I am tired of games that upgrades to the next version without improving the current game. It's also getting tiring, I really hope that they are just still with this game and keep on updating as it is.

 

Observadorpt mentioned the TSW3 as it was what he was thinking not that it would happen.

But let's play Devil's advocate here, say they were going to do TSW3 and soon, do you think DTG will tell us about it now and risk no-one buying any TSW2 content? Perhaps this is why they keep stalling the preserved route updates as they have no intention of releasing them?

I don't think this will happen as I said 'Devil's advocate' but IF it did I would not be that surprised.

 

Of course they stopped support of TSW2020 that was because TSW2020 could not be upgraded to unreal 4.16. So that is why we have TSW2 and it could have it's unreal engine updated without having to create a whole new game. Since then they have upgraded the engine to 4.23 continuing to call it TSW2.

There will not be a TSW3 at this point in time as there is no need to do so.

 

“Soon” in the context of the current preservation crew updates means before the next game update. That’s the hope from what Matt said in the stream. To get all the completed updates released before the next major core update, whenever that is, probably a similar time of year as the last one. So perhaps six months or so for the lot if it all goes well. By that time there will have been a further two years worth of content released (all of TSW2 releases to date by then) that hasn’t even started to be looked at by Adam’s team unless they move on to those routes in the meantime. No timeline, even estimated, was given for things like the PS5 DLC limit fix although for that fix it was said that it hasn’t worked out like they had hoped so a new approach is being taken to try and do a better fix rather than release a fix that might not work that well.

Nothing is going to happen quickly it seems.

 

While I still don't like the new format of the devblog (I'll keep calling it this way because it isn't a roadmap), I quite like the idea that there will be a public beta on Steam. I find that I launch more often the games which are updated frequently, to see what's changed, even if these updates aren't marked as 'stable', alternatively when there are some 'community events' to do, or for a couple of days in a row after purchasing a new DLC, than when the game simply sits 'stable' on the disk and is not going anywhere, so I can play it anytime 'later', while other games get updates in their 'public_beta', 'experimental', 'edge', 'unstable', 'untested', etc. branches and get my attention instead.

I'm hoping there will be at least a pinned thread listing changes in that branch both here and duplicated in the Steam hub, or a new subforum created specially for it.

 

It seems to me that Matt and Nat quite rightly spent most of their time addressing THE #1 CONCERN SEEN ON THESE BOARDS over the last two weeks (or more)- why the QA logjam, why have the updates been so delayed, and is anything being done to fix the problem?

I don't know why people are complaining about the lack of new DLC announcements, because
a) New DLC is NEVER announced in a Roadmap, and
b) There's nothing new to talk about besides Lucerne; I don't believe the next year's DTG routes have been chosen yet

 

Maybe at this time DTG chooses mostly Multiple Units such as Acela, ICE-T, Class 700 as loco addons which require significantly larger amount of development time (each with multiple unique carriages) than single locos (363, C40). Just be patient and see what's to come

 

i would really like to know if i can finally get a gold medal on one service in pennisula corridor. i don't think i'm being heard, as i've sent reports before, and i haven't seen specifically if the issue got fixed or not.

 

Matt did in fact spend a fair bit of time on the PS5 problem and the fix they have in the works.

 

i wish the same could go for me but unfortunately its not, i still can't get the next graphic card due to scalping price in rtx 3000 series, hopefully thing would be better around the summer time, so that i can update my graphic card. Still have the worst graphic card which is the gtx 760

 

Agreed. An article to summarize what's been said in a stream would be very useful.
DTG Natster anything your team can do to help us with such a thing?

 

Nothing much was said above and beyond what's on the RM document. Fixes are being QAed and will be released in due course

 

I posted this in another thread:

 

I wonder if DTG released some Loco's instead of routes would we see the pres updates any sooner?

 

Probably not . But that has made me realise , the last dtg loco to be added to the roadmap was the 313

 

And look how long that took..............

 

DTG have just released some updates for certain routes which is always welcome but I wonder why they didn't include any of the finished preserved updates considering they said slots had to be planned for?

 

Hopefully next week we will start seeing more updates.

 

TL/DR version of the stream:
Explanation of the dev to release process:
-Devs do work -> features compiled into a build of the game
-Build sent to QA for testing
-QA then sends back their findings (things rarely ever pass the first time)
This process repeats until QA passes (this is normal software development)
-Once QA passes, that build gets slotted for release. Part of releasing, is getting a certification from Sony and Microsoft that the game meets their platform standards. This is just another round of QA, but for completely different reasons than are tested internally, and is handled by Sony, Microsoft, Epic, and (sort-of) Steam themselves.
Once all of their 7 platforms are certified, it's able to be released.

Noted that the term "release slot" that comes up refers to this entire process and not just the release date itself.

Explanation of the QA process and why DTG got into this mess:
QA queue prioritization: basically work being sent to QA wasn't being optimally prioritized and certain things, like the preserved crew updates, were sitting around. They're looking at how they can better prioritize things. Additional QA passes on a specific release (like failing, more dev work, reenter QA) don't so much rely on priority, as there are fewer things to test - so additional passes are less of a hold up than it might seem

Compiling builds: Selecting which features are going into each build is a manual process (slower, introduces higher error risk). Basically, devs never work on the actual copy of the game (for obvious reasons), instead working on their own copies (essentially) that they make changes, fixes, and improvements to - which can contain multiple unrelated changes, depending on what they're doing. So unneeded stuff doesn't make it, you "cherry-pick" out what you want and place it into the master build of the game which is then used to generate a copy for each platform and sent down the release pipeline. Also worth noting, they build for 7 total platforms and as Matt has said in previous streams, it takes a while to cook and build the entire game. So if something goes wrong, there's a lot of time lost here. They're looking to make this semi-automatic to speed it up and reduce potential errors.

Speed of QA vs Release Certification: As platform certification is sent off to Microsoft/Sony, there is a delay in the process at this step - which doesn't always match up with the speed at which things are coming out of QA. You also can't send multiple things at a time (that's what I got from Matt's comments) for certification, so while this process doesn't slow QA down, it means there's another bottleneck between QA and release inserted at this step.

COVID: QA is outsourced to other countries where COVID has hit especially hard, reducing the number of people available to test the product. I don't remember the figure Matt gave, but it was pretty significant. They're assessing other options to ensure if outsourcing fails in the future, they have other alternatives to pick up the slack.

At the end of the day, Matt basically said he wants to fix the underlying issues that led to this, rather than just skim over it, get the work done and out, and be back into the same situation a year from now. That's a slower process and keeps the wait time extended for us, but should fix the problem long-term- which IMO is the correct move.

Also they're looking into the possibility of doing a open beta via Steam for bug fixes (only) to assist with this in the future, but this isn't a concrete thing yet.

 

It was said in the stream yesterday, the top of the preserved crew list in the roadmap just wen through its first round of QA, and is now back with Adam's team to fix any outstanding issues. There's still a bit to go - but my takeaway is that that stuff has finally hit QA, which should accelerate things. Since we like speculation on the forums, I'd guess if all goes well the first of the updates probably comes later this month or early next month.

 

That's an excellent summary wxtr7, thanks! I'll just add a couple of specifics regarding Pres. Crew and QA:

Mat & Nat stressed that the order of release listed on the roadmap is what's expected but will be flexible; if a lower-down fix is completed quickly, it will be released then rather than being held until its "turn". In general, the first half of the list is more likely to be stable while the second half is more likely to be flexible.

It looks like Peninsula Corridor and its two loco DLCs will be the first released. They have already gone through the first QA pass, and when Adam returns from vacation next week, he'll evaluate the results.

Also having completed their first QA pass: East Coastway, Great Western Express, MSB, and RSN. Oakville is currently in its first QA pass. This has all happened since the last roadmap, so things are moving. Again, when Adam returns, he'll review. The current priority release is PC.

Second QA passes, being much smaller, will be given priority, with the goal being to get things completed and out the door.

As far as the list overall, the current goal is to get everything on the list out before the next update to the game.

 

Why would it? They're not made by the same people.

 

It was interesting to hear that DTG, a modern software engineering company, must have no CI/CD and/or devops team working to make this process more efficient. Tools like automatic source control and build (git/svn, jenkins), or automated QA, testing, and deploy/package (codestar/codedeploy). Every time a developer checks in code, (complete with comments in their code, and any other changes properly documented in your documentation system), the pipeline should go to work. Validate dependencies, scan the source and dependencies for any known security vulns, build the artifacts from source, push those artifacts to some VM’s and run automated tests, or package/push to outside QA for final validation. These technologies have been in widespread use for over a decade if not more, and today you can easily and cheaply deploy an entire chain start to finish on a public cloud like AWS or Azure.

You cannot rely on developers themselves to make these changes. They often do not have the skillset to architect these solutions for production efficiency long term. Developers develop, let the sysadmins and devops guys build out the toolsets and make life way easier. It sounds like this is what Matt wants to do but I worry that without sufficient resourcing this is just asking for trouble.

I suggest reading up on Agile style software development, then bring on the right resources to implement and maintain such a style. Weekly scrums, a kanban board, an automated CI/CD pipeline…get r done. Dont allow items to move from in-progress to the backlog without sufficient justification. Otherwise you end in this exact scenario. Previous items that should have been completed in the weekly sprint have are now 6-12 months behind and no way to climb out of the hole.

 

DTG is a company which has grown from an indie to just-bigger-than indie, with total payroll of circa 100 including the receptionist and cleaning lady. "no CI/CD and/or devops team"? No. They also don't have an in-house legal department, arts patronage committee or a team looking into stadium naming rights.

 

Looking at DTG says about bottlenecks would it not be beneficial to employ someone who has the same skillset as Adam? It seems Adam has gone on holiday and everything must wait until he comes back.

 

For the next roadmap stream can you build a new slidshow with new screenshots? It´s feels old

 

I'm sure that things can wait a week or two for Adam to return, especially given that we've waited for a QA slot for months on end so all Adam's team would need is to "spit and polish" the last few bits

 

This is good context to know. But 100 employees is still plenty large enough to have someone doing DevOps, especially considering that DTG has a number of different software packages in active development. Frankly, it doesn't sound like they are using Agile system, and they should. That was more my point. Implement Agile and get the teams aligned then work on the tool chain.