1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Engine Upgrade And Preserved Collection

Discussion in 'PC Discussion' started by dreampage, Jun 16, 2021.

  1. dreampage

    dreampage Well-Known Member

    Joined:
    Nov 29, 2018
    Messages:
    450
    Likes Received:
    657
    In the last Roadmap live stream Adam talked a bit about the graphics engine upgrade TSW 2 will get soon. Namely the game will be upgraded from Unreal Engine 4.23 to 4.26. While it seems to be a minor bump, the way he described what it means for existing content is concerning in my opinion.

    In short, according to Adam, every single piece of content will have to be upgraded from the old engine to the new one, and every piece has to be re-released separately again. I'm concerned that this will, again, lead to an abundance of bugs, just like when the preserved collection was upgraded to TSW 2 and its then-new engine. Also, not every content will be available at the launch of Rush Hour and the upgraded TSW 2, but they will be released "over the course of around a week", which I bet right now will be much longer.

    I do think the very design of TSW 2 is wrong if everything has to be re-released every single time change happens to the core technology. This is not future-proof at all, and the more content is released the harder it will be to regularly upgrade everything, let alone doing it with quality results.

    Again, we may end up with broken routes, and we may have to wait months, again, for every single one of them to be updated and fixed. Not to mention that after the release of Rush Hour, who knows how much we'll have to wait for our existing routes to appear again in the game. This is completely unsustainable and the next time TSW is upgraded again (maybe in 2022), who knows what content will be kept, if any at all. For me as a player this makes no sense. I'd like to have a unified platform for which I can safely buy content, knowing it will be maintained and supported for many years to come (look at Train Simulator). Not re-released, broken, fixed, and re-released again like a vicious cycle. After a while, there will be simply too much content to handle like this.

    I'm getting more and more wary of the Rush Hour release, again not knowing what will happen, what will disappear or reappear in a buggy state. This just erodes my confidence to buy anything for the game.
     
    • Like Like x 31
    • Helpful Helpful x 2
  2. Ben_Broomfield

    Ben_Broomfield Active Member

    Joined:
    Mar 16, 2018
    Messages:
    115
    Likes Received:
    232
    Pretty much my feeling from yesterday’s roadmap stream.

    Hopefully given that the Preservation Crew are working on updating the routes for the release of Rush Hour, in theory they *should* be ready for the release and have no delay to them being available for players. Although given the fact Adam said there could be a possibility of some being delayed and added within the week of RH being released.

    I hope it doesn’t turn out like the start of TSW2!
     
    • Like Like x 1
  3. DTG Natster

    DTG Natster Community Manager Staff Member

    Joined:
    Sep 2, 2020
    Messages:
    1,203
    Likes Received:
    7,910
    When we first ported over the preserved collection this was a steep learning curve for us as it was the first time we were doing it.

    With any core update of this size we will of course run small risks that something might not work completely as we had intended. We are however more confident and more experienced than the first time we did this.

    If we were to stop updating the core from now on and remain in Unreal Engine 4, we would eventually run into limitations of what the game can achieve.

    With the new engine we have new opportunities and this allows us to do more with the game, we want to continue to upgrade so that we can offer you the best experience we can.
     
    • Like Like x 20
    • Helpful Helpful x 3
  4. Inkar

    Inkar Well-Known Member

    Joined:
    Dec 13, 2016
    Messages:
    707
    Likes Received:
    1,135
    I understand that you want to upgrade the game engine, I might even go further and say that pretty much you need to do it. But, there are a few things that you said above that I do not agree with.

    1. Game engine upgrades are not something that you learn to do. Of course you will have learnt something, but each upgrade is its own thing. Your 5th engine upgrade can be much harder than your 1st one, just depending on what needs to be done. Even if in this case the upgrade for each individual route is easier this year, the number of DLCs to update has grown.

    2. The risk of an engine upgrade to cause bugs is not small. It is pretty big, specially when DTG develops each DLC as its own self contained branch of code and design philosophy. Last year's engine upgrade caused lots of problems that are not fixed one year later. This should be a huge warning sign telling you that something is not going as it should.

    3. I know DTG wants to upgrade so they can give us the best experience but remember, the road to hell is paved with good intentions.

    4. Do not expect customers to be very happy if their previous content starts working worse than it did before the update, even if it means future content will be better.

    All of this goes back to what we have said several times in many posts already:
    1. Be careful with your software design decisions. Every single design philosophy and different way of working you add in a route is another headache for the next engine upgrade, and EVERY engine upgrade after that. You know more engine upgrades will be coming, so act accordingly.
    2. If the number of routes to be maintained is growing, the preservation crew needs to grow too. Right now it is not only not growing, but the core team sometimes "borrows" people from the preservation crew to do something else.
     
    Last edited: Jun 16, 2021
    • Like Like x 23
    • Helpful Helpful x 1
  5. dreampage

    dreampage Well-Known Member

    Joined:
    Nov 29, 2018
    Messages:
    450
    Likes Received:
    657
    I fully support the game to be upgraded so it can look and run better, utilizing new techniques, etc. What I'm afraid of, however, is that the whole TSW 2 platform is designed in such a way that after a while this will become unmanageable. There will be so much DLC content, routes and locomotives, that it will become humanly impossible to upgrade everything on a, say, annual or bi-annual basis. I would not like to either lose content as time goes on, nor wait much for preserved content after such an upgrade. However, as there is more and more content available every year, every subsequent engine upgrade will mean more and more work for the team. And this is something you obviously can't expand infinitely.

    And then there's the (potential) bugs, from graphics to game systems to sound, etc. We've seen many times how upgraded routes became different to their original forms, and many of them need additional fixes and balance passes. This burdens you, the team with yet another layer of additional work. Of course, I do hope everything will go smoothly for this year's upgrade, but if it becomes such a laborous process every time then it does make me wonder about the longer term future of the game, and how you can manage all of it further down the line.
     
    • Like Like x 8
    • Helpful Helpful x 1
  6. DTG Natster

    DTG Natster Community Manager Staff Member

    Joined:
    Sep 2, 2020
    Messages:
    1,203
    Likes Received:
    7,910
    I agree that every upgrade will come with it's own challenges. That is to be expected. But we are prepared for these challenges. And if we do come across any hurdles we will communicate that.

    The aim of the core update is to make the game better, not worse. And although the update last time came with it's own challenges we don't anticipate the same issues again. Your feelings are completely understandable after past experiences.

    If the preservation crew needs to grow in the future to handle the amount of DLC and the extra workload, this will be something that is considered at the time. For now we have a suitably sized preservation team.
     
    • Like Like x 5
    • Helpful Helpful x 1
  7. Inkar

    Inkar Well-Known Member

    Joined:
    Dec 13, 2016
    Messages:
    707
    Likes Received:
    1,135
    How many DLCs does TSW2 currently have? 19 (route DLCs, everything is 30+).
    How many DLCs did the preservation crew get to work on this year? 8
    How many DLC preservation patches were actually released? 4

    You might want to reconsider your previous statement.
     
    Last edited: Jun 16, 2021
    • Like Like x 20
  8. dbrunner#4864

    dbrunner#4864 Well-Known Member

    Joined:
    Mar 25, 2021
    Messages:
    162
    Likes Received:
    251
    Meantime we still have broken sound for preserved content from TSW2020 ported to TSW2. M7, BR 143, 146, 185, all of them have broken / missing in cab sound compared to TSW2020. If in this amount of time the team could not get this locos from TSW2020 preserved routes to sound properly in TSW2, how can we know history wont repeat itself and after the new engine update even more content will have issues like this?
     
    • Like Like x 16
  9. Cramnor

    Cramnor Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    693
    Likes Received:
    1,138
    I share your concerns. And I honestly have to say, nothing anyone from DTG says will help to dillute or change these concerns. Actions speak louder than words - and I am slightly optimistic, but in the end, it will be judged on what actually happens, and not on what someone might say now or in the future. So I am really curious to see what will happen and if DTG can live up to their promises.
     
    • Like Like x 7
  10. DTG Natster

    DTG Natster Community Manager Staff Member

    Joined:
    Sep 2, 2020
    Messages:
    1,203
    Likes Received:
    7,910
    I appreciate you all sharing your thoughts, and I do understand why you are feeling this way. I will make sure your concerns are shared internally.

    At the moment it is indeed a case of 'Wait and see'.

    I will have a discussion with Adam and the preservation team tomorrow to touch base with how their work is coming along and if there is any news to share with you.
     
    Last edited: Jun 16, 2021
    • Like Like x 10
    • Helpful Helpful x 3
  11. dreampage

    dreampage Well-Known Member

    Joined:
    Nov 29, 2018
    Messages:
    450
    Likes Received:
    657
    Thank you for the kind replies, I highly appreciate them. While I am concerned I am also cautiously optimistic and as you said let's wait and see, and let's hope it will all go well.
     
    • Like Like x 4
  12. zefreak

    zefreak Member

    Joined:
    Jun 7, 2021
    Messages:
    49
    Likes Received:
    97
    I agree 100% - I've been a software developer for over 10 years and this is obviously an unsustainable development model.

    Edit: since this is my first post, I love the game and really want it to be a continued success, but the amount of effort that goes into updating DLC individually makes no sense to me considering that the burden will only grow with time.
     
    Last edited: Jun 16, 2021
    • Like Like x 8
  13. fabristunt

    fabristunt Well-Known Member

    Joined:
    Dec 23, 2019
    Messages:
    332
    Likes Received:
    536
    Most of the preserved collection is still broken a year after the release of tsw2 and DTG is going to break it even more. Great, can't wait!
     
    • Like Like x 5
  14. praxidike.meng

    praxidike.meng Well-Known Member

    Joined:
    May 26, 2019
    Messages:
    109
    Likes Received:
    296
    The subject of this thread is increasingly relevant as the Rush hour nears release. Much of the Preserved collection is far from its original state and here we are already faced with another engine upgrade... I don't want to be a downer, but, honestly, it's difficult to trust this will be a smooth transition. And the sheer amount of time it then takes to patch everything up... If the past year is any indication, I'm not optimistic.
     
    • Like Like x 12
  15. hyperlord

    hyperlord Well-Known Member

    Joined:
    May 20, 2019
    Messages:
    818
    Likes Received:
    785
    I guess as long as future updates (any, not just the work of the PCrew) are churned out more frequently the situation might calm down. Big error? Broken? "Unplayable" - fixed in one week and noone complains (too much) and the feeling of "we trust them to fix glaring errors in a reasonable time" could shine trough right into this great community ;-)
     
    • Like Like x 1
  16. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    It's an interesting discussion - software is the game I'm in for my daily crust. Sometimes changes to an underlying architecture go smoothly and sometimes they do not. Success purely comes to a whole lot of planning, good execution, good QA and a pinch of luck.

    Comparing TSW2 to TS isn't particularly valid dreampage - TS has certainly had enhancements but essentially its the same engine as it had when it was MS Train Simulator in 2001 and the limited release Kuju game before that. There haven't really been any fundamental changes to the way it works. Matt even recently said on a stream that they actually cannot make any fundamental changes because if they did they would have to design the existing bugs into the new version as all the DLC would stop working.

    The concept of DLC between TS and TSW is quite different. TS is very simple compared to TSW.

    TSW2 being based on a 3rd party engine is at the mercy of that engine to a large extent, valid uses of the engine should be expected to work, but often developers find work arounds to issues or find techniques on forums other than the engine documentation on how to get things done. This has to have happened in TSW - Unreal has no concept of rotating objects like rapidly rotating wheels - getting this working in TSW is a hack likely developed by DTG and included with the engine. Just migrating that could be "fun".

    While there are certainly risks involved and everything you've pointed out along with the other posters could happen, I think Dovetail are taking the right risk mitigations
    • They wait at least six months before adopting the engine avoiding bleeding edge problems - Unreal 4.26 was released on Dec 4th 2020
    • They're upgrading from 4.23 to 4.26 (thanks to Anthony Pecoraro for the version correction)
    • The update isn't massive
    • If DTG stick with an annual major core update then they're likely to upgrade Unreal more often too
    I can't say anything on their game architecture except that it isn't unusual in the sense that older DLC are usually "left behind". What does make them unique compared to other games - in TSW the older DLC live alongside the newer DLC and the players use both - in most games new DLC tends to supplement or entirely replace the older DLC as the user progresses through journeys/levels.

    I'm not going to try and say what a right answer may be, it would probably take me three to six months of understanding their architecture and code base before I was willing to predict a right answer. However, I don't see great big glaring warning signs either. DTG's handling of both game and community is maturing. DTG have shown signs of adopting big developer practices rather than small game house practices. In the long run this will pay off.

    What will the core update bring? What will the pitfalls be? No idea, but we shouldn't assume based on history that the same issues will happen again. The one sure death for the game would be to sit still. Operating systems move on. Drivers move on. Graphics hardware moves on. Motherboards and processors move on. It would not take very long before the game would become unplayable. We don't have the stable luxury we used to have especially when you consider that this one code base functions across three major platforms with three utterly different operating systems. I would also expect although I cannot find any information anywhere that Epic most likely have sunset dates on older versions where they either don't support the version or they make you pay more to get support. This is normal in industry to force engine users to keep their products up to date.

    Paul
     
    Last edited: Jun 18, 2021
    • Like Like x 12
    • Helpful Helpful x 4
  17. hyperlord

    hyperlord Well-Known Member

    Joined:
    May 20, 2019
    Messages:
    818
    Likes Received:
    785
    • Like Like x 2
  18. Bertil Persson

    Bertil Persson Active Member

    Joined:
    Aug 23, 2017
    Messages:
    125
    Likes Received:
    63
    I was worried before, now I understand more. As long as we don't loose our DLC:s including Preserved Collection, I'm satisfied. Those old diesels make a difference as layers in the newer DLC:s. That goes for the german locos too.
     
    • Like Like x 1
  19. Anthony Pecoraro

    Anthony Pecoraro Well-Known Member

    Joined:
    Dec 11, 2016
    Messages:
    3,181
    Likes Received:
    3,814
    They're going from 4.23 to 4.26.
     
    • Like Like x 2
  20. roysto25

    roysto25 Member

    Joined:
    May 1, 2019
    Messages:
    70
    Likes Received:
    80
    I can appreciate the concern over changes deriving from UE updates, but my experience over some years with systems based on external engines is that it is imperative to keep up with engine patches (or updates whatever the description), in order to minimize big surprises with introduced bugs. Problems introduced in version changes tend to cascade and if there are big jumps in version numbers, the problems easily get out of control. A bug introduced by a single version change is easier to stomp. This is particularly true if, as is likely with DTG, there are customizations of the engine (rotating objects for example). I have had what diplomats call 'frank and honest discussions' with support personnel who do not understand the effect their changes have on customizations. As an aside, the habit of outsourcing development support by major players is problematic, since the outsourced personnel are often trained to the 'out of the box' product and have little in-depth experience of non-standard implementations.
    The real solution is for DTG to get serious about the frequency of updates from Adam's team. it is ludicrous to wait months for patches already completed.
     
    • Like Like x 5
  21. nockwurst

    nockwurst Active Member

    Joined:
    Apr 5, 2020
    Messages:
    121
    Likes Received:
    219
    Great first post, and fantastic points.
     
    • Like Like x 1
  22. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    That's even better then :). Not such a big jump after-all. Thanks for the correction. I'll edit my post and credit you.

    Paul
     
    • Like Like x 3
  23. Inkar

    Inkar Well-Known Member

    Joined:
    Dec 13, 2016
    Messages:
    707
    Likes Received:
    1,135
    Yeah, it is not a big jump. Here is the change list just for the audio from 4.23 to 4.24:

    Code:
    Audio
    [LIST]
    [*]Crash Fix: Fixed a crash on shutdown of Google Resonance module.
    
    
    [*]Crash Fix: Fixed a race condition crash between the audio engine suspend state and audio engine toggle on iOS.
    
    
    [*]Crash Fix: Fixed a race condition when loading DialogueSoundWaves that would sometimes lead to a crash or invalid data on the DialogueSoundWave.
    
    
    [*]Bug Fix: Fixed the null audio device pace. It was previously generating audio slower than it should have, since it was relying on a Sleep function to set the pace.
    
    
    [*]Bug Fix: Fixed edge cases that were not reporting back to AudioComponents that playback had finished (and failed).
    
    
    [*]Bug Fix: Fixed the MaxDuration sound calculation when cooking surround.
    
    
    [*]Bug Fix: Fixed an issue with SoundWaves being played via SoundCues set to PlayWhenSilent (flag wasn't being observed).
    
    
    [*]Bug Fix: Fixed a bug with OSC bundles not being parsed correctly due to reads not accounting for the header element offset.
    
    
    [*]Bug Fix: Fixed potential audio distortion when recording gameplay with GameplayMediaEncoder, caused by wraparound when playing loud sounds.
    
    
    [*]Bug Fix: Fixed a thread safety conflict with SeekToTime being called on the audio thread and StreamCompressedData on the decompression worker thread. The change forces all seek requests to be processed on the worker thread.
    
    
    [*]Bug Fix: Fixed audio stats not being available in test builds.
    
    
    [*]Bug Fix: Fixed a MaxDistance calculation being wrong for box attenuation, causing sounds to cut out when the listener moves toward diagonals of a box.
    
    
    [*]Bug Fix: Fixed an ADPCM bug where non-interleaved encoding was not reporting the correct number of samples per channel in the asset viewer.
    
    
    [*]Bug Fix: Fixed a bug with focus not interpolating correctly when enabled.
    
    
    [*]Bug Fix: Fixed an iOS audio bug where crackling could be heard during certain OS events.
    
    
    [*]Bug Fix: Fixed an incorrect sample rate for media playback.
    
    
    [*]Bug Fix: Fixed an issue with the "audiomemreport" command where some sounds were incorrectly reported as being fully decompressed.
    
    
    [*]Bug Fix: Fixed a bug where the submix envelope follower would appear inactive due to the temp result being optimized away on some platforms.
    
    
    [*]Bug Fix: Fixed a bug for audio crackling when suspending an iOS app.
    
    
    [*]Bug Fix: Fixed a bug where per-source, one-pole filters were interpolating during their first callback with a new source.
    
    
    [*]Bug Fix: Fixed an issue with Time Synth PIE crash caused by playing more clips than the synth was initially allocated for.
    
    
    [*]Bug Fix: Fixed issues on some platforms where seekable audio streams were not acknowledging seek requests.
    
    
    [*]Bug Fix: Fixed an issue with Dynamics Processor where increasing the Input Gain would decrease the Output Gain.
    
    
    [*]Bug Fix: Fixed a bug in Phaser source effect becoming potentially unstable at project sample rates less than 48kHz.
    
    
    [*]Bug Fix: Fixed an issue on Android where when audio mixer is enabled, audio would render at the incorrect sample rate on some devices, thus causing stutters and pitching.
    
    
    [*]Bug Fix: Fixed a bug with reverb parameter interpolation overshooting its target, creating a possible edge case that would cause audio artifacts.
    
    
    [*]Bug Fix: Fix for multichannel file reimport in the case where a file is reimported to be mono or stereo when it was previously multichannel.
    
    
    [*]Bug Fix: Fixed a bug with AudioMixerDevice incorrectly reporting buffer size to be used in logs.
    
    
    [*]Bug Fix: Fixed a bug with non-seamless LPCM looping.
    
    
    [*]Bug Fix: Fixed iOS audio engine run time settings that were not observing project settings.
    
    
    [*]Bug Fix: Added audio engine option for device callback to wait configurable number of milliseconds for next audio buffer to be rendered before underrunning, which fixes an issue an Android where audio became choppy after connecting bluetooth headphones. You can configure this by setting au.UnderrunTimeoutMSec from the console or a configuration file. The default is 5 ms.
    
    
    [*]Bugfix: Fixed attenuation box shape incorrectly reporting 0 when distance is 0
    
    
    [*]Bugfix: Fixed an issue where the first import of sounds was missing playback or playing back a stale version of a sound. if stopping a sound during import, sounds will now log to the display. Also, removed a prompt for a template update when reimporting a sound.
    
    
    [*]New: Added the ability to force synchronous audio decodes. This is useful with platforms with a low core count, particularly mobile devices. Enable this feature by setting the new cvar au.ForceSyncAudioDecodes to 1.
    
    
    [*]New: The Audio mixer is on by default for all platforms other than Switch. You can turn the new audio engine off for a specific platform, by setting UseAudioMixer (in the [Audio] section) to false in that platform's Engine.ini file.
    
    
    [*]New: Added debug draw cvar au.3dVisualize.Listeners 1 to help debug listener location visually (off by default when au.3dVisualize.Enabled 1).
    
    
    [*]New: Set the default log level to Display for LogAudio to avoid log spam.
    
    
    [*]New: Sound asset creation right-click menu is now broken into multiple sub-menus to allow a user to find types more quickly and easily.
    
    
    [*]New: Updated the Audio Submix Graph to be more intuitive visually, and fixed usability issues regarding having numerous editors open and editing parent/child relationships.
    
    
    [*]New: Added the ability to disable binaural audio if sound starts playing below a specified radius.
    
    
    [*]New: Steam Audio Beta 17 improves performance with support for multi-threaded environmental effect simulation, Intel® Embree, and AMD Radeon Rays. Each of these features provide developers with significant performance boosts compared to simulating environmental effects with the built-in Steam® Audio ray tracer on a single thread. This update also adds support for scene reinit on level transitions, improves geometry tagging, and improves overall stability. Visit the release blog post for more information and be sure to check out the transition notes for upgrading projects from earlier versions of Unreal.
    
    
    [*]New: New Signal Processing class FSilenceDetection—A processor class that only outputs audio above a certain threshold.
    
    
    [*]New: New Signal Processing class FSlowAdaptiveGainControl—Can be used as an automatic gain control system in conjunction with the FSilenceDetection class.
    
    
    [*]New: New Signal Processing class FFFTConvolver—An FFT convolver class. FFFTConvolver does not currently support partitioned convolution, so window sizes should be less than or equal to the buffer size used in the ProcessAudio callback.
    
    
    [*]New: New Signal Processing function CrossCorrelate—Performs an acyclic correlation of two signals.
    
    
    [*]New: The Audio Capture plugin now also supports Xbox One, PS4, and Mac.
    
    
    [*]New: Added a per-source LPF frequency parameter control to sound classes and sound mixes.
    
    
    [*]New: Added the ability to query the audio component play state from Blueprints.
    
    
    [*]New: Added cvar au.resonance.quality to reduce Google Resonance quality dynamically.
    
    
    [*]New: Added classes to support multithreaded audio rendering. See MultithreadedPatching.h for details.
    
    
    [*]Improvement: Added optimized Push/Pop Containers to Blueprint for manipulating OSC Addresses. Also added a Blueprint call to clear an OSC Address' Containers (instead of having to pop each off individually).
    
    
    [*]Improvement: Added support for additional device routing to our native VOIP engine. This can be done either by using IOnlineVoice::PatchRemoteTalkerOutputToEndpoint(DeviceName), or by using the console command "voice.sendRemoteTalkersToEndpoint ".
    
    
    [*]Improvement: Decoupled AudioMixer callback size and iOS device callback sizes to enable latency to be configurable.
    
    
    [*]Improvement: Improved GameplayMediaEncoder audio handling (timestamps and duration). It now also adjusts the audio clock if it falls behind, such as when we break into the debugger.
    
    
    [*]Improvement: The AudioCapture module has been split into several modules:
    
    [LIST]
    [*]AudioCaptureCore—Includes our core audio capture interfaces that don't rely on the Engine.
    
    
    [*]AudioCapture—Includes engine dependent classes such as UAudioCaptureComponent.
    
    
    [*]Platform-specific implementations of IAudioCaptureStream in independent modules (AudioCaptureRtAudio, AudioCaptureAudioUnit, and so on).
    
    [/LIST]
    [*]Improvement: The AudioMixer module has been split into several modules:
    
    [LIST]
    [*]AudioMixer—Includes any engine-dependent code used in Unreal Audio.
    
    
    [*]AudioMixerCore—Includes IAudioMixerPlatformInterface and other audio classes that don't rely on the engine module.
    
    
    [*]SignalProcessing—Does not rely on the engine, and is used as our general purpose audio DSP library.
    
    [/LIST]
    [/LIST]
    

    Can't post the whole change list because the forum does not allow more than 25K characters in a post, but I think you get the idea.
    The version numbers do not matter much. The change list does.
     
    Last edited: Jun 18, 2021
    • Like Like x 2
    • Helpful Helpful x 1
  24. londonmidland

    londonmidland Well-Known Member

    Joined:
    Nov 8, 2017
    Messages:
    2,181
    Likes Received:
    9,747
    Considering we still have unresolved issues from upgrading from 4.16 to 4.23 (ambient sounds, for example) I’m extremely spectacle this will go smooth sailing.

    I’m also expecting it to take significantly longer than ‘a few weeks’ to get all content upgraded and working as designed.

    Not being negative or anything but I’m really not convinced this will be a quick and easy process.
     
    • Like Like x 5
  25. JonnE

    JonnE Well-Known Member

    Joined:
    Jan 23, 2020
    Messages:
    324
    Likes Received:
    557
    To be honest, anything before summer 2022 would be a (welcome) surprise.
     
    • Like Like x 2
  26. solicitr

    solicitr Well-Known Member

    Joined:
    Jun 20, 2020
    Messages:
    4,754
    Likes Received:
    6,823
    It's my understanding that most Preserved Collection routes have already been ported to 4.26. The process started a good while ago, in fact well before RH was even announced.
     
    • Like Like x 3
  27. Mat_Jam_Ca

    Mat_Jam_Ca Well-Known Member

    Joined:
    Jun 4, 2020
    Messages:
    3,600
    Likes Received:
    6,951
    It’ll certainly be a little devastating if we suddenly take a massive step back. Adam’s team has barely made a dent in the existing route stock prior to 4.26 and we all know there are still some seriously annoying sound issues that need urgent attention. DTG cannot afford to take one step forward and two steps back. I’m not sure the community would be as forgiving as they have been since TSW 2 launched if this were to happen.
     
    • Like Like x 4
  28. solicitr

    solicitr Well-Known Member

    Joined:
    Jun 20, 2020
    Messages:
    4,754
    Likes Received:
    6,823
    Unless I completely misinterpreted the last stream, the 4.26 versions include all the fixes and upgrades that Adam's team have made. Remember, the issue with almost all of them has been releasing the patches; the programming has been done. IOW, Rush Hour (or more accurately the TSW2.1 update) will finally bring us all of those fixes which have been in limbo over the last several months; ergo it will be a step forward (use "massive" to taste).
     
    • Like Like x 5
  29. synthetic.angel

    synthetic.angel Well-Known Member

    Joined:
    Aug 27, 2019
    Messages:
    1,484
    Likes Received:
    2,257
    There is one thing that the community can do to help ensure that the engine upgrade happens flawlessly, and that is simply to think about it positively, offer DTG our faith, encouragement and best wishes, assume that only the best will happen and that everything will be just fine. Okay that's six things..... There are six things that the community can do to help make sure it all goes well, as stated above.

    If everyone could now just focus on having a Positive Mental Attitude, then I am sure that we all reap the rewards that we all deserve...
     
    • Like Like x 3
  30. londonmidland

    londonmidland Well-Known Member

    Joined:
    Nov 8, 2017
    Messages:
    2,181
    Likes Received:
    9,747
    With all due respect, no matter which way we think, it won’t change anything internally. I can only go off past experiences, and that certainly wasn’t plain sailing.

    I just don’t want the same issues we faced upon the first upgrade. Which was having to wait months, even over a year, for content to be updated. Even then, there were still issues with it.
     
    • Like Like x 4
  31. tallboy7648

    tallboy7648 Well-Known Member

    Joined:
    Apr 9, 2020
    Messages:
    4,130
    Likes Received:
    4,731
    I see what you are saying but I can understand why some have concerns after reading this thread. Look what happened when most preserved routes were moved to tsw2. Now I know they said that some routes that was being currently worked on will have issues fixed when rush hour comes out but we know what happened in the first preserved collection update when some things were supposed to be fixed but weren't. Actions speak louder than words.

    I hope things go smoothly but my expectations are low. It's not as if preserved crew is suitably sized. Most preserved routes still have issues or new features not yet implemented almost a year after release. Apparently one person moved up to the main team a while ago. The preserved team should be getting bigger, not smaller.

    I understand that updates like these are necessary but if dtg keep doing updates like these and older dlcs are gonna come out a hot mess and not be fixed for a year then this can't be sustainable in the long run if there are only 4 people working on updating old routes to latest standards, new dlc's coming out, having to port those new dlc's into a new engine upgrade and re-release it again and again and having to fix existing bugs and potentially new bugs over and over again. The burden will only grow with time. I want this game to succeed but I don't want history repeating itself again
     
    Last edited: Jun 19, 2021
    • Like Like x 2
  32. Doomotron

    Doomotron Well-Known Member

    Joined:
    Oct 24, 2018
    Messages:
    1,221
    Likes Received:
    1,383
    I was aware that TS uses the Rail Simulator engine but not that the Rail Simulator engine was itself based on the MSTS engine. Surely barely anything is in common between the engines?
     
    • Like Like x 1
  33. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    I would happily be corrected because I don't know this for sure but I *think* the lineage is

    2000 Kuju built a single line Japanese rail simulator (possibly an offshoot from Simis - Kuju's predecessor) - I kind of think this was a demo for the engine to get customers but have zero evidence for this.
    2001 Kuju developed Train Simulator for Microsoft (not with as is often said).
    2007 Kuju produced Rail Simulator for EA as a successor adapting the Train Simulator engine.
    2010 Railworks / Train Simulator the successor to Kuju's Rail Simulator was made by Dovetail and as far as I know used Rail Simulator's core. Note that Train Simulator and Railworks are the same thing - it was marketed as both. Train Simulator has been enhanced into its current state of TS2021.

    All of these games have the Kuju logo in their branding to this day.

    Just to confuse things there is also BVE which appeared in 2010 and Open Rails. BVE is no relation but I really don't know about open rails except it can use TS content. There is also the seeming wanna-be "Train Simulator" from "King Activision NetEase Games" which is a very confusing (deliberately maybe) set of brand names. It also claims to be the "first train simulator" lol. It's not.

    Paul
     
  34. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    That is what I heard in that stream too. It would have all the fixes done to date to preserved content and that those fixes are in testing now.

    Paul
     
  35. synthetic.angel

    synthetic.angel Well-Known Member

    Joined:
    Aug 27, 2019
    Messages:
    1,484
    Likes Received:
    2,257
    Well, yes, in reality there is absolutely nothing that we can do to prevent the disaster to come. And there is nothing that DTG can do about it either, because the approach that they will take is already...um.... "baked in". And the company appears to have almost zero capability to learn from past mistakes.

    Based on everything that has happened so far, it's obviously going to be a totally horrendous disaster. We won't be getting the same issues as the transition from TSW to TSW 2 - because this time the problems are going to be much much worse.

    But if feels much nicer right now to think about it positively, and have some forlorn hope, because some hope is better than none - you know - "Get Rush Hour Done" and all that..... All I am trying to do, by encouraging people to be positive, is to give people the chance to live a slightly happier life for a brief time, until the inevitable comes to pass. Then there will be plenty of time for the pulling out of hair, the tearing of clothes, and the crying. So.....

    - My suggestion to the community is that we should all bury our heads in the sand, and pretend that everything will be absolutely fine, think about good times from childhood, honeymoon and/or trips to McDonald's, and if you didn't have any good times as a child, think about the good times that you would have liked to have had as a child...... that most of the other kids probably had. Oh, and if your platform lets you, make sure that your PC doesn't accept an update to TSW 2 on the day of the update to UE 4.26, and if you are on PS4 or XBox, then switch off your internet for a few days.

    - My suggestion to DTG is to carry on as you are, ignore all of the concerns in this thread (which is going to happen anyway), let the disaster roll out as predicted, and try to have some fun picking up the pieces and trying to glue it all back together after someone hits that "Confirm" button. Oh.... and.... um......

    1. Don't allow any holidays to be taken in the two months after UE 4.26 is deployed;
    2. Make a contingency plan to roll back everything from UE4.26 back to UE 4.23; and,
    3. Make a contingency plan to refund anyone that buys Rush Hour, because it presumably won't work on UE 4.23, when the roll-back to UE 4.23 happens. It would be very very naughty to try and keep that cash.

    But in the meantime, everyone should think happy things. ;-)

    Think Happy and Carry On.
     
    • Like Like x 2
  36. DTG Natster

    DTG Natster Community Manager Staff Member

    Joined:
    Sep 2, 2020
    Messages:
    1,203
    Likes Received:
    7,910
    You are correct in what you heard. All of the fixes that Adam has completed have already made it over to 4.26. These will be available with the Rush Hour update. This means you will be getting some of those long standing updates.

    I had a discussion with Adam and a lot of the content is already working on 4.26 and simply needs further testing, but we are yet to come across any complications. If we do experience any challenges we will make you aware.

    For anyone concerned about sound issues happening again, this is extremely unlikely as there are no new requirements for sound with the 4.26 update.

    We are investing a lot of time to help this transition happens as smoothly as we can.
     
    • Like Like x 9
    • Helpful Helpful x 5
  37. schorni

    schorni Well-Known Member

    Joined:
    Nov 8, 2020
    Messages:
    205
    Likes Received:
    398
    That means it will take another few months until the old stuff is fixed?
     
    • Like Like x 2
  38. solicitr

    solicitr Well-Known Member

    Joined:
    Jun 20, 2020
    Messages:
    4,754
    Likes Received:
    6,823
    No, it doesn't. It means all the fixes will be released at one time, with Rush Hour, presumably in August.
     
    • Like Like x 2
  39. DTG Natster

    DTG Natster Community Manager Staff Member

    Joined:
    Sep 2, 2020
    Messages:
    1,203
    Likes Received:
    7,910
    Anything that Adam has fixed for the preserved collection and has been waiting to be released will be released with Rush Hour.

    Anything that is still a work in progress will be released after Rush Hour.
     
    • Like Like x 3
    • Helpful Helpful x 3
  40. schorni

    schorni Well-Known Member

    Joined:
    Nov 8, 2020
    Messages:
    205
    Likes Received:
    398
    And we don't know when Rush Hour will be released, nor exactly which fixes will come with RH or after.
    So we actually know almost nothing ;)
     
    • Like Like x 3
  41. solicitr

    solicitr Well-Known Member

    Joined:
    Jun 20, 2020
    Messages:
    4,754
    Likes Received:
    6,823
    We do know what fixes Adam's team has made. Just watch the streams.
     
    • Like Like x 1
  42. synthetic.angel

    synthetic.angel Well-Known Member

    Joined:
    Aug 27, 2019
    Messages:
    1,484
    Likes Received:
    2,257
    You seem to be only referring to the work done by Adam's team, with respect to the updates to "Preservation Routes". This was my understanding from 15th June, in your fabulous Roadmap stream.

    Can you advise what the position is with respect to any improvements being done by the core/development team to improve Cathcart (not in Adam's remit yet)...... say, the audio mix/level and the light source on the front of the Class 314.... will this get patched before UE 4.26 is rolled out with Rush Hour....?

    And if Cane Creek comes out before Rush Hour, and it requires an immediate hot-fix in UE 4.23 (for whatever reason), will it get hot-fixed, or will buyers of Cane Creek have to take a gamble, and maybe have to wait until Rush Hour for any necessary fixes to be implemented?
     
    • Like Like x 1
  43. solicitr

    solicitr Well-Known Member

    Joined:
    Jun 20, 2020
    Messages:
    4,754
    Likes Received:
    6,823
    DTG have never improved the lighting or headlights on any route. Why do you think that Cathcart, uniquely, is a rush priority?
     
    • Like Like x 3
  44. schorni

    schorni Well-Known Member

    Joined:
    Nov 8, 2020
    Messages:
    205
    Likes Received:
    398
    Why do I have to look at the streams for that? There is a roadmap in written form where something like this should actually be on it ;)
     
    • Like Like x 2
  45. junior hornet

    junior hornet Well-Known Member

    Joined:
    Oct 15, 2018
    Messages:
    792
    Likes Received:
    1,224
    There is one major difference between this core update release and the upgrade to from TSW2020 toTSW2. With TSW2, those of us who had the original game had the option to revert to this to play the TSW2020 content that didn’t work properly in TSW2. As this release will be an automatic update, we won’t have that option. If any existing DLC is broken after the upgrade, we’re stuck with it as the facility to download and play the previous version will not be available.
     
  46. dreampage

    dreampage Well-Known Member

    Joined:
    Nov 29, 2018
    Messages:
    450
    Likes Received:
    657
    This is true. What I'm afraid of is that 4.26 could introduce a completely different set of bugs. Remember, three platforms, lots of pieces of DLC. So while the bugs from the prior upgrade may be fixed, others could be introduced. The recent release of the Rapid Transit and LIRR fixes shows that even lengthy testing doesn't guarantee a smooth rollout. And this time we're dealing with much more content.
     
    • Like Like x 4
  47. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    If you're on PC create a windows restore point before it upgrades then you can go back if you want to.
     
    • Like Like x 1
  48. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    I think you're being a little harsh there mate, for starters, this is a game, not a nuclear reactor, the upgrade isn't going to hurt or kill anyone no matter what happens.

    From what I've gleaned from the streams and forum discussions, the software architecture behind TSW is modular, compartmentalised and replaceable. Here are a few examples:
    • The new sky, it turns up in some routes but the old ones have the old sky until that is deliberately changed;
    • RailDriver it doesn't happen until its implemented into the loco deliberately yet its likely to be core;
    • PIS has appeared in some routes but hasn't appeared in others until deliberately added, perhaps with localisation e.g. LIRR
    • Track rendering has improved recently and doesn't appear in the old routes until its deliberately added.
    This actually gives me some confidence in the underlying architecture in that problems that come up from the upgrade (it would be foolish to think there will be none) are likely to be fairly localised if they are not outright game breakers. You would hope game breakers that affect everything would get caught in QA.

    Each time there has been a significant issue with DTG content that affects a lot of people in a way that makes game play impossible, within a week or so there has been a fixing patch. It's been a bit slower with Rivet but the important things have been fixed when you consider the 80/20 rules - fix the things that impact 80% of the people. Sure there has been delay at fixing the things that affect smaller groups or are annoying to people who know the route or vehicle but the reality is they are not game breakers.

    Take a look at the release notes for everything from 4.23 up to 4.26 - there hasn't really been all that much - there is a load of bug fixes and a bunch of new things that largely affect film production - some of them could become useful in the game but the fact they're not in the engine in 4.23 means that TSW isn't using them.

    When there are issues, most of us have to depend on the forums, but DTG would be a mid range player to EA so are likely to get attention and actual support because through engine royalties they will be pouring cash into EA every day.

    The other thing that gives me confidence is the last few releases, in the days coming toward the release there have been one or more minor updates followed by the big one. This helps minimise issues by reducing what is changed and doing some pre-work which both minimises the download time and minimises the install time.

    Am I saying there will be no issues at all? Of course not. There are bound to be some, but if they're important they are also bound to be fixed.

    Paul
     
    • Like Like x 4
  49. schorni

    schorni Well-Known Member

    Joined:
    Nov 8, 2020
    Messages:
    205
    Likes Received:
    398
    Freight exchange? Affects 100% of players, has been known for many months, has not been fixed. Not to mention all the signal errors, sound problems, etc.... ^^
     
    • Like Like x 2
  50. paul.pavlinovich

    paul.pavlinovich Well-Known Member

    Joined:
    Aug 21, 2020
    Messages:
    1,229
    Likes Received:
    1,412
    Remember what I said about the 80/20 rule and game breakers. Freight exchange is one scenario (and yes there are a few others that don't always or even ever play out). When compared to everything that works they are pretty small beans. I don't mean to belittle your thinking in any way, but we do need to consider reality. No software is every entirely bug free. It simply isn't economically viable.

    Paul
     
    • Like Like x 4
    • Helpful Helpful x 1

Share This Page