Postpone Netplay Development

[Erik Rutins]

ORIGINAL: Jagdtiger14
Concerning Netplay, if I cant keep a wireless connection to it, then its fairly useless. Maybe I need a better router, or maybe Matrix/Slitherine do not have the necessary infrastructure. Even if MWiF Netplay was completely bug free and working 100%, if it means I have to pay more money just to play this game (router upgrade, or deal with cables running across the floor...which my wife already vetoed anyway) then forget Netplay. I don't see it as Steve's fault. Matrix is a terrible company for many reasons. I think they are forcing Netplay. I would rather see optionals, scenarios, and a manual convoy system added to 2.3.4

Just one note on this specific point - our server handles the initial matchmaking and connects the two of you up, but after that it's your connection and your opponent's connection and your computers that matter. The communications at that point are direct and not going through any of our hardware.

Regards,

- Erik

[Erik Rutins]

I've read through all the feedback here and I appreciate it. I'll be discussing all this with Steve in a few hours as well.

A few points in the meantime:

- It is time for another "state of the game" post and I apologize for keeping you waiting. Steve and I are in regular communication, but we were hopeful that we'd be able to get to a stable netplay release before now and were continuing the cycle of testing and iterating with the expectation that we were closer to a stable release than it turned out.

- An official update is also overdue, I agree. Again, we were hoping that we'd have one based on the netplay betas at this point, but there have been more iterations than we'd hoped for. Until we have a stable release, it doesn't make sense to release a new official update. I'll see if we can get something out sooner, perhaps with a slight change of plans, without compromising the progress we've made on netplay.

- Netplay is a shared priority, not just driven by Matrix, but it is certainly a great concern and priority for us - primarily because we expected it to be working soon after release, as did Steve and the community. As you may recall from previous "state of the game" updates, we did focus on the solitaire game for a while to try to make sure it was in a good state before putting the focus back on netplay. Netplay, unfortunately, has taken more time, more focus and caused more regression issues than we would have hoped. Both we and Steve see the solitaire game and netplay both as extremely important. I hear the feedback from gamers who feel we have spent too much time on netplay and would like solitaire to get another pass sooner. I don't want to compromise netplay progress as we have come a long way, but I will see what we can do.

Thank you all for your patience and feedback, we do appreciate very much your continued engagement in the game and your communication to us.

Regards,

- Erik

[RFalvo69]

ORIGINAL: 4personalbusiness
PS, from my perspective as a PBEM-only player, without MWIF (such that it is), I would still be pushing around counters and maniacally crunching numbers on Cyberboard like I did back in the olden days of 2002 when I came back to the game. OK, I probably would switch to VASSAL, but still...

That is how the game is/was played since its inception - along with an infinite number of other wargames. Right now I'm playing OCS' Sicily II and my brain is not melting from the need to "crunch numbers" and actually move cardboard counters around a paper map.

True, you can make mistakes, and you need friends to be available if you do not want to play solo. These are the two things that MWIF was supposed to fix. Fact of the matter is: the WiF game I'm organising now will use my physical copy, we will be able to choose (just to give an example) between all the optional rules, and if we make a mistake we will "fix" it by remembering what the correct rule is before moving on; no need to wait for "patches" (which will break other functions anyway).

How do you think that people on BGG are playing WiF?

Anyway, there is another thing to consider re: Netplay. People who play alone are bound to play according to a certain "frame of mind" even when they do experiment or try to think out of the loop. This leads to parts of the game which are never properly tested because, maybe, not every solo player chooses a specific option/tactic.

If Netplay helps people to play against each other, then the willingness to "fire all the cylinders", to attempt unexpected stunts, and the very fact that two players are testing the game coming from different "schools" will help to shed light on seldom tested parts of the code.

Does this mean that a lot of games need to be started and then abandoned when the next dire bug comes up? Yes. But this is already happening anyway - what would be increased is the tempo of the stress test. This worked beautifully for finding bugs in both in the original WitP and then in WitP: AE - this is why exp. the latter is so polished.

My 2 cents anyway.

[AllenK]

Erik, thanks for your reply.

I understand Netplay is important to Matrix. It is only when MWiF, with its current coded set of optional rules and scenarios, is playable in Solitaire, Hot-seat and Netplay modes, will it be the game as advertised and sold on release. That has to be high on the list of priorities and I agree with others who comment that in this online age, a game that doesn't have this feature is likely to limit its customer base. It's for this reason I can see why it was chosen over developing the remaining optional rules and half-map scenarios and/or a working AI mode.

Had the Beta's only been introducing changes for Netplay, we wouldn't have bothered updating until we needed Netplay. However, they also contained fixes for problems with the basic game, which we wanted in order to resolve problems we were getting. The frustration came when important features like manual convoy routing no longer worked (it hadn't been trouble free in 2.2.0), we couldn't repair factories and, the real game killer, couldn't get past CW set-up without MadEx's. You can't even begin to play a game you can't set up.

It was these plus other equally frustrated posts on Tech Support, reporting problems in other areas that had either previously worked or previously been fixed, that prompted me to post this thread. The game just seemed to be going backwards from a state where people could play and enjoy it (if not in the way they would choose) to one where nobody could unless they stuck to a pretty ancient (and still flawed) official release.

I'm sure all and especially Steve are equally as frustrated and hope a way can soon be found to get the lost Solitaire functionality back so the necessary development on Netplay can carry on.

[Erik Rutins]

Hi everyone,

A quick update now while we put together a longer "state of the game" news post for the future.

Steve and I had a conference today and agreed to put the primary development focus back on fixing and stabilizing solitaire play for the next couple of weeks with the aim of getting a new official update with improved solitaire play out sooner rather than later. We had always planned to do another pass on any new bugs once Netplay was stable, but we will now do that sooner as well. You can expect to see that as a public beta in the next week or two, then as an official version once we confirm it is a net improvement. As soon as that's been released, focus will return to Netplay with the goal of aiming for a stable Netplay release by sometime in May.

The entire update process throughout both pre and post-release has been two steps forward, one step back and it's only through continued iterations and testing that we can find those steps back and fix them while we head for a solid version.

For those of you currently playing in a game together and encountering bugs, the best thing you could do to help Steve is to make sure that any bug reports you've submitted or posted include save files and as much detail as possible on how to reproduce. He is very motivated to get the solitaire issues resolved over the next couple of weeks. As long as he has reports for the issues you've encountered, the odds are good he can fix them.

Regards,

- Erik

[warspite1]

Erik/Steve

Many, many thanks for this. This is most welcome news indeed [:)].

[AllenK]

Erik/Steve,

Thank you and much appreciated.

As always, I will continue to post problems to Tech Support with as much detail and a game save in order to assist with the development.

Look forward to the new Beta/Official upgrade soon.

[Sir Roland]

I'd also like to add my voice in thanks to Erik/Steve, for the prompt reply.

[Shannon V. OKeets]

Fixing Solitaire Bugs - Priorities

[page 1]


Here are the next (in priority) two problem areas I have on my Master Task List (aside from NetPlay bugs): Peace and Production Planning. This was last updated at the beginning of March, despite the date of March 21st at the top of the page.

The source of the bug reports can be identified by the MTL number I assigned to it. So, 2363 was from a beta tester. 6631 from from an email I received (i.e., a Mad Except error). 7027 was from a post in Tech Support. Within each source, lower numbers mean the bug report is older.

Note that this is a fragment of my Master Task List. Not shown are the sections before Partisans (e.g., Air, Land, and Naval Operations) and those after Quit (e.g., Supply, Vichy).

The color of the MTL number reflects whether there is a saved game available or not; those colored green have an associated saved game. They may or may not have accompanying instructions on how to reproduce the bug. The numbers following the section label [Peace (30 - 17)] indicate the number of bug in the section and the number which have associated saved games.

The colored text describing the bugs highlights for me what I believe I should look into next. There's no real difference between red and violet. I have not confirmed that all of these are actual bugs! Specifically, upon investigation I have often found that the older emailed Mad Except reports (3000 to 6999 and 9000 to 9999) have already been fixed.

I haven't highlighted/colored the Production Planning bugs yet (except for one). I would normally do that upon clearing the highlighted Peace bugs and moving on to Production Planning. This is all rather dynamic and not a rigorous process for setting priorities. For example, if a new report in Tech Support is similar to one already on my list, I will usually try to fix them both.

If you look at the details of the bugs in Peace, you will find that most of them at rare occurrences (e.g., liberation of Yugoslavia while Vichy units are present).

As should be obvious, I prefer to work on bugs that have saved games where I can reproduce the bug - or prove to myself that it no longer exists.

[Shannon V. OKeets]

Fixing Solitaire Bugs - Priorities

[page 2]

Perhaps I should have shown this page first. It is part of my spreadsheet for tracking bug reports from Tech Support. It's roughly the still outstanding bugs reported in the first ~2 months of 2017. Later today, I'll go through bugs reported in Tech Support since March 7th and bring this list up-to-date. I should note here that most bugs related to NetPlay don't make this spreadsheet. I have a separate text file where I record all the NetPlay bugs, which I have been giving a pretty high priority in February and March.

The items with an X before their Category, and colored green, have been fixed. The one in yellow (with an O suffix) I have deemed Opaque. I'm not sure whether it is or is not a bug but I do not want to dismiss it out of hand. On the other hand, I have investigated it as far as possible with the information I currently have. So it lingers in the spreadsheet and I read it over again from time to time.

You can see that I have fixed some of the February bug reports and investigated others, even though they are for Solitaire and Head-to-head play. Indeed, I have taken a first crack at all these items, trying to reproduce them and seeing if there is any obvious fix that I can make. Most of these entries have not been transferred to my Master Task List. Whenever possible I prefer to tackle new bugs from Tech Support immediately after I place them in this spreadsheet. About half the time (8 out of 19 in this case) I'll move them to the Fixed section of this spreadsheet and never make an entry in my Master Task List.

Actually, I don't move the individual line entries. I simply do a sort on the Category. That places all those beginning with X at the bottom of the page, but still sorted by category.

So, in summary, I work on newly reported bug in Tech Support every 2 to 3 weeks and process them all. Those that I can fix relatively quickly, I do. If the bug is fatal, then I do my best to fix those too. If that effort does not solve the problem, then the bug is moved to the Master Task List and I process those by 'section' (e.g., Peace). There are two other streams of bug reports: from emails and from beta testers. I also go through each of those every 2 to 3 weeks and handle them the same way.

Such us the view from my chair, ...

[Centuur]

I have currently three netplay games running. In all three we can't proceed. I would like to see a timetable with a date on which Steve will resume fixing netplay bugs. This decision looks to be a slap in the face of the people who are participating in the Netplay Challenge...

You can't ask for volunteers and than simply discard what they have been doing in the last couple of week...

[CanInf]

In a few months I look forward to restarting my netplay game! In the meantime, I was wondering what this means for hotseat. I suspect there should not be a lot of issues ensuring fixes that work for solitaire are equally effective for hotseat.

[brian brian]

some thoughts

fascinating bug list, thank you for posting Steve

War in the Pacific was mentioned, which makes me ask - how do other Matrix games handle things like the players using WiFi with their routers, and the ever evolving hardware ecosystem and thus inevitably evolving connection protocols? I learned via trading hobbyist type live music recordings how to do things like "port forwarding" but average computer users have no interest in that kind of thing. I would hope that some of the issues involved could be solved by Matrix with modular solutions from their other programmers who have to deal with the same thing.

The whole idea of Netplay seems to be asking Steve to put on another Wizard's hat, and he has already mastered wearing several of those.

Perhaps Netplay and World in Flames could be looked at from a Software Engineering point of view, more than a simple the-whole-game-has-to-be-playable-with-2-connected-computers. And by that I mean coming up with a framework for how the game will actually be played by 2-6 people all in locations remote from each other. I still maintain that I don't need or want to be waiting around with my computer and my time while my opponent does something like move every unit on the Eastern Front. Ditto for being the American player in 1940. If the hard coding of the rules can't be leveraged to the advantage of the player experience, I see no point in moving beyond VASSAL/Cyberboard.

So perhaps Netplay functionality can be prioritized to work perfectly during naval movement and any and all combat phases, which have so many interaction points. Non-interactive phases could be handled while some players are offline and Netplay / the Sliterine server would simply make sure the movement decisions and political and economic decisions are transmitted to the other players when they need them.

Perhaps that is asking for whole new features, I'm not sure. But I thought some of the NetPlay bugs were occasionally reports of units not moving correctly on other players' screens, or freezes etc. and perhaps focusing on NetPlay combat could keep development of every style of play moving forward, dunno.

[brian brian]

But to me, the biggest issue is not which "mode" of play is worked on when, is one that has problems across every "mode" - the Convoy system. I don't play MWiF because of this issue. That should ring pretty loudly somewhere.

MWiF is not intended to be a World in Flames "kit" to facilitate players doing whatever they want with the counters. Programming a no rules, counter&map management only mode would have taken time away from coding the rules. Fair enough I guess.

And yet some players hoped for an "automatic" convoy system because managing the Commonwealth Empire's economy is not a simple thing to walk in to when one starts playing WiF. Simultaneously, the final portion of the project - an AI opponent that can play Global War - has to be able to manage convoy lines by itself. So in some ways, a feature "crept" into the project, and in other ways it is just a feature that isn't truly needed, yet.

This is not a simple thing. It requires writing a routing algorithm to handle moving things (resources) across two separate networks (multiple, unconnected areas of land connected by a system of Sea Zones, both of which have routing constraints such as the Panama Canal or enemy ZoCs), with hard constraints beyond simple capacity limits at choke points - the Trade Agreements. Doing this would probably make for an excellent course assignment at a fairly high level of computer programming education.

I think what is needed is a do-over of the whole system. First, let the players manually set their convoy routes, click-click-click through the Sea Zones. Then check for legality against the movement restrictions and the Trade Agreements. If anything is in violation, do not allow the player to complete production. And usually in RaW 7 Oil Rules, the first thing you do is re-organize your oil units, then figure out everything else.

Without this, I just don't have any interest in playing this game. WiF simulates the constraints and frustrations of a Major Power's penultimate leader. I hardly want the program to add to these limits, in any way. If I can't control my own counters, I will use other software that allows me to do so. And that's the bottom line.

[warspite1]

ORIGINAL: Centuur

I have currently three netplay games running. In all three we can't proceed. I would like to see a timetable with a date on which Steve will resume fixing netplay bugs. This decision looks to be a slap in the face of the people who are participating in the Netplay Challenge...

You can't ask for volunteers and than simply discard what they have been doing in the last couple of week...

warspite1

Centuur I think this is a little unfair.

Firstly, if you look at the timescale Erik has suggested, then you will see that those proposing this course of action have only been given a short amount of time - your pain should therefore be short. Don't get me wrong, I am grateful for anything after such a long wait, in order to get another stable official version that actually (hopefully) gets rid of the convoy/production and remaining supply bugs.

Secondly, you ask for an actual timetable, but when has anybody - on either side of the fence - ever been issued with a timetable on this project, which by its very nature has proven unwilling to conform to timescales.

Thirdly, you have three netplay games on hold - well we have four non-netplay games on hold. This is exactly why some of us were calling for a re-evaluation of the flight path in the first place.

Fourthly, 'you can't ask for volunteers and then simply discard what they have been doing' - well I could counter that with 'you can't ask people to pay good money for a game and then expect them to be either beta testers or no longer be able to play the game pending another official version which shows no sign of turning up after 14 months'. Well that is the reality hence the request to make this slight - and it is a slight - deviation in flight path.

[Shannon V. OKeets]

ORIGINAL: Centuur

I have currently three netplay games running. In all three we can't proceed. I would like to see a timetable with a date on which Steve will resume fixing netplay bugs. This decision looks to be a slap in the face of the people who are participating in the Netplay Challenge...

You can't ask for volunteers and than simply discard what they have been doing in the last couple of week...

I am not abandoning work on NetPlay. It is just being delayed for a week or so. I should be able to make another Hot Patch available tomorrow (Thursday Hawaii time) that will enable you to continue your NetPlay games in progress. It took me a solid week to fix #s 17 and 18 below - which were solely for NetPlay.

===

Changes Since Version 2.4.1.0
(as of March 4, 2017)

6. March 10, 2017 - Fixed a problem with displayed Activity Limits were the US was not being shown.

7. March 12, 2017 - For NetPlay, made some changes to advance the subphase after the attacking player decides whether to use Snow units in an attack.

8. March 12, 2017 - For NetPlay, added some checks so a player can no longer change the actions and destinations of his opponent’s resources (unless they are being received in a trade agreement).

9. March 14, 2017 - Added code to prevent placing partisans in a non-frozen lake hex. This change also prevents having too many partisans to place. That is, the number of partisan units is never more than the number that have hexes in which they can arrive.

10. March 14, 2017 - Added a check for overstacking when the weather phase ends.

11. March 16, 2017 - Fixed a couple of problems with correcting overstacked hexes at the end of the Return To Base phases.

12. March 16, 2017 - For NetPlay, fixed several problems with correcting overstacked hexes. The way the program now processes multiple overstacked hexes is that each player gets to correct the overstacked hexes in which he has units. He may have to process them for several major powers. Once all players have processed all their overstacked hexes, the end of phase completes and the game advances to the next phase.

13. March 17, 2017 - For NetPlay, added code to report on US Entry markers’ draws, moves, and removals to non-US players.

14. March 17, 2017 - For NetPlay, added code to report when the US player decides to not execute any US Entry Options in a turn.

15. March 18, 2017 - For NetPlay, added a check to avoid a spurious fatal error when the game is restored at the subphase Choose Aligned Minor’s controlling major power.

16. March 19, 2017 - Fixed a cosmetic problem where the informative message about territories being converted to minor countries after their parent country is conquered had the names of the conquered and converted countries reversed.

17. March 19, 2017 - For NetPlay, fixed a problem where building new convoys, pilots, forts, offensive chits, and factories could have mismatched unit numbers on different computers. This would cause all kinds of problems, such as moving a convoy on one computer could cause a factory to be moved instead on the other computer.

18. March 20, 2017 - For NetPlay, fixed a problem with Production where the Axis computer was reinitializing the Production phase in the middle of reinitializing the Production phase. This was causing the Axis computer to set all resources coming from neutral minor countries to be set to Idle.

19. March 20, 2017 - Changed the order in which the program attempts to find actions, destinations, and paths for resources during Production Planning, so the player’s most recent instructions are attempted before trying to use the actions and paths from the previous turn. The sequence is now: Override settings, Default settings, Most Recent settings, Last Turn settings, and, finally, try to find a new route. This should correct some of the problems with the program ignoring recent changes and reverting to the actions, destinations, and paths used previously.

20. March 21, 2017 - On the Spend Surprise Points form, corrected the number of ships in a naval combat to exclude those that are aborting.

21. March 21, 2017 - On the Spend Surprise Points form, corrected the number of Air-to-Sea factors in a naval air combat to reflect the weather conditions in the sea area (e.g., halved during Rain).

[peskpesk]

Nice start on fixes! My five cents on what to fix: Work on bugs that have saved games(both netplay and Solitair/Head to head) and a correct bug report (so you can reproduce the bug) and make fixes faster, take game hindering bugs first. This way all testers/players also learns the importance of a correct bug report with a save game, if they want it fixed fast.

[rkr1958]

ORIGINAL: Shannon V. OKeets
19. March 20, 2017 - Changed the order in which the program attempts to find actions, destinations, and paths for resources during Production Planning, so the player’s most recent instructions are attempted before trying to use the actions and paths from the previous turn. The sequence is now: Override settings, Default settings, Most Recent settings, Last Turn settings, and, finally, try to find a new route. This should correct some of the problems with the program ignoring recent changes and reverting to the actions, destinations, and paths used previously.

I'm really anxious to see the impact of this change and hopeful that this will give us the ability to manual set convoy routes, one again, that aren't immediately overrided by the program.

[Erik Rutins]

ORIGINAL: Centuur
I have currently three netplay games running. In all three we can't proceed. I would like to see a timetable with a date on which Steve will resume fixing netplay bugs. This decision looks to be a slap in the face of the people who are participating in the Netplay Challenge...

You can't ask for volunteers and than simply discard what they have been doing in the last couple of week...

This is not a slap in the face to anyone, at least it's not intended that way. We expect to resume the Netplay focus in a couple of weeks and I'm sure Steve will also be able to hit a few Netplay bugs during that time, this is not a permanent change in direction.

Regards,

- Erik

[michaelbaldur]

remember, solitaire is the foundation of netplay, so any fixes in solitaire is also a fix in netplay.