Many things seem to be wrong at the moment

[jason oates]

Hi,
I upgraded to the latest version half way through playing "Crimea a River". Now my aircraft prefer to fly on ignoring Bingo and nearby tankers until the run out of fuel. Strikes head for the Baltic and the Sahara. I give up. let us know when the AI running the game is returned to playable status.

[thewood1]

Typically, things break in isolated ways that lets the player work around issues with specific types of units or around narrow time periods. But the last (almost) two months have really been tough. Flight plans, waypoints, RTB, WRA, etc. all seem to be having intermittent, universal, and serious issues. These are critical aspects of the game that players should depend on and AI must depend on for scenario design and execution. My hope when these issues were reported was that a short-term hot fix would come out to keep things rolling. I really hope nothing serious has happened to the game's road map.

[DWReese]

Jason,

I agree with you about the wonky play that you have described. It was suggested by one of the programmers that you could return to 1328.11, as that was a stable version. I took it back even further, to 1328.8. To me, that works even better.

There is one catch that I must mention, however. Apparently, each version has a max cap as far as the database that it can handle. 1328.8, for example, stops at database 501. So, if a scenario requires something greater (like 505), then it won't work.

Additionally, if you have updated the scenario, then there is no going back.

Finally, if you have INSTALLED a later update, then it likely automatically updated your PURCHASED scenarios (as it did mine), which means that if you went back to 1328.8 (or some other version), then it probably won't work because it is now expecting database 505.

I would like it if the scenarios could be accessed via the version that is being played (i.e. the ability to go backwards), but I don't know how that can be done. In the meantime, I guess I will just wait until the latest version has been repaired.

[stepbackjumper2]

It's frustrating that the current version of the strike mission is broken.

[Quark73]

I take this longer cycle as a good sign that it is possibly caused by a more diligent testing phase, there were some indications that Matrix Games is putting more focus on testing, what I totally appreciate. There were lots of cool features in the past versions, that unfortunately had partial impact on core functionalities. This is somehow normal for software projects of a certain complexity. So careful testing is key. I ,personally, could perfectly live with a feature freeze for some iterations and stabilizing, harmonizing and maturing all the helpful tool, planners and managers.

And have in mind all the updates coming for free for the commercial version, some features are introduced there even before they make their way into PE. That is not a matter of course I don't know another game where user have that much direct contact and impact on development and course of a game.

[thewood1]

Thats all well and good. It generally matches my philosophy both as a player and having run software companies. I'm all for more testing before release. But when you have an issue or series of issues that inhibit core game functions, my expectation of any game company is to get a quick patch out to fix or allow easy regression of the version. I stopped saving previous versions a couple years ago. I have learned my lesson.

[mikerohan]

I stopped saving previous versions a couple years ago. I have learned my lesson.

You're not alone there...

[Tcao]

Jason,

I agree with you about the wonky play that you have described. It was suggested by one of the programmers that you could return to 1328.11, as that was a stable version. I took it back even further, to 1328.8. To me, that works even better.

There is one catch that I must mention, however. Apparently, each version has a max cap as far as the database that it can handle. 1328.8, for example, stops at database 501. So, if a scenario requires something greater (like 505), then it won't work.

Additionally, if you have updated the scenario, then there is no going back.

Finally, if you have INSTALLED a later update, then it likely automatically updated your PURCHASED scenarios (as it did mine), which means that if you went back to 1328.8 (or some other version), then it probably won't work because it is now expecting database 505.

I would like it if the scenarios could be accessed via the version that is being played (i.e. the ability to go backwards), but I don't know how that can be done. In the meantime, I guess I will just wait until the latest version has been repaired.

I played a PBEM game with a friend since Dec, all these updates broke the game one way or another. it's frustrating to find out the new update solve some issues but introduce new issues. We have frozen the PBEM game

That's a good idea to go back an old version.

But do you have a list of known bug under 1328.8 and 1328.11 ?

[DWReese]

From what I remember, 1328.8 introduced 1.06, which was a change from 1.05. The testing had been done over the previous months, and when they made the move to 1.06 there wasn't anything that I know of that was broken.

It was suggested that revert to 1328.11. I believe that there were some issues there because that's about when those aiming "crosshairs" started to appear. To me, those were always off, especially with ABM. So, I didn't want to take a chance, and I reverted to 1328.8.

Ironically, while the newest version (1328.18) has many bugs (most of which are associated with the Flight Planner, etc.), it did seem to fix the problems with the ABM issues. FWIW.

[Quark73]

Thats all well and good. It generally matches my philosophy both as a player and having run software companies. I'm all for more testing before release. But when you have an issue or series of issues that inhibit core game functions, my expectation of any game company is to get a quick patch out to fix or allow easy regression of the version. I stopped saving previous versions a couple years ago. I have learned my lesson.

I am completely with you. Looking on the past quick fix releases, they came pretty often with some side effects. That's why I prefer that devs take the time to get the half baked parts done and nice before adding new stuff. As a former co-worker of used to say: If you in a hurry, you have to act slowly.

[Kobu]

With the current state of the simulator, the only thing that is needed is for the current bugs to be fixed and for other features not to be broken in the process. We have had many months of many new features added, now it is time to consolidate that, not add more features.

There is no point in continuing to add features when there are bugs reported that are more than 6 months old.

Regards

[thewood1]

With the current state of the simulator, the only thing that is needed is for the current bugs to be fixed and for other features not to be broken in the process. We have had many months of many new features added, now it is time to consolidate that, not add more features.

There is no point in continuing to add features when there are bugs reported that are more than 6 months old.

Regards

You'll never have all bugs fixed. If thats what you are waiting for, I'm not sure how you can enjoy any game that is more than moderately complex. But there are levels of bugs:

- Design issues a player doesn't agree with...not a bug
- single unit or small groups having db issues.
- bugs with no work around
- bugs that cross all units but are narrowed to specific time periods
- Bugs that impact specific functions across all units and time periods, but not main game function and the functions still generally have outcomes as designed.
- Bugs that impact major and core game functions so that those function do not provide outcomes as designed.
- Can't start the game for a wide group of people

The last three types would be expected to have an expedited hotfix. The rest can be grouped into different levels of patches. In the three software companies I have worked at, there would be escalation processes, mitigation actions, and then directed responses at each level that would set timing and resources. I don't really expect Warfaresims to have that type of complex corporate process. But I'd like to think there is a recognition that something has gone wrong.

[boogabooga]

+1 to the idea that a feature freeze to focus on quality control would be a good idea. CMO has so so so much potential, but it's in a frustrating state right now.

[zclark]

I do like the suggestion of a feature freeze until core functionality is fixed and a quality control process is put in place to prevent another release such as 1328.18. Looking back through the discussion threads, I’m not really seeing a lot of community feedback for the beta prior to release. Is the CMO team heavily relying on community feedback during betas to determine if a release should be launched? If so, I might revisit that approach. Where I do see the community assisting the CMO team is in the development of a test scenario library and testing steps. Each scenario /testing step could be built in close collaboration with the CMO team to test core functionality. A consensus and prioritization of every feature / function would need to occur first. Afterwards, specifications / requirements should be clearly defined for each function to help testers (internal and community) know what the desired state should be.

Dissecting the documented weather functionality as an example, here are some test scenarios / testing steps that could be created to test functionality:

1. Manipulate the weather sliders to change weather globally. Expected results, global weather matches selected settings. Note, if the Weather modeling level feature is not functional, can it be removed to eliminate end-user confusion until implemented?

2. Create a Custom Environment Zone and change the settings from global weather. Expected results, weather in custom region exhibit the desired weather conditions.

3. Create two Custom Environment Zones, one extremely hot and one extremely cold to test impact to sensors. Will need to know specifically what sensors are impacted and with what penalties.

4. Set global weather with Rainfall maxed out. Verify visual, IR, and laser sensors are impacted. Will need to know how much each is penalized.

5. Set global weather with Cloud coverage to max. Create a LGB strike mission. Expected results, strike package unable to deliver ordinance.

6. With global weather set with Cloud coverage maxed, test an aerial platform with only visual sensors. Expected results, platform unable to spot ground targets visually.

7. Set global weather with Cloud coverage to light coverage. Create an iron bomb strike mission. Expected results, strike package reduces altitude under the cloud base to deliver ordinance.

8. Set global weather with Sea state set to Hurricane. Expected results, Surface ducting is ineffective. Sonar ranges severely degraded. Will need to know by how much.

9. With global weather Sea state set to Hurricane. The minimum safe altitude for low-flying aircraft is increased. Will need to know by how much.

10. With global weather Sea state set to Hurricane. Accuracy of naval gunfire degraded. Will need to know by how much and by ship size.

Bottom line is the responsibility still sits with the CMO team for testing as community engagement can be unreliable.
Hope this sparks some ideas to find a solution to the current situation.

[boogabooga]

I suspect that the helpdesk ticket paradigm itself might be part of the problem. Yes, the community can submit bug reports with a demonstration scenario, and the developers can fix that one bug and show that it's fixed in that one scenario. But what seems to get missed is how that change might have unintended consequences somewhere else.

[Dimitris]

We are aware of the issues, and we are doing everything possible to ensure a steady update path that will restore the level of quality that our regular players are accustomed to. We can only ask for a little patience at this point. Unfortunately we are unable to provide a firm schedule for the next updates just yet.

[thewood1]

Can we at least get a retrograde version where missions can function consistently?

I got by somewhat when .18 showed up here by using Steam at .14. Then about a month ago Steam went to .18 and shut down that workaround.

[Nikel]

I have the exes of v12 and v14, may I upload them to the forum?

Or v8 that was commented as better, may it be uploaded?

[thewood1]

I think the issue with just uploading an exe or a simple update is newer dbs and updated scenarios

[Nikel]

Yes, the devs use this kind of protection for some reason.

You may use v18 to a edit a scen to an old db and then load it with a compatible exe, but I do not know if it breaks something.