Page 1 of 5

0.4.7 preparations

Posted: Mon Feb 13, 2017 7:47 pm
by Vezzra
I think it's time to get serious about getting 0.4.7 ready. I already adjusted the milestones, went over all the currently open issues/PRs and tried to decide if they should be considered required or optional for the release, or if they should preferably be postponed, and assigned them to the corresponding milestones accordingly.

I've set up the following milestones:
  • 0.4.7: Required for the release. All issues/PRs assigned to this milestone are considered release-blocking, and have to be fixed/finalized before the release branch can be created. They should therefore be everyones topmost priority.
  • optional for 0.4.7: Optional for the release. Issues/PRs assigned to this milestone are not considered release-blocking, which means they will go into 0.4.7 if fixed/finalized in time (that is, before the creation of the release branch), but we won't delay the release for any of them. These are of lower priority.
  • post 0.4.7: Not intended for inclusion into the release. Work on issues/PRs assigned to this milestone should continue after 0.4.7 is out, until then they shouldn't be given any priority.
  • Undecided/TBD: This milestone acts as a marker/indicator if we can't immediately decide which milestone an issue/PR should be assigned to. Issues/PRs assigned to this milestone are of course not considered release-blocking.
There are a few issues/PRs without a milestone, this is the case for things that actually don't affect the release cycle (e.g. technical documentation, all issues/PRs tagged as "component:infrastructure", etc.). Which btw. is why I didn't use "no milestone" for "undecided".

Here are direct links to the milestones and the issues/PRs assigned to them:

0.4.7: https://github.com/freeorion/freeorion/milestone/16
optional for 0.4.7: https://github.com/freeorion/freeorion/milestone/19
post 0.4.7: https://github.com/freeorion/freeorion/milestone/18
Undecided/TBD: https://github.com/freeorion/freeorion/milestone/20

As your assessment of which issues/PRs should be assigned to which milestones might differ from mine, I encourage everyone to go over these lists and change the assignments where you think that's necessary. If you assign an issue/PR to another milestone, leave a comment giving your reason for the reassignment. Alternatively, post a request here in this thread to reassign an issue/PR (especially when you think the reassignment should be discussed first).

What needs to be done:

1. The new fighter/carrier mechanic and all components/elements that have been introduced by it need to be polished/get their final touch. Same goes for the AI adjustments for the new mechanic. Please, everyone, think what might still be missing, what still needs to be done to get the new feature release-ready. And then of course playtesting.

2. All the issues/PRs assigned to the "0.4.7" milestone must be fixed/finalized.

3. Any of you who have something (a project, fix, tweak, whatever) they are currently working on, which they want to get into 0.4.7 and which aren't already covered by the issues/PRs assigned to the "0.4.7" milestone, please open an issue or PR (whatever fits) and assign it to the "0.4.7" milestone. Please include the reason why you think it's required for the release in the description of the issue/PR. You can also post a request here in this thread if it can't be done via an issue/PR (but I can't imagine where that could be the case).

What's important now is that all of you concentrate your efforts on getting 1 and 2 done, and put other projects/work on the back burner. A "status report" on all things related to the new fighter/carrier mechanic would be a very good thing, especially from the content scripting/balance and the AI department, to get an idea what still needs to be done, and then work from there.

It would be good if we could get all necessary preps done and create the release branch by the end of this month, then we have a good chance ot get the release out by the end of March.

Comments, opinions, feedback welcome. 8)

Re: 0.4.7 preparations

Posted: Thu Mar 23, 2017 2:18 am
by Dilvish
Rather than open a new thread for the final AI planning for 0.4.7, it looks appropriate to piggyback on this. Right now Morlic has a branch where he is working on dealing with the AI complications of fighters not being able to attack planets. I think that is our highest priority right now and might tinker on it with him. Morlic, if you have anything else you were wanting to cover for the AI for 0.4.7 please note it here. CJ, I am assuming that you are still too busy to have anything you are wanting to take care of yourself for 0.4.7.

In Morlic's Invasion Improvements PR we talked about a few things, most of which seems done to me, I just noted there a last bit yet undone, about some further improvements in coordination between the military dispatcher and the invasion fleet dispatcher; we'll see how it goes but it is not currently seeming to me to be critical that we finish that for 0.4.7.

I have also been continuing to tinker a bit with some of the AI fleet movement decision making, and some AI research priorities, and hope to finish up a bit in those areas, but I don't consider them critical for 0.4.7 either.

It would be nice to have a bit more on the diplomacy side, but that will probably have to wait.

Is there anything else I am overlooking, or does someone have a different take on any of this?

Re: 0.4.7 preparations

Posted: Thu Mar 23, 2017 1:01 pm
by Vezzra
As you've probably noticed, I went over all the issues and PRs assigned to the "0.4.7" milestone on github and requested a status report on some of them. Several issues/PRs are close to being merged/addressed/fixed, and most of the rest probably aren't really critical for the release, so I'm going to reassign those to the "optional for 0.4.7" milestone (unless someone objects and explains why a particular issue/PR should be considered critical). I've posted corresponding notifications/remarks in the comment sections of the issues/PRs in question.

This leaves the work which needs to be done for the AI as the major issue that has to be dealt with in order to get 0.4.7 out. Considering that there is still a lot of things that need improvement and could be done in that field, I suggest to concentrate on the things which are really critical for the AI si it's not to be crippled/can function properly. Postpone everything else until after the release.

With that in mind, @AI team, what do you think is absolutely necessary to get done for 0.4.7 that hasn't been properly addressed yet? Can you give us a rough estimate on how long it might take you to finish those things? I'd like to finally get some schedule down for the release.

Anyone else aware of anything that absolutely needs to be addressed for 0.4.7, which isn't covered by any of the open issues/PRs currently assigned to the "0.4.7" milestone?
Dilvish wrote:Right now Morlic has a branch where he is working on dealing with the AI complications of fighters not being able to attack planets.
I wonder if we could address that issue by simply allowing fighters to attack planets. I can't recall why we decided they shouldn't, although I dimly remember there has been a discussion about this... I guess I'll need to do some digging to refresh my memory.

Re: 0.4.7 preparations

Posted: Thu Mar 23, 2017 2:56 pm
by Vezzra
Vezzra wrote:I wonder if we could address that issue by simply allowing fighters to attack planets. I can't recall why we decided they shouldn't, although I dimly remember there has been a discussion about this... I guess I'll need to do some digging to refresh my memory.
Doh! Just found the discussion about fighters and planetary shields, and guess what, I have been one of those arguing for fighters and planets not being able to attack each other. The problem (as I now remember very clearly) is that if fighters could attack planets and vice versa, fighters would make planetary defences almost completely useless (as the high damage single shots of planets would be utterly ineffective against the high number of low structure targets fighters are).

So yes, we need to keep that. In that discussion it has also been mentioned that this is a challenge for the AI and needs to be addressed. So my question is, how far are we along with that specific project? Because that's definitely critical for the release, the AI has to be able to at least basically handle fighters vs. planets.

Re: 0.4.7 preparations

Posted: Thu Mar 23, 2017 5:00 pm
by MatGB
From a balance and gameplay perspective, I also REALLY like that fighters don't attack planets or break stealth for the carrier, it makes stealth viable for combat in an interesting way and means you can have "fleet killer" fleets and "planet killer" fleets, making things like the spinal antimatter cannon more useful.

But I 100% agree, this is the Fighters release, if the AI can't handle fighters properly (and it's got a LOT better recently) then that has to be the priority. If I can help and/or there's test commits I can check out LMK.

Re: 0.4.7 preparations

Posted: Thu Mar 23, 2017 7:20 pm
by Morlic
Vezzra wrote:So my question is, how far are we along with that specific project? Because that's definitely critical for the release, the AI has to be able to at least basically handle fighters vs. planets.
I think I will have a PR ready this weekend which provides the basic functionality.

Re: 0.4.7 preparations

Posted: Sun Mar 26, 2017 6:48 pm
by Vezzra
The list of issues and PRs critical for the release is down to 5 items (one of which is "compile changelog" ;)), so I decided it's time to finally propose a schedule for 0.4.7. This time I'm using a github issue to manage the checklist and track the progress of the release (suggested by Marcel).

The issue with the inital dates for the milestones (creation of release branch, RC1, declaring release) can be found here:

https://github.com/freeorion/freeorion/issues/1442

I propose to give us another 2 weeks until creation of the release branch, that should hopefully be enough (especially for the AI team) to address the last pending issues. One week would most probably have been a bit too tight. Everything "optional for 0.4.7" must be merged by then too if it shall go in. Should some serious issues remain unaddressed until then, creation of the release branch will be postponed accordingly.

@AI team: do you think you'll be able to get all the important/critical stuff done until then? Or do you need more time?

After that I've given us a few days to finalize the changelog, commit all the release specific stuff (like disabling super testers etc.) to the release branch, and fix any bugs left before releasing RC1. If that turns out not to be enough (e.g. if some more serious bugs turn up), RC1 will be postponed accordingly.

I suggest to give us 2 weeks of playtesting before declaring the official release. If bugs/issues turn up, they will need to be fixed, another RC announced and the date for declaring the final release postponed accordingly.

Comments, suggestion, objections, requests here in this thread, be it about the suggested dates, the procedure etc. Please check the release checklist issue regularly to keep yourself updated about the progress and any changes. That checklist issue will be updated to reflect any progress and changes, I'll also announce all important changes and completion of milestones here in this thread.

Once again I want to appeal to everybody to really focus on the critical/release blocking stuff now. It's time we get that release out! 8)

Re: 0.4.7 preparations

Posted: Sun Mar 26, 2017 8:44 pm
by Dilvish
Vezzra wrote:I propose to give us another 2 weeks until creation of the release branch... @AI team: do you think you'll be able to get all the important/critical stuff done [by] then?
Two weeks seems to me like it should probably be OK; it will help if we can get some more experienced players helping playtest Morlic's Fighters vs Planets PR and giving feedback on it.

Re: 0.4.7 preparations

Posted: Tue Mar 28, 2017 9:28 am
by Vezzra
My dear friends, please don't hijack this thread for reporting and discussing bugs/issues. The purpose of this thread is to coordinate the release. For feedback on playtesting we have the Play-Testing Feedback subforum. Posting a link to such a feedback thread here is fine, so you can open a thread on the playtesting feedback forum and put a link to it here if you want.

I'm going to split the posts discussing these AI issues with invasion off to a new thread.

EDIT: Done. New thread can be found here.

Re: 0.4.7 preparations

Posted: Wed Mar 29, 2017 9:30 am
by Geoff the Medio
Vezzra wrote:Comments, opinions, feedback welcome. 8)
Suggestions and comments about what to include in the list of key changes for v0.4.7 would be helpful.

Re: 0.4.7 preparations

Posted: Thu Mar 30, 2017 12:32 am
by dbenage-cx
Suggestions and comments about what to include in the list of key changes for v0.4.7
Other than fighters:
BeginnerBonus PR #928
DiplomacyAdditions (both for supply changes and for new diplomacy option) PR #1320

If key changes include non-user facing:
C++11 support
improved font rendering
improved listbox
improved pathfinding
improved app termination
IPv6 support

Other:
Multiplayer "Ready" status
Queue item pause/resume
System circles
Combat log improvements
Tech tree (list view) sort
New galaxy images

Re: 0.4.7 preparations

Posted: Thu Mar 30, 2017 10:56 am
by MatGB
AI improvements, I know we always list this but it's a noticeable jump this time, their ability to keep moving and not stop at every system is particularly different.

And while I haven't gone through it properly, the Pediatric rework and categorization is significant.

Re: 0.4.7 preparations

Posted: Thu Mar 30, 2017 1:58 pm
by Vezzra
Geoff the Medio wrote:Suggestions and comments about what to include in the list of key changes for v0.4.7 would be helpful.
My picks:
  • Fighters/Carriers (obviously... ;))
  • Diplomacy Additions & Allied Supply
  • AI improvements
  • Expansion & Restructuring of the Pedia
Not sure about the following, but maybe also:
  • Multiplayer Lobby seems to have received some attention
  • Ancient Guardians
  • Pausing/Resuming of production items

Re: 0.4.7 preparations

Posted: Thu Mar 30, 2017 2:20 pm
by Geoff the Medio
Vezzra wrote:
  • Ancient Guardians
Elaborate?

Re: 0.4.7 preparations

Posted: Thu Mar 30, 2017 2:31 pm
by Vezzra
Geoff the Medio wrote:
Vezzra wrote:
  • Ancient Guardians
Elaborate?
While the feature itself isn't a "big" one, I expect it to be felt quite strongly by players, as it addresses an issue that has caused a lot of complaints (guardian monsters at chokepoints near home systems making for bad starting positions). So from a players perspective in might be perceived as something "big" or "important".

But that's a very subjective assessment on my part, and might likely be wrong. Which is why I listed it as one of the items I'm unsure about if they should really count as "key change".