0.5 release

Discussion about the project in general, organization, website, or any other details that aren't directly about the game.
Message
Author
Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: 0.5 release

#91 Post by Ophiuchus »

Geoff the Medio wrote: Mon Jan 30, 2023 8:48 pm
Ophiuchus wrote: Mon Jan 30, 2023 2:45 pm
geoff would a one turn preview be possible for 0.5 ?
Not sure exactly what you mean, but probably not.
i am not sure how the next turn predictions for human players are currently calculated.

the suggestion was that to get the target predictions by actually simulating a single turn in the client. make a copy of the universe, do turn processing with that knowledge (also performing orders like invasion,colonisation etc), report back to the player the resulting values. Then one would have a good prediction of next turn influence (because it can actually count how many colonies exist).

so if that would be already mostly in place, maybe just some small parts need to be added (e.g. performing orders).
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: 0.5 release

#92 Post by Oberlus »

Ophiuchus wrote: Tue Jan 31, 2023 8:11 am the suggestion was that to get the target predictions by actually simulating a single turn in the client. make a copy of the universe, do turn processing with that knowledge (also performing orders like invasion,colonisation etc), report back to the player the resulting values. Then one would have a good prediction of next turn influence (because it can actually count how many colonies exist).
That sounds pretty heavy on computations. Huge lags in huge galaxies maybe.
I think all this is unnecessary.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.5 release

#93 Post by Geoff the Medio »

In one sense, the client already does simulate the next turn every time something changes client side due to orders like planet focus, policy adoption, production queue, etc. This causes updakes like effects and meter values being recalculated, production queues being re-simulated, and fleet movement being repredicted based on the current local gamestate, and probably / agruably a few other similar things.

In another sense, it's not currently possible to copy the full gamestate and running a server turn update on it without impacting the client's main gamestate. The main problems are global state accessors from various bits of game logic code that make it not possible to isolate a copy of the gamestate from the main client gamestate. Making that possible is a long-term ongoing project, and not at all feasible for v0.5.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: 0.5 release

#94 Post by Ophiuchus »

Oberlus wrote: Tue Jan 31, 2023 8:20 am That sounds pretty heavy on computations. Huge lags in huge galaxies maybe.
done right you will not have your UI lag (because of working in the background in a different thread). just the results would maybe show after a while.
there is also a lot of caching going on, so there is a lot reuse. maybe there could be a lot more.

a simpler version would be a "precalculate button", where lag would be expected by the player

the huge galaxy problem could only occur if the client knows about the huge galaxy (so late game).

generally it would be good to increase turn performance, that would help everybody.

i like geoffs current implementation for colonizing via colony ships. but all these have to be done per content. the one-turn-preview is quite generic for all kinds of content.
the precalculate would be also a good debugging tool. it could show mismatches between the fast computation and the thorough computation.

for feasability etc see geoffs post
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
LienRag
Cosmic Dragon
Posts: 2148
Joined: Fri May 17, 2019 5:03 pm

Re: 0.5 release

#95 Post by LienRag »

Oberlus wrote: Mon Jan 30, 2023 11:15 am The scenario that LR depicted is close to unbelievable:
Yep, it's the fact that it's indeed close to unbelievable that made me having a hard time to believe it even when it was unfolding in front of me, and makes me believe that similar things happening to not-yet-hooked-on players will discourage them and have them believe that this game makes no sense at all.

I guess that with an extra-game spreadsheet it's actually manageable.
Do we want to limit FreeOrion to people who want to play with an extra-game spreadsheet open in parallel ?

o01eg
Programmer
Posts: 2004
Joined: Sat Dec 10, 2011 5:46 am

Re: 0.5 release

#96 Post by o01eg »

Ophiuchus wrote: Mon Jan 30, 2023 2:45 pm
o01eg wrote: Mon Jan 30, 2023 1:28 pm I suppose including colony orders to calculation is fine as the player could order/undo order to see if he will manage result value.
yes, i think that would be good UI.
Additionally it better to include enqueued colony on outpost so player could do the same.
LienRag wrote: Wed Feb 01, 2023 4:49 pm Do we want to limit FreeOrion to people who want to play with an extra-game spreadsheet open in parallel ?
Does any strategy game provide player with multiple turn predictions?
Gentoo Linux x64, gcc-11.2, boost-1.78.0
Ubuntu Server 22.04 x64, gcc-12, boost-1.74.0
Welcome to the slow multiplayer game at freeorion-lt.dedyn.io.Version 2024-03-15.b3de094.
Donations're welcome:BTC:bc1q007qldm6eppqcukewtfkfcj0naut9njj7audnm

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: 0.5 release

#97 Post by Oberlus »

LienRag wrote: Wed Feb 01, 2023 4:49 pm
Oberlus wrote: Mon Jan 30, 2023 11:15 am The scenario that LR depicted is close to unbelievable:
Yep, it's the fact that it's indeed close to unbelievable
What I meant is that you were playing wrong, that you had plenty of turns to do it better (one doesn't invade 25 planets on a whim, but if you do, you already have some other 100 planets to milk extra influence).

In other words: LienRag, learn to play and don't blame the game for your lack of "careful long-term planning" (you always have the mouth full of that sentence but it is all bullshit in your case).

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.5 release

#98 Post by Geoff the Medio »

Oberlus wrote: Thu Feb 02, 2023 9:02 am What I meant is that you were playing wrong, that you had plenty of turns to do it better (one doesn't invade 25 planets on a whim, but if you do, you already have some other 100 planets to milk extra influence).
People can play (or otherwise run) FreeOrion more or less however they want... There no wrong way to do so as long as they get something like what they want out of it.

wobbly
Cosmic Dragon
Posts: 1879
Joined: Thu Oct 10, 2013 6:48 pm

Re: 0.5 release

#99 Post by wobbly »

I'm not sure how much of this issue is about colony IP costs being unpredictable and how much it's about people prefering less snowballing in colony influence costs.

If it's the 2nd and some players want to be able to colonize everything an option for influence cost scaling (similar to tech and production cost options) could be a solution.

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: 0.5 release

#100 Post by Oberlus »

wobbly wrote: Fri Feb 03, 2023 7:23 am I'm not sure how much of this issue is about colony IP costs being unpredictable and how much it's about people prefering less snowballing in colony influence costs.

If it's the 2nd and some players want to be able to colonize everything an option for influence cost scaling (similar to tech and production cost options) could be a solution.
I'm almost sure it's about the 2nd. And your suggestion is great.
Geoff the Medio wrote: Thu Feb 02, 2023 2:07 pm
Oberlus wrote: Thu Feb 02, 2023 9:02 am What I meant is that you were playing wrong, that you had plenty of turns to do it better (one doesn't invade 25 planets on a whim, but if you do, you already have some other 100 planets to milk extra influence).
People can play (or otherwise run) FreeOrion more or less however they want... There no wrong way to do so as long as they get something like what they want out of it.
If you mean they should be able to tailor the difficulty, cost factors, etc., absolutely. But I meant that the game should not be "improved" to allow for suboptimal gameplay (which is what was referring to with "playing wrong") to get optimal results.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.5 release

#101 Post by Geoff the Medio »

Oberlus wrote: Fri Feb 03, 2023 8:24 amIf you mean they should be able to tailor the difficulty, cost factors, etc., absolutely. But I meant that the game should not be "improved" to allow for suboptimal gameplay (which is what was referring to with "playing wrong") to get optimal results.
I mean nobody gets to decide what is (sub)optimal gameplay for someone else. If someone enjoys colonizing 50 planets on one turn with or without considering the game mechanics consequences of doing that, then they are welcome to do so. How much the content or UI is designed to support doing something doesn't change that, and they can complain if it doesn't work how they'd like. That doesn't mean they can expect any extra effort to be made to support their preferred play style, though.

Adding balance options is a reasonable request or suggestion, though. I'm hestitant to add such now for v0.5, though.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: 0.5 release

#102 Post by Vezzra »

LienRag wrote: Wed Feb 01, 2023 4:49 pmI guess that with an extra-game spreadsheet it's actually manageable.
Do we want to limit FreeOrion to people who want to play with an extra-game spreadsheet open in parallel ?
The fundamental issue here is, once a game reaches a certain "level" of complexity (which 4X games usually do), avoiding an extra game spreadsheet if you really want to play optimally is nearly impossible. Even big and well designed 4X games as Stellaris can't completely avoid that.

Everyone who plays 4X games knows that they come with a certain learning curve. Some steeper, some easier, but no 4X game worth their salt can be mastered by the 2nd or 3rd playthrough (unless you are unusually good at 4X games, that is ;)). So when newbies to FO mess up their IP management in their first couple of games, before they get the hang of it, I wouldn't consider that too much of a problem.

It takes time and patience to master a good 4X game, and become good at it. That's how it's supposed to be. ;)

o01eg
Programmer
Posts: 2004
Joined: Sat Dec 10, 2011 5:46 am

Re: 0.5 release

#103 Post by o01eg »

LienRag wrote: Wed Feb 01, 2023 4:49 pm Do we want to limit FreeOrion to people who want to play with an extra-game spreadsheet open in parallel ?
Small spreadsheet off-topic in EVE Online: https://store.steampowered.com/news/col ... 8565770705
Gentoo Linux x64, gcc-11.2, boost-1.78.0
Ubuntu Server 22.04 x64, gcc-12, boost-1.74.0
Welcome to the slow multiplayer game at freeorion-lt.dedyn.io.Version 2024-03-15.b3de094.
Donations're welcome:BTC:bc1q007qldm6eppqcukewtfkfcj0naut9njj7audnm

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: 0.5 release

#104 Post by Vezzra »

Major status update: It looks like we're finally close to be ready for the first release candidate. The remaining issues and PRs on the 0.5 milestone all are not considered release blocking (IIRC at least).

Which means I'm considering and suggest the next weekly test builds to be RC1.

If there are any objections to this, if anyone thinks that there are things (issues, bugs etc.) that should be addressed before RC1 and should delay it, please post a respective request/reply here in this thread.

If possible, the two currently remaining PRs on the release milestone (#4387 and #4388) should be merged and cherry picked to the release branch before RC1 (so, this weekend).

For further procedure, issues and PRs that should be considered release blocking must be tagged with the "priority:high" label, to distinguish them from issues/PRs that would be nice to get into 0.5, but are not strictly necessary.

@Geoff: Do you think you can update the changelog once more? There's been a lot of stuff cherry-picked into the release branch since you compiled the changelog for 0.5, I guess there are at least some things that should be added. If necessary I can delay RC1 until Thursday or Friday if Tuesday is a bit too close?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.5 release

#105 Post by Geoff the Medio »

Vezzra wrote: Fri Feb 10, 2023 11:06 am@Geoff: Do you think you can update the changelog once more? There's been a lot of stuff cherry-picked into the release branch since you compiled the changelog for 0.5, I guess there are at least some things that should be added.
Was mostly not notable for the log... https://github.com/freeorion/freeorion/ ... d9cb012aad

Post Reply