TripleA 3.0 - Design Proposal & Discussion
-
@ubernaut imo, it was the roll-out of the new lobby and all the bugs when the new lobby became the only lobby.
-
I'd like to see the task/to do lists broken up a bit more thoroughly and actionably. Not that it's vital or anything.
It'd be nice to see what exactly needs to be done for 3.0 and 2.6, and what's being waited upon. What the new architecture is going to be in terms of package structure and calls and so forth. In particular I'm not sure how separable the various parts of the proposal are, and how much has to be covered as part of one big unified rework.
I also hope we manage to fix the issue with global and tww at some point, even if only as an eventual result of replacing the bots with the new network code.
-
@zlefin 2.6: https://github.com/triplea-game/triplea/projects/25
The biggest item is thorough testing of the 2.6 engine.
-
@alkexr said in TripleA 3.0 - Design Proposal & Discussion:
With 3D graphics there would be no such thing as a mapmaker, I'm afraid.
There are lots of map makers out there.
Choosing the right basis for our own work may become a smaller project of its own. Some are a joy to play with, e.g. Mapgen 4. (A much simpler map maker would of course do.)Also, what do you mean, look at the list of top rated games on Steam, most of them are 2D.
I did. The most popular games are 3D now.(Apart from classics and cookie maker). For TripleA 3 I think about something like the total war strategy map - or the game of thrones board game online. Freeciv web is unsing 3d units.
-
@rainova i think you are oversimplifying the situation i don't think personally, there's any real benefit to making this game 3D and it would come with a shit ton of overhead. the game is a 2D game, 3D would only be a graphical change and would arguably make the boards harder to see. also, the mapmaker doesn't just make maps its also is used to wire them up into playable map files. we have a lot of enhancements that are much more applicable to this game than just making it another 3D game. personally i'd would also argue there a ton of 2D grpahical enhancements that would have better impact on user experince.
-
Nice. On that project sublist, there doesn't seem to be issues attached to it in a linked way that can be edited; ie if I were try to working on one, how do we note that and avoid duplicative effort? Do we make new github issues for work on of those and put our details in there?
Is there already a thread or something covering the direct map download details? I might try working on that as something small and easy. But someone may've already started on it, and I have a feeling there's been some discussion fo it already somewhere, but I'm not sure where.
-
@lafayette I would be down and I've sometimes asked to play my 270BC Wars in the 2.6 lobby, but no-one has ever wanted to so far.
-
@zlefin Either open an issue to discuss & coordinate or open a forum thread. Either one would be a good place to outline the design and approach.
In the 'triple-dot' menu on the project cards, there is a 'convert-to-issue' option FWIW.
-
for me I'm not seeing an option for convert to issue on the project cards. The triple dot menu only has copy card link.
I probably don't have something setup right on github.
At any rate I'll see about opening up an issue soon.
-
(3) Text-based save games. Rather than serializing java objects we would instead store text based game state changes.
This would fix a lot of the problems [...]
Second, a text based save would be far faster to serialize and we could get a 10x increase in AI performance by doing this. Most of the time AI spends 'thinking' is doing battle calcs, the long pull on that is serializing game save data. If we convert that to text it'll be faster, but also the data model would be much improved and perhaps that step could be cut out completely.Could the second reason be eliminated: Why does the battle calculator make a copy of the game save data in the first place?
(The first reason for text-based save games still being strong enough to make it an important issue to move forward.) -
@rainova said in TripleA 3.0 - Design Proposal & Discussion:
(3) Text-based save games. Rather than serializing java objects we would instead store text based game state changes.
This would fix a lot of the problems [...]
Second, a text based save would be far faster to serialize and we could get a 10x increase in AI performance by doing this. Most of the time AI spends 'thinking' is doing battle calcs, the long pull on that is serializing game save data. If we convert that to text it'll be faster, but also the data model would be much improved and perhaps that step could be cut out completely.Could the second reason be eliminated: Why does the battle calculator make a copy of the game save data in the first place?
You cannot rely on the game file (xml) only, because how units behave and what units every player has access to may change in the course of the game (because of technology and triggers). The point is that only what matters should be loaded, not everything. My best guess is that nobody bothered making it efficient because, as long as you only play the basic games or such, it is virtually timeless anyway (unless maybe in case you play for hundreds of rounds, which virtually never happens). I believe that currently the game which suffers the most from it is "270BC Wars": I have to wait about 6 seconds for the battle-calculator to become usable.
-
To be continued in Topic 3077 - fast-battle-calculator