With modern tooling, libraries and such, many of supcoms old technical issues can be address. (plus it now works on linux and mac!)
I think I'm close to a like for like version.
And of course the performance its leaps and bounds beyond what supcom could do, in the replies a few down, you can see a 64,000 unit air fight, and I'm pretty sure i can get that going at 60fps plus!
and end game vs the ai, is still 4x realtime, well beyond what supcom could do.
Definitely! Supcoms downfall was the poor threading support and IIRC the in-game AI path finding continuing for destroyed AI units which bogged the game down over time. At least one of them was never fixed properly.
SupCom FaF has a very long list of changes, I'd guess some of the people behind it already have access to source code. IMO only changes that are still missing is making it multi-threaded and maybe fixing some AI deficiencies such as pathing and the like.
The first would be great, but probably very difficult technically. The second would probably cause a holy war in the community, as multiplayer RTS enjoyers glorify fighting against the genre's jank as an expression of skill (admittedly it would completely change all micro).
PvE vs. AI on maps larger than 10x10 slows down to a crawl after ~40minutes. It's not FPS (which remains perfectly smooth), though: it's the simulation speed, influenced by pathfinding, trajectories of every single projectile, managing construction, managing units' scripted movement, and all the AI decision-making. Plus the performance penalty of using Lua (and its GC) for parts of game logic. Apparently, all of that is crammed into a single thread, and by the time you have 1k units, simulating 1 second of in-game time can take 3-4 seconds (or more, obviously, depending on the CPU; but single-core performance stopped improving drastically at some point). It's been a few years since I last checked, but I'd be surprised if anything beyond 20x20 maps would be playable against the AI, even today. And since it's a matter of codebase architecture, fixing it would require a rewrite (someone in this thread is actually attempting it, which could be fun).
I have actually been working on this!
With modern tooling, libraries and such, many of supcoms old technical issues can be address. (plus it now works on linux and mac!)
I think I'm close to a like for like version.
And of course the performance its leaps and bounds beyond what supcom could do, in the replies a few down, you can see a 64,000 unit air fight, and I'm pretty sure i can get that going at 60fps plus!
and end game vs the ai, is still 4x realtime, well beyond what supcom could do.
and of course higher player v player battles
https://forum.faforever.com/topic/10451/supcom-moho-engine-r...
Definitely! Supcoms downfall was the poor threading support and IIRC the in-game AI path finding continuing for destroyed AI units which bogged the game down over time. At least one of them was never fixed properly.
I've played the LOUD mod for years, seems to solve a bunch of performance problems.
https://www.moddb.com/mods/loud-ai-supreme-commander-forged-...
Awesome thanks for sharing!
SupCom FaF has a very long list of changes, I'd guess some of the people behind it already have access to source code. IMO only changes that are still missing is making it multi-threaded and maybe fixing some AI deficiencies such as pathing and the like.
The first would be great, but probably very difficult technically. The second would probably cause a holy war in the community, as multiplayer RTS enjoyers glorify fighting against the genre's jank as an expression of skill (admittedly it would completely change all micro).
man i played so much supcom every time we played at like 2fps in the end haha
The game still looks great, what's stopping you from playing it?
PvE vs. AI on maps larger than 10x10 slows down to a crawl after ~40minutes. It's not FPS (which remains perfectly smooth), though: it's the simulation speed, influenced by pathfinding, trajectories of every single projectile, managing construction, managing units' scripted movement, and all the AI decision-making. Plus the performance penalty of using Lua (and its GC) for parts of game logic. Apparently, all of that is crammed into a single thread, and by the time you have 1k units, simulating 1 second of in-game time can take 3-4 seconds (or more, obviously, depending on the CPU; but single-core performance stopped improving drastically at some point). It's been a few years since I last checked, but I'd be surprised if anything beyond 20x20 maps would be playable against the AI, even today. And since it's a matter of codebase architecture, fixing it would require a rewrite (someone in this thread is actually attempting it, which could be fun).
It has some major technical flaws which hold it back even on modern hardware. Something I’m sure a model can fix.