If I had to write a headline for this project for my tombstone (my partner is at the limits of tolerance for my AI spending), it'd probably be:
Exited Startup Founder Spends GDP of Small Nation to Rebuild 1990's Game with AI that started his career
In 2000 a games studio Cavedog Entertainment shut, stretched thin in a classic story - they'd released Total Annihilation in 1997, its expansions in 1998 - and then a slightly less loved sequel, Total Annihilation: Kingdoms in 1999/2000 before going under, as they buckled under the weight of a sequel that didn't hit targets and 3 other concurrent game ideas in execution that couldn't be carried. Imagine Ion Storm, but without the controversial marketing and a smaller budget.
I was an angry 15 year old kid playing a game "Total Annihilation". I went to GameSpy's sites and found PlanetAnnihilation.com - but it was basically dead quiet, so I wrote some flames on the forum saying the game has a huge player base, why is nobody updating the site. Roll on 3 days, and I'm introduced to Frank "DMFrank" Rogan, head of the planet network. He offers me the choice of taking on the site - see if I can do better. What I didn't know at time was that Frank was ex-Cavedog Entertainment. For a few years I ran the site (up to ~2006 or so) and worked with modders and the online community, particularly projects trying to rebuild the shuttered Boneyards.net gaming service. I was 15 and learning how to write editorial, build web-sites and apps for content etc - it was hugely formative.
In 2014 I tried to reverse engineer it all, once and for all, again - https://www.tauniverse.com/forum/showthread.php?t=44061 - made huge progress - but after having recently moved to Australia, work ramping up and getting a family going - it just got damn hard so the project died.
Roll on the advent of AI, I got back into it recently and managed to get things working - including an end to end playable game experience. I was able to fund the huge cost because I'd sold out my business to a US firm (Can google my details from the info I've put out there).
What I ended up with Basically the OpenRA equivalent for Total Annihilation:
Its all open source, 400,000-ish lines of C++ (and I don't, lets be honest, really know modern C++). I've probably put 30K+ (USD) into tokens, and another 20K+ into hardware, and more just to get this built.
I've done a press release page I'm sharing with gaming media - https://press.coreprime.net/ - wire compatible netplay with the original game is already working cross-platform.
Not complete yet, but now you can play it on a Mac, Linux, etc. I'm very happy with it. I've also got crossplay working between this and the original game, which is something I feel quite proud of - but was an absolute nightmare.
The whole thing stemmed out of trying to rebuild a dead online service by guessing/reverse engineering the client to design a new server, sight unseen - but then I switched to MacOS life - and then running a 1997 game in Parallels sucks - so I was "well, I'd go faster if I ported it to MacOS...." and then the rabbit hole opened.
Wow, that's commitment. You must really love this game!
I didn't realise quite how expensive tokens were until last week. I got $250 of Claude tokens for free and burnt through them in less than a couple of hours with just one session.
I've been tokenmaxxing across Claude (multiple subs, but I've now burned all of them - and also exhausted, for all of them, the discounted "Buy Credits" from 40% to 25% to 20%....). I used Astra, and it lasted less than hour before it was dead for a month - and Grok whilst did solid work, struggled with cohesive architecture.
DeepSeek, well, I've got 400+ dollars left in credit I will donate to my enemies, to slow their work - because it did nothing good for me in the scheme of things.
If I had to write a headline for this project for my tombstone (my partner is at the limits of tolerance for my AI spending), it'd probably be:
Exited Startup Founder Spends GDP of Small Nation to Rebuild 1990's Game with AI that started his career
In 2000 a games studio Cavedog Entertainment shut, stretched thin in a classic story - they'd released Total Annihilation in 1997, its expansions in 1998 - and then a slightly less loved sequel, Total Annihilation: Kingdoms in 1999/2000 before going under, as they buckled under the weight of a sequel that didn't hit targets and 3 other concurrent game ideas in execution that couldn't be carried. Imagine Ion Storm, but without the controversial marketing and a smaller budget.
I was an angry 15 year old kid playing a game "Total Annihilation". I went to GameSpy's sites and found PlanetAnnihilation.com - but it was basically dead quiet, so I wrote some flames on the forum saying the game has a huge player base, why is nobody updating the site. Roll on 3 days, and I'm introduced to Frank "DMFrank" Rogan, head of the planet network. He offers me the choice of taking on the site - see if I can do better. What I didn't know at time was that Frank was ex-Cavedog Entertainment. For a few years I ran the site (up to ~2006 or so) and worked with modders and the online community, particularly projects trying to rebuild the shuttered Boneyards.net gaming service. I was 15 and learning how to write editorial, build web-sites and apps for content etc - it was hugely formative.
In 2014 I tried to reverse engineer it all, once and for all, again - https://www.tauniverse.com/forum/showthread.php?t=44061 - made huge progress - but after having recently moved to Australia, work ramping up and getting a family going - it just got damn hard so the project died.
Roll on the advent of AI, I got back into it recently and managed to get things working - including an end to end playable game experience. I was able to fund the huge cost because I'd sold out my business to a US firm (Can google my details from the info I've put out there).
What I ended up with Basically the OpenRA equivalent for Total Annihilation:
https://github.com/open-annihilation/open-annihilation
Its all open source, 400,000-ish lines of C++ (and I don't, lets be honest, really know modern C++). I've probably put 30K+ (USD) into tokens, and another 20K+ into hardware, and more just to get this built.
I've done a press release page I'm sharing with gaming media - https://press.coreprime.net/ - wire compatible netplay with the original game is already working cross-platform.
It was 29 years to the day of the games release.
I did love that game, still got my copy sitting around in the shed somewhere.
Cool project, well done for having completed it!
Not complete yet, but now you can play it on a Mac, Linux, etc. I'm very happy with it. I've also got crossplay working between this and the original game, which is something I feel quite proud of - but was an absolute nightmare.
The whole thing stemmed out of trying to rebuild a dead online service by guessing/reverse engineering the client to design a new server, sight unseen - but then I switched to MacOS life - and then running a 1997 game in Parallels sucks - so I was "well, I'd go faster if I ported it to MacOS...." and then the rabbit hole opened.
Wow, that's commitment. You must really love this game!
I didn't realise quite how expensive tokens were until last week. I got $250 of Claude tokens for free and burnt through them in less than a couple of hours with just one session.
I've been tokenmaxxing across Claude (multiple subs, but I've now burned all of them - and also exhausted, for all of them, the discounted "Buy Credits" from 40% to 25% to 20%....). I used Astra, and it lasted less than hour before it was dead for a month - and Grok whilst did solid work, struggled with cohesive architecture.
DeepSeek, well, I've got 400+ dollars left in credit I will donate to my enemies, to slow their work - because it did nothing good for me in the scheme of things.