Caveman 3,0 is a solo gamedev project. open world. custom engine. 65 types of weapons, 50 types of monsters, 300+ types of objects, 1000+ actions, a 2500x2500 mile game world, 10,000 dungeons.

i have 26 years experience as a gamedev.

time spent so far: about 10,500 man-hours.

total estimated time to complete: about 12,000 man hours.

i try to work 18 hours a day, 6 1/2 days a week. but that pace can usually only be maintained for a couple years before burnout starts to creep in. These days i'm doing good if i manage 80 hours per week on it.

engines were still expensive when i started on the game. next time i'd use unity etc.

game is bigger and taking longer than expected. next time, i'd probably not do it as being "too big". IE the larger the time investment, the lower the chance of it selling well enough to make it worth the time.

can it be done? yes. more importantly: should it be done? probably not.

"A game that has no main story missions/quests and is progressed as the player does what

they want to do, everything un-lockable such as abilities, weapons and customization is locked behind an
experience system. Which is basically earned through doing the activities you want."

Caveman does all this already - i forgot to mention it tracks experience in 45 different skills.

Impressive project Norman!

It takes a caveman to create one. :)

meaning no story or setpieces, no manmade structures, and other similar constraints that keep the scale down a little bit.

no hard coded storyline quests yet, but i'll probably add them. thy only thing i don't like about them is the lack of replayability. the paleo-world simulation is like a blank page just waiting for story to be told on it.

huts, lean-to's, temporary shelters, storage pits, and landmarks are the only man made structures in the game. so only 7 building models required! <g>.

massive re-use is made of assets (meshes, textures, models, animations). the result is a world similar to Skyrim or The SIMs, executed with just 200 meshes, 386 textures, 171 models, and 165 animations (so far). By contrast, a single dungeon level in Skyrim probably has many more meshes, and as many textures. Skyrim has fewer monster types, but more people models, so many more models overalll, at least twice as many. The SIMs v1.0 shipped with over 1000 animations. Although Caveman has over 1000 actions (A la The SIMS), i'll probably be able to get away with about 40 animations for all of them by changing things like item in hand and items on ground nearby.

The underlying world map is procedurally generated, and terrain chunks and collision maps are generated on the fly as needed from the world map, so no need to model an entire world like in Skyrim.

Dungeon (cavern) maps are also procedurally generated. so no need to model "Bleak Falls Barrow" and the 200+ other dungeons, like in Skyrim.

At the end of the day, its content creation that's the real killer. And for that there's only a few ways to go:

1. an army of artisits

2. procedural content generation

3. massive reuse of assets

4. license available 3rd party assets for use (i figured about $10K to license just the skinned mesh assets for Caveman).

5. use of PD/free/existing assets

6. do it yourself and just take forever to make the game assets for a large title.

negazina,

here's what i would do:

based on your current skill level and free time, figure out what you can get done in two weeks. then give yourself a month to do it. engineers are always overly optimistic about time estimates. take your best guess, then double it. you won't be far from wrong. then go ahead and do it. doesn't matter what it is or how basic. drawing a mesh in unity, or just clearing the screen in DX or OGL. the important thing is, its the first step. then figure out what's the next step. if you aren't sure, post a question here on gamedev. continue to develop modules/features/capabilities in an agile/sprint fashion. soon you'll have enough parts to start prototyping and demoing things.

but begin with the end in mind. worry about the problems you haven't figured out how to solve yet. that's ALL that matters in game development. once you know how to do something, the rest is all grunt work. so in your case don't start on the game until you figure out the asset creation problem and ALL other issues related to its large scope and your small dev team size. if you can't AHEAD OF TIME solve all the problems that get you from project start to gone gold, DON'T START AT ALL! IE don't even try to build the game until you've figured out to a reasonable level of detail exactly how it _ALL_ can be done. emphasis on ALL. you need every piece of the puzzle, every link in the chain, every potential issue anticipated and solved before sitting down to the PC. after all, one doesn't go to war without at least some idea of how to win. and win not just the first battle, but ALL battles necessary for final victory.

As long as you consider the time it will take, sure, it is possible. Just remember that in 10 years, technology will have advanced a great deal. Consider the time it will take to learn those skills and do them well (It has taken me at least 5 years to get to the state I am at now, which is a novice).