3 ms·

I'm not so sure. I think it should be possible to get something done with 2-3 people in 3 years, not 10. But before I start a serious attempt I'd certainly want

by erikb 11y ago

I'm not so sure. I think it should be possible to get something done with 2-3 people in 3 years, not 10. But before I start a serious attempt I'd certainly want to learn basic things first, like how to make a fighting system. These things are only hard and time consuming when you figure them out the first time. The second or third time you are much faster. One example I have from own experience is that one topic like that for the program I wrote in my thesis: Took me 3-6 months depending on how you count start and end date. But last week I had to do the same thing for another project and it only took me half a day. I didn't need to google anything, I didn't need to compare different variations. I knew what worked, what were the disadvantages of the most common options, choose one in 5 seconds and got it implemented. I didn't run into any of the things that took days to debug in the first attempt, because I knew where the problem areas were. Therefore I would assume that while the first attempt of an RPG for a group of noobs might take a year or three for a prototype, a small team of experienced developers can put together a basic prototype in a few weeks.

louhike 11y ago
The problem is that a prototype is not equivalent to a real fighting system you will put in a game. You will need a lot of testing so the game is challenging is fun. And it must be challenging and fun during all the game (30+ hours). EDIT: To give an example, see a website. A small website to show some prototype will be quite quick to do. But when you need to do a website used by hundreds to thousands persons with a lot of use case, it gets far more complex.
- erikb 11y ago
  I'd still disagree but I can see your point. It's not just the basics. A real system needs some things you can't learn from a book.
gjig 11y ago
I agree with your general assessment re: experience, but I wish people would stop talking about these sorts of things in years and months. I understand why it's difficult to express it in hours, but even a very rough estimate of total hours spent is a far, far more useful metric than "years". My personal experience with studying Japanese has been that there's roughly two camps of people: those who become fluent-ish in 1-2 years, and those who know ~100 kanji after 3 years (1-2k for fluency). The primary difference between these two camps is that the former studies 6-8 hours a day, and the latter has Japanese classes thrice a week, that's it. Speaking in hundreds, thousands or tens of thousands of hours gives you a much better idea of what's actually necessary to make the magic happen, and makes it easier to judge how well you're doing: is making barely any progress in 1 week a bad thing, or a really bad thing? That depends on whether a week translates to 3 hours or 50 hours.
- erikb 11y ago
  I agree! I'd also wonder why so little information is given in hours. We have that likely horizon of mastery at 10k hours and we have a soft limit of "usability" at about 3k hours. But maybe the thing is, nobody really counts. Like, I can tell you when I started to study my major, which led me to spend a lot of time on programming, taking it serious for the first time. Now I work full time as a software developer but do I spend 30 hours, or 10 hours, or 50 hours programming a week? Can't tell. There are lot of things involved, no matter if you're a student or a full time worker. Not all of them count. And going back to your Japanese example. Sometimes the drunken two hours you've spent in a Karaoke bar trying to impress local girls was worth much more than the previous 10 hours study session because these two hours were marked as important memories by your brain and the previous ten hours were boring. It's really hard to calculate, right? Also that study which came up with that 10k hours for mastery stated that many people very much overestimate hours spent, which from feeling I would agree. Like, since I work full time I don't believe any more that people really spend 40+ hours working. It's closer to 20 or 30 hours depending on the person. The rest is spent smoking, chatting, fighting via email, reading HN. And from own experience I also know that 12 hours of doing nothing can be just as tiring as working hard. ;)
  - IronTowerStudio 11y ago
    We talked about years because that's what most people look at. I gave this simple breakdown before: "Let's say tomorrow an AoD-like game is announced by a "real" studio. I'd expect at least 3 years in development and at least 15 people working on it full-time. If you expect less, please explain. So, 15 people x 40 hours a week (a bit on the low side, I know) x 52 weeks x 3 years = 93,600 man-hours vs 4 x 15 hours a week x 52 x 10 = 31,200 man-hours (roughly a third). Some facts: Civ 5 has been in development for 3 years, 52 people team = 324,480 man-hours. Fallout 3 was in development for 4 years. Gothic - more than 4 years. NWN - around 5 years, Witcher - more than 5 years, fairly large team, but the quality is undeniable. Arcanum 2.5 years. KOTOR 3 years. Fallout 3 years (In 1994 Steve Jackson mentioned that Interplay was working on a post-apoc game using GURPS)."

by erikb 11y ago

louhike 11y ago

The problem is that a prototype is not equivalent to a real fighting system you will put in a game. You will need a lot of testing so the game is challenging is fun. And it must be challenging and fun during all the game (30+ hours). EDIT: To give an example, see a website. A small website to show some prototype will be quite quick to do. But when you need to do a website used by hundreds to thousands persons with a lot of use case, it gets far more complex.

erikb 11y ago
I'd still disagree but I can see your point. It's not just the basics. A real system needs some things you can't learn from a book.

gjig 11y ago

I agree with your general assessment re: experience, but I wish people would stop talking about these sorts of things in years and months. I understand why it's difficult to express it in hours, but even a very rough estimate of total hours spent is a far, far more useful metric than "years". My personal experience with studying Japanese has been that there's roughly two camps of people: those who become fluent-ish in 1-2 years, and those who know ~100 kanji after 3 years (1-2k for fluency). The primary difference between these two camps is that the former studies 6-8 hours a day, and the latter has Japanese classes thrice a week, that's it. Speaking in hundreds, thousands or tens of thousands of hours gives you a much better idea of what's actually necessary to make the magic happen, and makes it easier to judge how well you're doing: is making barely any progress in 1 week a bad thing, or a really bad thing? That depends on whether a week translates to 3 hours or 50 hours.

erikb 11y ago
I agree! I'd also wonder why so little information is given in hours. We have that likely horizon of mastery at 10k hours and we have a soft limit of "usability" at about 3k hours. But maybe the thing is, nobody really counts. Like, I can tell you when I started to study my major, which led me to spend a lot of time on programming, taking it serious for the first time. Now I work full time as a software developer but do I spend 30 hours, or 10 hours, or 50 hours programming a week? Can't tell. There are lot of things involved, no matter if you're a student or a full time worker. Not all of them count. And going back to your Japanese example. Sometimes the drunken two hours you've spent in a Karaoke bar trying to impress local girls was worth much more than the previous 10 hours study session because these two hours were marked as important memories by your brain and the previous ten hours were boring. It's really hard to calculate, right? Also that study which came up with that 10k hours for mastery stated that many people very much overestimate hours spent, which from feeling I would agree. Like, since I work full time I don't believe any more that people really spend 40+ hours working. It's closer to 20 or 30 hours depending on the person. The rest is spent smoking, chatting, fighting via email, reading HN. And from own experience I also know that 12 hours of doing nothing can be just as tiring as working hard. ;)
- IronTowerStudio 11y ago
  We talked about years because that's what most people look at. I gave this simple breakdown before: "Let's say tomorrow an AoD-like game is announced by a "real" studio. I'd expect at least 3 years in development and at least 15 people working on it full-time. If you expect less, please explain. So, 15 people x 40 hours a week (a bit on the low side, I know) x 52 weeks x 3 years = 93,600 man-hours vs 4 x 15 hours a week x 52 x 10 = 31,200 man-hours (roughly a third). Some facts: Civ 5 has been in development for 3 years, 52 people team = 324,480 man-hours. Fallout 3 was in development for 4 years. Gothic - more than 4 years. NWN - around 5 years, Witcher - more than 5 years, fairly large team, but the quality is undeniable. Arcanum 2.5 years. KOTOR 3 years. Fallout 3 years (In 1994 Steve Jackson mentioned that Interplay was working on a post-apoc game using GURPS)."