5 ms·

> ... the two biggest problems in game dev are building what people want and R&D. What's even harder again about point one, building something people want, is

by movedx 9y ago

> ... the two biggest problems in game dev are building what people want and R&D. What's even harder again about point one, building something people want, is this is a purely creative industry so it's REALLY hard to know what people want. Making a new coffee machine would involve simple market research to understand what people want, but painting a picture is a different story entirely!

timavr 9y ago
Yes and No. When Blizzard makes Overwatch or HotS, they have a good initial idea that it is going to be popular because games of similar gameplay already exist and popular. On top of that when they play internally and if it is fun, they can be pretty sure that it is going to work. Obviously, if someone like Jonathan Blow is making a completely new game like Witness, then all bets are off.It is basically up to him and people who he trusts with feedback to determine if this is the game worth making. Making Call of Duty version 21 at the same time as close to risk-free as it comes.
emmelaich 9y ago
>> ... the two biggest problems in game dev are building what people want and R&D. Surely that's where getting a production quality level one helps? You really need to know that as early as possible. You must resist the Siren's Song of rough prototypes. They could be great gameplay but awful to look at. Or vice versa.
- movedx 9y ago
  Yep, and I completely agree. I like producing an MVP, but even what the MVP offers should be production ready. I think people read "Minimal" in MVP and believe it means, "No monitoring, logging, error checking, or stability" which of course it does not :-)
- timavr 9y ago
  In terms of production quality, sometimes it takes years of iteration on art side and tech to get the look. You want to nail it down early, but in real life concept artist does the image, artists model and texture the scene and then art director comes back and says, those bricks don't look good. Engine guys say, if we put more polys to make them look good, it is going to kill performance. Then someone says, look we can write this shader using this new GPU feature which is going to make everything awesome, but it might not work. And in any game production, there are 1000s of cases like that, where the thing you want does not exist or you don't have people who have the knowledge. So you just roll into production hoping shader works out and that you are going to hire people who are going to solve XYZ.

by movedx 9y ago

timavr 9y ago

Yes and No. When Blizzard makes Overwatch or HotS, they have a good initial idea that it is going to be popular because games of similar gameplay already exist and popular. On top of that when they play internally and if it is fun, they can be pretty sure that it is going to work. Obviously, if someone like Jonathan Blow is making a completely new game like Witness, then all bets are off.It is basically up to him and people who he trusts with feedback to determine if this is the game worth making. Making Call of Duty version 21 at the same time as close to risk-free as it comes.

emmelaich 9y ago

>> ... the two biggest problems in game dev are building what people want and R&D. Surely that's where getting a production quality level one helps? You really need to know that as early as possible. You must resist the Siren's Song of rough prototypes. They could be great gameplay but awful to look at. Or vice versa.

movedx 9y ago
Yep, and I completely agree. I like producing an MVP, but even what the MVP offers should be production ready. I think people read "Minimal" in MVP and believe it means, "No monitoring, logging, error checking, or stability" which of course it does not :-)
timavr 9y ago
In terms of production quality, sometimes it takes years of iteration on art side and tech to get the look. You want to nail it down early, but in real life concept artist does the image, artists model and texture the scene and then art director comes back and says, those bricks don't look good. Engine guys say, if we put more polys to make them look good, it is going to kill performance. Then someone says, look we can write this shader using this new GPU feature which is going to make everything awesome, but it might not work. And in any game production, there are 1000s of cases like that, where the thing you want does not exist or you don't have people who have the knowledge. So you just roll into production hoping shader works out and that you are going to hire people who are going to solve XYZ.