- Mar 25, 2008
- 437
- 70
Before I start, I would like to say that this post is not about me promoting my map or bragging about it. In all honesty, I think my map is mediocre at best, and that is why I wanted to type this up. I want you guys to learn from my mistakes. If you don't want to read all of this, just scroll down and read the bold text, it is my mistakes that I think are very important to not make.
I started cp_timber in late September/early October. I knew then that it was going to be my senior project for high school. All the seniors at my school have to do a senior project, with a total of 20 hours recorded. I knew that I would have way more than 20 hours, but I decided to do it anyway. I wanted to do something that was important to me, and I was planning on starting a new TF2 map, so it fit perfectly. I knew that, although there was a deadline, I would work on it like I would any other map, not rush it, present what I had by the deadline, and continue work on it until I was satisfied with it.
(side note: I originally planned to create an XNA game for the project. Long story short, it didn't work out so well :blush
timber would become the first full map that I actually "completed." I say it in quotes, because although there is no final version, I am satisfied with what I have done with it, and it is time for me to move on. Before creating timber, I had made 2 portal maps, a scrapped TF2 map (never got past pre-alpha), and cp_backfire, a collaboration that didn't work out as well as planned.
I started out with a sketch. Originally, timber was going to be a 4-cap payload map, with a valley and a river in the middle. Man, things have changed. My sketch was very detailed, but it was obvious that it would be impossible to optimize (ironic, but more on that later). I scrapped it, deciding that the payload entities and logic would be too big of a step from doing no game mode logic besides a 2-cap cp map. Also, I realized the appeal of a Gravelpit-style map. There were hardly any custom GP-style maps, and I love GP, so I decided I would do that.
Mistake(s): Although I like GP-style, I also like payload. I probably should have gone with PL so that I could challenge myself and learn more about game mode logic, but the PL contest eventually showed up, so it was a moot point anyway.
Next step was to make another sketch; this time, it would be for the GP-style map. I created a fairly simple sketch, just to get my layout idea down on paper. I kept the idea of a river being a main point of the map, but I made sure to keep from going overboard on the amount of water.
Mistake(s): I definitely needed to account for classes more. I kept them in mind, but I didn't focus on specifics, which I should have.
Alpha
Now, onto the actual map. I started out with a very rough version for Alpha 1
as you can see, it is very wide open. Not a good start for optimization or gameplay. I can not stress enough how important it is that you make sure your map is optimizable. It was one of the biggest complaints during the beta stages of testing. Unfortunately, my map was so unoptimized that I couldn't compile it on normal vis once I reached the beta stage. No matter what version you are on of your map, if you don't know about optimization, go research it NOW. If you are unsure about optimization, go research it now. If you know everything there is to know about optimization, give yourself a cookie.
Here is a top-down view of the layout of the very first alpha:
Mistake(s): Too wide open, bad for optimization and gameplay. General layout was too similar to Gravelpit's
I had the first alpha version tested on a gameday. I got some good feedback, fairly positive and good, but it needed some major changes. Somehow, I didn't wrap my mind around this, because during the alpha stages, I did not make any drastic changes that were needed. By the time I got to beta, it was too late.
For each alpha version, though, I basically just made the C area smaller, changed B visually, and kept A mostly the same. That doesn't sound like much for 5 versions of alpha, does it? Somewhere along the way, I kept hearing that the path between A and B needed revising. It was too easy for BLU to get from A to B, so RED could never do a quick change-up on BLU. I added this tunnel for RED, so that it would be a dangerous move for BLU to go from A to B or vice-versa.
I think the tunnel added some nice strategy to the map, and some depth. It was a good idea in theory, but even with the signs, I don't think RED had much of a reason to use the path.
Also, I kept getting comments that the B building looked too much like Gravelpit's B building. In fact, I think a few people asked if it was that building. I planned on changing it drastically, but I made the dumb mistake of putting the building at an angle. I kept having to modify the building in a different vmf, copy it in, and then delete the old one. It got frustrating, and it was just a stupid design decision on my part.
By a5, I had fleshed the map out a little more, and made it brighter. The earlier versions were much too dark, and made the map look flat and boring.
a1:
a5:
Mistake(s): I did not make the important revisions to the map during alpha. Also, I think I should have done a few more alpha versions before continuing to beta. B building stayed mostly the same, but I should have at least made it look different than GP's B building
Other Mistakes: I didn't work on my map as much as I should have. I got arrogant, and was making a few changes and submitting my map nearly every gameday. That sounds great, but the changes were so small, that I could have got feedback in a much more efficient way. If I had spent more time on the map versions and made bigger changes, it would have been a much better outcome, I think.
Beta
I spent a lot of time working on the map to get it from alpha 5 to beta 1. This isn't a complaint. I got to finally see what my map was going to look like, and what visual theme it would have. I was satisfied with the significant changes from alpha to beta, especially because it included my first 3D skybox.
a5:
b1:
Along with adding little details and overlays, I fleshed out both spawns and the pit behind A.
Mistake(s): Because Lumberyard was the only alpine map out at the time, I could only look at that for a 3D skybox example. I think I should have explored the possibilities of actually having ground brushes in the 3D skybox, instead of just the illusion of mountains.
There's not much else to talk about with the beta versions. I improved my lighting, along with removing the side path from B to C, and changing the path from A to B. Also, I eventually changed the horrendous respawn times. They were way too long.
One of the most important topics I need to address is displacements. If you do not fully understand displacements, go through this tutorial:
http://forums.tf2maps.net/showthread.php?t=798
After creating my map, I went through this and was amazed. I learned some great tips from it that would have made my displacements so much better, if I had just known that information from the start. A special thanks goes to Youme for that one :thumbup1:.
Final version:
After many versions, my final version is b9. This is the version that I will be presenting for my project.
Making this map has been a great experience. After all my hard work, I put together this slideshow for my presentation:
http://dl.getdropbox.com/u/580413/Team Fortress 2 Level Design slideshow.pptx
A special thanks to:
-TF2maps.net, and all of the admins and members for their help, cooperation, and gamedays
-VALVe, for making an awesome game
-Youme, for his awesome tutorials and contests
-The Portal team, for making the game that got me into mapping
-Interlopers, a great resource
I won't be creating any full maps for a while. I have a lot of things going on over the summer. Also, I have been accepted into the Digipen Institute of Technology; I am fortunate enough to live an hour away from the school . I will probably frequent the forums still, along with mapping for some of the mini-mapping contests.
I hope you guys learn something from this, and have fun mapping! :thumbup1:
I started cp_timber in late September/early October. I knew then that it was going to be my senior project for high school. All the seniors at my school have to do a senior project, with a total of 20 hours recorded. I knew that I would have way more than 20 hours, but I decided to do it anyway. I wanted to do something that was important to me, and I was planning on starting a new TF2 map, so it fit perfectly. I knew that, although there was a deadline, I would work on it like I would any other map, not rush it, present what I had by the deadline, and continue work on it until I was satisfied with it.
(side note: I originally planned to create an XNA game for the project. Long story short, it didn't work out so well :blush
timber would become the first full map that I actually "completed." I say it in quotes, because although there is no final version, I am satisfied with what I have done with it, and it is time for me to move on. Before creating timber, I had made 2 portal maps, a scrapped TF2 map (never got past pre-alpha), and cp_backfire, a collaboration that didn't work out as well as planned.
I started out with a sketch. Originally, timber was going to be a 4-cap payload map, with a valley and a river in the middle. Man, things have changed. My sketch was very detailed, but it was obvious that it would be impossible to optimize (ironic, but more on that later). I scrapped it, deciding that the payload entities and logic would be too big of a step from doing no game mode logic besides a 2-cap cp map. Also, I realized the appeal of a Gravelpit-style map. There were hardly any custom GP-style maps, and I love GP, so I decided I would do that.
Mistake(s): Although I like GP-style, I also like payload. I probably should have gone with PL so that I could challenge myself and learn more about game mode logic, but the PL contest eventually showed up, so it was a moot point anyway.
Next step was to make another sketch; this time, it would be for the GP-style map. I created a fairly simple sketch, just to get my layout idea down on paper. I kept the idea of a river being a main point of the map, but I made sure to keep from going overboard on the amount of water.
Mistake(s): I definitely needed to account for classes more. I kept them in mind, but I didn't focus on specifics, which I should have.
Alpha
Now, onto the actual map. I started out with a very rough version for Alpha 1
as you can see, it is very wide open. Not a good start for optimization or gameplay. I can not stress enough how important it is that you make sure your map is optimizable. It was one of the biggest complaints during the beta stages of testing. Unfortunately, my map was so unoptimized that I couldn't compile it on normal vis once I reached the beta stage. No matter what version you are on of your map, if you don't know about optimization, go research it NOW. If you are unsure about optimization, go research it now. If you know everything there is to know about optimization, give yourself a cookie.
Here is a top-down view of the layout of the very first alpha:
Mistake(s): Too wide open, bad for optimization and gameplay. General layout was too similar to Gravelpit's
I had the first alpha version tested on a gameday. I got some good feedback, fairly positive and good, but it needed some major changes. Somehow, I didn't wrap my mind around this, because during the alpha stages, I did not make any drastic changes that were needed. By the time I got to beta, it was too late.
For each alpha version, though, I basically just made the C area smaller, changed B visually, and kept A mostly the same. That doesn't sound like much for 5 versions of alpha, does it? Somewhere along the way, I kept hearing that the path between A and B needed revising. It was too easy for BLU to get from A to B, so RED could never do a quick change-up on BLU. I added this tunnel for RED, so that it would be a dangerous move for BLU to go from A to B or vice-versa.
I think the tunnel added some nice strategy to the map, and some depth. It was a good idea in theory, but even with the signs, I don't think RED had much of a reason to use the path.
Also, I kept getting comments that the B building looked too much like Gravelpit's B building. In fact, I think a few people asked if it was that building. I planned on changing it drastically, but I made the dumb mistake of putting the building at an angle. I kept having to modify the building in a different vmf, copy it in, and then delete the old one. It got frustrating, and it was just a stupid design decision on my part.
By a5, I had fleshed the map out a little more, and made it brighter. The earlier versions were much too dark, and made the map look flat and boring.
a1:
a5:
Mistake(s): I did not make the important revisions to the map during alpha. Also, I think I should have done a few more alpha versions before continuing to beta. B building stayed mostly the same, but I should have at least made it look different than GP's B building
Other Mistakes: I didn't work on my map as much as I should have. I got arrogant, and was making a few changes and submitting my map nearly every gameday. That sounds great, but the changes were so small, that I could have got feedback in a much more efficient way. If I had spent more time on the map versions and made bigger changes, it would have been a much better outcome, I think.
Beta
I spent a lot of time working on the map to get it from alpha 5 to beta 1. This isn't a complaint. I got to finally see what my map was going to look like, and what visual theme it would have. I was satisfied with the significant changes from alpha to beta, especially because it included my first 3D skybox.
a5:
b1:
Along with adding little details and overlays, I fleshed out both spawns and the pit behind A.
Mistake(s): Because Lumberyard was the only alpine map out at the time, I could only look at that for a 3D skybox example. I think I should have explored the possibilities of actually having ground brushes in the 3D skybox, instead of just the illusion of mountains.
There's not much else to talk about with the beta versions. I improved my lighting, along with removing the side path from B to C, and changing the path from A to B. Also, I eventually changed the horrendous respawn times. They were way too long.
One of the most important topics I need to address is displacements. If you do not fully understand displacements, go through this tutorial:
http://forums.tf2maps.net/showthread.php?t=798
After creating my map, I went through this and was amazed. I learned some great tips from it that would have made my displacements so much better, if I had just known that information from the start. A special thanks goes to Youme for that one :thumbup1:.
Final version:
After many versions, my final version is b9. This is the version that I will be presenting for my project.
Making this map has been a great experience. After all my hard work, I put together this slideshow for my presentation:
http://dl.getdropbox.com/u/580413/Team Fortress 2 Level Design slideshow.pptx
A special thanks to:
-TF2maps.net, and all of the admins and members for their help, cooperation, and gamedays
-VALVe, for making an awesome game
-Youme, for his awesome tutorials and contests
-The Portal team, for making the game that got me into mapping
-Interlopers, a great resource
I won't be creating any full maps for a while. I have a lot of things going on over the summer. Also, I have been accepted into the Digipen Institute of Technology; I am fortunate enough to live an hour away from the school
. I will probably frequent the forums still, along with mapping for some of the mini-mapping contests.I hope you guys learn something from this, and have fun mapping! :thumbup1:
