Plan for BobTheCBuilder (John Kabus)
by John Kabus (BobTheCBuilder) · 10/07/2003 (10:24 am) · 7 comments
Trying to juggle content and code...
Our team is still feeling out its "process" for managing/designing/developing our content and code. Things are starting to fall into place, maybe not as quickly as I would like, however I have a habit of pushing a bit too hard. I have been trying to avoid pushing my team as hard as I push myself and I think I'm doing an alright job, but I feel like during the week (we usually meet on weekends) that I'm in a void (communication wise). Trying to find out what's done and what needs changed without sounding pushy or impatient is very difficult, especially over email.
About six weeks ago we chose one of our levels as a test bed for all of our future work. The level met all of our criteria for the perfect test level while still being one we would use in the game. Basically the level was one of the smaller levels, had a lot of background content but not too many interiors, followed the simplest game play rules, and content wise contained the majority of the effects and tricks we would like to have incorporated into the game.
In the past six weeks we've gotten only 10% of the level hammered out, but we've covered upwards of 60% of the learning curve. Part of the learning curve has been adding tools into Torque to speed up development of the levels. The latest (and only tool interesting enough to mention) is a compound object type. Basically I can add a compound object to the mission and add any number of any objects to the compound object, then add a second compound object that references the first. The second compound object now has a unique copy of all of the objects from the first compound object. How is this different from cut and paste? Changes can be made in the first compound and then if the second is refreshed the changes are reflected in the second, however transforms are made in compound object space (as in relative to the compound parent), so all refreshed changes are reflected relative to the second compound object. So effectively you can add a new compound object, move it to where you want it, tie the reference to another compound, and your cloned objects will all have the same relationship to each other as in the original compound, only relative to the new compound. Also refreshes do not affect objects in a referencing compound that do not exist in a referenced compound, so you can effectively do multiple layers of inheritance.
While building our test level I've added a whole slew of changes that really add depth to the engine. The first change was statics casting shadows (obviously statics can have much nicer geometry, but normally look out of place because of shading, shadows, ...). Some of the others changes are mission level baked lights (with spotlights), updated lighting model on the statics, detail mapping on interiors, and updating the volume light resource to derive from fxlight so I can point the volume 'beam' and spot with one object (fxlight only acts as a placeholder to register the existing LightInfo class, same as always).
We've finally nailed down our standards for 'texel : world unit' ratios based on object type (interior or static), this is a critical step in making objects blend together in a scene visually. Nothing stands out like a blurry wall with really detailed objects next to or on it.
On a different note; I was a little miffed about IGC. A month and a half ago Jay sent me an email letting us know that people might be interested in seeing our project at the conference. I sent an email confirming our interest in participating and a very short list of questions about participation. Well that was the last I heard from GG about the conference. I've sent countless emails and received several responses from Jeff that completely side stepped the issue and nothing from Jay. Two weeks before the conference I read that the conference is open to anyone, that it's basically a giant lan party. Not only is two weeks too short of notice, but the info wasn't even in a news post, it was in Jay's plan.
I know GG is a small company and that things are often disorganized around here, but for those of us that are relatively new to the community (< 6 months) some real info would be nice. IGC is a big deal here and the only info I've really seen in a news post is the seminar lineup posted about a month ago. Oh well...
Anyway time for a few screenshots. These are showing the detail mapping:
Our team is still feeling out its "process" for managing/designing/developing our content and code. Things are starting to fall into place, maybe not as quickly as I would like, however I have a habit of pushing a bit too hard. I have been trying to avoid pushing my team as hard as I push myself and I think I'm doing an alright job, but I feel like during the week (we usually meet on weekends) that I'm in a void (communication wise). Trying to find out what's done and what needs changed without sounding pushy or impatient is very difficult, especially over email.
About six weeks ago we chose one of our levels as a test bed for all of our future work. The level met all of our criteria for the perfect test level while still being one we would use in the game. Basically the level was one of the smaller levels, had a lot of background content but not too many interiors, followed the simplest game play rules, and content wise contained the majority of the effects and tricks we would like to have incorporated into the game.
In the past six weeks we've gotten only 10% of the level hammered out, but we've covered upwards of 60% of the learning curve. Part of the learning curve has been adding tools into Torque to speed up development of the levels. The latest (and only tool interesting enough to mention) is a compound object type. Basically I can add a compound object to the mission and add any number of any objects to the compound object, then add a second compound object that references the first. The second compound object now has a unique copy of all of the objects from the first compound object. How is this different from cut and paste? Changes can be made in the first compound and then if the second is refreshed the changes are reflected in the second, however transforms are made in compound object space (as in relative to the compound parent), so all refreshed changes are reflected relative to the second compound object. So effectively you can add a new compound object, move it to where you want it, tie the reference to another compound, and your cloned objects will all have the same relationship to each other as in the original compound, only relative to the new compound. Also refreshes do not affect objects in a referencing compound that do not exist in a referenced compound, so you can effectively do multiple layers of inheritance.
While building our test level I've added a whole slew of changes that really add depth to the engine. The first change was statics casting shadows (obviously statics can have much nicer geometry, but normally look out of place because of shading, shadows, ...). Some of the others changes are mission level baked lights (with spotlights), updated lighting model on the statics, detail mapping on interiors, and updating the volume light resource to derive from fxlight so I can point the volume 'beam' and spot with one object (fxlight only acts as a placeholder to register the existing LightInfo class, same as always).
We've finally nailed down our standards for 'texel : world unit' ratios based on object type (interior or static), this is a critical step in making objects blend together in a scene visually. Nothing stands out like a blurry wall with really detailed objects next to or on it.
On a different note; I was a little miffed about IGC. A month and a half ago Jay sent me an email letting us know that people might be interested in seeing our project at the conference. I sent an email confirming our interest in participating and a very short list of questions about participation. Well that was the last I heard from GG about the conference. I've sent countless emails and received several responses from Jeff that completely side stepped the issue and nothing from Jay. Two weeks before the conference I read that the conference is open to anyone, that it's basically a giant lan party. Not only is two weeks too short of notice, but the info wasn't even in a news post, it was in Jay's plan.
I know GG is a small company and that things are often disorganized around here, but for those of us that are relatively new to the community (< 6 months) some real info would be nice. IGC is a big deal here and the only info I've really seen in a news post is the seminar lineup posted about a month ago. Oh well...
Anyway time for a few screenshots. These are showing the detail mapping:
#2
The format was hardly cryptic since it seems identical to the last one, something you probably missed being new(ish). I think the co-ordination was for trying to gain some synergy in marketing efforts etc. I'm not sure it's fair to call the IGC a giant lan party, the ShowOFF format suits the style of the gathering perfectly.
My artist has let me down on a fair load of material I needed for my IGC demo, so you're not the only one missing out :) If you have anything to demo you can probably mail it to the GG staff and they'll make sure it's run.
I wouldn't worry about missing out on the publishing deal of a lifetime for not being at the IGC. You'll lose the opportunity for a load of kudos points though ;) The conference rocked last year though, i'd be annoyed about missing it...
Great detail maps.
10/07/2003 (10:59 am)
Bob (John),The format was hardly cryptic since it seems identical to the last one, something you probably missed being new(ish). I think the co-ordination was for trying to gain some synergy in marketing efforts etc. I'm not sure it's fair to call the IGC a giant lan party, the ShowOFF format suits the style of the gathering perfectly.
My artist has let me down on a fair load of material I needed for my IGC demo, so you're not the only one missing out :) If you have anything to demo you can probably mail it to the GG staff and they'll make sure it's run.
I wouldn't worry about missing out on the publishing deal of a lifetime for not being at the IGC. You'll lose the opportunity for a load of kudos points though ;) The conference rocked last year though, i'd be annoyed about missing it...
Great detail maps.
#3
-Jeff Tunnell GG
Edit: IGC is more than a LAN party. BTW, this represents an immense amount of work.
www.indiegamescon.com/schedule.html
10/07/2003 (11:23 am)
Miffed about IGC? Sidestepping what? I don't understand. IGC is open to anybody. Any project can be shown off to anybody. As I explained to the entire community, I was out of the office for two months, so emails went unanswered. I did reply to you when I got back regarding Cartography Shop. I thought we were fine. If not, please let me know what the problem is. You are doing great stuff and I don't want you to have a problem with GG.-Jeff Tunnell GG
Edit: IGC is more than a LAN party. BTW, this represents an immense amount of work.
www.indiegamescon.com/schedule.html
#4
Thanks, I loved your dev shots last week. I'm really looking forward to your demo, hopefully I can get something together you guys can checkout too. It may be a while though; we're still trying to get a dev shot together.
Gareth,
I didn't mean for 'lan party' to sound derogatory, on the contrary it's a good thing, I was referring to the conference being open to anyone to demo at (and I assume play the demos). Its not just about the demo, I was actually interested in meeting some of the people that I talk to on GG. Unfortunately I think it's a little late to make plans to go.
Anyway there is always next year. Hopefully by next year we will be more prepared for the conference anyway (hopefully).
Jeff,
There is not a problem, just a misunderstanding. I was unaware of the way IGC works and without knowing that my questions to you about participation must have seemed a little strange. I was only trying to point out that to the uninitiated the concept is a little foreign and the information is a little hard to stumble across. I originally assumed that the conference was by invitation only (for participation), so I was expecting a more formal follow up to Jay's email. When I sent you questions regarding IGC in the Cartography Shop emails, I received only Cartography Shop responses. I hope you can see why I was confused by this.
10/07/2003 (12:08 pm)
Davis,Thanks, I loved your dev shots last week. I'm really looking forward to your demo, hopefully I can get something together you guys can checkout too. It may be a while though; we're still trying to get a dev shot together.
Gareth,
I didn't mean for 'lan party' to sound derogatory, on the contrary it's a good thing, I was referring to the conference being open to anyone to demo at (and I assume play the demos). Its not just about the demo, I was actually interested in meeting some of the people that I talk to on GG. Unfortunately I think it's a little late to make plans to go.
Anyway there is always next year. Hopefully by next year we will be more prepared for the conference anyway (hopefully).
Jeff,
There is not a problem, just a misunderstanding. I was unaware of the way IGC works and without knowing that my questions to you about participation must have seemed a little strange. I was only trying to point out that to the uninitiated the concept is a little foreign and the information is a little hard to stumble across. I originally assumed that the conference was by invitation only (for participation), so I was expecting a more formal follow up to Jay's email. When I sent you questions regarding IGC in the Cartography Shop emails, I received only Cartography Shop responses. I hope you can see why I was confused by this.
#5
10/07/2003 (10:12 pm)
Ok one more screenshot, this time of an interior I'm currently working on. Clearly we haven't textured the statics yet, but I think the shot is impressive non the less:
#6
Looks great btw.
10/08/2003 (9:40 am)
I like this last shot a lot. =) It doesn't look like you are using the particle emitters to produce this steam. Care to explain to the curious? Looks great btw.
#7
A lot of what I'm running is still Torque. I have to admit I was not originally into the idea of buying technology for our game. I think this is a natural reaction for a developer, but technology development is a vicious cycle (no pun intended). Torque's framework has made it VERY easy for me to add or update portions of the technology, while still having plenty of time to work on content as I go. Well worth the $100. A lot of cool things can be done with Torque; it's just a matter of learning ways to exploit it.
If you have any specific questions on the steam let me know.
10/08/2003 (12:11 pm)
That steam is 100% Torque. The trick is to run Torque in windowed mode and have your particle system script open in Tribal or Notepad. Then dedicate the next hour to tweaking the script and re-exec-ing it in the console (be careful some data members don't reload until you exit the mission). The fire in the background of the first two shots is also Torque.A lot of what I'm running is still Torque. I have to admit I was not originally into the idea of buying technology for our game. I think this is a natural reaction for a developer, but technology development is a vicious cycle (no pun intended). Torque's framework has made it VERY easy for me to add or update portions of the technology, while still having plenty of time to work on content as I go. Well worth the $100. A lot of cool things can be done with Torque; it's just a matter of learning ways to exploit it.
If you have any specific questions on the steam let me know.
Torque Owner Davis Ray Sickmon, Jr
Default Studio Name
The ShowOFF room is an interesting deal - everyone loads thier games on there, and the players pretty much determine what runs. There's no organization for that part of it really. For the LAN party, pretty much same thing goes - the players are the ones calling the shots.