Disappointed in Advanced book
by Dave ("Pococurante") Scheffer · in Torque Game Engine · 11/27/2005 (12:18 pm) · 2 replies
Too much filler, too much emphasis again on using tools outside the TGE SDK, too many trivial example within the SDK.
100 pages alone on the TGE API reference? Full error dumps "just to show an example"? (not even with real commentary to sort of justify it) An entire chapter on basic PHP for the website? I could excuse the flimsy excuse to resell free third-party tutorials but at some point there has to be a game underneath all the graphics. I liked having overviews of the built-in tools but much was a rehash of the first book. Don't just show me how to tweak a setting in a skybox control, show me how I can link that to the characters in my world - show me how NPC/M behavior can be coded to adapt to the weather, to do more when recognizing another mob than follow or shoot it.
Show me how to make game mechanics. I don't need ideas for the mechanics - I need to know where it makes the most sense to plug into what part of the product to make it happen. I need to understand when a mechanic must occur on a single server or if it makes sense to split different mechanics across different servers. Ok this last part I don't expect from an indie publisher but I'd at least like to come away from a 700 page read understanding how an engine supports common game mechanics at least as well as I understand unwrapping a model to get to the textures.
For something marketed as emphasizing the backend when the first book concentrated on front end I must say I didn't see it. The A.I. discussion is bluesky with the closest thing to pragmatic use was an outline on swarming, something perhaps useful when building a Stargate Trooper type of FPS but no use at all in a VW.
I realize TG's roots as an FPS but it does the product a real disservice to not emphasize its potential in persistent "virtual world" games. Pure FPS won't be the flagship examples of Torque simply because they are too simple in principle and too easy to knock out. There should have been extensive review of classes/skills, of non-combat mechanics, even more to do with complex PC-NPC dialogue than the rudimentary GUI example. In fact what I find missing in all Torque resources are practical examples of AI for mobs doing anything more than hacking and slashing. And those examples don't even bother with such basics as threat memory/hate lists.
I can't recommend the second book. I recommend the first book mainly for the tutorials on using non-Garage Game tools. For anyone really wanting to learn the details of the Torque product beyond the graphical tools sadly neither book really is helpful.
Anyway ranting aside is there a resource someone recommends for creating rich NPC behaviors and player skill/class/stat systems? Two texts each the size of a phonebook couldn't help me.
100 pages alone on the TGE API reference? Full error dumps "just to show an example"? (not even with real commentary to sort of justify it) An entire chapter on basic PHP for the website? I could excuse the flimsy excuse to resell free third-party tutorials but at some point there has to be a game underneath all the graphics. I liked having overviews of the built-in tools but much was a rehash of the first book. Don't just show me how to tweak a setting in a skybox control, show me how I can link that to the characters in my world - show me how NPC/M behavior can be coded to adapt to the weather, to do more when recognizing another mob than follow or shoot it.
Show me how to make game mechanics. I don't need ideas for the mechanics - I need to know where it makes the most sense to plug into what part of the product to make it happen. I need to understand when a mechanic must occur on a single server or if it makes sense to split different mechanics across different servers. Ok this last part I don't expect from an indie publisher but I'd at least like to come away from a 700 page read understanding how an engine supports common game mechanics at least as well as I understand unwrapping a model to get to the textures.
For something marketed as emphasizing the backend when the first book concentrated on front end I must say I didn't see it. The A.I. discussion is bluesky with the closest thing to pragmatic use was an outline on swarming, something perhaps useful when building a Stargate Trooper type of FPS but no use at all in a VW.
I realize TG's roots as an FPS but it does the product a real disservice to not emphasize its potential in persistent "virtual world" games. Pure FPS won't be the flagship examples of Torque simply because they are too simple in principle and too easy to knock out. There should have been extensive review of classes/skills, of non-combat mechanics, even more to do with complex PC-NPC dialogue than the rudimentary GUI example. In fact what I find missing in all Torque resources are practical examples of AI for mobs doing anything more than hacking and slashing. And those examples don't even bother with such basics as threat memory/hate lists.
I can't recommend the second book. I recommend the first book mainly for the tutorials on using non-Garage Game tools. For anyone really wanting to learn the details of the Torque product beyond the graphical tools sadly neither book really is helpful.
Anyway ranting aside is there a resource someone recommends for creating rich NPC behaviors and player skill/class/stat systems? Two texts each the size of a phonebook couldn't help me.
#2
Generally an "Advanced" book would do more than simply provide additional entry-level walkthroughs of tools - that alone would have freed up half the book for the kind of in-depth discussions I was really looking forward to reading. Had they called this "volume 2" I'd rate it 4/5.
Thank you very much for the reference to Dreamer's tutorials.
12/01/2005 (10:48 am)
David thank you for your reply. I agree that was my perspective and I do appreciate the differences between FPS and MORPG.Generally an "Advanced" book would do more than simply provide additional entry-level walkthroughs of tools - that alone would have freed up half the book for the kind of in-depth discussions I was really looking forward to reading. Had they called this "volume 2" I'd rate it 4/5.
Thank you very much for the reference to Dreamer's tutorials.
Associate David Montgomery-Blake
David MontgomeryBlake
Now, I think where I see your bias is in your dissection of "easy to knock out" FPS's. Good realtime networked AI in a fast action shooter has logistical speed issues that most realtime RPG's (networked or non-networked) do not have to deal with. Plus, with the latest tech available to the off-the-shelf FPS market, you would have to have an amazing art team to make something that the common tech-heads would want to buy. Half-Life 2's physics and Far Cry's environments have spoiled us a bit.
But this is not saying that the book is perfect by any means. It is simply not oriented towards what you wanted. But it is a good starting point into the various questions that were asked here once the first book was released and didn't cover topic x. For you, it didn't cover topic x in as much detail or with the orientation you wanted.
Dreamer put together a number of MMORPG tutorials that might be helpful to you. Not so much on the AI front, but on the character front.