Author Topic: opinions on new game using idtech4.5  (Read 4093 times)

0 Members and 1 Guest are viewing this topic.

MrC

  • Jr. Member
  • **
  • Posts: 58
  • Karma: +8/-24
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #30 on: November 12, 2015, 08:33:48 AM »
Yes.

The Happy Friar

  • Happy Happy Joy Joy
  • Administrator
  • Hero Member
  • *****
  • Posts: 852
  • Karma: +41/-4
  • Semi-Newbie.
    • View Profile
    • EarthQuake
Re: opinions on new game using idtech4.5
« Reply #31 on: November 12, 2015, 10:51:03 AM »
Well, i understand that AAA assets cost a lot, im mostly referring to whether this engine could support assets that could be around today's standards. Doesnt haven to be AAA.

Ok, Capt Obvious time: if you don't intend to use super-high quality assets or pipeline, why worry if the engine supports it?

vladdrak

  • Newbie
  • *
  • Posts: 43
  • Karma: +0/-4
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #32 on: November 12, 2015, 11:12:15 AM »
I don't wish to hijack the thread, so please let me know if I should open a new topic and delete this post overall.

I've been planning a game, and similarly to rebarkillburst, I'm a bit unsure whether an idtech engine (particularly dp/idtech4/doom3bfg) would be a viable option, even tho I stil find idtech4's use of lighting/dynamic shadows hot as hell after all these years, so it's a strong candidate for me.
The game would basically be a turn-based dungeon crawler rpg with roguelike-ish elements and grid-based movement and varying themes, heavily inspired by Wizardry, Grimrock, Elminage Gothic and Paper Sorcerer.
Gothic and PS might be the closest points of reference of them all. Note how there's no enemy, only when a combat event triggers, which means there's no pathfinding and stuff like that, only combat AI, which shouldn't be insanely hard to implement even in idtech4 (wherein you said AI is practically non-existent aside from Doom 3). I'm also very much interested in grid-based movement and map layout randomization - that's where the roguelike part comes in, among other things, like random enemy placements, etc. So how hard would it be to add these features to the existing codebase? Or which of these could be done on the doomscript level?
Sorry if I'm being totally ridiculous with these questions, I played around with gtkradiant some time ago, but that's how close I ever came to idtech stuff, so.

« Last Edit: November 12, 2015, 11:42:31 AM by vladdrak »

motorsep

  • Hero Member
  • *****
  • Posts: 960
  • Karma: +74/-127
  • Artist
    • View Profile
    • Kot in Action Creative Artel
Re: opinions on new game using idtech4.5
« Reply #33 on: November 12, 2015, 06:54:04 PM »
I don't wish to hijack the thread, so please let me know if I should open a new topic and delete this post overall.

I am no boss here, but I have a gut feeling it's fine :)

...  I stil find idtech4's use of lighting/dynamic shadows hot as hell after all these years, so it's a strong candidate for me.

As if lighting is what makes or breaks video games  ::)

Note how there's no enemy, only when a combat event triggers, which means there's no pathfinding and stuff like that, ...

So, there are enemies then. And there is basic pathfinding. Pathfinding is how enemy find its way from where it is now to the player. So when you get into a room with an enemy, it needs to get from where it stands to you. Note that idTech 4 already has pathfinding solution, and that solution is pretty clever. I doubt an average programmer can make a better one (hell, even above average programmer will have hard time making a better pathfinding solution).

only combat AI, which shouldn't be insanely hard to implement ...

You think? :) Probably easier than fully fledged FPS combat, but by no means trivial.

... even in idtech4 (wherein you said AI is practically non-existent aside from Doom 3).

Behavior logic is primitive in Doom 3 (most of it is script based). Pathfinding and navigation are quite good. Not superb, but really good. There is room for improvement of course, but it's pretty solid for a lot of types of FPS/TPS indie games.

Pathfinding and navigation are implemented in C++. Behavior logic (states and animation control) are in scripts, which you have to write from scratch and it's not that easy as it might appear to be at first sight.

I'm also very much interested in grid-based movement and map layout randomization - that's where the roguelike part comes in, among other things, like random enemy placements, etc. So how hard would it be to add these features to the existing codebase? Or which of these could be done on the doomscript level?

That's not idTech 4's strongest point. Visibility culling is pre-compiled, meaning your random levels will have no culling, if you do it without some dark magic. So performance will be super crappy.

To get it done right, you would have to generate map and re-build it, before loading it for gameplay. That means you need to implement a way to manipulate brushes and groups of brushes (which you can't do anything with at runtime at this moment). Good thing Doom 3 has all the foundation for such task embedded, thanks to built-in Radiant. It's a complex task. Basically anything procedurally generated would be like making a game inside the game.

Sorry if I'm being totally ridiculous with these questions, I played around with gtkradiant some time ago, but that's how close I ever came to idtech stuff, so.

You should play around with DarkRadiant 2.03. It's the only 3rd party Doom 3 mapping tool that is quite advance. There is built-in DoomEdit, but DarkRadiant is better.

I hate to retype my lines, but if you have no clue about Doom 3 modding (and I mean on a level of making total conversion) and you want to make whole game from scratch, you might want to reconsider. If you know Doom 3 modding really well, and you just want to make a solid mod for Doom 3 that is in line with what Doom 3 has to offer, by all means - it's the best tech you can get today in terms of flexibility/freedom/iteration capacity/visuals/performance (I know Dota 2 and StarCraft 2 offer decent modding tools, but it's mostly maps and it's pretty rigid modding; I think anything Source-built is moddable).

vladdrak

  • Newbie
  • *
  • Posts: 43
  • Karma: +0/-4
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #34 on: November 12, 2015, 10:56:58 PM »
thank you for your reply!
Quote
...  I stil find idtech4's use of lighting/dynamic shadows hot as hell after all these years, so it's a strong candidate for me.

As if lighting is what makes or breaks video games  ::)

Obviously not, but it always gives me some ideas :)

Note how there's no enemy, only when a combat event triggers, which means there's no pathfinding and stuff like that, ...
Quote
So, there are enemies then. And there is basic pathfinding. Pathfinding is how enemy find its way from where it is now to the player. So when you get into a room with an enemy, it needs to get from where it stands to you. Note that idTech 4 already has pathfinding solution, and that solution is pretty clever. I doubt an average programmer can make a better one (hell, even above average programmer will have hard time making a better pathfinding solution).
only combat AI, which shouldn't be insanely hard to implement ...
Quote
You think? :) Probably easier than fully fledged FPS combat, but by no means trivial.

Sorry I wasn't very clear about this. The combat and AI that I "dreamt up" is pretty much identical to what's seen in Elminage Gothic. The player basically would just go around in the map, room-to-room, until it encounters an enemy, then the game would stop to switch on a combat state - similarly to Final Fantasy -, so the combat may begin, which would go in an oldschool RPG fashion. A solution like this would strip down the whole navigation/pathfinding business and I could get around some heavy C++ AI programming and might be able to script it up. If you see it that way then it really would be just combat/behavior logic.
Also using still and/or animated sprites for enemies. Maybe via GUI-scripting? Would that be possible?

Quote
I hate to retype my lines, but if you have no clue about Doom 3 modding (and I mean on a level of making total conversion) and you want to make whole game from scratch, you might want to reconsider. If you know Doom 3 modding really well, and you just want to make a solid mod for Doom 3 that is in line with what Doom 3 has to offer, by all means - it's the best tech you can get today in terms of flexibility/freedom/iteration capacity/visuals/performance (I know Dota 2 and StarCraft 2 offer decent modding tools, but it's mostly maps and it's pretty rigid modding; I think anything Source-built is moddable).

Could you please then recommend me an engine that might meet these demands and is also available on Linux and uses GPL? Would DP do the work? Thx in advance!
« Last Edit: November 12, 2015, 11:02:57 PM by vladdrak »

motorsep

  • Hero Member
  • *****
  • Posts: 960
  • Karma: +74/-127
  • Artist
    • View Profile
    • Kot in Action Creative Artel
Re: opinions on new game using idtech4.5
« Reply #35 on: November 12, 2015, 11:01:37 PM »
Could you please then recommend me an engine that might meet these demands and is also available on Linux and uses GPL? Would DP do the work? Thx in advance!

I only saw stuff like that done in Unity. I am sure UE4 can do it too.

The thing is that you will have to program stuff in C++ or C#. Regardless of the engine.

Unity is a clean slate. It's like id Tech 4 without gameplay code. And by gameplay code I mean C++ code. UE4 has templates, but it's pretty much the same.

So I can't really say go with that or this. It's entirely up to you and your skills/team. Any of those engine is free. Download it and try it. It will take time to learn at least foundation, but you will have decent idea about each engine.

vladdrak

  • Newbie
  • *
  • Posts: 43
  • Karma: +0/-4
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #36 on: November 12, 2015, 11:03:50 PM »
thanks!

The Happy Friar

  • Happy Happy Joy Joy
  • Administrator
  • Hero Member
  • *****
  • Posts: 852
  • Karma: +41/-4
  • Semi-Newbie.
    • View Profile
    • EarthQuake
Re: opinions on new game using idtech4.5
« Reply #37 on: November 13, 2015, 06:59:14 AM »
I'd still suggest you try out Torque too.  The purchasable AI has downloadable demo's for you to test too.   here

The specific AI you want might be doable w/o ever making a monster in D3 just by using level scripts.  Since it's all just if/then's & changing animations/etc.

bkt

  • Jr. Member
  • **
  • Posts: 96
  • Karma: +12/-2
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #38 on: November 13, 2015, 01:50:48 PM »
The TL;DR summary of this entire thread is: If you want to make a new game from scratch, use UE4.

Done :)

argoon

  • Sr. Member
  • ****
  • Posts: 267
  • Karma: +21/-80
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #39 on: November 13, 2015, 02:00:38 PM »
The TL;DR summary of this entire thread is: If you want to make a new game from scratch, use UE4.

Done :)

If this forum dies you only have this thread to blame and the ones that on it encouraged people to go away from idtech 4. 

motorsep

  • Hero Member
  • *****
  • Posts: 960
  • Karma: +74/-127
  • Artist
    • View Profile
    • Kot in Action Creative Artel
Re: opinions on new game using idtech4.5
« Reply #40 on: November 13, 2015, 02:09:51 PM »
If this forum dies you only have this thread to blame and the ones that on it encouraged people to go away from idtech 4.

This statement is silly.

motorsep

  • Hero Member
  • *****
  • Posts: 960
  • Karma: +74/-127
  • Artist
    • View Profile
    • Kot in Action Creative Artel
Re: opinions on new game using idtech4.5
« Reply #41 on: November 13, 2015, 02:11:33 PM »
The TL;DR summary of this entire thread is: If you want to make a new game from scratch, use UE4.

Done :)

I would say rather this: if you want to make a game from scratch - become a good programmer or find one that will stick around to see the game completed.

vladdrak

  • Newbie
  • *
  • Posts: 43
  • Karma: +0/-4
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #42 on: November 13, 2015, 02:16:41 PM »
The specific AI you want might be doable w/o ever making a monster in D3 just by using level scripts.  Since it's all just if/then's & changing animations/etc.

that's what I was thinking, thx

bkt

  • Jr. Member
  • **
  • Posts: 96
  • Karma: +12/-2
  • Doom Newbie
    • View Profile
Re: opinions on new game using idtech4.5
« Reply #43 on: November 13, 2015, 03:20:10 PM »
The TL;DR summary of this entire thread is: If you want to make a new game from scratch, use UE4.

Done :)

If this forum dies you only have this thread to blame and the ones that on it encouraged people to go away from idtech 4.
Last year I released a mod for idTech 4 and if the circumstances were right I'd happily make another.  I'd never make a new game from scratch with the intent to sell it using idTech 4.  In every aspect I value UE4 is a better prospect for a fresh project.  There's no need for me to elaborate either, because if you're familiar with both idTech and UE4, the distinctions are clear.  If you're not familiar with what UE4 is capable of, go do some research to find out.  Then make an unbiased comparison and tell me which is more suitable.


The Happy Friar

  • Happy Happy Joy Joy
  • Administrator
  • Hero Member
  • *****
  • Posts: 852
  • Karma: +41/-4
  • Semi-Newbie.
    • View Profile
    • EarthQuake
Re: opinions on new game using idtech4.5
« Reply #44 on: November 13, 2015, 08:19:11 PM »
If this forum dies you only have this thread to blame and the ones that on it encouraged people to go away from idtech 4.

I am not going to lie and say the d3 gpl engine is necessarily worth it to make everything from scratch.