opinions on new game using idtech4.5

[motorsep]

PC isn't dead for AAA and III (AAA indies as they call them nowadays). For indies it's pretty much dead due to a need in actual marketing that costs a lot of money. Of course there is always a chance to break through without investing into PR / marketing, but it's an iffy gamble (which we will be taking with Phaeton).

The problem is that there is no alternative to Steam on PC and since Steam is a monopoly, they do a lot of shit that hurts indies. Plus their support became soooo horrible. And there is no way around it Releasing on Steam is the only option.

Mobile is problematic. Mobile VR is what I was talking about. It's entirely different market, a new frontier. In a few years it will be harder to get in. Now is a golden opportunity (more or less, we'll see when Gear VR CV1 is released this month).

Game dev is tough field to be in, regardless of the engine. Quake has game code w/ primitive AI, but no tools, no docs, no programmers to help you with it. I don't know much about Quake 2 scene and what's left of it, but Quake 2 has game code w/ better AI. Quake 3 has no tools (except mapping), no programmers, but solid gameplay code with bot AI and good networking. Doom 3 has tools (except ), some docs, but no programmers and no actual gameplay code (including AI/etc.) I has solid gameplay framework in C++, but that's all there is. UE4 has no gameplay, not whole a lot of programmers willing to work with you. Unity has no gameplay framework, a ton of programmers of different level of qualification. I don't know much of Torque and other engines. Source 2 (Dota 2 Reborn) seems to be pretty awesome from what I know, but not sure how many programmers are available and I do know it's tied to Dota 2 at this time. CryEngine is pretty but evil engine.

So either way, you need to engineer a lot of code (AI for sure, since gamers want to play against something better than what Q1/Q2/Q3/D3 offer), unless you are willing to stay with core mechanics of Quake, Quake 2 or Quake 3, which is still an option.

[caedes]

PC isn't dead for AAA and III (AAA indies as they call them nowadays). For indies it's pretty much dead due to a need in actual marketing that costs a lot of money.

I don't think this problem is PC specific.
The problem is that there are too many indie games so it's hard to get noticed in that mass.
When Steam opened for indies, just being on Steam gave you exposure and sales - this is not the case anymore, because so many (indie) games are there.

But, as far as I know, it's pretty much the same on mobile appstores that are also full of games, so it's (at least!) as hard to stand out there.

One other thing that should be kept in mind: Many types of games that work well on PC or consoles don't work well on mobile. Developers should accept that simulating a joypad or similar on touchscreens just sucks. So egoshooters, platformers etc don't really work that well on mobile.
I guess that's part of the reason why games like candy crush are so popular there: Their interface doesn't feel horrible on a touchscreen.
(Of course other kinds of games like point-n-click adventures, strategy games etc should also work well)
So I think the "target Windows and Android" advice should be taken with a grain of salt. So keep input in mind.

About VR: I don't know. If you can do something great there it might be an opportunity (especially in the near future while the market is small), but doing that right is probably hard and requires a very competent team.
I also wonder if VR is gonna be a success or if it turns out that in the end it isn't that awesome for most games (like 3D doesn't work that well for most movies).

[motorsep]

About VR: I don't know. If you can do something great there it might be an opportunity (especially in the near future while the market is small), but doing that right is probably hard and requires a very competent team.
I also wonder if VR is gonna be a success or if it turns out that in the end it isn't that awesome for most games (like 3D doesn't work that well for most movies).

That's the thing about VR - you don't need to do anything great VR isn't about just gaming it traditional sense. You could make Doom 3 in VR where you just wander the base and learn about the lore. You can make a puzzle in VR and it will work wonders. When I tried it, I was blown away momentarily, even with Cardboard (although I got motion sickness really quickly in it).

With Unity, you don't need a team to create simple experiences. Oculus has been doing pretty good job with their Unity Utils. Basically you drag and drop assets into your project, place player controller into your scene and off you go. At this stage, VR doesn't require any more effort than traditional games.

If you would make a new game engine for VR or add VR support to Doom 3 BFG, then yeah, it will not be easy. But with Unity and UE4 (more with Unity, since that's Oculus's officially supported engine and they work closely with Unity on VR) it's really more about building gameplay that doesn't eat up a lot of CPU time, and optimizing art assets (kinda like what game artists had to do in Quake 3 days).

I am not sold yet on desktop VR, due to high cost and being tethered to PC, but I am sold on mobile VR (Gear VR in particular), even though visuals in mobile VR are not as great as on desktop VR.

[The Happy Friar]

I like GOG for my indie games.  As far as I can tell, it's the biggest competition to Steam.  There was a great article I read about how for the older games they actually modify the code (if available) to make it work on modern OS's.  If no code then they start hacking away as the executable. 

[argoon]

I don't comprehend Motorsep sometimes, he is working on a idtech 4 engine and game but don't waste any opportunity to bad mouth idtech 4 and make people run away from it, then starts to complain that there's no coders that want to work on idtech games anymore ,so he can't find help to make his dream game on it.

Motorsep sometimes i wonder how you made a game on DarkPlaces that has even less documentation and even less user friendly and modern tools.



If you are a coder no matter what engine you use you will not have any problem making a game, even on idtech 4, if you are only a artist then idtech 4 is not really artist friendly, compared to UE4 or Unity that is, but if you can't code or don't find a coder to help you will never make a game on any engine. 
Making a game using idtech 4 has cons but also some pros, like, you don't need to pay anything (if you don't want to release your engine source code modifications to the public you will need to pay id) and you have the full source code for the engine for free, and anything you make from sales is yours (and the publisher if applicable).

I like GOG for my indie games.  As far as I can tell, it's the biggest competition to Steam.  There was a great article I read about how for the older games they actually modify the code (if available) to make it work on modern OS's.  If no code then they start hacking away as the executable.

That is not entirely true, more times than anything they just use a wrapper around the game, like a emulator, for example they many times use NGlide to make their games run on modern PC's, for thief they even implemented newdark (that is a user made mod for the thief engine).   

[motorsep]

I don't comprehend Motorsep sometimes, he is working on a idtech 4 engine and game but don't waste any opportunity to bad mouth idtech 4 and make people run away from it, then starts to complain that there's no coders that want to work on idtech games anymore ,so he can't find help to make his dream game on it.

You don't need to comprehend me. Take my words/experience with a grain of salt and do you own thing

idTech 4 (or rather BFG engine in my case) is not a bad PC engine. It lacks a lot of optimizations unless you are willing to stick with Doom 3 type clone. There is a lot of work needs to be done for the engine to allow making something that is beyond Doom 3. Again, if you go Doom 3 route, you'll be just fine (even with multiplayer, if you are using BFG engine).

There are no coders for idTech 4 (in a sense of a pool from which you could hire/recruit some, as it's the case with Unity), and not because I "bad mouthing" the engine. All savvy programmers either moved on with their life outside of modding, or switch to more modern and easier to work with engines long time ago.

I give people a piece of advise, considering the circumstances. When I started with DP engine, Unity was utter shit and UDK had really bad terms. Plus I knew Quake 1  / Quake 3 modding on the level design / art end. DP's author was of a great help guiding us, fixing bugs and so on.

When I went with idTech 4, I already know art pipeline (not completely, but a lot of workflow can be carried from Quake 3 to Doom 3). I didn't know that AI, weapons, etc. are all in script and I didn't think it will be such a pain to rewrite it from scratch. When I already invested a lot of time into idTech 4, I realized it's a dead end for me. However, BFG engine was released and it seemed that on performance side it offered a lot more freedom than idTech 4. I found my team and we began working on it. Several years and thousands dollars down the road we have more or less production ready engine. I had to throw away most of the original design for the game due to engine's limitations. We still have a few major things on the list to get engine more optimized to allow just working on the game and not fighting the tech. The problem is that with all this R&D time, I don't have time to actually make the game assets.

That's why when someone comes across asking about making a commercial game with idTech 4, I make sure to be as open about as I can, so people make a right call. Commercial game dev is not modding. It doesn't even come close to it. If someone is starting from scratch, clean slate and looking to assemble a team, getting it done (along with the game) is possible with Unity / UE4 and maybe Source. It's almost impossible with idTech 4 / BFG (especially BFG in the state it is in at this time).

I'll tell you this - I had a long and very productive conversation with several id Software former employees, who worked on the tech up close and personal. Every one of them suggested to leave idTech 4 behind and use either Unity or UE4. Note how Carmack pushes Unity and UE4 and not fiddling with his own engine? That's for a very good reason.

So if you think I bad mouthing idTech 4 and id Software guys, including JC, don't know what they are talking about, then by all means you are the smartest developer on the planet and I shall leave it at that.

Motorsep sometimes i wonder how you made a game on DarkPlaces that has even less documentation and even less user friendly and modern tools.

Oh, it took me a lot of R&D time, then server side QuakeC coding a basic prototype of Prophecy game (3rd person hack and slash), then showing it off at QuakeExpo 2008 (yes, there used to be such online trade show, before Unity / UDK / UE4 choked Quake modding scene). After that I found a QuakeC programmer who was interested in helping me. We planned the project and figured out it was too much work. So we went with plan B, Steel Storm. Which took us 3 years to complete, including 5 rewrites of AI. The community was strong back then, and word of mouth was a way of documenting things. Tools were there, made by Quake community and DP engine author. Some where developed along the way, during production. It wasn't easy, but there was continuous support, because a lot of community wanted to see Steel Storm released, as it was first commercial game using DP engine (maybe even first commercial game using GPL engine).

As there were no ragdolls, no particles as in Doom 3, no entity defs, no bitmap fonts, no complex materials, the only thing that needed tools was models. So, Blender exported into ASE and OBJ, and I happened to have Doom 3 MD5 exporter, so we had a tool to convert MD5 into DPM. Eventually IQM model format came to DP engine, along with nice IQM exporter. So we really didn't need any tools. It would be nice to have WYSIWYG tools back then, but the complexity allowed to work without those.

If you are a coder no matter what engine you use you will not have any problem making a game, even on idtech 4, if you are only a artist then idtech 4 is not really artist friendly, compared to UE4 or Unity that is, but if you can't code or don't find a coder to help you will never make a game on any engine. 

That is a common misconception. As I said, id Software programmers much rather work with Unity than with any if id Tech engines, including id Tech 5. That should tell you something.

If you already familiar with id Tech 4 code base, sure it would be much easier and less work to add stuff to existing gameplay code base. However, there are soo many unfinished features that it entirely depends on what kind of game mechanics you are pursuing.

We got our engine to where making art for it is almost like making art for UE4 / Unity. As in until you get to import your art into the engine. Getting your art to show inside idTech 4 is a way more work than with Unity (haven't worked much with UE4 yet, so I can't say much about the process there).

Making a game using idtech 4 has cons but also some pros, like, you don't need to pay anything (if you don't want to release your engine source code modifications to the public you will need to pay id) and you have the full source code for the engine for free, and anything you make from sales is yours (and the publisher if applicable).

If you don't want to release you code, you are screwed - ID no longer licenses engines. Publishers stay away from GPL engines (Steam is not a publisher, they are distributor). They much rather have an ability to release on consoles than be stuck with PC only.

You don't pay anything if you are using Unity 5 either. All proceeds from sales of your game are yours (and Steam's or GoG's or Origin's; except only Steam and GoG would allow GPL engine).

You pay 5% from your sales if you use UE4. 5% is nothing, compare to the time savings, pool of available talent (who are willing to work on your game), available platforms and promotions you get with UE4. Plus, if your sales are only so many dollars per month, you pay nothing at all.

[rebarkillburst]

Torque is open source (MIT) and it doesn't include AI, but you can buy AI for it for under $100 (maybe $70).  I like torque.  I am slowing learning it because it does a lot of what I want and doesn't cost royalties. 

I like D3 tech and know it pretty well, but the lack of AI/weapon/script code vs other id engines turns me off to it.  IE you could take Darkplaces and make a whole game w/o touching the code. You could take Knightmare Q2 and remake HL1 if you wanted.  You could take Q3A (not familiar with those engines) and make a game from that.  With D3 you need lots of coding experience, or an experienced coder, to get off the ground.

That's the reason for the recommendation of other engines if you're not doing a mod, because the other engines have more support on the coding end.

Would using either darkplaces or those other engines be reasonable to make a commercial game? Heard that darkplaces is very versatile but isnt it based on the first quake?

[MrC]

Torque is open source (MIT) and it doesn't include AI, but you can buy AI for it for under $100 (maybe $70).  I like torque.  I am slowing learning it because it does a lot of what I want and doesn't cost royalties.

I like D3 tech and know it pretty well, but the lack of AI/weapon/script code vs other id engines turns me off to it.  IE you could take Darkplaces and make a whole game w/o touching the code. You could take Knightmare Q2 and remake HL1 if you wanted.  You could take Q3A (not familiar with those engines) and make a game from that.  With D3 you need lots of coding experience, or an experienced coder, to get off the ground.

That's the reason for the recommendation of other engines if you're not doing a mod, because the other engines have more support on the coding end.

Would using either darkplaces or those other engines be reasonable to make a commercial game? Heard that darkplaces is very versatile but isnt it based on the first quake?

Some more information about your project might help aid you in this process, if I understand correctly so far you want to make an something along the lines of an id game, is that old id or new id? I.e Doom/Quake fast combat or Doom 3 slower linear survival horror or hub based FPS with complex story etc...?

DarkPlaces is rooted in Quake 1, yes. It's also a mish-mash of Quake 2 and 3. If you're going to use DP you might as well take advantage of Q3BSP and Q3Map2. The use of lightmaps and realtime lighting (edited in-game) make it more performance friendly. Programming will be done using QuakeC which has been criticized for being limiting but it has also been greatly extended upon, it's also pretty easy to use and a good starting point for learning procedural programming. As for starter code, you get all the Quake 1 QC code with server side AI so if you're doing a coop game this is a big plus. Speaking of which, there are lots of cool QC resources to get you on your way, https://www.quaddicted.com/webarchive/minion.planetquake.gamespy.com/tutorial/main.htm
http://www.insideqc.com/qctut/
A few areas you'll find lacking compared to D3 is a more robust system for dealing with skeletal animation (blending layers) and root motion movement. Also, single threading vs. that nice multi-threaded jobs system that BFG does. Physics works through ODE but so far ragdolls are the topic of much debate, as in they're technically supported but there's no real system or anything past some early tests to actually take advantage of anything that complex - mostly a tools / docs / code related issue. You'll also have limited shader support, Q3 shader that is compared to D3's material scripts as most of the materials in DP are done through a naming convention of the image files themselves. I.e: rock1.tga, rock1_norm.tga, rock1_gloss.tga. If you want to reuse textures then you'll end up with a lot of duplicate image files sitting around taking up unnecessary memory - this has been one of my biggest gripes with the engine. Anyway for more information read dpextensions.qc included pretty much in every distribution.

Is DP a viable option to create a commercial product? Sure. The engine might not be the issue here, are you yourself ready to undertake such a task? Or do you think perhaps you should get in some modding time first?

If this is one one your first commercial projects I would say aim small and this is where the whole Unity or UE4 recommendation comes in. Follow some of their starter tutorials, perhaps take them a bit further and ship something. Unity has some very nice starter tutorials for all kinds of different games that help get you familiar with different types of controls and a game manager like a Space Shooter or Roll-a-Ball although with Unity 5 there's some outdated code in those tuts but a quick search will get you going again on some of the changed API.

[bitterman]

Hey, motorsep,

first you dream then you die

life is in motion, do not care achievements

I think idTech4 like first love - charm passes, but memories remain.
And begin to better understand women gamedev.

[rebarkillburst]

Torque is open source (MIT) and it doesn't include AI, but you can buy AI for it for under $100 (maybe $70).  I like torque.  I am slowing learning it because it does a lot of what I want and doesn't cost royalties.

I like D3 tech and know it pretty well, but the lack of AI/weapon/script code vs other id engines turns me off to it.  IE you could take Darkplaces and make a whole game w/o touching the code. You could take Knightmare Q2 and remake HL1 if you wanted.  You could take Q3A (not familiar with those engines) and make a game from that.  With D3 you need lots of coding experience, or an experienced coder, to get off the ground.

That's the reason for the recommendation of other engines if you're not doing a mod, because the other engines have more support on the coding end.

Would using either darkplaces or those other engines be reasonable to make a commercial game? Heard that darkplaces is very versatile but isnt it based on the first quake?

Some more information about your project might help aid you in this process, if I understand correctly so far you want to make an something along the lines of an id game, is that old id or new id? I.e Doom/Quake fast combat or Doom 3 slower linear survival horror or hub based FPS with complex story etc...?

DarkPlaces is rooted in Quake 1, yes. It's also a mish-mash of Quake 2 and 3. If you're going to use DP you might as well take advantage of Q3BSP and Q3Map2. The use of lightmaps and realtime lighting (edited in-game) make it more performance friendly. Programming will be done using QuakeC which has been criticized for being limiting but it has also been greatly extended upon, it's also pretty easy to use and a good starting point for learning procedural programming. As for starter code, you get all the Quake 1 QC code with server side AI so if you're doing a coop game this is a big plus. Speaking of which, there are lots of cool QC resources to get you on your way, https://www.quaddicted.com/webarchive/minion.planetquake.gamespy.com/tutorial/main.htm
http://www.insideqc.com/qctut/
A few areas you'll find lacking compared to D3 is a more robust system for dealing with skeletal animation (blending layers) and root motion movement. Also, single threading vs. that nice multi-threaded jobs system that BFG does. Physics works through ODE but so far ragdolls are the topic of much debate, as in they're technically supported but there's no real system or anything past some early tests to actually take advantage of anything that complex - mostly a tools / docs / code related issue. You'll also have limited shader support, Q3 shader that is compared to D3's material scripts as most of the materials in DP are done through a naming convention of the image files themselves. I.e: rock1.tga, rock1_norm.tga, rock1_gloss.tga. If you want to reuse textures then you'll end up with a lot of duplicate image files sitting around taking up unnecessary memory - this has been one of my biggest gripes with the engine. Anyway for more information read dpextensions.qc included pretty much in every distribution.

Is DP a viable option to create a commercial product? Sure. The engine might not be the issue here, are you yourself ready to undertake such a task? Or do you think perhaps you should get in some modding time first?

If this is one one your first commercial projects I would say aim small and this is where the whole Unity or UE4 recommendation comes in. Follow some of their starter tutorials, perhaps take them a bit further and ship something. Unity has some very nice starter tutorials for all kinds of different games that help get you familiar with different types of controls and a game manager like a Space Shooter or Roll-a-Ball although with Unity 5 there's some outdated code in those tuts but a quick search will get you going again on some of the changed API.

Well, yes, both old and new id as in fast combat, but with additional perks like different melee attacks, an additional iron-sight like look, and such. I'm trying to pull off a fps but with magic, spell casting, and such replacing weapons.  Like hexen, and such.

[The Happy Friar]

Ideally, with so many engines out there now a days, the "best" one is the one you know. 

With what you described want to do, Quake 1/Dark Places can do that and has everything you need minus art.  Motorsep started game based on a RPG-style game with Darkplaces.  I enjoyed several hours of MP in just the alpha that was released. 

[rebarkillburst]

Looks interesting, but it still looks kinda similar to quake. This is a commercial game, and thus it would look like a AAA-looking game. Does this engine support art assets at AAA or close level?  Does this engine support an expansive, large levels? It'll have mostly melee combat, but there's ranged weapons by means of spells, it'll have full player awareness, as in your character isnt just a floating gun or floating set of hands. That will allow you to finish off enemies when theyre down.  It'll have inventory, items to pickup, lore to read, which i am working on intensively. But it'll still be a mostly linear fps, any other kind of game like an rpg is too overwhelming to make at least as of now.

[The Happy Friar]

So you mean you want the graphics to be on par with the new Doom?  How many people are you planning on working on this? 

Darkplaces can support the texture quality of any modern engine.  Not sure about poly counts but I'd figure it would support relatively high.  After all, Source is built of the Quake engine.  :p

[motorsep]

Looks interesting, but it still looks kinda similar to quake. This is a commercial game, and thus it would look like a AAA-looking game. Does this engine support art assets at AAA or close level?  Does this engine support an expansive, large levels? It'll have mostly melee combat, but there's ranged weapons by means of spells, it'll have full player awareness, as in your character isnt just a floating gun or floating set of hands. That will allow you to finish off enemies when theyre down.  It'll have inventory, items to pickup, lore to read, which i am working on intensively. But it'll still be a mostly linear fps, any other kind of game like an rpg is too overwhelming to make at least as of now.

Anything with AAA class art assets will be overkill for you to make. Hell, Steel Storm has stylized assets and it tool me 3 years of mostly part time work on it to get it all done. Phaeton has stylized assets, but more complex than Steel Storm (more complex shapes, animations, some normal/spec maps, ragdolls, damage effects, etc.) and it's freaking close to impossible to work on part time.

And you think AAA-looking assets are not going to be overwhelming...  Good luck with that.

P.S. You'd be better off making interesting game (better off in a sense that you would get more people buying/playing it) with non-AAA look, rather than yet another linear AAA-looking game.

[rebarkillburst]

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.