How come the Source engine never caught on with Devs like the Unity & Unreal 3 did...

RPGMaker as shitty as it is today requires some scripting skills, SJWs can't onto that.

RPGMaker doesn't require scripting skills at all. Everything most RPGMaker games use is 100% done through the wizards/UI, and you can install third-party scripts that do complicated stuff for you, there is no need for end-users to do scripting, and very few people do.

I tried making a test map for TF2 once, the thing about the maps needing to be closed rooms was confusing, not to mention it doesn't seem to have a basic terrain editor, you literally need to model your terrain on a 3rd party program and then import that to source.

Source is surprisingly unoptimized for just about anything. On my computers that should be able to handle just about anything source throws at them they still get frame drops. Just look at fucking Vindictus. Source is the fucking reason I don't touch that damn game.

RPG maker is so easy. Even people with fart fetishes and edgy hipsters could use it.

Come to think of it.
Shocked that Fart Waifu game not steam greenlight yet

No, not yet, this is Valve after all
DOTA 2 is using an early build of it right now that changes nothing visually.
The VR games they made also use it.
Lots of fanfare but nothing substancial

I heard it was one of the shittiest editors out there.
No wonde rmost games still used vanilla assets.

Dark Messiah gave me hope and not just for the engine, but alas nothing good lasts long.

...

Fart waifu"s has a lot good original art. Cute original art that too good for a game about girls farting. Than again Fart Fetishism is the one fetish that don't get a lot of love in anime.

Terrible support and documentation for development.

Vampire: The Masquerade was built on an unfinished version of the engine and launched the same as HL2, and has always suffered bugs.

Even the games that did kind of OK in development like Zeno Clash ended up moving on to Unreal for their sequel.

The engine really is only good for making shooters in the vein of older game design methods involving lots of indoor hallways made out of static meshes. Once you try to go to big open spaces and fill levels with lots of high poly models, the limitations of the engine and the development tools show themselves.

Source 2 seems to have been built specifically to give better support for games that aren't hallway shooters. Who knows if it can actually compete with Unreal 4, CryEngine 3, or even Unity, but I think we will soon find out.

Somehow the farts are less disgusting than rpg maker

The engine really is only good for making shooters in the vein of older game design methods involving lots of indoor hallways made out of static meshes. Once you try to go to big open spaces and fill levels with lots of high poly models, the limitations of the engine and the development tools show themselves.
There's EYE doe.

Because post 2005 Valve didn't care so they destroyed the engine with further updates with no planning beyond what they needed for that one game they were working on.

EYE has some large locations but hey aren't very geometrically complex and detailed. Not compared to most modern games anyway.

Ur a faggot

Ask Ross Scott

Such tools can only go so far.

source has a pretty shit licensing terms for budget devs and the tools for the engine have been shit for a long time.

Unless you're putting 50 particle effects in the same room or not making hint brushes to cut areas off from each other then it's fucking easy to get a stable frame-rate. Also the compiler taking a long time is a result of the map NOT being optimized because if it was optimized the compiler would have a clear idea of which portals can see what.

That's a holdover from Quake/Quake 2 engine. They never bothered to update the BSP system.

This is due to no one but Valve actually fucking understanding how to make maps for it.

Hahah,,, heheh… oh god you don't even know.
You think you can just throw a couple of cubes and a skybox together to make a basic map, right?
Oh god no, not at all. You will need to optimize the fuck out of those five cubes if you want VVIS to render the map in less than a hundred hours.

There is a reason Streum On abandoned it when working on Spacehulk.

Oh I do, oh god do I do. The thing is that they only go so far. You want to make a cyberpunk city, you want to make a large outdoor area? That shit will still take multiple hours to go through VVIS not to mention VRAD.

Confirmed for never making a large map.
Furthermore unreal uses similar BSP architecture and doesn't need to take three years for its VVIS equivalent. What's Source's excuse?

Because it's ugly as shit? That's all I know about it. Well that and it runs like shit too.

Hammer does have a terrain generator, but it is absolutely fucking garbage.

You know how to properly define detail brushes right?

So that you don't spend 48 hours on VIS, right?

It has shit performance that's embedded into the very core of it, for it uses technology that goes by the name "Binary Space Partition". Without going too technical about it, it uses clever search algorithm to only find polygons that actually show up on the screen, saving on GPU rendering. And it actually works very well, it can figure out if the polygon is blocked by another polygon and it will not continue search if it knows none of further results will be visible, so only a handful of polygons are sent to be rendered in a busy scene and especially indoors. Back in the day, GPUs were fucking slow and you had to be really deliberate with what you were rendering, doing shitload of CPU work was offset by huge boost in GPU performance. However nowdays GPU rendering speed is enormous, the time it takes to render a polygon is a minuscule fraction of the time it takes for CPU to ask GPU to render it. And since BSP naturally submits polygons to GPU one by one, it's performance is hugely bottlenecked on the CPU side. Even batching all of that shit together doesn't save the situation. Today it's far superior approach to simply use large models and make simpler, less sophisticated polygon culling algorithms, because the more time you spend culling that shit off rendering queue, the more time GPU just sits there doing absolutely nothing - not only does your framerate tank because you're not fully loading the GPU, the CPU performance also suffer.

Everyone and their mother can use a func_detail.

source has used several different iterations of updated bsp's, but the 2010 BSP2 is the best version of BSP around. Valve sticks to their own standards instead, but Source 2 looks like a regression in this regard. Much less brush focused, much more prop and mesh focused.

It's called fog with v clip brush, using vis-clusters to combine useless portals also dividing the outdoor area with area-portals at choke-points

toplol

tbh if your vvis is taking retard amounts of time, chunk out your map better because that's just a sign of poor optimization on the designer level. vvis is flawed but its easy to make it go from 3 hours to 3 minutes.

Except for the fact that Unreal engine 3 ran better and the current version of Source is from 2013.
Furthermore:
Once again it can only take you so far. Oh look now I can have a huge amount of empty space that will lag all to hell the second I add more complicated brush geometry!
One thing I will give Source though is at least the fog does look nice.

then where's the sequel ?

Exactly

It was merely a cash grab testing the waters and both the gullibility of the audience and the influence the gaming press holds over them ,i'm not sure any of those one-hit-wonders thinks the same trick would work a second time around and they seem to be aware of it thus the lack of sequels on any of these best-selling tenouttaten emotional rollercoaster of interactive experiences

Also remember Dear Esther is actually switching engines to Unreal 3.

...

Wrong. They switched to unity because it's cheaper.

Daily reminder that you're not supposed to make visible geometry with BSP brushes.

The compiler is the same as it was in 2004 only with support for the extra graphical enhancements
Also you're forgetting the most important thing with hammer to make the compile a breeze… it's to get good.

...

...

I need some fucking sleep man

Despite the fact that most Valve games are built near entirely with brushes?

Yeah, func_detail or walls to block off LOS, if you look at de_dust in CS:GO and take away everything that isn't bsp then you're left with the most blocky level in existence

Yes, it's a shitty approach, see

The only way to get Source games run at adequate framerates with respect to amount of geometry displayed, is to use 3d models only for geometry, and only use invisible BSP brushes to cull invisible models.

Source does not scale well on modern hardware and has serious problems with higher texture loads

Its the same shit you deal with when you play blizzard games, run great on toasters, terrible on powerful hardware at highest settings

Unity/Unreal engines are designed to scale with whatever you want to fucking throw at them, which is why they've had untold millions thrown at them in order to develop them, and they're easy to program for/create content for

Too bad using models in Source is complete shit because you can only have like 7 hitboxes applied to them so anything sufficiently complicated ends up having full retard interactions with players and projectiles.

Gunsmith Cats

It's okay, user. We'll all sleep eventually.

Furthermore unlike literally every other current engine Source uses a proprietary model system that requires a shitload of work just to import one model.

Oh yeah, forgot about that shit too. What a pain in the ass.

Not quite as bad as Torque 3D, but close.

Goodnight user

I didn't need that feel.

thechineseroom went on to make other stuff you retard

Here's how to make a model for Source.
It should now work fine! Thank you, Bravo Newell.

Making games with a modern Quake sourceport is now officially easier and more ideal than using source. You even get not shit licensing if you're okay with the GPL.

The SDK wasn't really flexible.

You got a source on that?

Years of making Quake mods. If you need anything more than that I don't know what to tell you but none of these problems are a thing. Please jump back over the border Carlos.

I think the issue is with map making. Sourceas far as I recall accepts .bsp which is kind of shit.

Would it be possible from a leagal standpoint to make a full-conversion game with the free quake engines and sell it?

BSPs are an old but not bad method of developing games. It's very good for quickly making lots of progress in a short amount of time, while sacrificing a little bit of geometrical complexity/ This can be remedied by mixing with meshes. Not ideal for a large studio but great for a small team that just wants to get shit done.

Do you have any documentation or information on quake modding that you could maybe provide me? Would making a third person game in it be possible?

...

Yes but the source code must be GPLed. If you want to be a jew you can write 100% from scratch Quake C code and not release your gamecode but having sourcecode available for people to use will greatly extend interest with whatever you're making, especially considering mods are fucking DEAD in gaming now.

This is an interesting question, I think it's legal to sell the complete package but you still need to give source code. I know Forty flights of Nothing was sold and that's based on a sourceport of the Quake 2 engine. I think the open source Idtech engines were all released under GPL too so I also guess any derivatives are also GPL?

Unrealfags needs to die. Fuck you, your shitty lod, mess of collision meshs and price hiking 3D artists.

Yes, firstly because it's open source and secondly because someone has already made an X-Com clone using Idtech 2.

Because the bullshit that went on with Troika

The thing that complicates it further is the question of third party tools. I don't know anything about Quake modding, but I know that Doom modding is heavily dependant on free third party stuff like DECORATE and SLADE besides the map editor.

Slade is GPL, but even then it shouldn't matter what tools you use to make your content right? as long as you have a licence to use those tools and they are generally free to use for things like this.

There's lots of great 3rd party tools you can use. It doesn't matter what tools you use though.

GPL works well for games because unlike programs you can't just compile it and then play it because the assets are copyrighted.

Yeah, but I like to make my structures in AutoCAD then have a first-person walk in a quick build. And going from AutoCAD to functional .bsp is hell in Source.

Unity on the other hand accepts .fbx, no questions asked. Honestly, that was my determining factor in choice of engine to fuck around with. Might not be for others, but I suppose there's more than few people that ran into this issue.

Source is limited by the scope of games intended to be possible in it, and the engine licensing terms.

This is partially due to the fact that it was so heavily developed with a focus on FPS games that don't need much scalability by design (FPS or MOBA games with mostly small, static maps, combined with primarily close combat and small quantities of players).

Whenever I see other game genres explored in the Source engine, it looks like quite a bit of tedious work would've been required to achieve that, compared to other engines.

Unity, GameByro, the newer Unreal, etc, focused on a larger scope right from the beginning.
The incentive for people to use these engines is higher because they see that it is designed with a far larger range of game genres in mind.
Not only that, but with things such as the asset marketplace for Unity, people are shown that they can achieve a certain type of game with far less effort, just by using someones pre-made plugin. This is a key detail that comes to mind, because I noticed that as soon as there was a voxel engine addon for Unity, a shit-load of voxel games started popping up.

I think Valve are very aware of this difference. It's very evident in the fact that they already have games made using Unity instead of Source, and they're focusing on Unity quite heavily for the Vive…and not even just the generic marketing of "hey Unity works great with Vive", they're actively giving Unity devs free Vives and developing a better render engine for Unity.

Personally, I think it would be interesting if Valve purchased Unity, and then merged the Source engine with the Unity engine.

Valve have engine developers and Unity and Source are completely different engines with different goals in their design, merging them would probably achieve nothing. What features does Source have that Unity doesn't, BSP? wew laddy.

You make a good point. I should've been more specific.

I don't think it would be a "make Unity more like source engine" sort of gig.
There's a few things Valve have done better in their rendering side of things (which they're already developing a way to integrate into Unity), but primarily a lot of their stuff is dated.
I think it would be "make it easier for us as a company to transition over to Unity by merging some of our engine in it" sort of gig.

Side note: I have nothing wrong with BSP as a map format. Many wonderful maps were made in it, but yes of course, there's other more scalable solutions that exist now that make BSP look very dated and primitive. It still does a good job though, for those more primitive scenarios.

Valve is already working on Source 2 an engine we know almost nothing about that probably fixes a lot of peoples problems with Source, even then Source and Unity are opposed in that Source is designed for professional developers who will end up fiddling with the engine's source code to achieve results whereas Unity is script intensive and users rely on it's ultra versatile C#/Java based scripting engine to achieve results.

Although my use of Source engine is limited so if it is scripting based and I just don't know, umm, I still disagree in that it would be a waste for all the work on Source 2.

And even given that it would probably be easier to just upgrade Unity without integrating Source Engine's code.

I think you're right actually, I'll have to re-evaluate my thoughts on who would be the likely company to buy Unity if that were to happen.
The scripting side of Source, eg. Lua, seems mostly prominent when people want to make mods of mods…eg. all those custom game modes and assets for Garrysmod. So yeah it does seem that most commercial devs just work directly along-side with the source engine code, rather than not touching it and only ever throwing scripts at it.

This is the part I'm interested about, as I am curious whether this is just them wanting to push VR, or if the new engine truly will be able to live up to the track and jog along-side other engines in terms of scalability and being able to approach a good range of game genres.

I'd have responded earlier but I had to go to the bank.

1) Learn to love the BSP. Really, they're limited in comparison to modern technology but just making a brush or two and then copy/pasting them with a few keystrokes is really nice.

2) Pick an engine. Your best option is to pick Darkplaces or FTE and stick with it.

3) Gitgud at Quake C. THere is no way to learn Quake C, you basically have to know C and then fuck around with miscellaneous things in Quake enough to the point where you know what you're actually doing. There is a reference manual online though if you want to see that. pages.cs.wisc.edu/~jeremyp/quake/quakec/quakec.pdf

4) Trenchbroom is the greatest thing that has ever been made. Learn it to make maps quickly but it's also a good idea to learn Radient. If you want to work with Q3 format BSPs Radient is the way to go.

5) Scour IRC channels and wait hours for responses because very few people jnow a damn thing about Quake C.

6) If you're actually competent at programming and by that I mean better than 99% of the faggots that make video games these days say fuck it and use Idtech4 instead.

Ganbatte~!!!!!!!!!!!!

I never found it then.

Actually the VR games are running in Unity at the moment

...

How is this on youtube?

I don't get it. Can someone explain to why Fart fetishism exist and what so Sexy about Girls Farting?

Literally every game feels like a Half Life 2 level. That's why. Same with the xbox 360-era Unreal Engine. Shit felt like cheap crap.

...

...

Yeah, but wheres your game though?

Valve supported their engine very poorly early in development so it didn't catch on.

Not to mention Unreal engine is way more versatile than Source.

...

That easy, but first let's talk about parallel universes.

If it's shit it's not free.

Who needs a game when you have a sweet engine?

who needs a engine when you can have game