Search This Blog

Thursday, September 29, 2016

The long road to ruin... or something better?

So as you are aware my on again, off again blog is *MOSTLY* centered around the Game-Guru engine.  I specifically became a backer during their kickstarter-style launch when it was called 'FPS Creator, Reloaded'.

I'd heard bad things about FPSC in the past but figured the demo videos were good.
I covered that about two years ago in a blog post on my old FPSC Reloaded blog here:

At the time, the cost was 80 dollars/100 dollars. I got in for 80, I think. At the time, I was short on cash; this was a big deal for me to spend on a hobby.  Now I think Game-Guru retails for something absurd like twenty dollars on steam.  Is it worth it? Well, that's a complicated question.

Let me bring you up to speed first on a few events.  First back in August Lee went on vacation.  Lee Bamber is the primary coder for Game Guru.  It's his baby and he is really the only one qualified to do significant amounts of work on it.   After the vacation we got a token update around early/mid August then.. nothing.  No communication.  No nothing.  Just a blank hole with no updates.  The natives got restless.  They made thread after thread asking for an update.  Nothing.

A lot of us have been real patient; it's been almost two and a half years for me and the engine's development has clearly stalled.  Early limitations were hit due to bad decision making (using Dark Basic, for instance) resulting in big setbacks (A complete recode in C/C++), incomplete feature sets like AI, Lighting, Particle effects all leave us reeling and having to scramble to make a functional game or demo.

Then... Lee came back.  About two days ago.  The discussion has been very lively, to say the least.  I encourage you to take a look at these two threads and get involved.

There's a lot of real honest evaluation by Lee, some critical introspection.  Some good questions and a lot of struggle over what's the right answer.

One thing I agree wholeheartedly on is getting rid of the community voting board.  That thing is a curse.  It has way too many options, often many of which are flatly unnecessary!  I mean come on; what's the point of having things that can be done by community scripting taking up valuable cycles when we don't have a good core engine to work from?

I would love to see this product actually reach the goals it intended.  We're way behind and I realize that at this point they made enough mistakes to probably not achieve what is needed without a major influx of capital.  So could it be the end of Game Guru? Possibly.

It could also get a renewal.  Often with hard questions come good answers so I hope Lee and his team find the ones they need to continue this project and bringing it to fruition.

Monday, September 12, 2016

Technique Training: Fog and Environmentals

I was originally going to bake this in with my Advanced lighting series but this is more of a subject unto itself.  One of the more powerful elements within Game-Guru for creating environmental effects is the fog system.  The fog system was originally... laughable:

Picture courtesy of  Lee Bamber's now extinct dev log.
As you can see it was just as simple coloring of everything in the background.  By comparison, this is the fog system from the way past dead Game Creation System from the mid-late nineties:

Literally the same system in practice.

That said, at some point, Lee realized this was a titanic error and fixed it to provide a significantly better fog system.  We ended up with something vastly better:

This is an old picture, but shows the lower limits of what's possible.
So as you can see this has a major impact on the function of environmentals.  This game engine has very few 'modern' features.  It's important to stretch everything to it's limit so that's what I'm aiming to teach you to do!

Realistic nighttime!

So with that we approach our first technique; realistic nighttime.  In game-guru, nighttime effects are actually quite difficult to achieve.  This is primarily because the sky system has a constantly active light which will frequently circumvent your efforts.  This light is fixed in space and you can turn down ambient/shadow values to zero but your game will still fall short of achieving that desired 'nighttime' look.

The trick, the real trick, is to use fog.  If you set all the values to 1 for RGB, set the range start up close and the far distance close, then modify the intensity as desired it will create a black fog which accurately mimics the feel of night and it's creeping darkness.
As you can see from this picture:  it looks silly with a daytime sky.

But if you add a nighttime sky, it all falls together.  The best part is it accurately sets up flash-light fall-off.

As you can see, it looks far more like night.

Distance simulation with Fog

One common mistake I see in Game-Guru games is that they very frequently don't factor in standard atmospheric effects.   This in turn, makes their own job HARDER.

Not a bad city depiction, texture misalignment aside.

So this scene looks good.  It's a basic city done with some simple city models someone made.  Overall, not terrible.
Compare this against GTA IV:
Can you guess the difference?
The difference is atmospheric haze.  A slight atmospheric haze provides a level of reality to any city scene as most any place in the world has at least SOME slight atmospheric haze.

Haze is defined as: "Traditionally an atmospheric phenomenon where dust, smoke and other dry particles obscure the clarity of the sky."

What you'll find is in almost every scene, from top to bottom at least a slight tinge of fog on the outlying distance (use your total visible range in an average scene, add fog to the last 15-20%) will have atmospheric haze.

Don't overdo it, or you'll look like silent hill.
Trivia fact: did you know the first Silent Hill's trademark fog was actually a way to disguise the draw-distance limitations of the PlayStation 1 and the respective game engine involved?  By implementing a heavy fog they accomplished both an incredible atmosphere and also improved game performance.  That's win-win!

When implementing your haze, you want to keep in mind that time of day matters; remember haze is effectively a reflection or light being blocked.  As the time of day changes, so too does the haze.

Not a game-guru picture.  Just some random image from google.
As you can see above, the relatively nearby trees have a SLIGHT orange tinge (even in the upper branches) caused by atmospheric haze and ambient lighting.  The lightrays further illustrate the haze, though lightrays on Game-Guru are totally dictated by the aforementioned fixed sun.
This is more what you're striving for:

From Twin Worlds - Note the orange haze, matching the sky perfectly.
As you can see, these settings are pretty simple stuff. You can do quite a lot with them.

Further examples(settings included):

My city test - day, with Atmospheric (normal) coloring:
If you further modify your depth of field settings you can achieve a more pleasing result, but this fairly decent for a quick example.

Preparing for a more night-oriented effect.

This is a longer range darkness, but still dark.  I left the cloudy sky in for contrast.

Now is a really dark version - high intensity.  Distance remains the same.

Notice how the red light is no longer visible.
Maximum darkness.  Close range.  No flashlight.  Dark sky.  

I had to move close just so you could see anything!!
Sunset Sky, Sunset fog for atmospheric haze.

My favorite, personally speaking.

 Silent hill, just because :P

Just add pyramid head.

As always, feel free to comment.  If this has been useful to me, please post a link to your results below!

Current status

Uhoh, I'm getting lazy again.

Ok so things have been really intense.  I blame Legion, the newest Xpac for WOW (World Of Warcraft).  Now I've been out of love for WOW for quite some time.  Draenor (the previous expansion) was like the final nail in the coffin, I thought. I thought wrong.

Legion I had NO hopes for; it was, at best to me a novelty.  A bad novelty by a failing franchise.  It is, however, a massive surprise.  It's done exceptionally well.  Class balance is superb.  Class feel is superb.  Class order halls - superb.  Daily quests are random and interesting - they scale as well.  Terrain is impressive and varied.  As a part time game dev, I see this stuff and take mental notes.  I've played more than my share of WOW.  I know what the previous expansions were.  They bored me to tears, in many cases.

I don't know what happened over there at Blizzard, but hell yeah.  That said, it's affected other things.  That and other life stuff like having to clean out my basement and ~8+ years of literal and figurative shit (chicken coop in the garage is gone, finally!).

I have done a few things though. I've written up my next installment on lighting in Game Guru but also have had a few other things hold my interest in no particular order:

  • Master of Orion.  A classic.  I've been part of the early access stuff and watched it come along.  It's a great game, buy it :)
  • A review will be forthcoming on Dagored's sci-fi kit.  Let's just say that I had a realization when watching the video after nearly a year of being on the fence and wham, instantly pressed 'buy now' when it clicked.  This is an AWESOME kit.  I'll try to get this done in the course of the week.
  • A lighting write-up is nearly done.  I need to take screen shots though to supplement.  It will cover things like 'making a night level that looks like a night level', 'creating the illusion of distance with fog', 
  • An epic shitstorm in the Star Citizen community.  Interesting to read and watch.  Especially from a gamedev/indy perspective.  I hate to say I was right to my close friends who heard me predict this happening years ago but - cheers, I was right
So there we go.
Incoming stuff.  Eventually.  If I ever stop playing wow (The ARMS WARRIOR CAMPAIGN IS FUCKING AMAZING ERMAGERD DROOOL) ...