EGADS #1

Solar2D General Questions/Discussion
#41

Oh I’ll have to try that as I’ve never played with shaders before.  

I have made an interesting discovery…  I apply an ellipse to indicate to the player the area of effect of the building (green in this screen shot)

Bizarrely, with this removed and with the colorMatrix code unchanged the frame rate only drops to 25 fps. Much better than 12 fps. I then tried inserting the circle into different display groups and I found that the lower down the display object stack I inserted it the slower the game ran!  If I didn’t place it in a display group at all the frame rate remains unchanged.

#42

Update: your shader shows NO slowdown and runs at 30 fps.  But it doesn’t handle transparency very well on 32 bit pngs and seems to really highlight edges.  Is this simple to change?

The colorMatrix filter handles semi-transparency.

#43

Just multiply something by texColor.a. Should be easy enough :wink: Try playing with shader in my last link as well.

#44

AH fixed it… It would appear I am somewhat out of my depth with shaders!

I’m trying to use this in my game - https://goo.gl/igLAlp.  But I am really struggling.  The code I have is this can you see where I’m going wrong please?

graphics.defineEffect { category = "filter", name = "water", isTimeDependent = true, fragment = [[P\_COLOR vec4 FragmentKernel( P\_UV vec2 texCoord ) { P\_DEFAULT vec2 uv = ( texCoord.xy / CoronaVertexUserData.xy ); uv.y += (cos((uv.y + (CoronaTotalTime \* 0.04)) \* 45.0) \* 0.0019) + (cos((uv.y + (CoronaTotalTime \* 0.1)) \* 10.0) \* 0.002); uv.x += (sin((uv.y + (CoronaTotalTime \* 0.07)) \* 15.0) \* 0.0029) + (sin((uv.y + (CoronaTotalTime \* 0.1)) \* 15.0) \* 0.002); P\_COLOR vec4 texColor = texture2D(CoronaSampler0,uv); return CoronaColorScale( texColor ); }]], }
#45

Works for me. Wavy wave. I think your problem is that you are not setting CoronaVertexUserData. It is set to 1,1 in link you provided doesn’t have definition of parameters… Here’s documentation on it: 

https://docs.coronalabs.com/guide/graphics/customEffects.html#vertex-userdata

Quick and dirty way to fix it - just delete “/ CoronaVertexUserData.xy”

#46

Thanks, yes I just discovered that…   I am trying to simulate water in my isometric environment and I thought that might work as it looked good in the sandbox.  Unfortunately, it doesn’t scale well to thousands of tiles and halved the frame rate.

Does the shader continue to use resources when the iamge it is allocated to is hidden with alpha=0?

What I’m trying to achieve may not be possible unless you have any genius suggestions?

#47

I’m not sure about that… Also, if you want to synchronize all your water, you may attempt to somehow use world position for wave displacement, rather than UV. It’s little more complicated. May be someone would be willing to help you with that.

#48

Good luck! We can’t wait to see what you come up with.

Rob

#49

Best of luck Ed.  Looking forward to seeing what you come up with.

#50

Dude you have way too much time on your hands… but much kudos for you!

#51

It has been a week and I am about half a day behind schedule.

I will be releasing the first of the games (" Limited Ammo’) on Google Play tomorrow.

I ended  up prototyping about 30 games and ran into a weird issue.  I had a number of winners that I felt needed more treatment and polishing than the schedule would allow.  So, I had to pick and choose the lesser of the prototypes to take to full game.

The good news is there are several real gems in my prototype batch that I am looking forward to giving extra attention before I publish them.

Again, first game tomorrow “Limited Ammo”.  I’ll try to get back on schedule, but I may end up staying one-day behind per-release.

-Ed

PS - @adrianm - I only have time on my hand because I don’t have enough paying clients right now.  A huge bummer and difficult financially.  So, I figured.  Why not throw myself at making games really hard.

#52

Hey Ed, if you have a spare 10 mins then a real simple plugin with a single function that returns total RAM and available RAM would be extremely helpful. Although this really should be part of the SDK.

#53

+1 for the memory plug-in, definitely worth $5-10. I’d like to be able to store as much of my game DB in memory as possible on devices that can handle it - no point restricting it to 150MB on a 16GB desktop for the sake of steering clear of trouble on a £50 android tablet.

#54

EGADS #1 - Color Flick (Google Play) 

colorflick.png

This game is a very simple one-touch color matching flicking game.  Try it.  I think you’ll understand.  I think it starts a bit easy so I may ramp up the difficulty on a future update.  

This game utilizes a basic ‘jump right in’ game framework (no main menu), Vungle ads (waiting for ids to propagate so these may still show test ads till tomorrow), AdMob banner, and IAP for no Ads.

It also uses a ‘watch this ad’ to continue mechanism when you fail/die.

EGADS #2 is coming shortly and it will be “Limited Ammo”.  Sorry for the order mix up.

#55

EGADS #2 “Limited Ammo” is working its way through the approval process now: (Google Play).

limitedammo.png  limitedammo2.png

Next up, “Momentum Defense” (working title):

#56

Color Flick is a fun game, probably could be a little more challenging from the start, but I didn’t want to put it down.

Rob

#57

@everyone

I have decided to terminate this experiment.  I prototyped about 30 game mechanics, published two of seven games, and nearly finished a third.  However, I have realized that while fun and interesting, this effort was a mistake.

If you have played the games I produced so far, you’ll probably agree, that while there might be a kernel of goodness in them, they are not very fun and they don’t suck you in.  

They are a success, for the fact that I completed them with several notable features (game mechanics, interfaces, sound, ads, and IAP).

They are also failures because they won’t make any money.  Sure, this effort wasn’t really about money, but contrary to @adrianm’s quote…

Dude you have way too much time on your hands… but much kudos for you!

I don’t really have any free time at all.  That is, I don’t have time I should be spending w/o the thought of earning a living.

So, again, I have decided to terminate this experiment.  However, I want to close this thread on a positive note.  So, I will share two videos below.

EGADS #3 - Momentum Defense

This is the game I was working on before I terminated the experiment.  As with the other two, it has something interesting going on, but it just isn’t fun.

https://www.youtube.com/watch?v=cEYivr7WPZo&feature=youtu.be

EGADS Completed Games & Prototypes

This video shows game play from the first two (finished) EGADS games, plus about half of the mechanics prototypes.  Cheers!

https://www.youtube.com/watch?v=vsXqNsmhgeY&feature=youtu.be

PS  - Yes, I experimented with almost every transition in my video editor’s feature set.  Sorry, but I wanted to try them out.

PPS - I flipped a couple of the prototypes on their side (into Portrait layout) in order to record the video.  So if you were thinking, “That looks wrong…”, you are correct. :slight_smile:

#58

Hey Ed,

I think you’ve nailed everything that is right and wrong with game development.  The upside is Corona makes it real easy (especially if you have existing libraries) and games can be knocked up/reskinned in a matter of days.  The downside is that leads to disappointment as the games get almost zero downloads and no revenue.  

This is not just a Corona issue: unity, game maker, cocos, etc. also have this issue too.  This gives indie development a bad rap and we are often portrayed as not being real and/or serious developers in the media.

I’m sure someone of your considerable talent and experience could create a great (and most importantly profitable) game.

Adrian

#59

Thanks!  I should say, I’m not sad about this.  It was an experiment and it failed in some ways and succeeded in others.

That said, many gems came out of this and deserve more time than 24 hours each.  So, I decided to stop early and get to the business of game development.

Cheers,

Ed

#60

The whole purpose of an experiment is to learn what works and what doesn’t and there are many failures before you get to a success. It’s like the old saying “If at first you don’t succeed, try, try again.” 

Maybe something that’s more marketable is 7 mini-games, one release.

Rob