next »

[marcgfx]

Thanks sbeast! No, I have enough music for now.

What I really need is marketing skills
Greenlight started out ok-isch but it's pretty much dead already. Although every one has been saying its easy now. So either my game sucks, there is no interest or no one knows about it. 0 media presence is probably not that helpful

[marcgfx]

I'm still working on bosses and tons of other stuff. For this boss I invented circular laser beams, to go along with the standard lasers. Biggest issue here is collision detection with the circular laser. I could not think of a way to do this efficiently, so I ended up just checking for the players units. This of course means that anything non-player is not affected, including the hexagonal structures.


I've updated the hexagonal structure to be more dome like. The original version (top image) was more or less a random distribution, but it did not look interesting. I feel the new look is visually more appealing. To do this, I also needed to add more images for the structures, going from 25 to 40 images. I've also added a lot more dust when damage is caused.


Greenlight is not really doing very well, but it's likely just due to no views. So.... I've got my own twitter account (@falkenbrew), but have yet to post something. I am not the accustomed to using twitter, so I am a bit scared of messing up

something like this maybe?

Devader needs your help on steam greenlight #gamedev #greenlight
http://imgur.com/a/TfmEd

edit: I made a gif, uploaded it imgur, but I dont think you can link to it from tigsource??

[io3 creations]

edit: I made a gif, uploaded it imgur, but I dont think you can link to it from tigsource??

Actually, yes you can.  Just make sure you're using the image url and with the img tag:


After uploading an image, you can either Right Click > Copy link location or there should be Direct Link on the right side (might need to click the image to see the links)

[marcgfx]

@io3 creations: thanks man, I could not get it to work, but now I can see what you did by quoting you! very much appreciated!

http://i.imgur.com/6qdReEI.gifv

edit: removed the gif image...

[io3 creations]

Oh, looks like the forum formatted the tags that I didn't meant to.  Anyways, it's good to hear that you got it working.

[marcgfx]

@io3 creations: I understand the trick now, it's just no longer a gifv, but a gif and that means the file is huge

new gifv showing off the assistants and turrets
http://i.imgur.com/lxpaBAh.gifv

[marcgfx]

I have been in a bit of a slump lately, just not quite happy with a lot of things. I've been wondering for a while what I can do about the power-ups, but maybe someone here can give me a good kick in the right direction. Some players have complained that the extras are not obvious enough (how they function).

Currently I have one type for weapon-power ups (the oval one) and one type (barrel) for health/shield/turret pickups. The only difference to signify the content is the hue I use for the power-up.


I quite like it this way, as its obvious after a short while, that it is actually something you can pick up. You have to learn what the colors mean based on the pick-up sound effect and the change to the weapon. All weapon power-ups can be combined with each other and are visible on the back of the players robot.

barrels:
red: turret
blue: shield
green: health
white: attachment (little drones next to the bot)

oval/weapon power-up:
yellow: spread shot (+2 bullets slightly to the side)
orange: explosive (+2 bullets on impact, random direction)
magenta: homing shots (enemies close by to the shot are detected and homed in on)
cyan: power (larger, stronger bullets)
purple: bounce (shot bounces off anything it hits)
lime: tree (the shot splits into 2 after a short time, multiple upgrades of this type result in a tree like structure)
brown: bend (shots bend a little in one direction and have larger range, a bit of an anti-power-up sometimes)

Any ideas how I could improve this?

[marcgfx]

I needed to do something more fun. Really not what I should be doing.

I started out by making a grid like structure, built up out of seperate objects. They are initialized together, but there is not grid structure holding the data. I used a sin function to alter the height based on position/time.


To make it look more organic, the carpet was rounded. The height value is now calculated using the position in the non existant grid and time. I also added some horizontal movement.


The objects now know their neighbours, there is still no grid structure. It was a bit more complicated than anticipated to get this to work. If neighbours are destroyed the hue is changed.


Every edge node can fire. The fire command itself is sent from the central node and transmitted to all neighbours. The transmission is not instant, so you can see different nodes firing at different times. This also means that destroying nodes changes the firing pattern.

[marcgfx]

Currently mostly working on GUI stuff...

The new selection screen for Nuke or EMP


Selection screen for choice of Turret


I still need to add some text, so the player knows what the different choices really mean

[marcgfx]

Still working on Devader, even if progress seems terribly slow. I am still slaving away at GUI issues (html should be easy, but it's just not meant to be handles by gamepads...).

Game play wise the changes have been rather subtle. Today I changed the weapon power-up handling. So far the power-ups just timed out after you picked them up, no matter how often you fired. Now I've switched to an ammo based usage. This forces the player to think more about how to use the weapons he has.

[marcgfx]

I received two new images for the startscreen

Original


Version 1, silhouette


Version 2, eyes


Version3, larger logo with small changes


What do you think? I'm still a little on the fence about the new graffiti stile, but I like that it's not as sterile as what I had before.

[Zorg]

I'm voting for #3 with a less hard gradient in the logotype.

Quick overpaint:

[marcgfx]

Thanks zorg, that might be a good idea  . Will have to do some experiments, as I will be animating the whole thing too.
The graffiti style text is not to everyone's liking, so that might be revised too.

[marcgfx]

opinions to the new revisions?

@zorg I didnt manage to make the logo lighter like you did... how on earth did you do it, without the black getting lighter?

[Zorg]

I just used the fill tool without soft edges and scaled the image down so the bicubic reduction would cover my crappy work.



What i wanted to say: it does not matter which colors you use for the gradient in the letters, but the darker value should not visually merge with the outline. Test how your logo looks in grayscale and black/white in small sizes.

I think it looked more interesting with the gradient, just balance it with the readability (grayscale, black/white).


PS: Die letzte Revision finde ich am besten. Aber pack ruhig wieder einen Verlauf rein.



Muss auch nicht unbedingt grün sein.

[marcgfx]

thanks zorg!
I missed that I posted in german , has been corrected
Adding a gradient is probably a good idea, so far most people prefer the less playful cleaner font better.

[marcgfx]

I'm still working on Devader, even if there has not been much activity here. I'm pretty much done with the basic game, but I seem to be saying that to myself for a long time already. The newest feature that creeped in is screen-shake for explosions. Nothing overly wild, but it does mean updating nearly every single object in the game...

Art Update
I received some new art and this makes the end-of-level-screen look a lot nicer


I also have a start screen that I am actually happy with now


Laser Issues
Then I did a lot of bug hunting. For some reason my lasers sometimes passed through objects and it took me quite some time to figure out the culprit.


I first accused the bresenham algorithm https://en.wikipedia.org/wiki/Bresenham%27s_line_algorithm

As you can see, there are small gaps where the line misses the squares. So I updated my algorithm to use "Supercover" http://eugen.dedu.free.fr/projects/bresenham/

But this did not cure the illness. At least, I now knew what it wasn't and thanks to the change lasers and line of sight have improved. I eventually managed to narrow down the problem to an accuracy issue. What I had been doing: traverse the grid with supercover, find the first object that gets hit. Now find the point the laser actually hits the object. In some cases it didn't actually hit it... just because of some minor calculation errors. After adding some error correction, the lasers now work perfectly.

Localization
I added capability to localize my game from the start, using my own custom code. I am actually quite proud of it, but I don't know if there is some kind of standard for javascript localization atm. I might consider a github project, but my git game is not very strong.
So currently I am writing everything in english and german, I will have to find someone to localize in other languages though. If anyone has good prior experiences... :D

Feature Creep
I've added a hard mode for the game, as I am finding it easier and easier myself. I am no pro-gamer, so that might be a red flag.
I've added the original robot with a tank body. It requires to be driven like a racing car, making for an interesting change in game play. I removed it a while ago, as most people will just get nowhere, but it didn't require much to add back in.

I am thinking about adding support for multiple controllers. The game can currently be played by keyboard/mouse or gamepad. I was thinking about letting someone else join in, by simply pressing start on the gamepad. For this to work, I need to identify if someone is using the keyboard. Otherwise a player switching from keyboard to gamepad on his own, would suddenly have two bots to control. Am I overthinking things?

Target Platform
I am considering releasing on itch.io , GOG and Steam, mainly because I kind of want to go through the entire process... Not sure if that is a good idea though.

[io3 creations]

I might consider a github project, but my git game is not very strong.

You know what they say:  You gotta Git Gud!

I've added a hard mode for the game, as I am finding it easier and easier myself. I am no pro-gamer, so that might be a red flag.

I am thinking about adding support for multiple controllers. The game can currently be played by keyboard/mouse or gamepad. I was thinking about letting someone else join in, by simply pressing start on the gamepad. For this to work, I need to identify if someone is using the keyboard. Otherwise a player switching from keyboard to gamepad on his own, would suddenly have two bots to control. Am I overthinking things?

I haven't been playing much but don't recall anyone mentioning how they'd handle this situation.  The only similar thing is right before the game but then it's obvious that there's only one player.  If the game is already playing and the player presses the Start button on the controller is to show a menu:
-One player mode: switch controller
-Two player mode: Second player joins game
-Cancel (in case someone accidentally presses the Start button)

I am considering releasing on itch.io , GOG and Steam, mainly because I kind of want to go through the entire process... Not sure if that is a good idea though.

That's a good question.  Each will probably have their own API so depending on your engine, you'll likely need to keep track of three separate version.  Might be better to start with two and then add the third later?

[marcgfx]

I probably should git gud  Maybe someone would be interested in my localization code, or my gamepad HTML-GUI support. If I'm lucky it can generate some interest in Devader I've been neglecting marketing terribly, so it would be desperately needed 

I've decided to move co-op to later. If by any chance there is a demand for co-op I can still add it. Showing a menu sounds like a fairly sensible choice, thanks for the suggestion!

Platforms: You're right I should probably concentrate on less, first thing I want to do will be Steam. I can add GOG, Itch or any other platform later.

I just created a new test build and am hoping for some feedback from my small group of testers. I'm mainly worried about Gamepad implementation, as I only have a XBox360 Controller and I really suck at using it.

Thanks for coming by

[marcgfx]

I just came back from Gamescom in Cologne on Saturday. First time I've ever been an exhibitor and I was a little nervous to start with. I did not know what to expect. The first day was really hard on me, instead of sitting in silence for 10 hours, I was standing for 10 while talking. But it was a good start and I met and talked to a lot of interesting people.

Not everything went well. First off I was presenting on an iMac from 2011 with similar GPU power to my MacBook. So I ended up presenting on the lowest settings I had available. I also had to hack in some code to reduce the amount of scorch-marks as the fill rate of the graphics card seemed to be reached rather quickly.
There were also the occasional bugs and I spent much too much time explaining to people what to do. This resulted in further midnight fixes and improvements in minified javascript code. Not the most enjoyable way to pass the time.

Most notable changes since I last wrote here are sand and multiplayer. But I don't have a screenshot of the multiplayer, so at least I will show you the sand.