Games that had the most profound/immersive experience for me are the ones I go back to. If I could one day make a game that gives someone the same feeling I had when playing Pokemon Blue, Lufia, or Freespace for the first time, I'd consider myself very accomplished.

Mm, I had a killOrphans() function in my first physics engine. That one gave me a good chuckle as well.

I like the motion blur version most, feels more arcadey.

I also agree that the difficulty curve is backwards, the first couple of waves are frustratingly difficult. I believe that the circle you're defending could stand to be a bit smaller too.

The main complaint I have is that bullets don't act how I expected them to; you have this orbit mechanic where you're tethered to the circle, but your bullets fly exactly where you point them. You're completely missing the potentially awesome mechanic where you could slingshot bullets around the circle using its gravity to hit enemies that are otherwise out of reach. I feel that the game would go from being just "neat" to being very entertaining if you implemented this.

I'd really be surprised if that cities issue was enough overhead to cause slowness in this sort of game, but depending on your setup you could early out by the maximum possible size of a city, or barring that, maintain a hash table of territory occupation as a grid, and just check which cities overlap with new sections of the grid that have been expanded into.

I actually had a game concept once that did the opposite of this. Players weren't informed that the other players were actually controlled by people, and it would be described as a single player game.

Haven't touched this piece of code for months and it's never given me any trouble. Suddenly starts behaving weirdly when I try some new input for it.

Tracking the issue to the source, I find this.



Drunk me strikes again.

I laughed but it was bitter.

Heh, reminds me of the source code to one of my really old projects. There were badly made dry humor everywhere.
By the way, is that Visual Studio (other than 2012)? If so, where can I find that theme?

It's actually Flashdevelop, sorry to disappoint! It really is a delicious theme though, I've been using it on every fresh windows install since the start of 2009.
You can get the config file here if you want to harvest the colours and set up your own visual studio theme!

Also the piece of code in question? It literally blitted the entire text box to a buffer, cut it out, then pasted it lower so it appeared to be aligned to the bottom of the text box.

So if you put too many lines in at once, it just overran and printed half the text only.

Haven't touched this piece of code for months and it's never given me any trouble. Suddenly starts behaving weirdly when I try some new input for it.

Tracking the issue to the source, I find this.



Drunk me strikes again.

I laughed but it was bitter.

One thing that I really like doing, is every time I finish or scrap a project, I transplant useful code out of it into a framework I start each project on. That way, the framework always gradually expands with things I know are useful. It started out as just a finite state machine, but now it has physics, graphics, memory management, debugging, netcode, a math library, all manner of string manipulation, a text renderer, input management, etc.

This way even failed project feel like they yielded something productive!

Unless you're planning on building an engine or designing something enormous (which you shouldn't be at this stage), I'd suggest you prototype your game in the quickest and dirtiest manner possible.

Once you've got it prototyped, you can start thinking about what the code actually needs structure-wise, and work from there.

I personally find this works a lot better than wasting time making EVERYTHING expandable and dynamic and optimized, and then finding out that most of it is not necessary or even useful for the project at hand. It stretches development time unnecessarily.

A debugger for my game framework. Pressing ` hijacks the game onto the debug viewer regardless of what state it's in. The debug view has an input section for commands.

This is a great exercise it becomes really useful when you add watches, breakpoints and some form of scripting

Yeah, it's been really fun over the weekend hooking all the other parts of the framework into it (for example, you can now change and reinitialize states in the FSM), and then at the end of it opening up another project which depends on the same framework and suddenly it's ludicrously easy to debug.

Today I Created: A save-log-to-disk function to save all the output (might be useful for tracking bug reports from players) and two additional flags on the hook command; one causes the hooked method to only be available if the game is compiled in debug mode, the other causes the hooked method to not show up in the "help" command. Centralized cheatcode support: done~.

That looks fun, but those drums are so default superior drummer samples it hurts.

Recent monster database update:

I agree with John, that font needs a lot of work. I read "helder" and "mone".

I've been stalking you around the board a bit, as well as reading up on your project, and I'd like to offer my observations.

First up, your writing style reminds me of a few people I've worked with and learned to avoid working with; all of them have crashed and burned. Essentially, there's no need to take everything as a challenge or brag about your project's outcomes. It's unspoken that you take on a lot of challenges as an indie dev, and why would anyone work on a project that they don't believe has potential? You end up writing yourself into a corner of expectations with very few allies to help you out of it.

Second, as a (previously) avid WC3 gamer, I think you're overstating your potential fanbase a bit. I've heard a lot of people say that their map is the "second most popular" to DotA, but I've never even heard of Battleships. Maybe it wasn't popular in Asia/Pacific, I guess, but the map's boards only have about 500 threads on them. The DotA boards probably get that in three days.

I'm not trying to bring you down, here. I'm trying to point out how much wiggle room you have, and that you don't need to do this whole carrying-the-world-on-your-shoulders thing.

Reading your game's rules and mission statement, it looks like it's quite similar to other DotA-esque games on the market: Kill the enemy heroes, buy items, kill the enemy base, etc. The big runners in this field are League of Legends, DotA 2, and Heroes of Newerth, and all of them have something in common -- iterative development.

When each of these games came out, there was almost nothing to do. I remember being in the alpha for HoN and there was almost no hero variation between games, because there were so few picks. There's a reason for this. This is what the fanbase of these style of games is used to. When they first came out as WC3 maps, there was little to do in them, and it was gradually patched in.

You can reduce your scope. All you need to release a beta is enough ships to have a game and enough items to fill an inventory. I can see that Battleships has at least some kind of fanbase because there are competitive matches on youtube, which means all you have to do is attract this fanbase to your game and they'll let you know what the priorities are.

Tl;dr drop the NDA crap, drop the me-against-the-world act, get a basic version of your game out that's playable and iterate to make something. It worked for LoL, HoN, DotA and DotA 2, there's no reason it can't work for you.

Sorry that I haven't read the whole thread but this looks fantastic. I played a game with a similar idea quite a long time back, a Java multiplayer game called Astro Battle, but it was far too glitchy to be enjoyable. I'd love to see it done right!

Your movement model makes me nostalgic for Subspace Continuum -- which is definitely a good thing.

A debugger for my game framework. Pressing ` hijacks the game onto the debug viewer regardless of what state it's in. The debug view has an input section for commands.

Anywhere in the code, you can trace out data to the debugger by calls to the static Console class. Also, anywhere in the code, you can hook local methods onto commands for the input section, complete with arguments.



EDIT: I also made the font and font renderer from scratch because why not.

A game I was thinking about making before I got started on my current project. I may come back to it after this game is done.



1985: The space race between the USSR and the US continues to intensify. Both countries set their sights on Mars. Before a human colonization is attempted, The Soviets initiate Operation Medved Galaktiki, a program to send a massive ship full of 300 bears to Mars, so that they can be studied to determine the feasibility of a human colony. Sadly the ship loses contact with Earth as it leaves the atmosphere. With no method of controlling the craft, it drifts far into space.

1999: The bear ship crashes into an unknown planet. Although the ships hull is torn apart, the atmosphere is breathable. The bears venture forth on to their new home.

4063: Millenia later the bears have evolved into sentient creatures. Curiosity leads them to establish their own space program and return to the blue dot rumored to be their ancestors home. A crew of three of the most talented pilots begins their journey, not knowing what, if anything will await them on this mysterious planet.

The game would be an adventure, somewhat like Zelda, set on a ruined Earth, with the player learning more and more about what happened to the human race as the game progressed.

With the right atmosphere this sounds amazing. It could be so mysterious and bittersweet.

EDIT: I think for maximum impact, earth should have collapsed sometime shortly after the present day, to make sure as much is relateable as possible.

There is one RTS like that already - Stronghold.

Yeah, and it suffers from the problem of too many resources. Most Stronghold players just lay it heavy on the catapults.

The best Stronghold player I've faced simply created a lot of apple farms, sold all those apples, bought armor and swords, and sent a few dozen swordsmen on me within 5 minutes. He simply reduced the game into a few optimal resources and everything else was just noise.

It's pretty rare for competitive RTS play to not end up like that, though. See: Zerg rush, Paladin rush, et al

Spoilers for the Freespace series here:

Defending the GTD Galatea in Freespace 1 just about had me in tears for the whole mission, you just knew that there was no way out of this one. Then the debriefing after the mission when your squadron commander tells you that Wolf went down with his ship.

The Freespace series is a really great example of telling the story of a losing battle, everything just feels so hopeless no matter how well you win the minor victories.

Also, Freespace 2, the Chekhov's gun set up when Kappa wing goes missing during a patrol mission, and then during a testflight weeks later, one pilot comes back out of nowhere utterly hysterical.

I think anyone who wants to know about atmosphere in storytelling needs to play through this series.

I really love Kira Kira


And Deardrops


Two unrelated (until the crossover) but set in the same universe novels about indie rock/pop/punk bands. The latter of the two was the first game I completed entirely in Japanese!

But yeah, I love the characters, I love the sound tracks, and they go a long way to inspiring me in how to (and how not to) act in my real-life bands.

Too bad the sex scenes have a relatively developed case of Give Me The Chocolate, Hisao syndrome.

http://www.medicaldaily.com/articles/9496/20120411/alcohol-solving-skills-analytical-thinking-creativity-study.htm

I highly suggest this book: Thinking, Fast and Slow

Basically, people repress their instincts with active thinking. First few chapters point out why thinking can be blinding.

Alcohol represses this urge to thinking about it, leaving only the part of your brain that wants to do stuff. Someone who has manners and good grammar drilled into them becomes even more charming with alcohol. Someone who is instinctively crude will have their manners suppressed by alcohol.

So, it can make you more creative if you're already good enough at it that you can do it without thinking. A good writer becomes better under alcohol. A drunk 10 year old won't be able to write anything.

I'm glad there's some science behind it. My drunk code is always confusing as fuck but works perfectly.

Also I have some of my best ideas while smashed.