next »

[ANtY]

Probably not gonna have any enemies, it's not a combat-focused game. Even for stuff like jumping on enemies' heads you need the pure control of something like Mario. I'm planning more environmental hazards, and just general exploration. Not dissimilar to Knytt.

Would like to do some wildlife though. Lots of interesting animated plants, and maybe some small animals and stuff. The player won't want to explore if the environment feels lifeless.

Want to play this already 

[SolarLune]

Probably not gonna have any enemies, it's not a combat-focused game. Even for stuff like jumping on enemies' heads you need the pure control of something like Mario. I'm planning more environmental hazards, and just general exploration. Not dissimilar to Knytt.

Would like to do some wildlife though. Lots of interesting animated plants, and maybe some small animals and stuff. The player won't want to explore if the environment feels lifeless.

Sounds good. I'd think that maybe just a single type of enemy that chased you, though, would add a little haste to your actions. You wouldn't have any means to counter-attack, after all, and a single hit would mean death. It'd be an interesting, "Get out of the room quickly!" kind of challenge. Really, any kind of hazard like that would be interesting.

[Ishi]

Want to play this already 

Excellent, thanks for saying so!

Sounds good. I'd think that maybe just a single type of enemy that chased you, though, would add a little haste to your actions. You wouldn't have any means to counter-attack, after all, and a single hit would mean death. It'd be an interesting, "Get out of the room quickly!" kind of challenge. Really, any kind of hazard like that would be interesting.

I'll bear that idea in mind. Environmental chasing is also a possible idea along similar lines (rolling boulder, rising lava, something like that). The game will probably need at least one such sequence to provide a bit of a change of pace. Hostile creatures could provide a good puzzle element (similar to the Oddworld games - the creature will back off from you unless you corner it, then it'll attack. That kinda thing.)

I'm not sure how prevalent I want death to be in the game yet. A major feature of dying that I've not decided on yet is how to handle respawning. I could go for Knytt-style save points (maybe a statue or something), or invisible continue points, or put Trixie back at the start of the "room" (if I do end up splitting the environment into obvious rooms). I welcome any thoughts on this matter.


Update

Unlike yesterday, today was really productive. Feels good.

I refactored a lot of the entity stuff - XML definitions, placement system, save data, in-editor rendering, etc. Before it was all designed only for scenery entities, because it was quicker at the time. Now it can handle all kinds of entities.

There's a new tab in the editor palette for Interactives. Used the flower platform as the logo because I couldn't think of anything else.



And, started doing the flower as you can see. It's just a first pass at the sprite for it. The stem will be automagically created out of multiple sprites based on where the floor is.

So now it's all ready for me to code the actual flower entity, with behaviour and collision. Future entities will be quicker to code, today has been mostly foundations.

I also did a bit of painting -> pixel conversion practise, on a pic of Trixie I did a while ago.



A bit messy in places, but I only spent maybe a couple of hours on it, so consider it a first pass. I could have done more anti-aliasing, but I think for such a low resolution game, AA can make things unnecessarily blurry.

I don't have a particularly special method for the conversion at the moment; I just scale the image down, start pixelling over it, and eventually I've covered most of it. Then I weedle out all of the slightly-differently-coloured pixels that I haven't overwritten.

For this image I then put the pixel and the original side-by-side and tweaked the pixel one to make it feel the same, as some of her expression had been lost (not that she has much of an expression in this pic).

[happymonster]

Lovely colours Ishi, and the progress sounds great. Are you now using XML for all the data then after your refactoring?

[SolarLune]

As for your save points, were you thinking about splitting up the world a lot, or going for a more open-world type of experience? Invisible save points is fine if the world is a little more linear; I think it might be a little confusing if crossing a certain line changed where you would respawn from when you died.

The pixel portrait of her is outstanding. It's quite good - the only problems I have with it are that the green of her pupils are quite bright in comparison to your sketch - maybe just add a little more black at the bottom (or top) of them? Also, the pink on her cheeks starts very harshly - I think I would dither that a bit, but the rest is fine.

Great colors indeed.

[Ishi]

Tech-Tastic

Are you now using XML for all the data then after your refactoring?

I was using XML before. For example, the various scenery objects are defined in SceneryDefinitions.xml:

Code:

<?xml version="1.0" encoding="utf-8" ?>
<SceneryDefinitions
	spriteDefinitions="Texture/Scenery/Scenery.xml"
	>

	<Definition UID="3">
		<Strip name="Grass01"/>
	</Definition>
	<Definition UID="4">
		<Strip name="Sand01"/>
	</Definition>
	<Definition UID="7">
		<Strip name="Reeds01"/>
		<Strip name="Reeds01" type="single"/>
	</Definition>

	<Definition UID="1">
		<Strip name="Bush01"/>
	</Definition>
	<Definition UID="2">
		<Strip name="Bush02"/>
	</Definition>

	<Definition UID="5">
		<Strip name="HangingPlant01"/>
	</Definition>
	<Definition UID="6">
		<Strip name="HangingPlant02"/>
	</Definition>
	
	<!-- Next UID = 8 -->
	
</SceneryDefinitions>

(Strip refers to an animation, which is defined in a seperate XML that's used by the cAnimatedSprite class. Reeds has two strips; the first one uses the first frame of the anim as a static sprite. The second is a single play of the animation. This makes them animate at random intervals - because that's hardcoded into the cScenery entity at the moment. Eventually proper delays and stuff will be defined in this XML. Anyway...)

The refactoring today involved making a base class cDefinitions, so cSceneryDefinitions (which loads and stores info from the XML above) now inherits from that. So I can easily add more classes - for example, cInteractiveDefinitions which loads InteractiveDefinitions.xml:

Code:

<?xml version="1.0" encoding="utf-8" ?>
<EntityDefinitions
	spriteDefinitions="Texture/Interactive/Interactive.xml"
	>

	<Definition UID="1" paletteStrip="Collapse" />
	
	<!-- Next UID = 2 -->
	
</EntityDefinitions>

The two sets of definitions are used a bit differently. SceneryDefinitions specifies various appearances and animations for just the cScenery entity. On the other hand, each entry in InteractiveDefinitions refers to a totally different entity type (just cCollapse in this case, the collapsing flower platform).

So they share a bit of functionality, but it's enough to make it worth inheriting them from a base class. The main bonus of that is that adding a new tab to the editor's palette becomes really quick, because the palette never interacts with cSceneryDefinitions or cInteractiveDefinitions - only with the base class cDefinitions, and the same will be true for new tabs - so the code's already written.

(As a matter of fact, the palette doesn't interact with cDefinitions either. It interacts with another class called cBrushPaletteCategoryInfo, which cDefinitions inherits from and implements. This is so the class that handles the tile materials can also implement cBrushPaletteCategoryInfo, so they can be displayed in the first tab in the palette. But that doesn't affect the bonus mentioned in the previous paragraph. cBrushPaletteCategoryInfo just provides the functions necessary for displaying, and allowing the user to select, all the items in one category/tab of the palette

Code:

class cBrushPaletteCategoryInfo
{
public:
	virtual int BrushPaletteItemCount() = 0;
	virtual int BrushPaletteItemWidth(int index) = 0;
	virtual int BrushPaletteItemHeight(int index) = 0;
	virtual void BrushPaletteRenderItem(Bitmap &bitmap, int x, int y, int index) = 0;
};

)

I bet you're glad you asked!

--------------------------------------------

As for your save points, were you thinking about splitting up the world a lot, or going for a more open-world type of experience? Invisible save points is fine if the world is a little more linear; I think it might be a little confusing if crossing a certain line changed where you would respawn from when you died.

I'm aiming for an open-world type thing at the moment. I also think invisible save points would be confusing; the player would want to be sure of what the penalty for dying is when doing risky stuff, so they need to know where they'd be returned to. Good to know you're thinking is the same!

Thanks for the comments on the portrait too, think I'll do another pass on it now.

[baconman]

A little more size in the pupils would make pixie-Trixie looks far less zombified. And curious, what would constitute the losing conditions of the game? Or are there any at all?

[Relix]

How about save anywhere but get teleported to nearest "save" point (like in Zelda)?
And if you die, you get thrown there too.
But it really depends what kind stuff you want to include.

Also,

Want to play this already 

this.

[Ishi]

And curious, what would constitute the losing conditions of the game? Or are there any at all?

Nah, no game over or anything. You just return to the respawn point when you died and carry on.

How about save anywhere but get teleported to nearest "save" point (like in Zelda)?
And if you die, you get thrown there too.
But it really depends what kind stuff you want to include.

I think it'd be simpler to save when actually at the respawn point just to avoid extra UI work. Save points seems like the best way to go; they're clear, and it gives me control over where the player can save, so I can balance difficulty better (compared to an automated per-room save or something).

I'd actually not really thought about actually saving your progress in that sense. Visible save points is definitely preferable when it comes to that.

[Ishi]

Update

- Improved the flower graphic
- Entities can now have more than one sprite
- Flower uses multiple sprites to create a stem (but - the stem length is hardcoded at the moment, instead of detecting where the floor is)
- Started adding collision between entities. Very basic solid collision at the moment to allow Trixie to stand on the flowers. I need to add some event functions and stuff so the flower can react when something lands on it.



And since I haven't posted it in this thread yet, the slightly tweaked pixel portrait from yesterday.



(The pupils probably should be a bit bigger yeah baconman. A bit small relative to her eyes at the moment. Will keep that in mind on future drawings)

[happymonster]

Thanks for the technical breakdown! 

Trixie standing on the flowers looks nice indeed. Have you tried making the blue sky more of the pale blue in the portrait background? It looks slightly too overpowering at the moment to me.

[Ishi]

I haven't really thought about the background much yet, it's just a full-screen placeholder image that I threw in at one point. The pale blue is nicer though, yeah.

[Ishi]

Update

Haven't done anything the last few days for various reasons.

Tonight I started a collapse anim for the flower platform.



Just a first pass, I need to look at it freshly to see where I can improve it. Quite pleased with it so far though. I want to make the animations on big interactive objects quite fancy so they're fun to interact with.

[Relix]

I was expecting the flower to totally collapse, not flip out :7


Sorry if I sounded too negative ;__;

[Ishi]

It needs to un-collapse too, so it seemed better to tilt it over than to destroy it. It's like the flower's defense against predators, or something.

[happymonster]

In that case, the tilt does make sense..

[Ishi]

Update

Not a massive update today, wasn't working for most of it. I tinkered with the flower anim a bit, but didn't get very far. I think I need to start it again and plan it out better to get the feeling I want from it. I should feel fairly floaty and big and soft. The previous WIP anim I posted is very stiff and too immediately bouncy.

Sketched some potential interactive object ideas.

Did a Treasure Treasure drawing for fun.

[baconman]

^ That made my morning.

[Ishi]

Update

Sorry for the lack of updates, I've been busy with other stuff recently (mainly getting Johnny Platform Saves Christmas ready for PC release, and starting an 8-films-in-8-days Harry Potter fest). Tonight though I did a tiny bit of work on the collision stuff, so I could do a post.

Will try and get back into the coding groove over the next few days and make some progress worth showing.

I've kept my sketching up even when not coding. I find it easier to sketch if I'm tired after work (since I program all day), or if I'm watching something at the same time. It helps keep me involved in thinking about the game. Just thought I'd mention that in case anyone wonders why I have a pretty high art output when there's plenty of coding to be done.

[Ishi]

Update

Having another go at the "Morning!" cutscene pic.



Not convinced by the background shading yet, though it works as a quick and easy way of covering a large surface.