next »

[Ishi]

Also this week...

World Map facelift

I was bored of looking at the world map's extremely placeholder artwork, so I spent some time polishing it up nicely. Last time I showed you the world map, it looked like this:



And now it looks like this:



I'm really happy with this, and can't wait to create more of the map in the future!

More Music

As I mentioned before I've requested more music for the game from Woofle! I forgot how long it takes to write up a spec for all the music I need. I also wanted to provide her an updated build of the game and inevitably came across various bugs while putting that together.

This week I'll be getting into some new level design at last. I also want to plan out tasks for the next few months of work, but right now I'm just going to play some Zelda.

[JobLeonard]

That new overworld map is  

EDIT: and the scankey support is fantastic! One of those "levels of polish" things that few developers seem to bother with, but that makes all the difference to people on international keyboards.

[ProgramGamer]

Map facelift is indeed pretty dope!

[Ishi]

Sinking Platforms

A nice simple devlog entry this week. Here's a new gameplay mechanic!



If anything is on the platform, it starts to sink. This includes Leilani, enemies, powerups, basically anything that's a physical object in the game.

This really makes good use of the game's moving platform code, I'm really happy I got that system working nicely!



I also improved support for attaching objects to moving platforms - for example the spikes here. I can attach spikes, blocks, collectables, and probably other stuff to moving platforms, which offers a lot of flexibility.

[io3 creations]

If anything is on the platform, it starts to sink. This includes Leilani, enemies, powerups, basically anything that's a physical object in the game.

So, if there's a powerup on a "sinkable" platform and the powerup isn't picked up, the platform could potentially completely sink?  That could be an issue if Leilani can't progress because of that.  If powerups disappear after a while or when submerged, then that could solve that issue.

[Ishi]

If anything is on the platform, it starts to sink. This includes Leilani, enemies, powerups, basically anything that's a physical object in the game.

So, if there's a powerup on a "sinkable" platform and the powerup isn't picked up, the platform could potentially completely sink?  That could be an issue if Leilani can't progress because of that.  If powerups disappear after a while or when submerged, then that could solve that issue.

Yeah I should watch out for this. The lava does destroy powerups so it's not an issue - apart from the Fire powerup which floats on the surface and would currently prevent the platform from going back up. So I do need to be careful that it doesn't cause any weird situations. Thanks!

[LyricalReverie]

How long do you think it will take for a beginner to beat Leilani's Island, Ishi?

[Ishi]

How long do you think it will take for a beginner to beat Leilani's Island, Ishi?

That's an interesting question! It depends on so many factors that it's really hard to give a proper answer - the major unknown being how many levels there'll be in the game But, a vague target I've had in my head for a while is 40 levels, mainly because Wario Land has 40 levels.

So for fun let's imagine the game has 40 levels. Maybe the standard path to the end of the game would go through 30 of those levels? Typically the levels I've made so far take 2-3 minutes if played through quickly, so a figure of 5-10 minutes seems a reasonable guess for a first play through a level. Multiplied by 30 levels that would result in a playtime of between 2.5 - 5 hours, which would be pretty good I think. You could add on extra time for travelling around the world map, playing the combat challenge levels, etc. And then even more time for finding collectables and secret exits and stuff.

All hypothetical of course

[JobLeonard]

Done any playtesting yet?

EDIT: with people who haven't played the game before I mean

[Ishi]

Done any playtesting yet?

EDIT: with people who haven't played the game before I mean

Yep I have, I recently started sending out builds to people. Very small scale at the moment - reading and responding to feedback, and implementing changes, can quickly take a lot of time! So giving the game to tons of people feels like it would be a bad idea for me. But have got some great feedback just from a couple of people playing through the current build, making some little tweaks that will improve the game. Mostly things like areas where I can give better feedback about what's happening to the player, or just make certain hitboxes more fair, etc.

I'm not gathering any stats such as play time per level or anything, for now am just happy to hear people's thoughts and ideas about the game.

[JobLeonard]

Is the game deterministic? If so you could try logging inputs to analyze replays

[HiddenNatureDsn]

Love the new world map.  Been interested in this one for a long time.  Looks wonderful.

[Ishi]

Is the game deterministic? If so you could try logging inputs to analyze replays

The game isn't deterministic thanks to variable frame lengths and floating point numbers generally. But I do have some recording + playback functionality, which I tried to make closer to deterministic by fixing the frame time to 1/60s both when recording a playing back. Fixing the frame time could make the game run at the wrong speed depending on monitor refresh rates and stuff... there are ways around it, I could disable v-sync on the game so it's not tied to the monitor refresh, or I could avoid fixing the speed during recording and just make sure I play the recording back using the same frame lengths that it was recorded with.

That was a bit rambling (it's late!) but I guess the point is, yeah I could record stuff, and it's probably quite a good idea! But requires a bit of time investment to make it work nicely, and then there's even more time investment involved in actually watching the recordings once they exist. Certainly worth considering though, thanks!

And thank you HiddenNatureDsn, also thank you everyone else who posts encouraging comments, each one is read and appreciated even if I don't get around to replying to each one individually!

[io3 creations]

Regarding recording time, other than finding out gameplay, there is another benefit.  If selected, Unity can collect user data.  One of the uses is creating usage heatmaps that can be used to track player progress.  That way, you can find out if many/significant number of players spend time (or may get stuck and give up) at a certain location without them having to report directly. 

[JobLeonard]

Yeah I guess games that involve physics are rarely deterministic, unless explicitly set up that way from the start

The side-effect of recording stuff is that you could re-use it an in-game feature too.

For example, Mario 3D Land (and presumably other recent entries in the series too) had this thing where you could replay a level and see replays of other people's as Mii ghosts. It was basically a pseudo-multiplayer for racing the course. However, the more interesting part of it was discovering alternative routes or secrets this way.

[LyricalReverie]

How long do you think it will take for a beginner to beat Leilani's Island, Ishi?

That's an interesting question! It depends on so many factors that it's really hard to give a proper answer - the major unknown being how many levels there'll be in the game But, a vague target I've had in my head for a while is 40 levels, mainly because Wario Land has 40 levels.

So for fun let's imagine the game has 40 levels. Maybe the standard path to the end of the game would go through 30 of those levels? Typically the levels I've made so far take 2-3 minutes if played through quickly, so a figure of 5-10 minutes seems a reasonable guess for a first play through a level. Multiplied by 30 levels that would result in a playtime of between 2.5 - 5 hours, which would be pretty good I think. You could add on extra time for travelling around the world map, playing the combat challenge levels, etc. And then even more time for finding collectables and secret exits and stuff.

All hypothetical of course

Sounds good. Thanks for responding! I dig short games, so that sounds sweet for me.

[Louard]

Regarding recording. Probably more feasible to simply ask testers to record themselves playing on video. Possibly even on their phone such that the screen is visible. You could also have them "think out loud" while doing it.

I think input recording would be too time consuming a feature to implements if the main use would be watching tester replays. Every game I've helped on which has had such a feature turned out to be more work and more error prone than expected.

Alternatively, if you did want to do ghost replays, etc. A less error prone approach might be to log the position of all dynamic entities at each tick. That way, the length of each tick doesn't matter (though playback might look funny during framerate dips.. but the tick length could also be logged)
This approach would also be less prone to invalidating replays because of physics tweaks etc.

All I can say for sure, is that video captures of gameplay from testers has been invaluable in the improvement of levels for Suzy Cube.

[Ishi]

Regarding recording time, other than finding out gameplay, there is another benefit.  If selected, Unity can collect user data.  One of the uses is creating usage heatmaps that can be used to track player progress.  That way, you can find out if many/significant number of players spend time (or may get stuck and give up) at a certain location without them having to report directly. 

That kind of telemetry could potentially be useful, although I'm not using Unity and don't have time to spend looking into those sorts of systems really. I don't think Steam has anything like that built in, though would be happy to be proven wrong!

...

...

Thanks for your thoughts! That kind of ghost replay mechanic isn't really on my radar, especially as it would involve some online functionality which I'm extremely averse to programming :D

Since I have the input-logging replay functionality already in the game I may spend a little time to make it save replays during tests and see how well it works. I'll give it some thought. Louard it's good to know how useful you found the videos, seems like I should at least request that people record their playthroughs once I start doing some more serious playtesting.

I quite like the idea of saving out replays automatically as then people won't be conscious that they're being watched

[io3 creations]

I quite like the idea of saving out replays automatically as then people won't be conscious that they're being watched

Yeah, that can be a valid reason for some people no to record.  That, or not even having recording software/system/etc. 

Not sure if it was mentioned, nut even those people would likely provide various feedback (bugs, questions about certain things) so it's good to have an easily accessible way to provide feedback.  Either an on-screen button or "Press X button to provide feedback" type message.  It's also good to keep it on screen or perhaps in the Pause Menu if the game screen is small and you don't want to have it crowded.

[Ishi]

Devlog

A quick proper devlog update since I didn't post one at the weekend.

I've completed the level with the sinking platforms. By completed, I mean that it's playable from start to finish - but will surely need improvements, removals and additions! I'm always happy to get a full level blocked out in a basic form though, I find it much easier to analyse and tweak a level that already exists (whether it's based on playtesting, or just reviewing the level myself after taking a break from it) than it is to create the level in the first place. Those blank canvases are really tough to fill.

I've begun work on another level, the second one for December. I'm hoping to fit a third level in after this too.

Here's a screenshot, which doesn't give much away about the new level.



Apologies if you're bored of seeing the same old beach + ocean backdrop, most levels won't be using it in the final game, I just haven't got around to making more backdrops yet. Staring at placeholder stuff is a big part of game dev