[GM] Lunacy Star

[Taizen Chisou]

Hey, this is my first post here, and I'd like to advertise my game, Lunacy Star?

I'm an avid Touhou fan, so it's going to play similarly. If you don't like that, well, then, sorry.

--- COMPLETE GAME CLEAR ---
http://www.youtube.com/watch?v=u9tKlnTKY8I
Audioswapped for a number of reasons.

Latest distribution 0.8 click here

~THE GRAPHICS ARE GOING TO STINK THOUGH BECAUSE I CANNOT ART AND I AM SORRY~
~I AM ALSO AN AVERAGE COMPOSER AND DON'T BLAME YOU FOR MUTING YOUR COMPUTER~









Arrows to move, Z is shoot, X is bomb, Shift is focus, and C is hypers mode.

I did a little bit of digging in the "developers look here" thread and I just wanted to say that this probably isn't completely typical of GM shmups :V
A few notes on things from the "New Developers" thread:

I have recieved quite a bit of feedback before that it is not too easy. Then again most of the people playing it don't shmup. But Jaimers did manage to clear Standard Mode on his first try, so we'll see. The main deal is that the bullets are rather speedy.
Inappropriate ship speed: It suffers from that "diagonal is faster" error that I'm trying to remove to little success.
No healthbar. No inertia.
Slightly out of whack scoring system. More details here.
There aren't any sort of outright "popcorn" enemies until the fifth stage, so there might be a slight hiccup on enemy durability.
The longest boss battle is the final boss, at about 6 minutes combined. Every other boss takes two to three, and stages quickly tap out after two minutes. Stage 1 has a 40 second stage and a 1 minute boss battle, while stage 5 takes two and a half minutes and the boss also takes two and a half minutes.
This adds up rather well to a total game time of about twenty seven minutes.
No cutscenes.
Bullets don't all entirely mesh with the background, and backgrounds are absent from the fifth and sixth stages anyway because like aforementioned I can't draw very well.
Dead air is mostly caused by players dying at odd times. Otherwise there will be little of it.
Weapons are across a spectrum. Homing missiles to me are a waste of time, but they cleave through the third stage really fast. And then there are lasers and missiles and rapid needles and satellite bullets.

Something to take note of:

You can change the ship and shot type with the left and right buttons. There are three ships with different bombs and different shot patterns and two different subweapons.

And finally:

There is a changelist in progress, but the forum I have posted it on has temporarily disappeared. I will add it somewhere here once it comes back.



Please tell me what you think!
Open to all criticism / comments / complaints, etc.

[mjclark]

"Lunacy Star" is without doubt the best name for a shmup I have ever heard...

[Taizen Chisou]

Why, thank you :V

It's sticking. The sequel is called Lunacy Star Second and will be more like a Cave game.
And if there's a third, it'll be a horizontal shooter...


Changes today include Release Ruin edits, as well as Stage 4- no more zigzag bullets, variable speed bullets instead!
Also, Converge Cube is not totally bullshit and the Idiot Box on the whole is easier to defeat.

[LoudBlast54]

Yes; Lunacy Star is a very good Name.
I'm not good in this game(normal mode), because he must be very precise in his movements.

Wow ! you are the Prince of the shot patterns !!! Very refined, high art these patterns.
there are incredible and really good realized, congratulations.

[Taizen Chisou]

Yes; Lunacy Star is a very good Name.
I'm not good in this game(normal mode), because he must be very precise in his movements.

Wow ! you are the Prince of the shot patterns !!! Very refined, high art these patterns.
there are incredible and really good realized, congratulations.

A friend came up with it
The hitbox is 2x2. Certainly doesn't feel that way, does it :<
Most people play Easy mode anyway ;-;

Lol, I'm curious as to which ones you saw.
I personally don't feel as though shot patterns get good until stage 4, maybe stage 3.

Of course, I'm trying to fix that, as well as nerf everything.

[BPzeBanshee]

I did a little bit of digging in the "developers look here" thread and I just wanted to say that this probably isn't completely typical of GM shmups :V

That's because most GM shmups you see on places like Yoyogames aren't really shmups but stuff like Mactabilis - Oooo look at me I'm gonna revive the genre with my 1280x1024 HD res and healthbars, shield and bright-ass glow/surface/blending particle effects. Fear not, this forum has none of that.

As for your other concerns in relation to the Developers Look Here First thread, I wouldn't worry about diagonals being faster as a high-priority issue - in fact I don't like diagonals being slower at all which is why GMOSSE by default doesn't do it (but the option is there with the in-game code if you wish to add such). In fact most of the mentions in the thread are biased by people who would rag on GM because it's the weapon of choice used by developers who mess up way too often, but so long as you have most of the general idea in the back of the head it shouldn't be much of an issue.

By the way: I'm loving the debug info in your screenshots, quite detailed in letting the user know exactly where and when an error or slowdown is happening. Is there an option to switch it off though? Some people unlike me don't like seeing debug code everywhere.

I'm on OSX right now but I'll have a crack at this later on tonight. That being said I can already think of a few areas to consider for 'nerfing the difficulty':
- I see you have some kind of 'Rank' variable which does nothing. Consider using it for bullet density, enemy placement and bullet speeds and set absolute lower-upper limits for it per stage if some of the stages are out of control. Having it score-based like your article states could amplify the game to the point of ridiculousness even for pros if you're not careful.
- More is not necessarily better. You could take a bit of influence from the Raiden Fighters games instead of Touhou's cluster@#$( of bullets and reduce bullet count in favour of bullet speed being increased. At the same time, it could be that you may want/need to go the other way around.
- Same goes for the score too - more scoring mechanics can make the game convoluted. A simple proximity-based mechanic or speed-killing enemies could be a useful way to balance scoring to difficulty.

Like I said, these are just concepts and ideas for you to consider. I haven't played the game yet and some of these may not apply so take it with a grain of salt.

[Taizen Chisou]

I did a little bit of digging in the "developers look here" thread and I just wanted to say that this probably isn't completely typical of GM shmups :V

That's because most GM shmups you see on places like Yoyogames aren't really shmups but stuff like Mactabilis - Oooo look at me I'm gonna revive the genre with my 1280x1024 HD res and healthbars, shield and bright-ass glow/surface/blending particle effects. Fear not, this forum has none of that.

As for your other concerns in relation to the Developers Look Here First thread, I wouldn't worry about diagonals being faster as a high-priority issue - in fact I don't like diagonals being slower at all which is why GMOSSE by default doesn't do it (but the option is there with the in-game code if you wish to add such). In fact most of the mentions in the thread are biased by people who would rag on GM because it's the weapon of choice used by developers who mess up way too often, but so long as you have most of the general idea in the back of the head it shouldn't be much of an issue.

By the way: I'm loving the debug info in your screenshots, quite detailed in letting the user know exactly where and when an error or slowdown is happening. Is there an option to switch it off though? Someo people unlike me don't like seeing debug code everywhere.

I'm on OSX right now but I'll have a crack at this later on tonight. That being said I can already think of a few areas to consider for 'nerfing the difficulty':
- I see you have some kind of 'Rank' variable which does nothing. Consider using it for bullet density, enemy placement and bullet speeds and set absolute lower-upper limits for it per stage if some of the stages are out of control. Having it score-based like your article states could amplify the game to the point of ridiculousness even for pros if you're not careful.
- More is not necessarily better. You could take a bit of influence from the Raiden Fighters games instead of Touhou's cluster@#$( of bullets and reduce bullet count in favour of bullet speed being increased. At the same time, it could be that you may want/need to go the other way around.
- Same goes for the score too - more scoring mechanics can make the game convoluted. A simple proximity-based mechanic or speed-killing enemies could be a useful way to balance scoring to difficulty.

Like I said, these are just concepts and ideas for you to consider. I haven't played the game yet and some of these may not apply so take it with a grain of salt.

Oh, okay, then? I really just came here and have no idea how other GM!shmups do things.

Let's see:
Although that information is displayed for my purposes firstly, I never bothered to remove them because I felt it might have had use to some players. I'll make a note of it.
You know how in rhythm games you receive a grade based on your performance? Rank follows the same principle. Although I might bump it up because 1CCing the game usually comes with a high rank gratis. Or maybe that's just me and the end-of-game resource bonuses.
Anyway, the game is really too far along to add that effectively to more than a handful of places. The main 6 stages constitute 236 patterns- and I've got an extra 33 or so to make for the other modes.
But it will be implemented in the sequel, and in the sub-game within Lunacy Star. (What subgame? Music Room #25 for details!)

Since GM can really only shove out 600 bullets at a time fluently, I've had to resort to this.

Have you seen the Guide page on scoring in this game? Convoluted is my middle name- I'm worse than Zun! :V
Although it all follows the basic idea of risk+skill = reward.
All risk, no skill, you overdrive a lot and die repeatedly.
All skill and no risk you don't cash in on all the bonus to be had for all your ability.

*shrug*

Thank you though for checking it out! Hope you enjoy it~

[Ebbo]

This game sure is hard. I'm really struggling with Standard - even the easiest difficulty setting isn't exactly walk in the park. Although my skills are pretty mediocre I'm partially going to blame somewhat poor visibility. Everything from bullets to enemies looks so small and featureless that it's way too easy to lose sight of your own ship. Especially the third player ship (Wanderer something) is ridiculously tiny. Bland backgrounds don't really help things either. I'm sure that with a bit of practice you could improve your art skills, although it's going to demand some determination.

The level design is just kinda there, lacking cohesiveness: nothing really caught my interest or made me think about possible scoring routes. Scoring seems pretty convoluted as you said and while there are plenty of shooters with complex scoring systems, I'd have personally preferred something more simple and intuitive. Maybe it's because I found the presentation sloppy which makes it hard to tell when I'm actually doing well scoring-wise.

Anyway that's my two cents, hopefully you'll find some of it helpful and good luck with the project.

[BPzeBanshee]

TLDR version: get rid of precise collision checking and use boxes smaller than sprite size for everything + what Ebbo said + use an external dll for ini writing/reading cause GM sucks.

Detailed version:
Now that I've had a go at the game I can definitely see some of the issues regarding it being 'too hard' at times. Here's a few suggestions:
- The boss fights felt too long, not so much because of the actual duration, but because you've got a bonus-points-for-quick-kill timer showing up on the screen which does not fit with the duration you have to put up with the patterns. Maybe consider either boosting the attack of the player for the boss specifically or reduce the health it has in its phases.
- I didn't quite realise until after playing the game several times that you had SHIFT assigned to the focus key. I think you could probably kill two birds with one stone in regards to accessibility to newbies and the 'oh i didnt know you could pick your ship' syndrome by adding large arrow key graphics under the descriptions for the ship in that menu while showing a really basic outline of the controls.
- I realise this is probably out of the question, but I think one of the big banes of shmups is when they go with using mindfuck bullet patterns at the expense of difficulty. If you can, nerf some of those particular bullet patterns.
- I couldn't see how Stage 6 was doable with these huge bullets around the place and no gaps between them. In Cave games there's also a lot of bullets around the place but the hitboxes of the bullets are smaller to compensate. Yours are using precise collision checking, which is not only a bad idea for performance but makes for impossible patterns when said bullets are overlaying eachother in rings.

In addition, I had a few ideas regarding your technical matters too:
- The major lag at the end of the highscore screen had me convinced the program had stopped working, and Task Manager saying it's Not Responding doesn't help. This lag comes from two factors - the usage of Game Maker's INI file handling which is very slow for the amount of stuff you're writing/reading from file, and the ultracrypt.dll that you're using. There are external DLLs available for INI file parsing that are much faster, so consider using one to cure the lag issue there. Alternatively, you could just use the file_text_*** stuff instead but you'd have to redesign the system away from being like how it is with INI-based stuff so that might not be ideal.
- You're not wrong about the music being terrible, but I attribute this more so to them being MIDI files using Windows' terrible MIDI library more than just bad composition. Maybe consider using an external sound engine to run them as XM modules or something similar where you have more control over your instrument choice.
- Since you're using MIDIs, I believe it is possible to actually pause the music upon pressing of ESC with sound_background_tempo(). That should solve the problem of pausing 'ruining the timing of the music' and all that.

Hope this helps.

[Taizen Chisou]

This game sure is hard. I'm really struggling with Standard - even the easiest difficulty setting isn't exactly walk in the park. Although my skills are pretty mediocre I'm partially going to blame somewhat poor visibility. Everything from bullets to enemies looks so small and featureless that it's way too easy to lose sight of your own ship. Especially the third player ship (Wanderer something) is ridiculously tiny. Bland backgrounds don't really help things either. I'm sure that with a bit of practice you could improve your art skills, although it's going to demand some determination.

I tried to balance difficulty with loads of resources, primarily. :V
Not sure how exactly to add feature and distinction to bullets, but enemies are mainly like that because, well, each one had some three minutes of work done. I could go through and redo a lot of the sprites, but it would take me a long time and it wouldn't be that much better anyway. I think I'll commission some graphics guy sometime soon for it.
Wanderer 6D-4 is called "6D-4" because it's phonetically close to "sixty four" and because the ship graphic (sans outline) contains 64 pixels.
Backgrounds were also something I added with a little effort. I don't like to strain a lot of graphical things
I might ask someone sometime to redo a lot, again.

The level design is just kinda there, lacking cohesiveness: nothing really caught my interest or made me think about possible scoring routes. Scoring seems pretty convoluted as you said and while there are plenty of shooters with complex scoring systems, I'd have personally preferred something more simple and intuitive. Maybe it's because I found the presentation sloppy which makes it hard to tell when I'm actually doing well scoring-wise.

Anyway that's my two cents, hopefully you'll find some of it helpful and good luck with the project.

Did you end up using the Overdrive Mode?
The enemies' inflated amounts stack up in the text indicators and you gain Ranking much faster.
I was considering changing the point item graphic during it, just to reinforce it, though. Yea or nay?
I guess in terms of simplicity scoring could be boiled down to

1: Collect charge, hit maximum
2: Collect point items with maximum charge
3: When a lot of enemies are about to appear, use Overdrive
4: Don't die during Overdrive
5: Graze a bunch so the point items collected would be worth more

You can get some 150,000,000 in Stage 5 as a result, as opposed to some 50,000,000 without using Overdrive.

TLDR version: get rid of precise collision checking and use boxes smaller than sprite size for everything + what Ebbo said + use an external dll for ini writing/reading cause GM sucks.

Detailed version:
Now that I've had a go at the game I can definitely see some of the issues regarding it being 'too hard' at times. Here's a few suggestions:
- The boss fights felt too long, not so much because of the actual duration, but because you've got a bonus-points-for-quick-kill timer showing up on the screen which does not fit with the duration you have to put up with the patterns. Maybe consider either boosting the attack of the player for the boss specifically or reduce the health it has in its phases.

Which... ones? All of them?
Boss fight time for each stage goes about 1:00, 2:20, 2:50, 2:40, 2:40, 7:00, with 4, 6, 7, 7, 7, and 12 patterns to use respectively. (Using 6D-4 A shottype)

- I didn't quite realise until after playing the game several times that you had SHIFT assigned to the focus key. I think you could probably kill two birds with one stone in regards to accessibility to newbies and the 'oh i didnt know you could pick your ship' syndrome by adding large arrow key graphics under the descriptions for the ship in that menu while showing a really basic outline of the controls.

I can totally do that.

- I realise this is probably out of the question, but I think one of the big banes of shmups is when they go with using mindfuck bullet patterns at the expense of difficulty. If you can, nerf some of those particular bullet patterns.

You mean those zigzag bullets like everyone else, right?
Yeah, a lot of those are being changed.

- I couldn't see how Stage 6 was doable with these huge bullets around the place and no gaps between them. In Cave games there's also a lot of bullets around the place but the hitboxes of the bullets are smaller to compensate. Yours are using precise collision checking, which is not only a bad idea for performance but makes for impossible patterns when said bullets are overlaying eachother in rings.

Firstly, there is absolutely no "precise collision checking" anywhere, and the bullets all have smaller square hitboxes somewhere in the middle of them.
I had recently modified those bullets' hitboxes because when creating them, I forgot to change the hitbox from what they were copied from. As a result, they had the same hitbox as those zigzag / leaf / rain bullets.
They're being made slightly smaller.

In addition, I had a few ideas regarding your technical matters too:
- The major lag at the end of the highscore screen had me convinced the program had stopped working, and Task Manager saying it's Not Responding doesn't help. This lag comes from two factors - the usage of Game Maker's INI file handling which is very slow for the amount of stuff you're writing/reading from file, and the ultracrypt.dll that you're using. There are external DLLs available for INI file parsing that are much faster, so consider using one to cure the lag issue there. Alternatively, you could just use the file_text_*** stuff instead but you'd have to redesign the system away from being like how it is with INI-based stuff so that might not be ideal.
- You're not wrong about the music being terrible, but I attribute this more so to them being MIDI files using Windows' terrible MIDI library more than just bad composition. Maybe consider using an external sound engine to run them as XM modules or something similar where you have more control over your instrument choice.
- Since you're using MIDIs, I believe it is possible to actually pause the music upon pressing of ESC with sound_background_tempo(). That should solve the problem of pausing 'ruining the timing of the music' and all that.

Hope this helps.

I actually have an INI handling DLL, but I haven't used it for this project, seeing as my other project literally has over 150x the amount of data to save.
I'll implement it for the newer version.

Ow, thanks, "music is terrible," "just bad composition"
I'm still learning, you know?

I'm actually producing MP3s for the sequel, but the first game is having MIDIs mainly because the game wasn't made entirely flexible enough to allow for it.
Also so the game wouldn't be 100 MB and no one would play it because I didn't have anything beforehand released to prove that I can produce things of... "quality."
Have you seen the Music Room? There are 25 songs to convert and at about an hour combined, that's like 70 megabytes on top of a 10 MB game. :V

The sequel will have a lot more graphics and sounds though, it having more stages, but that just means I won't release it until 2016 D:

Oh, and thank you for the advice on the sound tempo thing. I didn't realize that I could pause it that way

Still, thank you for the advice and comments, though!

I ended up losing my flash drive, so all my difficulty editing for the past week went out the window, meaning the wait is just that much longer now >_>

[BPzeBanshee]

TLDR version: get rid of precise collision checking and use boxes smaller than sprite size for everything + what Ebbo said + use an external dll for ini writing/reading cause GM sucks.

Detailed version:
Now that I've had a go at the game I can definitely see some of the issues regarding it being 'too hard' at times. Here's a few suggestions:
- The boss fights felt too long, not so much because of the actual duration, but because you've got a bonus-points-for-quick-kill timer showing up on the screen which does not fit with the duration you have to put up with the patterns. Maybe consider either boosting the attack of the player for the boss specifically or reduce the health it has in its phases.

Which... ones? All of them?
Boss fight time for each stage goes about 1:00, 2:20, 2:50, 2:40, 2:40, 7:00, with 4, 6, 7, 7, 7, and 12 patterns to use respectively. (Using 6D-4 A shottype)

I was playing with the ship with the piercing lasers, and while I did not check the actual times, it felt quite irritable that not only could I not hit the bosses properly due to small size but also it felt a lot longer than it should relatively. I've seen good boss fights last 5 seconds, others a bit over a minute or 2. 7 minutes is probably overkill even for a final boss.

- I couldn't see how Stage 6 was doable with these huge bullets around the place and no gaps between them. In Cave games there's also a lot of bullets around the place but the hitboxes of the bullets are smaller to compensate. Yours are using precise collision checking, which is not only a bad idea for performance but makes for impossible patterns when said bullets are overlaying eachother in rings.

Firstly, there is absolutely no "precise collision checking" anywhere, and the bullets all have smaller square hitboxes somewhere in the middle of them.
I had recently modified those bullets' hitboxes because when creating them, I forgot to change the hitbox from what they were copied from. As a result, they had the same hitbox as those zigzag / leaf / rain bullets.
They're being made slightly smaller.

It certainly gave the impression there was precise checking on the borders. I guess I'll just learn2play more.

Ow, thanks, "music is terrible," "just bad composition"
I'm still learning, you know?

I'm actually producing MP3s for the sequel, but the first game is having MIDIs mainly because the game wasn't made entirely flexible enough to allow for it.
Also so the game wouldn't be 100 MB and no one would play it because I didn't have anything beforehand released to prove that I can produce things of... "quality."
Have you seen the Music Room? There are 25 songs to convert and at about an hour combined, that's like 70 megabytes on top of a 10 MB game. :V

The sequel will have a lot more graphics and sounds though, it having more stages, but that just means I won't release it until 2016 D:

What I mean is that the standard MIDI library is what's holding you back here. I understand your complaint about filesize too, which is why I suggested using module formats like XM to really get the most out of your composition skills while keeping the filesize down. GM's default sound handling is pretty terrible, but I do have to congratulate you on being able to make good use of it too. Didn't mean to sound too mean .

[Taizen Chisou]

Lol, I am working with Touhou in mind.
Touhou final bosses have ranged from three to five minutes.

And I'm probably going to have single-digit-second encounters in the sequel, which I will be making a post about here in the coming weeks.


Oh, yeah, those hitboxes (and that pattern by extension) are being nerfed.


Oh, it's fine. I don't know how to use trackers very well so I resort to FL Studio :<
I've moved LS2's Stage 1-1 theme to MP3, and things are looking up.
Especially in terms of filesize

[Taizen Chisou]

Debating whether or not to open a devlog/topic on the sequel.

http://www.youtube.com/watch?v=MfykxRVqtoQ